text
stringlengths 83
79.5k
|
|---|
H: How do I successfully stabilize or trim a TinyWhoop Nano?
I'm a newbie to drone flying and decided to start with a TinyWhoop Nano RTF. I know just enough about model aircraft controllers from using them in robotics and animatronics that fine-tuning is necessary to get stable or centered actuators.
In the case of the Nano, I am not sure about the right procedure to trim out the rotors. My expectation was to get it to hover and then use the trim buttons to get it to stop drifting in a particular direction. It would usually drift and hit something before I could manage to figure out what direction needed to be trimmed. I finally put it in a very large open-top cardboard box to stop having to chase it all over the room.
I've gotten it to drift much less, but it still seems to be all over the place when I throttle up more. I've had problems with one of the props scraping the canopy ever so slightly, which is annoying thing to try to fix.
So I guess my question is twofold:
What's the specific method to trim a TinyWhoop Nano using the TBS controller that comes with the RTF version; and,
Is there a general process for trimming out any new quad, such as how much throttle to use and a clean environment in which to more carefully check it?
AI: Often the really small/simple drones don't trim to completely stable, as their controllers are very basic and the 'step size' on the trim isn't fine enough. And, if you get them to trim, then they will usually drift again as the battery voltage changes - again, due to the simple design.
I usually just get it as close as possible, and accept there will be drift - its good object avoidance practice.
Finally, a tip - if such a small drone is going to crash, just stop the props ... this usually results in less (if any) damage to the drone or whatever it hits. |
H: Is superglue (cyanoacrylate) a viable means to repair quadcopter frames?
One of the tiny struts holding the rotor on my TinyWhoop Nano broke (surprise!) and I've used superglue (cyanoacrylate) to bond it back together. I don't know yet if this fix will be suitable or not, because I haven't even gotten the quad properly flying.
As fragile as some small drone frames seem to be, I wondered whether it's possible to glue them together, or if there's irreversible damage due to having the angle of a rotor changed slightly.
I know some frames are carbon fiber and I wouldn't expect superglue to work well.
Is it a "standard practice" to repair minor damage on frames, or is it recommended to always get a new frame should non-superficial damage occur?
AI: It depends on the risk you are willing to take, but for such a small drone being flown indoors I'd give it a go1 and accept it may crash - after all, it has already done so once so it can't be made worse! The repair will, however, be a weak point in future crashes.
How well it flies will depend on how accurately you reposition the part, but you're likely to get 'close enough' (again, probably 'worth a try'.) I would also, once the glue has fully set, try to hold the drone only by the broken bit and give it a shake - if it can hold the whole weight, it should be OK when the weight is divided between the propellers.
For larger drones you should replace the part to manufacturer specs or get a professional to take a look.
1 Subject to the normal indoor flying considerations - avoid pets, other people and anything expensive or valuable, etc. |
H: How does Amazon's new drone land itself into a station?
I am fairly interested in drone building and thought about a drone which automatically recharges itself by flying in to his station, but thought it was close to impossible without using extremely sophisticated tech that extremely pricey (D-GPS).
How can this low cost drone reliabely land in its small station? Can anyone guess how?
(this is the drone im talking about https://www.youtube.com/watch?v=EIQTR4qCoHU)
AI: Remember - it only takes 3 points to get a solid fix on location in 3 dimensional space. I have been thinking about this a lot, I will also have a unit on day 1 and possibly the unit disassembled day 2 and posted on line (maybe)
For the Amazon device there is most likely a sonar/echo or similar system that creates a feedback loop for position. I also suspect use of RFID or similar "near proximity tech" to get the last few inches.
There are a lot of systems available and affordable, LIDAR in particular is really cheap (https://www.getfpv.com/matek-optical-flow-lidar-sensor-3901-l0x.html)
The tech is pretty standard and accessible, Roomba has been doing this in 2D space for almost 20 years with self docking features for the floor cleaning system.
Once you have experience with auto-leveling and GPS technology the task becomes trivial. The hard part is universality and compliance - will it work anywhere and also be legal/interfering.
There are still a number of unanswered questions - but form what I know here a few of the approaches that come to mind:
Echo location indoors (bouncing signals to determine position, similar to how the blind or bats can "see" using echos and reflection of sound)
LIDAR with some GPU strength and the addition of hyper- or hypo-visible light can create an invisible beacon.
Various systems for mapping, used commonly in drones for recovery operations such as mineshaft exploration, possibly offset to cloud computation.
Wifi stereo has a lot of research in phasing and image control of sound worth considering.
RFID or Bluetooth beacon technology, possibly with other AWS devices in house?
Augmented indoor GPS (using what signal is available + other systems)
Final landing could be simple optics targeting a unique pattern, key for orientation (downward simple camera looking for landing beacon)
Sonar, like a bat, can help with direction, distance, and obstacle avoidance beyond aforementioned echolocation.
Additional off-device processing - such is the case for Alexa, Siri, and other services.
Also note how the female receptacle is wide and allows the unit to dock with +/- 1 inch accuracy.
Also important, the FAA claims to have NAS authority indoors, but they have never been able to asset any enforcement action.
This device is a HUGE blow to the FAA's weakening grasp on NAS policy. The laws that regulate this device are NOT settled and the legality is open for debate. Per the Congressional charter, the FAA controls all airspace "0inch and up" and there has never been any official clarification for indoor operation. A lot of questions remain unanswered.
FWIW, if f you fly this unit out of the house it is $40k minimum fine for violation of key BVLoS rules such as remote operation w/out spotter and BVLoS operation without an exemption. There are many other rules, each $20k each per violation should anyone take this out to the porch. See and avoid is not met by the device and that one is huge and is the massive thorn FPV faces for proper recognition. I will be following this close. |
H: What is the best drone for beginner fliers?
My brothers would like to start drone flying, as they haven't ever flown my Syma before. I am looking for a small, inexpensive pocket or medium drone for beginners that will probably get thrown into our magnolia tree.
AI: There's no single 'best' starter drone, as new ones are released every few months. However there are lots of good options under \$30, and some as cheap as $15.
What you're looking for is something small and light, so you can fly it indoors without breaking things. It should have rings around the props to protect them and anything it hits.
Syma, Eachine and Hubsan are popular brands but there are others.
Do get more than one battery. They typically only last about 5 minutes and take an hour to charge. A charger that can charge more than one battery at a time is a bonus too. That said, 10-20 minutes is enough for most people - at least after the first time.
'Headless mode' is quite a useful feature for beginners, it allows you to ignore yaw. Forwards is always 'away from you', whatever direction the drone is facing. It's not essential though, and can be confusing or lead to bad habits.
'Altitude hold' should be useful but I've found these drones always need an experienced pilot to trim them first. It's good for an experienced pilot to teaching kids under 10, but not for teaching yourself.
Many cheap quads have tiny 'jelly bean' transmitters, which can be difficult for those with big hands. If you think this might be a problem, you can spend a bit more for a proper sized controller. There are various hacks for extending the sticks too.
Don't buy one with a simple recording camera. Until you pay enough for one that can transmit the signal to you in real time, the footage will be rubbish because you can't point the camera and frame the shot. The first time you do it, it's cool to see a blurry, poorly coloured picture from the air but the novelty quickly wears off. View your first drone as a trainer and indoor toy that will take the knocks of learning while you save up for something better.
Don't bother buying loads of spares, these little things don't break very often, and it's as likely to be a motor burning out as a broken prop, and a pack of spare motors can cost more than the drone.
That said, they are very prone to picking up stray hairs and winding them around the props and motors. Try not to spend all your time bumping around on the floor, and check for hairs after each crash.
When you're learning, try to pop up into a hover a foot or so off the ground and concentrate on holding the right height. When it has drifted half-way to an obstruction, cut the power and land. Don't try to fight it because you'll just crash harder. Put it back in the middle of the room, review what you should have done and try again. When you can control the height, try correcting the drift but still land and review when it gets half-way to an obstruction. If you struggle, find a larger space to practice for a while. |
H: What type of connector is used on the DJI manifold 2?
I'm looking for a certain connector's specification that is used in the DJI Manifold 2. Unfortunately, there is no information about this connector's specifications in the data sheet, it only talks about the communication port.
Could somebody help me?
I need to know what the connector's specification is to buy for my search.
RJ-45
USB-A
HDMI
Antenna ports
connector XT-30
Power indicator
USB micro B
UART PORT, what is connector specification?
CAN PORT, what is connector specification?
I/O PORT, what is connector specification?
AI: Based on some searches, I think this connector is purposely undocumented as DJI states clearly on sites that sell the official cable:
Note: Only use a DJI UART cable such as this one to connect your own device to the Matrice 100's UART port, as other cables can cause the flight controller or your device to be damaged from excess voltage.
In any case, I think I've figured it out. Using this image:
And the dimensions from the manual:
I was able to establish that the connector pitch is probably 1.25mm. I use a lot of JST brand connectors in my work, and I think the GH series are a spot-on match:
Based on this, the connectors would be part numbers:
4-pin header (device socket): SM04B-GHS-TB
4-pin housing (cable end): GHR-04V-S
10-pin header (device socket): SM10B-GHS-TB
10-pin housing (cable end): GHR-10V-S
Contacts (crimp pins): SSHL-002T-P0.2
This is only a best-guess and I must reiterate once again that DJI does not appear to support nor endorse third-party cables, let alone DIY solutions. |
H: How do DJI drones achieve such long flight times compared to traditional FPV drones?
DJI is able to achieve ~30 minute flight times with their dones. Whereas, a typical FPV drone is only ~7 minutes
I see this as a huge achievement by DJI, I am interested to understand how they were able to achieve this advancement
The flight characteristics of these two crafts are quite different. A DJI drone is flown at much slower speeds than FPV drones So, if a FPV drone was flown slowly and cautiously, would the battery extend to be comparable with DJI's?
I have read elsewhere that the energy cost of FPV drones is due to the small blades, the motor needs to spin faster to compensate. This would imply that if you increased the blades (and the arm length) you would get longer flight times...?
Or has DJI really advanced that much in ESC performance, motor performance, battery chemistry to reduce energy losses?
AI: There's no magic technology, it's just a difference in design goals. FPV drones are designed for speed, performance and robustness, DJI drones are designed for duration and carrying a gimbled camera.
Larger, slower turning rotors are more efficient at producing lift but have a lower top speed. That's why helicopters can hover and propeller planes with the same size engine can't - but the plane is faster. Photography drones (such as DJI) use large rotors and exploit that efficiency for duration. Race drones use smaller rotors because they want to go as fast as possible.
The other factor is that photography drones typically carry larger batteries for longer duration, while race drones want to keep their weight down for maximum agility and acceleration. A photography drone will fly so long as it can produce more thrust than its weight, typically they have less than a 2:1 thrust-to-weight ratio and can't maintain altitude when angled more than 45 degrees from horizontal. Race drones can have a 10:1 thrust to weight ratio, and commonly fly at pretty extreme angles.
So long as a race drone has enough duration to complete the race, any extra battery capacity is just dead weight, slowing it down. For their pilots, changing the battery every few minutes is a small price to pay for the performance. |
H: Beginner looking to try a DIY plane, any suggestions on a kit?
Title pretty much sums it up, but I would like to try to make my own RC plane. I'm looking for a kit or bundle I can get that would basically include all the supplies to make a plane, so I guess that would be a controller, receiver, a motor or two, some servos, and a battery(I think?). Built in FPV camera is a plus but not necessary. Price is where I'm mainly concerned, and unless the circuitry is pretty indestructible I only really want to pay twenty dollars for it. If I can crash it without too much damage then I can do up to 50 dollars, but preferably you would have some experience with the kit so I don't have just amazon reviews to go off of. I can solder any connections that are needed if the kit isn't preassembled.
If you need any more details, just ask and I'll add it.
I am in the US, state of Tennessee.
AI: Probably the easiest way to get started building planes these days is with foam board models, such as those popularised by FliteTest. Foam board is a 5mm thick non-beaded polystyrene foam sheet with a layer of paper on either side for added strength. You can buy sheets from craft stores for a dollar or two. It's easy to cut with a sharp knife (small box cutters are ideal) and it can be glued with hot glue or UHU POR (but some other adhesives will melt the foam). The paper layer makes it easy to paint, with almost any paint.
You can build a good starter plane with two or three sheets of foam, and with practice it only takes an afternoon or so. Your first attempt may be a bit scruffy but you will get better with practice, so plan to make several. Smallish, light models are fairly tough and easy to patch up, but knowing you can completely rebuild for very little cost is a huge confidence boost and will help avoid the common story of buying an expensive model from a shop, crashing on your first flight and giving up.
Download and print out a 3-view drawing of your favourite plane and stick it to the foam. Cut out a profile of the fuselage, and the plan view of the wings and tail. Stick them together, add some nose weight so it balances about 25% from the front of the wing and it will probably fly. Experiment with the nose weight, the angle of the tail and dihedral. You'll also learn a lot about the location and weather required for flying.
Next find a plan for a beginner RC plane in foam board, such as the FliteTest Tiny Trainer, Simple Cub or Scout. These models are relatively light and simple, and all have the wings angled up (called dihedral) which causes them to self-level, vastly increasing your chances of success.
These all have free plans under the 'resources' tab in their shop. Print them out, lightly glue the paper to the board and cut them out. Feel free to simplify the shapes slightly as they were drawn for a laser cutter.
Now for the electronics. There are some toy planes under $20 that are quite fun for a 10-year old under adult supervision but they're quite limited and not a good way to start the hobby. You don't have much control and they're very slow, so will blow down wind. You also can't reuse the gear in anything else so they're not a good investment.
A basic radio will cost at least \$30, but you'll want to upgrade that within a year so it's not really a good investment. For \$40 you can get a cheap computer radio with multiple model memories (for when you have more than one plane) and channel mixes (for flying wings, deltas, V-tails or flaps) - that's all the features you really need for a lifetime in the hobby. Most people will spend more for a 'nice' transmitter, but you don't need to. Transmitters generally come with a receiver.
The recommended sizes for the various bits are listed with the plans. You'll need to do a bit of research to work out what all the numbers mean.
Next you'll need a motor and Electronic Speed Controller (ESC). These are \$5-10 each. You will break props, so get some spares.
9 gram servos are about \$3 each. You'll need 2 initially, but it's worth having a couple of spares for crashes or more complex models.
You'll need multiple batteries, as flights are typically 10 minutes or so.
Basic chargers start at about \$20. Get one that can handle a range of sizes and cell counts, and it'll last you for years.
You won't get that lot for \$50 but you should manage to keep it under \$100. I suppose the stuff that's in the plane should be under \$50, as the expensive bits are the transmitter, charger and spare batteries!
The good news is that the gear can be used on multiple models. When one gets too damaged, or you want to try something new, just move the gear across. The transmitter, charger and batteries can all be shared between multiple models.
It's worth shopping around a bit but avoid anything that's super cheap with no reviews. |
H: Learning to fly: One crash at a time
Question
Is there a comparison of model types versus the milestones that reflects which plane is best for learning in ?
Background
I bought an EPP flying wing some time ago and have taught myself and a few friends and family members, as well as any stranger with some curiosity, to get up into the blue yonder. My experience has been that it takes around 200-300 flights before you can hold a figure 8 pattern. It seems that the initial flights are the largest learning block since the flights are quite short and the window for learning the smallest as the person gets longer flights in they tend to learn quicker. Roughly the learning curve goes as follows :
The first 50-100 flights, with flight times of 5-15 seconds, one gets the control down (Dive, climb, left and right).
The next 100-200 flights, with flight times of 15-30 seconds, one can start transitioning to powered flight.
The 200-300 thereafter, with flight times of 30 seconds to a few minutes, one gets comfortable with it flying back at you and across your view.
Basis
The general claim is that a slow flying, top wing is best for learning but I have had fair success with a flying wing. I intend to use a discuss launch glider next as you can give the person a longer flying time for the initial learning curve and I was wondering what others experiences/learning curves were with different model types.
Rationale
I'm curious about the folk who selected the "wrong" thing to learn on and how they got along none the less. Perhaps it turns out that one type of craft is better for learning on then some other; granted a top wing is the likely result.
AI: The big problem with learning to fly is that some of characteristics you need for a good trainer are the opposite of other requirements.
Large models are less affected by wind, because their flying speed is proportionally higher relative to the wind speed. However small models are a lot more robust in a crash. and cheaper, so less discouraging when you crash.
Highly stable models return the model to straight and level flight but neutrally stable models (like a flying wing) are less affected by gusts.
Traditionally this leave you with the option of a small, stable model and waiting patiently for calm weather, or a faster, less stable model and an instructor (or lots of crashes and repairs!)
These days it's possible to use electronic stabilisation to tame a neutrally stable aircraft, and to reduce the effect of gusts on a small model (although it won't stop it disappearing down wind!)
IME it takes an hour or two (on a simulator, or accumulated flight time) before a beginner can keep a model under control and steer the right way when it's flying towards them. Most people can only concentrate for 5-10 minutes at a time, so that's 10-20 flights (on a reasonably stable trainer, suitable weather and an instructor avoiding crashes). It then takes few months to a year before progressing on to an aileron model that isn't stable in roll - maybe 10 hours of flight time, say 100 flights of 5-10 minutes each.
Obviously if you don't have an instructor and are just doing short glides, those first couple of hours will translate to a lot of flights. If you're starting with an aileron plane (like a flying wing) you'll need to reach that 10 hour mark before you're really flying.
Older people take longer to learn, and missing a few weeks due to bad weather or other commitments puts them back a few steps. It's common for people to struggle through weeks of poor conditions then suddenly 'get it' when we have a nice calm day because they're not fighting the gusts.
I've taught a couple of people who practiced extensively on a simulator and flew a trainer competently on their first flight and an aerobatic model 5 minutes later. |
H: Is there a list of milestones to check off before claiming one can "fly"
Is there a generally accepted set of milestones that one should "tick" off before claiming that they can fly ? Are there different guidelines based upon the craft e.g. fixed wings, quadcopter versus helicopter ? Are these published by the modelling associations or by legal bodies like the countries aviation authority ?
My checklist is roughly :
You crash beyond the first patch of khakibos in the veldt we fly in under glide
You crash beyond the second patch of khakibos in the veldt we fly in under power
You crash when you can fly across the veldt, where left is away from you and right is towards you or vice versa, and even return, where left is really right and right is really left and you understand the term "spiral of death"
You crash in style reminiscent of "Those magnificent men in their flying machines" that you feel you could claim was a controlled decent to your friends and family without too much blushing after doing a series of "Uppity up up ups and down di down down downs" (or due to another attack by the resident bully of a hadida)
I doubt this is authoritive in any way whatsoever and was wondering if there was a better set of guidelines to work against.
AI: In the UK, the British Model Flying Association offer an Achievement Scheme which recognises a few levels of competence:
The Basic Proficiency Certificates provide a measure of competence for pilots operating fixed and rotary wing aircraft fitted with
stabilisation technologies.
An FPV extension is also available to holders of any Basic Proficiency or ‘A’ certificate.
The ‘A’ Certificate, which may be equated to a basic standard of safety and flying competence.
The ‘B’ Certificate, which is designed to recognise a more advanced pilot’s increased ability and knowledge, and demonstrates a high level
of safety.
The ‘C’ Certificate, Aerobatics.
The ‘C’ Certificate, Scale
The ‘C’ Certificate, Gas Turbine
As well as covering theory (air law/safety) questions, the flying skill required increases as the candidate progresses through the levels - from assisted flight, through holding level figure 8's and other manoeuvers, to being able to fly in aerobatic displays. |
H: Do I need CW or CCW props for my plane?
So I'm making an airplane with a motor at the front, do I use CW or CCW props?
Pretty sure this is what counts as a puller plane (is that the right term?).
AI: The decision between CW (clockwise) and CCW (counterclockwise) propellers has to do with the direction your motor/engine will spin, not really in regard to where you're putting the engine on the plane. CW props will create thrust behind them when spun in a clockwise direction, and CCW props will create thrust behind them when spun in a counterclockwise direction. It's that simple. :)
For an electric motor driven plane, it probably doesn't matter which kind you get because it's highly likely that you can reverse the motor direction in the ESC (electric speed controller), but I'm not aware of any ability to do this with model plane gas engines. If you're using a gas engine, inquire to figure out either what kind of propeller the specs say to use (if this is listed) or what direction the engine spins and get the appropriate CW/CCW propellers.
If the model airplane plans/kit you have calls for a specific kind of propeller, like for example if the plane was designed to fly better with one propeller rotation direction, I'd recommend abiding by their recommendations. There are quite interesting aerodynamic and physical phenomena like P-factor which the designer may have taken into account in the aircraft layout, resulting in the design requiring a specific kind of propeller. |
H: Do laws on 'drones' in Australia apply to fixed wing model aircraft?
I've been looking at the CASA RPA rules (specifically under 250g and under 2kg non-commercial) and there doesn't seem to be a section on fixed-wing aircraft. Do the rules for 'drones' (for instance these) apply to fixed-wing aircraft. I assume they do but I just want to be sure.
I'm trying to get into model aviation, so expect more of these small questions from me.
AI: Strictly speaking, "Drone" refers to any unmanned aircraft that can fly autonomously. Usually, however, a drone is classified as any unmanned aircraft even if it requires a remote pilot's control. By this definition, yes, a fixed wing would be classified as a drone because it is an unmanned aircraft. |
H: Budget drone capable of lifting 0.5 - 1 kg and flying ~2km radius
I'm looking for some budget multirotor drone (up to $300) which can lift 500g-1kg of weight and has long enough range to fly a ~2 km radius, it should also have a small camera just for control. Is there a ready to fly model which I can buy, or can I somehow easily construct a drone with the following parameters?
AI: The flight you describe is 100% illegal in the US requiring 107/commercial license + BVLOS waiver. BVLOS waivers are very rare, 61 issued searching the DB for "107.31" here: https://www.faa.gov/uas/commercial_operators/part_107_waivers/waivers_issued/. The rules are similar in most countries.
Your budget is off by a zero as well. Depending on your duty needs, I think minimum \$1k for the core + $ for peripherals.
A Phantom could probably limp that payload - but you need to dig into the powertrain math:
What ratio of thrust do you want? Are you carrying a load form point A to B or will you require some performance?
Weight of unit AUW desired? Payload and performance are correlated to thrust and other performance features.
Flight time desired?
Camera for control at 2km requires high power VTX (FAA HAM license required). A 1 watt VTX in an urban environment will struggle to get you signal without a lot of extra gear (signal tracking from a ground station). Also, 1w is the legal max for a HAM / FCC licensed person, else you are 25mw (0.025 watt) limited and can't mod anything legally. To get more watts you need to dip into FCC Part 15 regulations, not cheap.
I see you are in Armenia - I don't know the local there laws but my international flight experience informs me that most of the world is either unregulated (3rd world) or as strict or more strict than the US. |
H: FPV camera and beeper seems to be interfering with each other
I recently changed my VTX from the Tbs unify pro 5G8 HV to the unify pro 5G8 V3. Because of this I had to move the cameras 5v input from the VTX to the FC (The HV model has a built in regulator, and the V3 doesn't). I also mounted the VTX on a 5v output on the FC instead of vbat.
Now to the problem: When I turn on the drone everything is fine until the beeper beeps once. This causes the cameras image to go gray momentarily.
When I then turn on the transmitter and start the beeper shortly, it never turns off, it just starts beeping with shorter and faster beeps. This also causes the camera feed to go gray.
My hypothesis is that the 5v regulator becomes overloaded since all peripherals are connected to 5v. Could this be correct?
I should also mention that the OSD is visible at all times, so the problem is definitely with the camera and not the VTX. The FC is a HobbyWing Xrotor fc f4 g2.
Update: I desoldered the beeper, and the camera image still becomes partially gray, but only when I turn on the transmitter.
AI: The problem was indeed the 5V regulator. I bought an external 5V 1.2A regulator and hooked up the camera and VTX to it, and all my problems have disappeared. |
H: Can you charge 1s and 2s batteries at the same time with the tinyhawk 2 included charger?
I just got a Tinyhawk 2. It came with 2 batteries (2s and 1s) and a battery charger that you plug in to USB.
I know that you can't parallel charge batteries with different cell count, but it looks like there is separate logic for each port on this charger.
Does anyone know if it is possible to charge both batteries at the same time? I could try it myself, but I don't like burning batteries.
AI: According to this source, yes you can charge both 1s and 2s batteries at the same time. This is because, like you mentioned, there are separate charge controllers. You can also choose whether you want them to charge to 4.2v or 4.35v per cell. |
H: 30 year old water-slide decals. Are they still usable?
Not strictly a drone or model aircraft question, but I have seen multiple questions here about painting models, and this is also about the aesthetics.
When clearing my deceased cousins house, I found several plane and car models he was in the process of building 30 years ago, but he never finished them.
In 1992 he got in a motorcycle accident, which shattered his right hand leaving him with severely restricted movement of that hand. He never got back into building models, because of the loss of dexterity.
As a tribute to him I would like to finish those models.
What has me worried are the water-slide decals.
There are a bunch of them to be placed over painted areas (Humbrol enamel paint).
They have been stored at room temperature in a closed (dark) cupboard all this time.
I have used some 5 year old decals in the past without any issues, but these are 30 years old.
Would these still be usable after 30 years or is it better not to try to put them on at all?
AI: It's probably safe to assume that they will no longer work. This source source suggests that water-slide decals have shelf life of about 3 years. "After a few years the glue holding the decal onto the paper breaks down, fusing decal and paper together."
Because there is some variation between different decals from different manufacturers, they may last slightly shorter or longer than that in storage (like 3-5 years), but 30 years is probably too long. I think it's worth trying out one of the smaller decals just to be sure of whether it will work or not. |
H: What size screws do I need for this motor
I am using this gimbal motor.
I need to know what size screws do I need to use. I am developing egg and a frame that has 5 mm thick walls, and based on the chart I know I need to use 4m screws. I want to know how much longer do I need to make the screws to perfectly fit this motor. I need help with this because I do not want to buy the wrong size screws and then have to buy more because I don't want them to be so short with a camera or far off mid-flight or I don't want them too long to short them.
Here's a picture of the chart I'm referencing to may provide some further help.
AI: 4-M3 means that you need four screws (on each side), each with a metric M3 thread. M3 is 3mm diameter.
The length of the screw obviously depends on the thickness of the thing you're bolting the motor to, plus a little for the base of the motor.
The amount of bolt in the motor is important. Obviously you need at least a couple of mm so that it grips, and there's another couple of mm of clearance, but any more and it'll touch the coils. It won't stop you spinning the motor by hand but it'll break the insulation on the coils, short the coil to the case and the motor won't work.
So if you're bolting through a 5mm thick frame, order 8mm long M3 screws. They tend to only come in 2mm steps. The length quoted is the specification is the length of the thread, not counting the head.
If you don't fancy cutting and filing a slightly long screw, you can put extra washers under the head.
Also, when ordering screws, you want 'cap head' not 'dome head' or 'countersunk' in order to use the largest possible Allen key. The small ones tend to round off and slip. |
H: How do I change GPS elements on the Betaflight OSD for my Acrobee Lite?
Right now I'm on Betaflight trying to change the OSD for my AcroBee Lite and it's being really weird, I can't seem to get more than three little thingies to be shown at once. Is there some sort of secret code I need or something? This is really confusing. I'm trying to have the altitude, artificial horizon, battery usage, compass bar, crosshair, GPS speed, home direction, home distance, RSSI value, and warnings shown.
I got everything laid out on Betaflight, but when I clicked save and then checked it on my drone it wasn't right. Currently, I have the horizon, crosshair, compass bar, and altitude showing but I don't know how to get the rest on. Sometimes when I change a box something will appear and another disappears.
How do I change my OSD?
EDIT: apparently some of the elements were just getting cut off the edge of the screen so I moved everything to the center and a few more appeared but I still can't get home distance/direction and speed.
EDIT2: I have the following elements on the OSD; compass bar, battery usage, crosshair, altitude, RSSI value, and I think warnings. I want to add the GPS speed, home direction and distance but am unsure how. Every element I have enabled right now has boxes 2 and 3 checked, but I can't get the GPS speed and home direction/distance to show up. What do I need to change?
EDIT3: AcroBee Lite apparently does not have GPS so problem solved i guess.
AI: Turns out the AcroBee Lite does not have gps functions so adding gps functions to the osd is impossible. |
H: ESC power up beeps sound twice: du-du-du...du-du-du beep-beep
Just recently, my power-up beep sounds have changed. The normal sequence is just du-du-du beep-beep for ESC power, then throttle signal, right?
Now I get du-du-du, pause, du-du-du (again) beep-beep.
What could this be indicative of? This is an NBD BeeBrain BL AIO FC/ESC running BF4.1 and JazzMaverick 16.77 48k.
This changed without any change in config or setup.
I might have crashed it a bit harder than usual?
AI: From what I can tell, this occurrence sounds like something causes the flight controller to reset after a second or so, triggering it to reconnect with the ESCs as it initializes itself again.
The "du-du-du beep-beep" noises you point out are generated by the ESC (through the movement of the motors) to signify that they're receiving control inputs from the FC and are ready to go. More specifically, I believe that the "du-du-du" sounds (with ascending pitch) signal that the ESC is receiving signal pulses from the FC and the "beep-beep" sounds mean that the ESC is done with its own initialization process and ready to fly. (some ESCs also use the number of these pulses to indicate the number of cells in the battery they think they're connected to)
If the ESCs are detecting that the FC connects and then reconnects to them after a second or so, this likely means that something on the FC caused it to hard reset and reinitialize itself. This could be caused by anything from broken solder connections on the board to a voltage regulator that finally kicked the bucket. (or has begun to)
With respect to troubleshooting the issue and verifying that the issue is with the FC and not something else, diagnoses aren't very clear cut because the symptoms can vary from undetectable to glaring. Some dead giveaways are:
Something on the FC is getting abnormally hot
There are solder connections on the FC board which are visibly broken/intermittent
As you say, a particularly hard crash can kill/degrade FCs
Something just looks off with the FC during startup
As @Kralc helpfully mentions: (slightly paraphrased from original comment)
it can help to try unplugging parts to see if the problem changes or goes away. (Remember, it might also be a combination of parts that cause the issue, so it is good to keep notes of what you've tried) If the symptoms remain even with nothing else connected, it is more likely to be the FC.
Other symptoms are far more subtle and might not be reliably detectible by the eye. If you are able, I'd recommend trying to replace the FC if none of the above symptoms are present. |
H: How to determine the minimum thrust to weight ratio for a fixed-wing aircraft?
I am considering building a fixed-wing aircraft with foam board, and I'm unsure what the necessary thrust to weight ratio is. With quads, it's easy to determine the necessary thrust. Just find a motor/prop combination that can provide more thrust than the weight of the quadcopter. With a fixed-wing, it isn't as simple because the aircraft can fly with less thrust than the total weight of the aircraft.
Is there any way to determine the minimum amount of thrust that can keep a fixed-wing airborne? The reason I'm trying to figure this out is because the goal of this aircraft is to be as efficient as possible.
AI: I'm afraid it's not that simple. As you mention, planes can fly with a fairly low thrust to weight ratio. Obviously gliders can fly with no power at all, so there's no critical lower-limit.
The other issue is that pitch-speed is important - the prop needs to be producing thrust when the plane is moving, not stationary. Measuring static thrust only tells you half the story.
Instead of thrust, it's better to look at power. Assuming that your motor and prop are reasonably efficient (and most are) measure the power going into the motor with a watt meter (or run it for a minute and see how much energy your charger puts back into the battery). Then weigh the complete model, including the battery.
50 w/lb is about the minimum for gentle cruising and slow climbs.
100 W/lb is 'sporty', say for a warbird. You don't have to worry about conserving energy and can easily climb out of trouble. It might not have unlimited vertical climb though.
200 W/lb is unlimited vertical climb, hovering and climbing vertically out of the hover.
For ducted fans, you should double these numbers. i.e. at 100 w/lb you'll need to be careful to conserve energy.
However foam board planes aren't super-efficient or super-fast, so you can usually do a simple thrust test to check it'll be OK. Dangle it by a wing tip and gradually apply power.
If the model has enough power to support its own weight, it's probably around 100 w/lb
If it can only reach about 45 degrees, its about 50 w/lb and will fly if you're careful.
About pitch speed
If a prop has 6" of pitch and is turning at 1000rpm, under no load it would advance 6000 inches per minute. Or to put it another way, if the model reaches that speed, it won't be producing any power!
'Slow' props have a pitch of about half their diameter (e.g. 8x4")
'Sport' props are 2/3 to 3/4 of the diameter (e.g. 8x6")
'Speed' props are 'square' with the pitch equal to the diameter (8x8)
If you find your plane accelerates well from the launch but has a low top speed, you might need a higher pitch prop.
If it doesn't have enough thrust to take off, go to a larger diameter.
If you change the prop, you should measure the current drawn to check you're not exceeding the limit of the ESC. Also check that the motor and battery don't overheat. |
H: Does the FS-R6B receiver work with the JJPRO EM-16 transmitter?
If I need a new receiver and can't find one that is exactly the same as the one that came with the radio, can I get one with the same amount of channels and from the same company and use it with the radio?
Specifically, I need to know if this receiver works with this radio.
AI: There are various digital protocols, and many receivers are able to handle more than one protocol (and so can a few transmitters).
It looks like that transmitter uses "AFHDS 2A", so you need a receiver that can handle that.
That receiver (FS-R6B) says "AFHDS" on the label, and there's a big complicated chart (the 5th image) showing which receivers handle AFHDS 2A, but I can't see the FS-R6B on the chart so it's not much help. The binding instructions imply that it might do both. |
H: What is this 4th row of pins on some receivers?
I was looking for a new receiver(think I fried mine) and noticed that some receivers have a 4th line of pins. I've only seen this on receivers where the connections are on the side instead of the top.
What does this 4th line of pins do? Will it work for my plane just the same as a receiver with only three pins? Is there a reason they only appear when the connections are on the side?
Receiver with 4th line of pins
AI: Modern RC receivers that support PWM/servo output will also support other output modes, which is what the fourth row of pins are usually for.
The extra row of pins you notice on the FlySky FS-iA6B receiver is for extra features of the iBUS signal output. The regular three-high 6-channel PWM outputs are present too.
The FS-iA6B has three different major output modes:
PWM: Each channel (1-6) is output on its own servo channel
PPM: Each channel (I think either 1-8 or 1-10) is output sequentially on one servo channel in accordance with the CPPM (combined PPM) protocol
If enabled, CPPM is output on the PPM/CH1 servo channel
iBUS: Each channel is output over a serial UART connection
If enabled, serial iBUS is output on the PPM/CH1 servo channel
Special auxilary features of the iBUS protocol are available on the 4th row of pins, but these are rarely used. The iBUS signal channel provides all the features usually required fy RC aircraft.
Of the three, iBUS is preferred for use with modern hardware if its supported because of the error-resistant and precision qualities of digital signaling protocols over "analog" ones. |
H: What would a basic fixed-wing electronic setup look like?
I'm building an FT Tiny Trainer. Currently it's just a chuck glider, but now that I have tested it, I want to add electronics. At first, the plane would be a controlled glider, but then I'd probably add a motor. So far, I've figured out what transmitter/receiver I want.
However, I am clueless about motor/servos/battery. FliteTest have "power packs" on their store, but I don't live in America and the shipping would be horrendous. I also don't want to just buy the parts in these packs from other places - my local hobby shops wouldn't have the specific ones I need, and I'd rather avoid buying online altogether. Instead, I want to find what parts I need (servos, motor, etc) and then go down to my local hobby shop and buy them.
My question has two parts, but they link together into one answer.
The first half of my question is: What electronic components do I need to make a plane fly?
The second half is: How are they linked together? What plugs into what?
A diagram would be lovely.
Also, I presume I'd need to learn to solder to do this, please correct me if I'm wrong.
EDIT:
Two months after I asked this, I went to my hobby shop and bought all of the components. I managed to avoid soldering (for now) by getting the guys at the shop to attach the connectors. Thank you to everyone who answered this question - you are the people who convinced me to make the leap and buy this stuff. The plane flies great and has enough power to hover on the prop.
AI: The main things you need are a motor, battery, electronic speed controller (ESC) and servos.
The ESC sits between the battery and motor, and it also has a thinner cable that goes to the receiver. The wires from ESCs are often labelled, but if not, there will be two big wires on one end that are the input from the battery (red for positive, black for negative), and three big wires on the other end that supply the motor. The motor wires can be any colour and it doesn't really matter how you connect them. If the motor runs backwards, switch any two motor wires. The cable to the receiver will be three wires joined together in a ribbon, similar to the cable on servos (and with the same plug)
The ESC also provides 5v to power the receiver and servos. All the positive and negative pins in the receiver are connected. The part of the ESC that does this is known as a BEC or 'battery eliminator circuit' because 'back in the day' you needed a separate 5v battery for the radio.
Servo cables have three wires, positive, negative and signal. Positive is in the middle so that plugging it in backwards doesn't cause short circuit. Servo plugs are pretty standard, although some have a little tab on one side to prevent them being reversed - but this requires that the receiver has a slot for the tab (and many don't, and the tabs get cut off).
Servo plugs are small and need a special crimp tool to make, so they are nearly always done for you - and they all use the same pins, so different brands are compatible.
You might also want plugs to connect the motor and ESC, either because the design routes the motor wires through the firewall, or so you can move components between models. 3.5mm bullet connectors are pretty standard here, with the male plugs on the motor (so you can safely power up an ESC without a motor).
If you shop carefully, you can probably find an ESC and motor with all the relevant plugs already attached but many only have the servo plug, either to reduce cost or to provide you with more choice. If it comes without plugs, or you need to change a plug, you will need to solder.
The Tiny Trainer is just about able to use the little red JST battery plugs that use the same pins as servo plugs, but it's right at the limit and old plugs tend to overheat, melt and short circuit. XT30 plugs would be a safer choice. XT60 plugs are much bigger than you need but a common standard for chargers and larger models. There are lots of other types of plugs these are the most common now.
Beware that the connectors such as the motor plugs or XT60 battery plugs require a larger soldering iron than the 20w irons that are usually sold for hobby electronics. Also, if you're soldering, you'll need some heat shrink tube to cover the joint. 5mm diameter fits XT60s and 3.5mm motor plugs. Remember to put the tube on the wire before you solder the joint, and slide it away from the heat or it'll shrink before you are ready.
If you are lucky enough to have a local hobby shop, they are usually very helpful in finding parts that will work together. Take a list of the recommended gear and they'll find equivalents, and may even do the soldering for a small fee.
When you're in the hobby shop, buy some spare props, some piano wire for the pushrods, control horns and swing keepers to make the attaching the push rods easier, and some Velcro strap to hold the battery in place. And a battery charger (with power supply). And some screws to hold the motor to the firewall. |
H: Does the RC transmitter control power consumption?
This might be a little off topic, please be understanding.
In case of an RC car toy (specifically, for children) you can control the speed of the car via the transmitter. I am wondering, if I were to customize the transmitter and limit the maximum speed of the car, let's say, to half of the original, do I (or the kids) get to enjoy playing for double the time with the same batteries?
(for the sake of simplicity, let's say that the throttle on the customized transmitter can only be pushed half way)
So does the transmitter control the power consumption of the RC car, or it's only a torque to rpm conversion?
Cheers,
John
AI: Yes, sort-of.
The speed controller switches the motor on and off in very rapid pulses, and the width of these pulses (and the width of the 'off' period) is controlled by the throttle. It's so fast that the motor just sees it as a varying voltage.
So if you cut a piece of card and stuck it on the transmitter to prevent full throttle being used, the speed controller would only run the motor at partial power. The car would accelerate more slowly, have a lower top speed and the battery should last longer. It should break less when you crash too.
However, voltage supplied to the motor is only part of the story. The current drawn from the battery is what's actually important, as battery capacity is measured in Amp-Hours (i.e. a 1AH battery can supply 1 amp for an hour, or 2 amps for half an hour, etc). The current decreases as the motor speed increases. A motor running at full throttle with no load will spin quickly but draw little current but the same motor struggling through grass will be running more slowly and drawing more power.
So if you want longer run time, drive on a smoother, lower-grip surface or change gear ratio to a lower gear so the motor can reach its top speed with less power. Or buy some extra batteries... |
H: Connect mavlink (on RPi) with QGroundControl
I am trying to connect Mavlink (installed on RPi) with QGroundControl via 3DR Radio Telemetry as to transmit telemetry data from RPi without dependency of Ardupilot/Px4 flightcontrollers. I want to achieve transmitting data such as GPS or Postion using the message format of mavlink without dependency of ardupilot/PX4.
So far, I have taken the example code from mavlink git and have built the code as per instructions. I start the QGroundControl and was able receive the System ID and component ID from QGC. I modified the code to transmit a heartbeat from RPI-mavlink to QGroundControl.
mavlink_msg_heartbeat_pack(system_id, 200, &msg, MAV_TYPE_HELICOPTER, MAV_AUTOPILOT_GENERIC, MAV_MODE_GUIDED_ARMED, 0, MAV_STATE_ACTIVE);
int len_hb = write_message(msg);
After building and executing this code, I was expecting the QGroundControl to connect/show received message, but nothing was available at the QGroundControl.
I need to understand if what I am doing is the right way to do or if there is predefined method to establish such kind of communication. Kindly, help me out with documentations (if any) available to accomplish message transmission without ardupilot/Px4. If someone could guide me on this issue, it would be really helpful.
P.S: I have posted the same Question on StackOverflow
Thanks
AI: After referring to lot of documents and mavlink message format, I was finally able to send heartbeat to the QGroundControl from the sample code without Ardupilot/PX4. I modified the code
mavlink_msg_heartbeat_pack(1, 1, &msg, MAV_TYPE_HELICOPTER, MAV_AUTOPILOT_GENERIC, 0, 0, 0);
and was able to receive the data at QGC end. It seems like default system id and component id values was key to communicate with the QGC. Now, since the first step is accomplished would look in transmitting other various parameters to QGC from mavlink RPi.
Thanks |
H: Motor will not go above half thrust. Why and how to fix?
I fixed the esc of my aircraft, but now it won't go above half thrust for whatever reason. Here is a short record of the electronics in the craft;
Crashed twice, broke esc, soldered 3-pin ground connector back to the esc, wrapped in electrical tape, worked fine, all seemed good, now won't go above 1/2 thrust.
When I go above half thrust, the motor will stop spinning, and the plane will beep a few times. After about 3-5 seconds, if I pull the throttle back down to 0 and then push back up, it will resume spinning but still will not go above 1/2 thrust.
I do have a new esc on the way from amazon so that might fix it but still have no idea why the plane won't go above half thrust. Any idea why and how to fix it?
Plane has 1 motor, rudder, and elevator in use. Ailerons were not connected for the flight when I noticed it would not go above half thrust.
AI: This sounds very much like like a PWM/analog ESC miscalibration issue to me. PWM ESCs required signal calibration to ensure that both the RC receiver and the ESC are on the same page about what kind of pulse length constitutes both 0% and 100%, as there can be clock speed discrepancies and other issues with the protocol.
I can't exactly find a source I like for this procedure to explain it better, so here's my rendition: (modified from those listed here)
With ESC power unplugged, connect the ESC signal to the receiver's throttle channel.
Turn on the transmitter and receiver and move the throttle to 100%.
Enable power to the ESC and wait for it to beep a few times indicating it's in calibration mode.
Briskly lower the throttle channel to 0% and wait for the ESC to beep to indicate it's now calibrated
Try to operate the ESC like normal and see if the full-throttle range has been restored |
H: What is the power draw of an average 5 inch FPV-drone?
I'm doing a project in school where I'm comparing different batteries for use with an FPV drone. To do that I need to know a how many watts the battery needs to be able to provide.
I found this motor test, but if I understand it correctly, the motor draws about 30 - 40 A on average depending on the propeller, and 150 - 190 A maximum. If this test is with only one motor (I assume it is, correct me if I'm wrong) that seems like a lot, considering that my drones draw about that much in total (4 motors + VTX, camera etc.) during flight. The voltage is about the same as my drones (4S), so that would result in a much higher power draw (About 9600 W vs. 2400 W).
Could someone help me figure out what is correct? Considering a typical 1550 4S battery can deliver about 200 A burst, it seems like I'm wrong about something.
AI: I think those figures are looking at very short spikes as a motor accelerates from stationary. It's typical for permanent magnet motors to have a 'stall current' that's several times higher than the normal current at their maximum power speed.
Most 5" FPV quads use 30 or 40amp speed controllers on each motor. These would melt and catch fire at 150Amps unless it's only for a tiny fraction of a second. Motors, wires and battery plugs have a fair bit of mass to them, and can absorb the heat from being overloaded for a few seconds or even a few minutes. The solder connections and circuit board traces in an ESC do not, and overheat very quickly.
The 'average' column in those tables looks like the average current for a motor at full throttle on a test stand in stationary air, not the average current during flight. You don't use full power very often, or for long, and as the quad flies faster the motors don't have to work as hard to pull air into the props.
The proof of that is that a typical 1500mAh battery typically lasts 3-4 minutes (and we aim to leave 20-30% capacity in the battery for maximum lifetime). A 3 minute flight is a 20c discharge (30amp average). A 4 minute flight plus 25% safety margin is only 12c, which gives an average current of 1.5A * 12 = 18amps - and that's the total for all four motors.
FPV batteries need to be able to handle short spikes of power, up to the maximum of the speed controllers, but most of the time they're not working all that hard. |
H: Why does motor not want to go above 2/3 thrust?
This is related to my past ESC troubles. See link for some flight history.
I tried the recalibration as suggested by ifconfig, which sadly didn't work. But now the motor will not go above 2/3rds thrust(slight improvement), no matter the calibration range. I did do one calibration where I used full throttle as max and the center as none, and the ESC would stop working once I got above 2/3rds of the calibration range.
What it is not;
ESC overheating/overloading(it is cool to the touch and I even ran it in our deep freezer, same results)
broken ESC(I used a new ESC as well as an old one I fixed. Both have the same problem)
What I'm not sure about;
the receiver has been in some crashes, perhaps it got shaken around and is messing with the signals?
the transmitter is cheap so it might be messed up, but it was working fine before.
motor might be broken but I don't think so, it's not even scratched up.
battery(charger) might be to blame. The charger I have (also cheap) has 3 lights for the cells, and the cell 1 light never turns green(ive left it charging overnight, still red). I just thought the charger was broken, but maybe cell 1 isn't getting charged?
hardware specs;
ESC is rated for 30A, uses T-plug to connect to battery.
Battery is a 3S 2200 mah 30C from Mirahobby
Motor is 2200 kv, is using a 9050 prop(slightly outside plane and motor recommendations)
Plane is an FT Spitfire from flitetest
The only modifications I have made to any of these pieces is when I globbed some hot glue onto the antennae on the receiver to strengthen it some
What do I need to fix to get the motor/esc to work properly? Do I need to replace any of my hardware? Should I just build a lighter plane that only needs 2/3 thrust?
AI: It was the battery.
Bought a new battery, charged it, all 3 lights turned green, and motor/esc has no trouble going to full thrust. Lots of power now, so that should really help the plane fly.
Problem solved!
And on the bright side, I get to tear apart the broken 3s battery and see how to fix it or turn it into a 2s. I'll probably put a question on electrical engineering SE about that later. |
H: How to keep my prop from flying off my motor?
After finally getting every piece on my plane to work, another piece decided to be a problem. Great.
My problem is pretty much already summed up in the title, my prop keeps flying off my motor, and I want to stop it from doing that because no prop pretty much means you crash.
Here is a picture of the motor pod and prop unattached from each other.
Here is a picture of the prop hardware that goes over the motor.
Sorry if the pics aren't working, I don't use them a lot and haven't quite figured out the syntax and stuff. Any suggestions on how to keep the prop on during flight?
AI: It looks like you are missing a piece of the prop adapter.
There should be three pieces, the nut, the central shaft with a thread for the nut on one end and a taper on the other, and a collar that matches the taper. You're missing the collar.
The nut forces the collar onto the taper, which squeezes the slots together, gripping the shaft. They're normally very secure but won't work without all three parts.
Here's a picture (I get an error when I try to embed it) or you could just look for pictures of 'collet prop adapter'.
https://shop.pichler.de/Collet-Cone-Prop-Adapter-60mm |
H: Video being indexed on SD card, but no duration time and no actual video recorded
I am using an HS-Series Holy Stone drone. I was able to get one video recording, but every subsequent video recording indexes the recording, but no duration time, and no actual video in it even when it says it has 23 Mb of video.
I am using VLC to view it since my MacBook Pro does not support .avi files.
What is going on?
Here is the VLC UI:
Notice how the first recording shows 00:59 seconds, but each subsequent recording I tried, shows no Duration, yet at least one of them has 25MB...of what I have no idea.
AI: Given that the drone camera is a cheap model from China, I think it's entirely possible that the camera modules is broken and could be outputting incorrectly terminated .avi files.
In earlier comments, you explain that no files are created when a known-good SD card is used to record video with the drone. My prior experience with camera modules for cheap Chinese-made RTF (ready-to-fly) drones has shown that there can be serious quality issues with the recorders. I used to have a Syma X5 model which I upgraded with the OEM camera module. The whole camera solution was shoddily made and implemented.
Based on the observations with the files created on the first SD card, one of which is unplayable but >0 bytes in size and the others which are entirely empty, this seems likely. Zero-size or otherwise corrupted video files are a common symptom of an incorrectly-terminated video stream. The .avi file format supports many kinds of video streams (wiki), but the drone camera most likely uses some kind of MPEG codec with interframe compression, which is more susceptible to corruption or incorrect stream termination than a full-frame uncompressed stream. |
H: How do I attach a propeller to this motor?
As many of you already know, I am getting into RC planes. I went to my hobby shop today and bought a motor and some props, but I can't figure out how to attach a propeller to the motor. Unlike what I've seen on YouTube videos, there is no shaft on the front of the motor. The motor is a Dualsky 2304c 1850kv brushless motor.
Apologies for bad pictures, I only have my laptop with me.
The front of the motor:
The back of the motor:
The motor accessories:
The propeller and accessories:
AI: You use one of the provided rubber rings to attach the propeller to the motor. The 2 protruding screws with their heads are the anchor points for the rubber before it is wrapped over the prop to the other screw.
The propeller motor connection is not rigid but has a little give and comes loose on impact, hopefully saving the propeller and the motor that way. |
H: Help finding ESC for brushless motor
I'm very very new to hobby electronics however I have studied a bit about basic electronics in school. I have four EMAX 0802 Brushless Motors and I want to drive them with an Arduino Nano. I know that I need an ESC, however the motors came with a specific connector. They are also very very tiny motors and I would like my ESCs to be fairly small too.
Here is a picture of the motor:
Does anyone have some advice on the best way to control this brushless motor? I assume that I am going to have to re-solder the connector since it is designed specifically for a prebuilt drone.
AI: ESCs for these tiny motors are always built into a single board that also contains the flight controller (and often the receiver and video transmitter). I'm afraid you won't find them individually. The quadcopters that they're designed for just don't have room for separate components, and everything is so tiny that the plugs and wires would be a significant weight penalty.
That said, there's no reason that a larger ESC won't work. The smallest generally available is about 6amp, but you might find 10 or 12 amp units for the same price and size - small stuff is rare and doesn't get updated as often as the more common sizes.
FWIW, I think that's a 3 pin JST 1.25mm connector. Measure the pin spacing to check. |
H: Wiring a 4-in-1 ESC to an Arduino Uno
I have four brushless motors that I want to control using an Arduino Nano. I was looking at getting a 4-in-1 ESC controller similar to this one on amazon.
At this point in my project, I'm more interested in toying around with components and understanding how to control them using an Arduino rather than having a fully functional drone. I was wondering if anyone had a good resource on how to communicate with and control ESCs using an Arduino. Is this something that can easily be done?
It seems pretty straightforward how to solder the actual brushless motors to the ESC, however I was mainly wondering how this 8-wire connector can be used to control them from an Arduino. If I directly connect the M1/M2/M3/M4 wires to pins on my Arduino, are there libraries which can be used to control the motors?
AI: If you must use dshot then try this library https://github.com/gueei/DShot-Arduino
However, for just playing around, it would be much easier to use PWM protocol and the servo library for Arduino. Most ESC's are backwards compatible so I recommend trying using PWM first and if you must, use DSHOT. |
H: My esc cuts the motor when I move control surfaces a lot
I finally built my plane and started testing it on my workbench. If I moved the servos a lot (using my transmitter), the motor would cut out, the ESC would play a tone similar to when I first turn it on, and after 3-5 seconds the motor would restart. During the reset, the servos still responded. I flew the plane with issue as I knew the extreme control inputs I had given in testing wouldn't be needed in flight.
I think the problem is probably that my BEC is too small, but I'd just like your confirmation.
Details about my setup:
I am using a 3 cell battery
The ESC has a 1-amp BEC.
I am using a Spektrum AR410 receiver (although I don't think this is relevant)
I have 2x 6 gram servos on extensions.
Depending on which direction you rotate the control surfaces, they do need quite a bit of force.
AI: I think it's a combination of the BEC and stiff controls.
1 amp BECs are often just linear regulators, effectively a transistor that acts like a variable resistor. It takes in battery voltage and turns some of the power into heat so that the remainder is 5v. If you were running a 2 cell battery, it would have to dissipate 7-5=2v but with a 3 cell battery it's 11-5=6v. That's three times as much, so it can only manage a third of the current.
2 servos are normally fine with that set up, but if the stiff control surfaces cause the servos to draw more power than normal they could cause the BEC output voltage to drop to the point where the receiver or ESC shuts down. Or maybe it overheats and there's a thermal cut off - I'm not really sure of the failure mode, but at least it didn't catch fire!
If you're curious, you could check to see how hot it gets, and measure the voltage at the receiver when the motor cuts. There are also servo power meters, or you could, HobbyKing seem to have discontinued theirs.
You may be able to solve the problem by freeing up the hinges and pushrods, but a better BEC is cheap and it's not a good idea to fly with a set up that you know could fail.
The better BECs are known as 'switching' because they switch the power on and off rapidly and then smooth the output, so there's much less power wasted as heat. Sometimes these are known as UBECs, because some early ones were sold under the name 'Ultimate-BEC'. |
H: What is Commutation Enhancement in BLHeli_M and what should i set it to?
In the BLHeli_M version of the BLHeli Configurator, theres an Option for Commutation Enhancement. By default its set to Async-PWM but it can be set from 3%-CE to 100%-CE. What does this Parameter do and what should i set it to on a Mobula 6?
AI: The BLHeli_M Commutation Enhancement feature is intended to increase motor power response times by "overdriving" (my words, may not be the most appropriate analogy) the coils to charge/discharge them more quickly.
Source: (rgroups.om)
Commutation is a job to charge two of three coils, A B and C, sequentially.
Enhancement is an attempt to make the charge time shorter.
Sync-PWM try to apply power at the very right moment without delay.
Enhancement try to make the charge speed faster.
Source: (rcgroups.com)
Commutation Enhancement(CE)
Use more power when charging the coils to speed up settling time of motor current. RPM will be higher with the same throttle.
Many request for normal RPM after disarm. So CE will be off right after startup. Then is on after a certain dshot transistion detected. After that, CE will be off after arming(dshot=0).
As for advice on whether or not to enable it on your Mobula 6, I'm not quite sure. From reading the development descriptions written by JazzMaverick (the primary maintainer of BLHeli_M), it appears that the feature was built to "accelerate" motors/ESCs in the 5" class, although I do see a couple mentions of applications for whoop-sized drones in the RCGroups thread.
I'd try leaving the defaults as-is and then coming back to investigate your options for changing the commutation enhancement setting later on if you feel like the quad's motor performance is lacking. Granted, I've never used the feature before, so take my recommendation here with a grain of salt. :) |
H: AliExpress drone vs DJI Pro 2
I can buy a drone in US or Canada like the DJI Pro 2 for over 1000$. But when I search professional drones on Ali they advertise many that share many features such as 4K video, dual camera, 20minute flying time. The one main thing I don’t see on many drones on Ali is the long range ability which DJI has. Is that really what I’m paying for with DJI, the range? It would be great if someone here has bought a drone from Ali and can lend some advice. And if you know of any decent Ali brands of drones please recommend.
AI: It's relatively easy (for you or a factory in China) to buy a set of components and put together a drone, but just looking at the raw specs doesn't tell you the whole story.
For a start there are some very cheap 4k cameras, with poor dynamic range, colour balance and low-light performance. Image quality is quite hard to quantify, and even if they do give you any figures, they are hard for a beginner to understand.
DJI drones have a set of sensors to detect obstacles and prevent you flying into them. Cheaper drones generally don't, but even if they can claim more sensors that doesn't necessarily translate into better obstacle avoidance.
Then there are various software modes, usually with fancy but not very descriptive names. These may be really useful for whatever sort of video you plan to take, or you may never use them. There's no kind of industry standard so it's hard to tell if it's just a gimmick.
In this sort of market, where there's one big established make, there may be a small, innovative company with a superior product at a lower price that is struggling to be noticed - but there are also a lot of copycats who will throw together the cheapest possible thing that will look the same on the advert (and with no after-sales support).
Be careful not to "judge a book by it's cover" - look for reviews and instructional videos, and try to work out which are just thinly disguised adverts. See if there's a big fan/user forum, and if the manufacturer ever answers their questions. Or just buy a popular product that you can see works well for lots of people. |
H: recommendations for a DIY/dev platform lifter drone
I was mainly wondering if there was something in the drones world, similar to what the ender3 is to the 3d printing community -- that is, a cheap kit massively owned by makers and open source enthusiasts.
I'd like to get a list of components I can, afterwards, highly customize. I can make 3d prints.
For now, what I've found for the features I want, are :
a thingiverse post of the mavic air/pro cloned foldable frame.
However I don't know if it's as worth as the "open flat frames" used
in plenty of youtube videos for DIY racer drones... In the sense of
payload carrying ability and speed limits/agressive maneuvers.
pixhawk as the flight controller board and px4 firmware. I would need at least one extra pin (more is appreciated) for controlling a
servomotor for custom experiments. Is there any better solution?
found out that the more blades I get on a propeller the more silent it will be. However I didn't found any foldable propeller
with more than 2 blades... You guys did?
some sliding design for swapable battery, such as on the autel evo 2, that I can't afford because it costs 2k$ and I don't even
know if it's customizable... Any thingiverse post you guys saved**?**
But my BIGGEST wonderings are for the FPV feature :
All the tutorials I've found use some DJI digital "fpv unit" to cast what the drones sees into operator's goggles. But if I don't want to wear goggles, but instead, still cast onto the remote controller screen, is there any build logs/tutorial for that?
How to get the remote controller screen to be displayed on a smartphone? Is there such famous couple <RC remotes, android app> that not only work with DJI models or specific brands? Or some generic opensource project I could flash on micro controllers and buy appropriate antennas. For now I've found the 300\$ skydroid T12 or the 800$ herelink but they are too expensive for me on a budget.
I heard about betaflight for assigning maneuvers to keys of a custom controller (such as gamestation ones), but these devices aren't radio-enabled (long range + anonymous), nor able to redirect the videostream into a smartphone, isn't it ?
Is there android apps for fully controlling a drone through smartphone (virtual joysticks on screen) over wifi while getting its point-of-view livestream on the phone screen?
By thanking you
AI: To answer your main question, I'm afraid there's nothing as widely used the Ender3.
It sounds like you want to build something with more lifting capacity and space than a race quad, with GPS.
There are two main styles of frame that might suit you.
One is based on the old 'DJI Flamewheel', which has plastic arms attached to hub of two flat plates. You can build these with as many arms as you want but the arms are only available in one size.
The other option is the many designs that use carbon fibre tubes. The tubes are very light and strong, and fairly cheap. Commercial designs use injection moulded or CNC aluminium fittings, but these would be perfect for 3d printing. Here's one I found but I'm sure there are others.
Unlike a 3d printer, the mechanical design of a drone really isn't all that important. You could use bits of broom handle and plywood, and it would still fly. That's lead to a wide range of designs and no clear market leader.
I'll have a go at the other questions:
sorry, I don't know much about the options here.
more blades is less efficient, which is probably why you can't find any multi-blade folding props. I'm not sure that more blades are quieter, unless you're comparing it to the same size 2-blade prop spinning faster.
most people use Velcro straps, which are lighter than sliding mechanisms, and work with a wide range of battery sizes. Batteries need replacing every few years, and you can't guarantee getting the same exact size.
the DJI video unit is the only real option for high quality digital video at the moment.
This is different to the video used in quads like the Mavic, that transmit via wifi to a phone.
Racers use much cheaper analogue gear based on old CCTV signals, but the picture quality is limited to old PAL/NTSC TV resolution. It's easy to send an analogue signal to a screen, and you can buy screens with built-in receivers. You can't always see screens in strong sunlight, which is why people use goggles.
Systems that transmit to a phone are generally proprietary and only available as part of a complete drone. They also have a noticeable lag, due to the video compression/decompression required, and other processing on the phone. That's why they're preferred for taking video, but not for high-speed flying. At the moment you can't have both (although the DJI air unit is close).
betaflight is the firmware run on race drones. It's not aimed at autonomous flight like the Pixhawk.
No, because there's no common wifi module for drones. |
H: Betaflight attitude update rate only 100hz?
In the tasks.c file, this line in particular
[TASK_ATTITUDE] = DEFINE_TASK("ATTITUDE", NULL, NULL, imuUpdateAttitude, TASK_PERIOD_HZ(100), TASK_PRIORITY_MEDIUM)
If I understood this correctly, attitude is updated at 100hz only by default. Why then do we need gyros which update at 8kHz rates?
AI: I haven't quite dug into the source code to confirm my suspicions, but this file lends credence to the idea that this task facilitates accelerometer-based attitude calculations and not gyroscope-based calculations.
From line 274 to line 280 in the file you link to (tasks.c), the TASK_ATTITUDE task is said to be enabled if the accelerometer device is enabled (in the Betaflight Configurator):
#if defined(USE_ACC)
if (sensors(SENSOR_ACC) && acc.sampleRateHz) {
setTaskEnabled(TASK_ACCEL, true);
rescheduleTask(TASK_ACCEL, TASK_PERIOD_HZ(acc.sampleRateHz));
setTaskEnabled(TASK_ATTITUDE, true);
}
#endif
Given that these are the only times that the TASK_ATTITUDE task is mentioned in this file, I doubt that this 100 Hz task is related to gyroscopic sampling. |
H: Can I re-flash the Hubsan fpv1 controller with built-in fpv screen?
the only radio transmitters, also called remote controllers, that embed a fpv screen, for seeing what the drone sees, near the joysticks, instead of having to get obstructed vision by goggles,
that I found, are :
https://www.youtube.com/watch?v=df0Cvv3T69Y called fpv1 (i'm not interested into h7000 as it runs android and requires wifi rather than RC)
https://www.youtube.com/watch?v=XBbptOuXXdA called fpv2
However they only seem to be able to bind Hubsan drones...
Would it exist a way to flash them in order to make them run openTX firmware, or anything that could make them become universal controllers?
Alternatively, what parts from Hubsan should I buy if I want to put in my own custom drone build and be compatible with the DIY ecosystem ?
Can I build from scratch a controller with arduinos so that I could use openTX, embed an fpv screen, while still getting decent range (>5km) ?
AI: I'll start by warning that you may have misconceptions that are leading you in the wrong direction. The reason that FPV screens built into transmitters are not more popular is that screens aren't bright enough to compete with sunlight. The primary function of goggles is to cut out bright reflections, so you can see the picture.
Also, transmitters are comfortable when held down at your waist, which would make the picture on these transmitters too small. You'd need something the size of a laptop or tablet computer screen to work well at arms length. The picture from an FPV camera is pretty low quality by modern TV standards, and needs all the help it can get.
The transmitters you link can't be flashed to run OpenTx. They just weren't designed to be modified like that. They may not even have enough flash memory or processing power, as OpenTx has vastly more features. Similarly Hubsan doesn't sell a stand-alone receiver for their transmitters.
You probably could buy an OpenTx radio and transplant the electronics into one of these cases, but I think you'd have to find something clever to make their TV format screens display information that was designed for a dot-matrix screen. You could look at the boards used for OSD (On Screen Display) but I imagine it would get very technical.
I think your best bet would be one of the many FPV monitors, mounted to the top of a normal transmitter - just be sure to stand in a shady spot. |
H: Can something be mounted underneath a quadcopter rotor?
I am a bit of a beginner, so I don't fully understand the way the air flows around the rotors in a quadcopter. My question is in the title, and I am basically wondering if something can be mounted adjacent to the motor, and directly beneath a part of the rotor, or if it will cause poor performance on that rotor. I have a very long camera which I can mount to the front of my drone body, but the edges of the camera lie directly underneath two of rotors. Will this cause performance problems/is it a better idea to mount it above them? Thanks!
AI: If you block the airflow into or out of a rotor, you reduce the lift it can generate (approximately) in proportion to the area you block. That means the motors will have to work harder, and your flight time will be reduced.
Work out what percentage of the total rotor area your camera will block. If it's more than about 20% you should probably look for a larger frame.
Obstructions above the rotors are less critical when hovering (as the rotors are sucking in stationary air from all sides) but in forward flight the airflow is more linear - and rotors don't like working in disturbed air so it's worse for high speed flight. |
H: Should VTX ever be lower freq./higher power than the control transmitter?
I can find video transmitters from 5.8GHz down to 433MHz with the latter being ideal for longer range, but I don't see nearly as much diversity as far as control transmitters. Am i correct in saying that a lower frequency transmitter will almost always go much further than that of a higher frequency, with transmit power being secondary in affecting range? The extension of this would be that the control link would be lost far before the video if the video is TX'd on 433MHz and control on 2.4GHz. I assume that this would be dangerous since, for example, if video is gradually lost, you have a chance to return the craft to a location if better signal, but you don't have obvious warning signs for control being lost.
Edit - I know that there are 72MHz control transmitters, but i don't know of any way to use them on a quadcopter.
AI: There's a few factors that you're missing. The first is that the RC world is dominated by traditional line-of-sight control, and 2.4GHz provides more than enough range to fly a model plane to the limits of vision. There aren't many long range system because few people need them. Even for FPV, long distance flying means a long walk or losing your model when you crash.
The other is that control is very low band-width compared to video. A typical 2.4GHz protocol includes error detection and can reject a lot of packets and you won't notice so long as one or two get through every second - maybe even less when you're at extreme range and not manoeuvring hard. A video signal needs most of the data to get through to be flyable. Analog video has no error handling or rejection, and while digital systems do, there isn't enough spare bandwidth to ignore any data. Even a corrupt picture is better than black screen!
Good RC systems these days do include RSSI (Received Signal Strength Indication) in their telemetry, so you know when you're reaching the limits of your range - and models with GPS can be set to return-to-home automatically if they loose the RC connection.
Finally, it's entirely possible to use 72MHz on a quad, with a FC that takes individual inputs for each channel - however the range is no greater than on 2.4GHz, and there was rarely any interference rejection on the old analog radios, we used to use the glitches as an indication that we were flying too far away! |
H: Connecting 4 in 1 esc with power module
I have a 4 in 1 iflight esc currently connected to my motors. I am attempting to interface it with my flight controller (Pixhawk 4), which comes with a power module (https://docs.px4.io/master/en/power_module/holybro_pm07_pixhawk4_power_module.html) in order to provide power to my escs and flight controller. This module has individual power outputs for every motor (each output is supposed to connect to a single esc, while the 4 in 1 esc only has one power input. Could I provide power to the motors by wiring all of the positive outputs together of the power module, and using them as the input to the esc, or would this create issues beyond losing the ability to track the power going to each motor? I am using a quadcopter. Thanks.
AI: I think yes, this is doable. In this diagram you can replace the ESC Motor 1 with your 4 in 1 iflight ESC. Note that you will probably need to redirect the PWM outputs to your 4 in 1 ESC signal inputs.
Could I provide power to the motors by wiring all of the positive outputs together of the power module
No, no need to do that. Use a single power output of the power board (it is simpler and in certain cases more reliable).
EDIT
Following the discussion in comments, your 4 in 1 ESC already acts as a power distribution board. Depending on your situation, you may not need the power module at all. For example, one could split battery power output into two: one to go directly to 4 in 1 ESC, and another one to an UBEC to get 5V (to power the flight controller). |
H: How high can the Mavic Mini fly?
I was wondering how high the DJI Mavic Mini could fly before the air became too thin. I know that there are restrictions in the app (and probably in the drone/controller itself) but if you were able to override those things, how high would it go (not including battery life)? Thanks in advance!
AI: The maximum altitude is 3000m (about 9800 ft.)
(Note that this is relative to mean sea level, not the ground.)
Sources:
DJI Mavic Mini "Specs" page (under "Max Takeoff Altitude")
DJI Mavic Mini "FAQ" page (under "Propulsion" > "What is Mavic Mini's maximum service ceiling?") |
H: Diy drone Digital control/communication
Is there a solution for digital drone communication through Wifi for diy fpv drones?
I'm searching for a way to build a drone that have multiple cameras and communicate with my computer through wifi, for control/autonomous drive/etc, i need something like the dji tello communications, but for a diy drone since i need the customization with diy drones (small form factor, long flight time, range, etc) which i need to control those parameters.
a diy solution is ok even preferred.
AI: There isn't an easy solution, because this isn't a common thing to do.
The main issue is that real-time control for FPV requires very little lag. Wifi requires several layers of processing, on a computer where it has to compete with other processes for processor time. Your video needs to be encoded and compressed, and still work if some data is missing. Most digital video isn't real-time, and relies on a buffer to smooth out these delays - but you can't fly with images that are a few seconds old.
That said, it's pretty simple to stream video over wifi with a small computer like a Raspberry Pi, and there are lots of guides on how to do it. Similarly there are lots of ways to send other data over wifi - it's just networking, and there are lots of guides on how to use UDP or TCP/IP.
You probably can't use the same computer to actually fly the drone. Again it's a real-time system that needs to adjust the motors hundreds or thousands of times each second. I think you would be best to use a normal flight controller running iNav, and to send it control signals from the Pi, via S-Bus. Effectively the flight controller attempts to hover at a GPS position, and you send it a new position occasionally.
I've not actually done this myself, or even read about anyone doing it, but I suspect it has been done. |
H: Why can't quadcopters handle failure of one rotor?
Ardupilot can detect (and it logs) motor output mismatch. I think detecting in flight motor failure is possible with modern firmware and flight controllers. Then if one rotor fails, why can't we have the quadcopter fly like a tricopter, since we now have ESCs that can change rotation directions?
AI: It depends on what you mean by "fly"
Raff d'Andrea's team at EPFL wrote a stunningly good paper on this subject, with practical demonstrations indoors and outdoors. The paper, Relaxed hover solutions for multicopters: application to algorithmic redundancy and novel vehicles, can be found at https://www.research-collection.ethz.ch/bitstream/handle/20.500.11850/155683/2/ETH23137.pdf#page=108
They accomplish this feat of flight by relaxing the constraints on the vehicle. If the goal is to simply stay put in XYZ space, and allow as much banking, pitching, and yawing as required, then this can be done all the way down to one motor.
If, however, you would like your quadcopter to be able to lose a motor and retain full control over roll, pitch, yaw, and thrust, then it cannot be done.
Normal flight with all four motors
I'll pull from an answer at How does a quadcopter yaw?, only I'll work the example of what happens when we lose a motor.
Let's start with the driving equation:
$$
\begin{pmatrix}
\tau_x \\
\tau_y \\
\tau_z \\
F_z
\end{pmatrix} = \begin{pmatrix}
0 & 1 & 0 & -1 \\
1 & 0 & -1 & 0 \\
-1 & 1 & -1 & 1 \\
1 & 1 & 1 & 1
\end{pmatrix} \begin{pmatrix}
\omega_1^2 \\
\omega_2^2 \\
\omega_3^2 \\
\omega_4^2
\end{pmatrix}
$$
What this does is it maps rotor speeds (squared) to torques about the roll, pitch, and yaw axes, as well as the net vertical thrust.
What we need to do for any desired torques and thrust is to solve for the required propeller RPM. So that means we need to multiply by the inverse of that 4x4 matrix:
$$
\begin{pmatrix}
0 & 1 & 0 & -1 \\
1 & 0 & -1 & 0 \\
-1 & 1 & -1 & 1 \\
1 & 1 & 1 & 1
\end{pmatrix}^{-1} \begin{pmatrix}
\tau_x \\
\tau_y \\
\tau_z \\
F_z
\end{pmatrix} = \begin{pmatrix}
\omega_1^2 \\
\omega_2^2 \\
\omega_3^2 \\
\omega_4^2
\end{pmatrix}
$$
For reasons we won't get into here, the inverse of that matrix exists and thus we can find the solution.
"Normal" flight with three motors
Let's eliminate a motor. It doesn't matter which one, so let's arbitrarily remove $\omega_4$.
$$
\begin{pmatrix}
\tau_x \\
\tau_y \\
\tau_z \\
F_z
\end{pmatrix} = \begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & -1 \\
-1 & 1 & -1 \\
1 & 1 & 1
\end{pmatrix} \begin{pmatrix}
\omega_1^2 \\
\omega_2^2 \\
\omega_3^2
\end{pmatrix}
$$
Note that we still have the four reactions: roll, pitch, yaw, and thrust. But we only have three motors to control those four concepts, and we wind up with this 4x3 matrix. And we cannot invert a non-square matrix. So there is no ability to find three motor speeds which will independently control the four variables.
A different perspective
The takeaway from the above might not feel completely satisfactory, because it's not necessarily intuitive. So let's think about things in terms of trying to work out what happens when we try to stabilize the craft with three motors.
For the sake of argument, let's say that we are flying a + configuration where we lose the forward motor. If you look at the equations in How does a quadcopter yaw?, then you see that the reason a quadcopter can be stable in roll and pitch is that as one motor in the pair speeds up, the other slows down.
Let's allow for the remaining pitch motor to spin backwards and forwards so we can always produce a stabilizing pitching force. In this case, I think we can both agree that the average speed is going to be 0 (or close to 0). However, our roll forces remain unaffected, and so those two motors will have identical speeds, W. So what happens to yaw?
torque_z = 1*(W^2) - 1*(0^2) + 1*(W^2) - 1*(0^2) = 2W^2
So in other words, when we keep the quad from rolling and pitching, we get a really bad yaw moment!
And thus this is how Raff's team keeps their quad in the sky. They give up control of yaw. |
H: What are controllers of Parrot AR drone 2?
I have a few questions about Parrot AR drone 2. Here is the official website
Q1:What are the controllers for the AR drone 2? Is there a position controller (GPS), velocity controller (accelerometer), attitude controller, heading controller (compass), and/or altitude controller (barometer) in the microchip of AR drone 2?
Q2:If there are some of the controllers that I mentioned above, and these controllers may need the current states for feedback control purpose, do the current states form the estimation based on IMU data?
To be more specific, if there is a PID position controller, we need the desired position and current position information for feedback controlling. Is the current position from the estimation algorithm (like EKF) based on IMU sensor data?
Q3:Is there any way that I can control the motor speed for AR drone 2 directly and use my own controllers?
any suggestions and comments would be helpful! Thanks in advance!
AI: After reading relevant materials, i think I have find the answers for this question.
Q1:Since AR drone2 is a proprietary system, this company didn't publish any documentation about the internal structure.
Q2:According to my understanding and guess, the controllers will use the estimated states of AR drone2 and the estimated states may from EKF(I am not 100% sure)
Q3:Since AR drone2 is a proprietary system, it seems there is no way you can control the motors directly by using your own controllers. But there is an alternative way:
using Simulink and Optitrack to control AR Drone
using Simulink and Vicon to control AR Drone |
H: Ftd driver for 3dr telemetry radio
I have been attempting to connect my pixhawk 4 to qgroundcontrol using an old 3DR telemetry radio set. The connection works fine over usb, but qgroundcontrol shows disconnected when I plug in both telemtry radios. I know the radios are connected properly, and both show a solid green light and flashing red when connected the telem 1 port. When connecting to telem 2 it shows solid green but no red. According to the docs, solid green means the radios are connected, and blinking red means data is being transmitted. Therefore, I thought the issue resided in the usb connection. I'm pretty sure I was correct, as when opening the devices menu (I am using Windows 10), there was a yellow triangle next to the device icon, and it could not find a driver. The device was titled F231x usb uart, so I tried to find drivers online, but am struggling to install them. I read they were the same as arduino usb drivers, so I downloaded that. However, there are two inf files, 'ftdibus' and 'ftdiport'. Which do I choose, and how do I install it properly so that the radios will work?
AI: FTDI drivers can be downloaded from the FTDI Website.
You should then be able to install the driver using "Update Driver -> Browse my computer for driver software" and selecting the folder containing the extracted files. |
H: Proper rotor fastening (Nyloc nut)
I am trying to figure out how to properly fasten my rotors on my quadcopter. I am pretty sure the nuts are Nyloc, but here is a picture for reference:
I read online in many places that you should just tighten the nut until you can't anymore with your hands, and some people recommended using pliers or socket wrenches. My problem is, no matter what tool I use, the nut will not tighten past the point where its top is flush with the shaft (I don't know much about Nyloc but this seems by design). Here are the motors I am using (https://www.amazon.com/iFlight-2750KV-Brushless-Racing-Quadcopter/dp/B07Y9JK2MW). This first result is what I achieve almost every time after tightening as far as the nut will go:
But this causes issues because I can tip the rotor in different directions with my hand, and so it seems like it would wobble when turning, and the fit is definitely not secure. (I at first thought this was because the 5mm hole in the rotor was too large for the 4mm shaft, but I was told that the threading in the shaft was 5mm so it would be fine). This second result was achieved after inserting two washers between the nut and rotor, tightening down the nut, untightening and removing the washers, and the tightening the nut once more.
I removed the washers because while I had read that some people tighten their rotors with washers, the gap seemed too large for this to be the solution, so I assumed I was doing something incorrectly in the mounting. Where did I go wrong, and how should these rotors be mounted?
AI: I've had the same thing happen when using these same motors and props (and others). The propeller seems to get stuck at a point where it is not flush with the motor. From my experience, the solution is as simple as tightening it further. I use a tool like this that allows me to apply more force than a typical prop-tightening tool.
It should look like this when fastened correctly, with the prop sitting flush with both the motor bell and the nut: |
H: XM+ bound to FRSKY x9 lite, but commands not showing up in betaflight
I'm very new to making drones, so sorry in advance for lack of proper terminology or if this is a dumb question. I've bound my xm+ to my transmitter (solid green light showing on receiver), but for some reason my inputs aren't showing up in betaflight. I have the correct serialRX activated on the correct UART (and have also tried both) and my receiver mode and serial receiver provider are set correctly (to serial based receiver and SBUS, respectively). This problem arose after I accidently connected my receiver to my flight controller backwards, and have already tried replacing the receiver, as I thought I might have fried my XM+, but that didn't solve the issue. Any help is much appreciated!!
AI: Since you connected power to the SBUS pad it is possible that the inverter that sits between the SBUS pad and the UART pin on the microcontroller is fried.
To get around this, I would suggest connecting SBUS to another RX pad. However, since SBUS is an inverted protocol, you will first need to get the uninverted signal somehow. There are two ways to do this with the XM+ (other SBUS receivers have similar methods). The first option is to solder the wire that goes to the RX pad to the leg circled in red in the image below.
source
The other option is to build your own inverter using a transistor and two resistors.
source
The first option is probably the easiest if you feel confident enough to solder to that small pad, but it does come with an increased risk of damaging the receiver. |
H: DJI Mavic Mini 2 can't record 2.7k 60fps after update
As announced, DJI Mavic Mini 2 should support 2.7k 60fps.
I have the latest update of 01.02.0300 but still do not see the option.
Does anybody have the same problem ?
Thank you!
AI: After a bit of research it seems like the problem is likely to be that you don't have the latest version of the DJI fly app. As per this forum thread you will need version 1.2.4 or later to see the option for 2.7k 60fps.
On android you will need to download the latest APK, while on iOS it can be downloaded from the App Store. Links to both version are availible on the dji website. |
H: What types of cameras don't have wide angle view? Are those a good fit for a racing drone?
I am looking to make an fpv racing drone but I want to find a small, lightweight camera without the wide angle lens. Should I use this camera as my main fpv camera or mount a wide angle camera for my racing drone?
I am thinking of just mounting one camera (the one without a wide angle lens equipped), but if a wide angle lens is better than I will mount both cameras. One for recording faraway footage (essentially use it as a hobby camera when not using it for racing), and the other one as my primary camera.
so, should I aim of using the wide angle lens or a normal angle lens camera? Also: what's the name of cameras without a wide angle lens?
AI: I don't know if there is a special name for cameras that don't have a wide angle view, but such a camera is not the best fit for an FPV camera.
If you do want a "small, lightweight camera without the wide angle lenses" anyway, and you don't care about the quality, look for an analog FPV camera with a long focal length. As I understand it you want to record footage as well, and depending on your needs and use cases there might be better options.
Why should you get a camera with a wide FOV, and what options are there if you want to record as well?
All the information about where you are flying comes from the FPV camera (unless you fly line of sight), so you will want to get as much information as possible from it. A wider field of view will give you that.
Most FPV cameras are not well suited for recording (high quality) footage, and most of them can't record onboard (meaning that you will have to record the lower quality footage that is sent to your video reciever). If you only want one camera, and you want to record acceptable quality footage, the DJI FPV camera is probably the best option since it can record onboard the drone. That camera (like a lot of other FVP cameras) has a FOV of 150°, so the fish eye effect isn't that noticeable. There are a few drawbacks with the DJI camera if you want to race though, since it is a digital system (See the question Difference between analog and digital FPV systems?), so an analog camera might be better if you are okay with recording low quality footage.
If you want to record higher quality footage than what the DJI camera or an analog camera provides, you will need to mount a GoPro or similar on your drone, alongside the FPV camera. This will significantly increase the weight, so you might want to avoid that, or just remove it when racing. |
H: Working on a Quadcopter. Unsure if this flight controller supports an FS-IA6 receiver
I am really new to drone building and I have had some awful luck trying to figure out what is compatible with what. Here is the link to the AIO flight controller I chose to use for my build. It won't be a tiny build but I will still use this flight controller --> The Flight Controller
I am also using this Fly Sky Remote and Receiver --> The Remote and Receiver
My main concern is using the FS-IA6 receiver with the SPI RX setup on the flight controller. Does anyone have advice on such a configuration? Any advice is greatly appreciated.
AI: So, I don't believe the components you have are entirely compatible with one another. The FS-iA6 RX only has PWM/PPM outputs, while the BetaFPV F4 AIO 1-2S FC requires some kind of serial RX. The "SPI RX" module mentioned in the spec sheet appears to just be a small FrSky RC module that BetaFPV sells and not something that would allow you to connect to your RX.
I think your cheapest mode of recovery is to get a <$15 USD FS-RX2A receiver, like this listing on Amazon. This model is featured on your FC's Amazon listing, including a wiring diagram:
It has a serial port output (Flysky iBus), which should be entirely compatible with the FC and Betaflight. |
H: Help with finding a motor to rotate the plane of a propeller
I don't really know if this is related to this community but it's for a drone so it's worth a shot.
I'm looking for some sort of motorized hinge or a motor that can move a propeller on the Y-Axis, a gif will be below on how it might move a propeller. (Pretend The Torus Is A Propeller)
AI: I think what you are looking for is a servo. At least, that's what it's called in the hobby, officially it's a servomotor. Each unit consists of a motor, gears, a sensor to measure the position of the output shaft, and a comparator circuit. The electronics compares the requested angle to the actual angle, and drives the motor as required to minimise the difference.
They're commonly used to control the steering for RC cars and boats, or the control surfaces in RC planes. They are specified by their movement speed and torque. Better servos have ball bearings on output shaft, and you can get metal gears for tough applications.
You might want to look at the servos used for retractable landing gear on RC planes. These servos only have two positions, and are generally geared a bit lower for extra torque.
Servos generally run at 4-6v and are controlled by a pulse width modulated signal from an RC receiver. Alternately you can buy a 'servo tester' for about $5 to generate this signal, or it's easy to generate from an Arduino. |
H: Please recommend a soldering iron for hobby use
I'm frustrated with cheap irons that take ages to heat up, don't hold a consistent temperature, only have one size of tip and break after a year or so.
What sort of thing should I buy that can handle small wires on a flight controller and larger power connectors?
I think I want something with some sort of active temperature control so it heats up quickly and holds its temperature accurately, but does that mean I need one with a digital temperature setting and a bulky box? I'd rather it wasn't too large, if possible.
I only use it occasionally so I don't need industrial quality, but I would like it to last years.
AI: Another product from the same OEM, I'd recommend getting the MiniWare TS80P. (OEM product page, Banggood product listing, Slightly-more-expensive Amazon product listing) MiniWare makes a whole bunch of different nifty electronics tools, including soldering irons, pocket oscilloscopes, etc. I personally have the older version of this product, the TS80, and I have no complaints from my year and a half of casual usage.
It's a 30W iron powered off of a Quick-Charge 3.0 USB connection and has a wide variety of different tip shapes. This also means you can use it with QC 3.0 battery banks. The power rating may seem tiny, but I can vouch for the exceptional performance of the older TS80's 18W power output, which has been more than enough for almost anything I've ever needed it to do.
Like @conditioHumana mentions about the TS100, you can also flash custom firmware to the STM32 MCU inside it, but I've never felt the need to because the interface has been just fine for my purposes. |
H: How to connect ESC to serial?
I am looking for a way to access to configuration of Blheli-S firmware running on a Diatone Mamba F30Mini:
It seems possible using :
blheli-configurator
BLHeliSuite
There is explanation on different site about FC Passthrough. This seems to use connection to a Flight Controller.
As I am planning to use an Arduino nano to connect to the ESC, I am wondering how to make the connection.
Is it possible to connect more or less to connect the ESC to a PC ? or connecting through the arduino ?
AI: Thanks to informations given by @ifconfig, using BLHeliSuite connecting :
GND to Arduino GND
ESC1 to Arduino D11
It allow to upload to Arduino the "Arduino Bootloader"
Then selecting "ATMEL BLHeli Bootloader (USB/Com)", and go to "ESC Setup" tab
Finally quite straight-forward, with a tool that may be more user-friendly. |
H: How to get a licence to fly a drone in the city of Prague for a week?
I would like to take my drone and go fly in the city of Prague - it's super beautiful there during lockdown. (I'm close by anyway so don't worry, I'm staying safe) But -
According to Czech law it's illegal to fly a drone within city borders. You need to get an extra licence/allowance for that. My own shortcoming is, I cannot speak Czech.
Does anyone of you have any prior experience in getting this licence? I read that you can apply for such a licence, but I neither know where nor how. On the other side, I read that people already got into trouble, especially when flying in Prague (residency of the president, etc. etc.)
I do have a registration and the required European licence to fly a drone. I'm really looking for specific allowance for Prague.
(I already asked in travel and they were kind enough to forward me here)
AI: Firstly, in addition to the common European rules, the whole territory of the Czech Republic is located in the LKR10-UAS restricted airspace. That basically means that the older stricter national rules remain in power in addition to the default lax European rules.
Secondly, most of Prague is a densely built area. The UA operation is forbidden with the exception of UAs in the SPECIFIC or CERTIFIED category that are licensed by the CAA (ÚCL). The OPEN category is not allowed. Toys that do not require registration are allowed.
Additionally, the territory of Prague is located in CTR Ruzyně where the rules for UAs in the controlled airport area must be observed. Thankfully, the operation of drones is possible up to 100 m AGL if farther than 5500 m from the airport.
Also, most interesting part of Prague is located in the LKR9 restricted airspace. You cannot operate a UA without an operation approval from the CAA (ÚCL). Do note that you will need the approval for each operation, you will not get a blanket approval to fly there for a week, at best for some specific repeated flights that follow some common pattern regularly.
So realistically, you need to get your drone running in the SPECIFIC or CERTIFIED category in the Czech Republic and to get yourself certified to operate UAs in one of these categories. You wrote you have "a license" but from a certain naivety of the question I have to assume it is actually just a registration of a pilot and drone in the OPEN category, but do correct me if I am wrong.
Be prepared for a lot of paperwork. It is basically meant for professionals conducting for-profit aerial work in the area.
Then you can get approval to operate them also in the LKR9 airspace. |
H: How common is it for drone operators to actually follow all the regulations?
If you look at the list of rules for drone operators, it seems like the hobby is getting harder and harder to enjoy. Registration requirements, ID broadcasting, bans on flying over buildings of any kind, bans on flying over national forests/parks, professional licensing requirements, etc. But in practice, how common is it for people to actually follow all these rules for casual drone operations? Are there any surveys available that measure the compliance rate?
AI: As with any new technology, at first things are quite chaotic. Take my country for example, a few years ago, the regulating agency of flight introduced a scheduled drone flying program. This meant you had to make an appointment in advance with 60 days (!!!) and specify where and for how long you were flying. Also it was necessary to have your vehicle registered which at the time was a terrible process. Of course, people would still fly their vehicles and eventually this implementation has been removed and now people can fly freely.
Latest generation drones have a no-fly zone in their memory, so whenever you try to go in those areas they won't go. Flying over buildings in cities is quite challenging due to various factors I won't go into. Flying over forests while inside forests is also challenging. If you build a custom drone using a different transmission channel you can theoretically fly over building, forests and so on without any technical issues.
These rules are made for the future. As the technology will become more available for the general public, there needs to be an controlled environment so that anyone can fly their vehicle safely. Take aerial transportation for example: they have very specific routes and timeslots when each plane can fly. They most likely said the same thing as you about planes back in the 40's or 50's.
Also certain bans are against people that cannot breathe and think at the same time and would cause serious damage to their surroundings. This why I support the idea of pilot testing and registration. Pilots and decision-making people are expected to know what they are doing.
People in my country still fly these vehicles without any permits or registrations, but also the governing body doesn't care and the city police is clueless about required paperwork when flying a UAV.
Yes...people in the city I live in fly drones in the city...and post it on our favorite social platform...and I know some of them, and no, they don't have registration or permits...they also don't care... |
H: Is Psychiatric History an Obstacle to Obtaining a Part 107 Certificate?
Certain psychiatric conditions will prevent persons from obtaining a commercial pilot's license. Do the same restrictions apply to Part 107? If so, what means are there of obtaining clearance?
Background Reference: Guide for Aviation Medical Examiners: Item 47. Psychiatric Conditions
AI: There are no specific medical restrictions placed on drone pilots in the US, however the pilot should not have reason to believe their state of health (physical and/or mental) will affect the flight:
§ 107.17 Medical condition.
No person may manipulate the flight controls of a small unmanned
aircraft system or act as a remote pilot in command, visual observer,
or direct participant in the operation of the small unmanned aircraft
if he or she knows or has reason to know that he or she has a physical
or mental condition that would interfere with the safe operation of
the small unmanned aircraft system.
Source: eCFR 14.I.F.107.B.17 |
H: why do drone frames have spaces and holes
This may sound like a very dumb question, but I am building my first quadcopter and I just realized that some frames have these perfectly symmetrical holes/spaces.
I used up all the screws on mine, so I don't think I have to actually use those spaces, but why are they there?
Is it to allow the user to make more modifications? Are some of the spaces/holes deliberately made to fit certain screws for modifications?
This is the drone frame that I am specifically referring to, but this applies to a lot of other frames (and even electronics in general) as well.
AI: A major consideration is weight reduction; if material can be removed without significantly reducing the strength of the item, this will result in an increased payload or flight duration. This is particularly evident in the motor arms of the product you linked, as the triangular shapes are renowned for their relative strength.
Another advantage of this is that it allows additional equipment to be attached easily, with cable ties or similar; in some cases I expect the cut-outs will be designed with this in mind - for example; a selection of parallel cut-outs can allow for a wider range of battery sizes to be securely attached to a frame. |
H: Using DJI mavic min in Iceland during July
I was wondering if someone had the chance to fly a DJI mavic mini/2 in Iceland?
Im asking because it is small and im not so familiar with drones, all I read is that it can fly up to 13/16 m/s winds but of course nothing can beat experienced answer.
Thank you
AI: The drone regulations for Iceland can be found here and here (full PDF); they include an FAQ which suggests recreational use is permitted - I have quoted a few of the points below, but I recommend you check their site for full details:
Do I need formal training and/or licence to fly a drone?
There is no requirement for special training/licence to fly drones under 25 kg, provided they are used for recreational purposes.
...
Is it mandatory to register the drone?
Recreational flights: No need to register.
...
Is insurance for drones mandatory?
...
Recreational use: Insurance is not mandatory.
...
Where are drones prohibited?
Over a crowd of people
Close to airports with scheduled traffic
Close to private houses and public buildings
Above 120 meters
Beyond visual line of sight of the pilot (BVLOS)
...
Can I fly over National Parks?
For flying over National Parks you need to apply for a permit...
As to the safe wind speed for operating a DJI aircraft, you should check the specifications of your drone on the DJI website; look for the "Max Wind Speed Resistance" number on the "Specs" page (e.g. DJI Mini 2.) |
H: What is the advantage of using more motors?
I am at the beginning of a drone project, which will have a 25 kg MTOW, and I am analyzing whether I use four, six or eigth engines.
Is there a good advantage of using more motors or does it really depend on the project design?
Does anyone know a simulator that can help?
AI: Adding more motors will generally give you more power, meaning you can lift more. You will however need a bigger battery to get the same hover time, since more motors will use more power as well. So there is a trade off between being able to lift a lot, and flying for a longer time.
Another advantage of more motors is that you can maintain maneuverability even if you lose a motor. So if you are carrying some expensive equipment or doing something critical, six or even eight motors might be preferable.
I'm not aware of any simulators that can test this accurately, but you can use this online calculator to test different configurations. I assume that is close to what you want? Most of the options require a payment, but you can probably get an idea of what you want without that. |
H: One of my motor don't turn if throttle is raised too rapidly
I've recently have an issue with my quad.
One of my motor don't spin if I raise throttle too rapidly.
If i raised it slowly i the motor turn correctly.
Only one of my motor have this issue and i've already change it but the problem is still here. Maybe it come from the esc but maybe someone have a solution ?
My Esc are configurate to use DSHOT300 (i've tried 150 but same probleme)
AI: Thanks to the different answer
After the proposition from ifconfig in the comment, I've swapped 2 motors and have the issue have remained on the same arm. So the issue was comming from he esc.
I've tried to change conf on the esc as Robin have recommended without result.
So i've buyed new esc and replace it and everythinh work fine.
But by curiosity I cut the plastic film of the esc and I discovered that a component was broken (a light grey cube, I think it's a capacitor but I let the experts correct me). It fell out when I opened the plastic.
I don't know what it's for but maybe someone will have the same problem and can use my experience |
H: what is the voltage output from the ESCs of drone motors?
the flight controller i am using is a custom one, which is a teensy 4.0 microcontroller. it cannot have more than 3.3V into its GPIO pins. I know I can use the 5V from the BEC of the ESC to power my teensy, but will the digital control pins from the ESC apply more than 3.3V at the pins of the teensy?
this the drone motor (and frame) that i am using.
AI: As you point out, the Teensy 4.0 online documentation mentions that neither the digital GPIO nor analog I/O are 5v-tolerant pins. It appears that the ESCs in the kit you linked to only have a PWM signal input and no bi-directional communication bus (e.g. Bidirectional DSHOT), so the only voltage present on the PWM input line would be one that you create. (thus there is no real danger of the ESC putting 5v on the signal line, though the 5v BEC pin will still have it)
However, you may run into another issue when trying to issue commands to the ESC from the Teensy. Being a 3v3 logic-level device, its logic high may not be sufficient voltage to "trigger" the ESC, depending on how the ESC's software interprets incoming signals. You can address this with a "level converter" (essentially just a set of small BJT transistors to switch 5v lines with 3v3 ones and vice versa) like this one from Amazon. |
H: How are IMU used to detect orientation with respect to ground?
I have seen some interesting videos of quadcopters floating autonomously and recovering from free-fall flight using only IMU.
Can someone tell me how gravitational acceleration is detected and distinguished vs acceleration from wind or control actuators?
Conceptually there's no clear way to distinguish between the acceleration due to gravity and acceleration due to actual acceleration but I suspect there is some hacky way to do it for aircraft.
AI: The core idea regarding the accelerometer is the following:
if you take any flight of an object and calculate the average experienced acceleration over the whole flight (ignoring gravity as a factor for a second), that average should add up to zero. Here I'm assuming that the velocity at the beginning and end of the flight is the same, which is usually the case: the drone powers on while still and powers down also while still. If you add acceleration due to gravity back in, then that will be the average: g + 0 = g.
Additionally, most utilitarian/autonomous drones spend most of their time hovering or moving in a straight line at a constant velocity, during which there is also almost zero measured acceleration besides the gravitational component.
Of course, while the aircraft is actively maneuvering, the acceleration readings will be way off and can't be trusted for getting an accurate attitude(=orientation) estimate.
However, the IMU has two different kinds of sensors; in addition to the accelerometers it also has three solid-state gyroscopes built in, which measure angular velocity (rotation rates) directly. Those rates can be integrated to get the current orientation very accurately in the short term. In fact, gyroscope-based attitude readings are actually the main source of attitude information for any aircraft, and accelerometer-based readings are secondary.
The biggest problem with gyroscopes is error accumulation: since their readings are being integrated (i.e. added together) and every sensor has a small persistent error, the errors add up over time and and in the long run the accumulated error can become quite large.
To get the best of both worlds, you could, through a process called sensor fusion, combine data from both of those systems into one "fused" measurement that is accurate both in the short and long term.
There are multiple different sensor fusion algorithms in use; Arguably the most popular ones are the Kalman filter and the Madgwick filter. Both involve high-level math, but the basic idea could be approximated this way: at each step, you take the current gyro-based measurement and correct it a tiny bit towards the current accelerometer-based measurement. This way the gyro integration error is being constantly compensated and doesn't grow, and the accelerometer-based error can't throw the measurement off too much at any given moment since the corrections based on it are small.
All that said, if you do maneuver a lot, the algorithm can be fooled. If you turn on the respective feature on a betaflight quad, fly several minutes of hard freestyle and then turn on angle mode, the quad is likely to tilt hard to one side and think that's the upright orientation. So that (in addition to energy conservation and simplicity) is another reason why autonomous drones tend to fly in a straight line most of the time, and also why some acro pilots turn off the accelerometer entirely on their drones.
Other sensor fusion algorithms can be used for different kinds of readings: for instance one could fuse GPS readings with accelerometer readings and get a position estimate that is more accurate than GPS (because the IMU has better resolution) but doesn't drift (because GPS readings compensate for it) and can continue to function if the GPS fails in flight for some reason. |
H: Pixahawk compatibility with high voltage digital servos
I am a novice RC member whom is doing a university project that uses a flight controller to actuate high voltage digital servos. It is part of a ducted fan rocket. The servos : https://hitecrcd.com/products/servos/digital/coreless/hs-7235mh/product ideally would work with 6v-7.4v. They are part of a thrust vectoring mechanism whose actuation is governed by a Pixahawk flight controller: https://www.rcworld.co.za/product_details.php?proid=1507 . We intend on using a 6s Lipo battery to power the components. I would like to confirm if this flight controller is capable of handling up to 7.4V to these servos (the stall current is 1.6A). I had seen in a documentation that the pixahawk can handle up to 10V, as seen in the image below . I would like to confirm if this is true, as I am not 100% if servo power refers to this. Additionally, I had read something about the servo rails being limited to 5V. I have not used a digital servo or flight controller before. Any advice is appreciated.
AI: Yes, it is possible from the Pixhawk V2 using the Dual Power input.
See the sections "Power Architecture" and "List of features changed on Pixhawk 2 from Pixhawk 1" from the documentation.
I recommend you test the Pixhawk without components and measure the outputs to ensure the FMU power supply is separate from the Servo rail. |
H: Compact Gimbaling propeller
does anyone know how would you create a compact gimbaling propeller, like how a rocket engine gimbals. Similar to that! Thank you for the help!!!
AI: There are some commercial units (single axis or two axis) or there are a variety of designs you can 3D print.
These designs are mainly based around a large ball joint behind the motor, but there are some designs that use two sets of pivots like a traditional gimbal.
If you're DIYing it, a simple way to do it is two motors that pivot in one direction only, using differential thrust for the other direction.
I've also seen models where the motor mounts are attached directly to the elevons, sharing a hinge and servo. |
H: Can you connect a mamba405-mk2 flight controller with a FrSky R-XSR receiver?
Spec for mamba405: https://ardupilot.org/plane/docs/common-mamba405-mk2.html
Spec for R-XSR: https://www.frsky-rc.com/wp-content/uploads/Downloads/Manual/R-XSR/R-XSR%20ACCST%20-Manual.pdf
Looking at the two specifications the R-XSR requires both a SBUS In and Out however the mamba405 only has one SBUS connection. Moreover, the mamba405 spec says it only supports "unidirectional protocols", does this mean I can just connect the SBUS out?
Thanks for any feedback
AI: RSSI = Received Signal Strength Indicator
The S(Smart) Port connects to the RSSI Pin on the flight controller for the telemetry information.
Only the SBUS out is necessary for most Kwads. |
H: FPV camera not working
I've bought the following two items:
https://droneshop.nl/emax-transporter-fpv-bril
https://droneshop.nl/star-power-vm275t-5g8-25mw-48-channel-vtx-with-osd
Both are 5.8ghz 48-channel. They both support NTSC and PAL.
I'm only getting video "noise", in the form of random horizontal lines mixed with snow. I am certian that I'm at the right channel, as when I mismatch channels I see "traditional" snow. See https://i.imgur.com/RdtNHhA.mp4.
I've hooked up the camera to a (chinese) CC3D, to the main port. The main port is configured for telemetry at a baud rate of 57600.
The manufacturer indicated the camera could be broken. I already received a different camera from the store (same model), but seem to be getting the same problem.
How do I continue to diagnose and fix this problem?
Mismatched channel giving traditional snow:
Matching channels. I'd expect this to work, but it gives this:
Connections couldn't be simpler.
AI: I think what's happened here is that the camera and flight controller are from different generations of gear, and although the plug fits, they aren't designed to work together like this.
All-In-One Cameras with 4 wires tend to have two for power, and one wire to send the raw video out to the flight controller, which overlays an On Screen Display (OSD) and sends it back on the last wire, where it goes to the video transmitter.
The CC3D is an old flight controller, from the days before OSDs were common. The ports are digital UARTs, not for an analog TV signal. They're for digital receivers, GPS modules, etc.
I think that to get this to work you should find the manuals for the CC3D and the camera. Check exactly what that port on the CC3D does. If it's what I think it is, check which pins on the port are power, and make sure they match up with the power wires on the camera. Then disconnect the video wires from the CC3D and connect them to each other.
Also, the lens cap is on in the last photo. I don't think that's the problem as I can see it in another photo, but that's actually the most common reason for 'my screen just shows black' questions ;-) |
H: Installing Spektrum receiver onto AtomRC Dodo
I'm a noob who just received an Atomrc dodo, and I have a Spektrum SPM4650 receiver I need to install. The wiring diagram (below) is clear enough, but I'm new to soldering and would prefer to avoid soldering onto the tiny pads on the FC if possible.
The tan box labeled RC corresponds to a 3 pin connector. The quad came with the corresponding male component of the connector, with 3 wires with exposed ends.
My question is this: is this connector an alternative to soldering the receiver directly to the FC? If I solder the wires on the connector to the receiver and plug it into the FC, will the FC be able to tell which kind of receiver it is and use it? Or am I misunderstanding what this connector is?
The above image is the only documentation available from atomrc's website, which seems a bit... lacking. It's like I'm expected to already know a bunch of conventions related to these components, but I don't where I'd learn that information if not from the manufacturer.
AI: Sorry, but soldering to tiny pads is unavoidable when building a quad, especially with a tiny 2.5" quad. (unless you buy one with the receiver already installed, and never need to replace anything)
I think that that receiver comes with solder pads and the option of adding a socket for that wire, or just soldering wires directly to the board. The web page for the receiver shows it both with and without a socket. However it has holes for fitting the socket, but the FC only has pads.
On a quad this small, plugs take up a significant amount of space, and there often just isn't room - or for enough slack wire to allow them to be disconnected. Also plugs can develop intermittent connections as they age (and as a result of crashes), so solder joints are generally preferred.
Get yourself a good soldering iron (they're a lot cheaper than they used to be) and a practice board (or just an old circuit board) and practice until you're confident. Or find someone who can do it for you. |
H: Brushless Upgrade for RC Car
I was told that if my question was about motors and ESCs than it would be okay to put it on here even though it has to do with an RC car...
I am wanting to do a brushless upgrade for my WLTOYS 12428 RC car... I looked on ebay, but all of the ones that say that they are compatible with my RC car, are a lot more expensive (between 10 to 30 dollars more) than others (most say that they are compatible with the WLTOYS 144001 RC car.
What makes a brushless motor upgrade compatible?
Here are the package list of two products: one that says it's compatible with my car, and another one that doesn't
Compatible with 12428
Compatible with 144001
120A Brushless ESC
120A Brushless ESC
Brushless Motor1
Brushless Motor2
2.4G 3CH Remote Control
2.4G 3CH Remote Control
Metal Steering Servo
Metal Steering Servo
Metal Rocker Arm
Metal Rocker Arm
Heat Sink
-
Would this be compatible? They pretty much have the same thing (with the same specs), isn't that all that matters?
I could just buy a heat sink separate, they are too expensive... though, if the brushless motor has an aluminum heat sink casing, would I need a separate heat sink?
1 From the pictures on the listing, we can see that it is a 4300KV brushless motor
2 The pictures from the listing don't show an more details on the motor, should see if it is a 4300KV motor?
AI: The important points are:
The physical size. One of these is labelled 'BL3650', which probably means that the can is 36mm diameter and 50mm long. Of course, it doesn't have to be exactly the same size, just fit in the car.
Output shaft diameter - so it fits your gear.
Mounting bolt pattern, size and spacing.
The Kv, or speed constant, which sets how fast the motor will turn at a given voltage. This doesn't have to be exactly the same but should be within about 10%.
These are usually all specified when you buy a motor on its own, but when you buy a plug-in replacement they often don't bother to list them. If it's not listed, there's a fair chance that motors for the same size car will match the first 3 points, as cars used to use industry standard brushed motors.
If there's no Kv listed, you could compare the cars they're designed for. If the cars look like they drive at a similar speed, have similar sized wheels and gear ratios, the motors are probably a similar speed. |
H: Is it possible to attach I2C and USB based sensors to the flight controller and receive some outputs via telemetry?
Readily available flight controllers have pins/pads available usually for UART ports where you can attach various modules like ESC, RX, VTX, OSD etc.
I wanted to know if it is similarly possible to attach I2C as well as USB based sensors to the flight controller? For I2C sensor (like a I2C temperature sensor) I'd like to have it relay back its reading in the OSD telemetry.
For the USB sensor, since OTG USB is not present on any FC that I know of, can I attach the D+ and D- lines to UART/USB bus on the FC and also receive that data via OSD telemetry or from the analog VTX back to me?
The most important part of this inquiry is that, is it readily configurable in the configuration softwares to attach external modules like this and have them working? Are there any examples of this in action on the popular projects like Ardupilot, iNAV, Betaflight, Emuflight etc?
AI: i2c
Yes. This is how flight controllers talk to the already available sensors on board*. The addition of barometer and magnetometer sensors (and any other sensors that these flight control softwares support natively) is, and should be, done via the i2c bus. Save some sensors like GPS which is generally UART (though i2c/SPI GPS chips may exist).
USB
Short answer - no.
USB-otg is a complex software stack that probably wouldn't even fit on the stm32 flash. It requires pretty intensive processing and handling of a wide range of devices. While this support may be possible, it certainly isn't natively supported. Especially on chips where the USB port actually passes to a USB->serial converter to talk to the main MCU on the FC.
*Modern flight controllers talk to the primary MPU (motion processing unit) over SPI as opposed to i2c because it is faster, lower latency, and allows for higher frequencies. Peripheral sensors are still UART or i2c in this case. |
H: When flying, how can I identify the orientation of a distant fixed wing model aircraft?
I'm learning to fly fixed wing model aircraft via an RC simulator. I can control the aircraft well in Chase mode, where my viewpoint follows the aircraft. However, in Fixed mode, where my view is from the pilot on the ground, it's hard for me. I believe the cause is that it's difficult to tell the direction of the aircraft's velocity.
Mathematically, from the ground, I see only two dimensions, azimuth and elevation. There's not enough information to determine the x, y, and z positions. The only way to do this is to judge the distance r from the aircraft by its size. This is hard to do when the aircraft is distant.
How do pilots flying RC planes tell the direction angle? How do you tell it if it is parallel to runway, getting futher from runway, or getting closer? How do you tell roll, pitch, yaw to keep it stable?
AI: The main way is through 'situational awareness'; if you are flying the aircraft, you should have an idea of what is it doing based on what it has done, and what you have commanded it to do. For example, if it is flying from right to left as you look and you command a right turn, then when the plane appears to have no horizontal movement it is going away from you.
However, it is common and easy to briefly lose this awareness. In this case, I would take what I can see and attempt a manoeuver that I know the outcome of:
If the aircraft looks like it is either coming towards me or going away, I will roll slightly - if I command 'right wing down' and the right wing goes down, I am looking at the aircraft from behind. If the left wing (as I see it) dips, it is 'backwards' - i.e. coming towards me. (Of course, you could also use yaw and see which way the silhouette goes.)
If the aircraft is possibly flying inverted, I will use either elevator to lift/dip the nose or rudder to yaw the plane:
If it is flying left to right and I try to push the nose down but it goes up, the plane is inverted. (I would not expect to see a yaw movement very well in this case.)
If it is flying away from me and I try to yaw left and it goes left, it is the right way up. (Up/down movement would probably be visible in this orientation, but I prefer maintaining altitude.)
I've also used simulators, and find that it is also slightly easier to work this out in real life as things like depth perception help a lot.
Finally, it is worth mentioning that if you are flying in a country that requires you to keep the aircraft in "Visual Line of Sight" (or VLOS), if your aircraft is so far away that you cannot easily identify it's orientation visually, it is probable that you are no longer flying in VLOS conditions (as you cannot see what your aircraft is doing, even though you can see the aircraft.) |
H: Are drone batteries more prone to swelling than other li-ion batteries?
I have lithium ion batteries in my cell phones, laptops, portable chargers, and all sorts of devices. Sometimes I'll keep them off for a week or two and when I turn the device on, the charge may be lower but it's not significantly lower.
But drone batteries otoh... after 5 days DJI batteries, save for the Mini 2's batteries, discharge to 60%.
Are drone batteries more prone to puffing / swelling / whatever than other types of batteries?
Drone batteries are probably discharged at a much higher rate than cell phone or laptop batteries. Maybe they're higher amperage too? Maybe those two things together make drone batteries more prone to swelling?
Maybe when other batteries say that they're at 100% they're really at 60% and they never actually let you charge the battery up fully to 100%?
In any of these cases, what about the Mini 2's batteries? The ones I have don't self discharge to 60% after 5 days, and I haven't seen any swelling either, so what gives?
AI: I think that puffing is generally a result of overheating, or over-discharging.
As you say, drones push their batteries hard. They can use a battery in 5 minutes and can draw very high current peaks during that flight. By comparison, a phone or laptop battery typically lasts several hours.
Similarly drones generally allow you to over-discharge. Discharging below about 15% starts to cause long-term damage, and discharging to 0% just once can kill a battery. However it's safer (and cheaper) to sacrifice some battery life rather than to have the whole drone fall out of the air - and when you have an FPV drone set up to fly for 5 minutes or less, an extra 30 seconds makes a big difference.
I believe the BMS (Battery Management System) in phones, laptops and even electric cars are designed to cut off at around 15%, for maximum battery life.
Also, DJI batteries have a built-in BMS that discharges them to 60% after 5 days.
Don't charge your batteries unless you're going to fly. The batteries are programmed to automatically discharge down to the storage level (about 60%) after left unused
Storing batteries fully charged also shortens their life, so this feature should extend their life. Normal RC batteries do not have this feature, and rely on the user to do it. Phones and laptops generally don't do this because they expect to be used daily and not left fully charged for days. |
H: What is the standard setting for the Ele stick in FPV mode
I've been flying quads for quite a while both line of sight and FPV. I started with Liftoff simulator FPV where I had the settings set the way that when I move the Ele stick backwards, the drone moves forward (inverted settings?). The inverted setting in FPV made more sense to me back then because the camera followed the stick.
Then I moved to line of sight flying and found the inverted stick counterintuitive. The observed drone moves in a different direction than the stick. So I learned to fly LOS in normal Ele setting.
Now I prefer normal Ele setting in LOS and inverted Ele setting in FPV and I'm wondering if I should force myself to learn to fly in non-inverted Ele setting in FPV. I think the normal setting might give a better control over a drone if you imagine it's position in space, but that's just my assumption. On the other hand in this setting camera moves the other way than the Ele stick which feels kind of weird.
What settings do professional drone pilots use for a better control?
AI: Short answer
The standard is that pulling the elevator stick backwards lifts the front of the quad, the same as an airplane (pull up). This is the same regardless of LOS/FPV/otherwise.
Longer answer
REalistically, whatever works for you is fine. Back in the days before electronic reversal, people would notoriously learn with the elevator backwards (this is fixed wing, so "push up"). This was either done because they thought this was correct (as you are thinking here), or because they couldn't reverse the mechanism and so had no choice. Some people still fly that way, however I would not feel comfortable handing someone the transmitter to one of my expensive machines if they were wary that their instinct might kick in and they push the machine nose first (this has happened to me, surprisingly with a military pilot who learned on real aircraft).
If it's more intuitive to you, and allows you to move forward in the hobby, then fine. However, it limits your ability to pick up someone else's machine and fly it. Same goes for "mode 2" transmitters which most of north america uses - their are arguments for and against any mode, but unless you have a compelling reason to switch, the disadvantages are negligible to comply with a majority-backed standard |
H: Which frequency band protocol to be used for fixed wing gliders?
I am designing an autonomous glider. I am also adding a provision to manual override the glider. I want to understand which frequency would be perfect for RF module or transmitters and receivers, so that I can send control signals to the glider. Please do consider the USA regulations by FCC and help me.
AI: There are many different currently leading control frequencies, and you can also deviate from these and design your own radios if you'd like to pull specific advantages/disadvantages
2.4 GHz
Advantages
Higher frequency generally means lower latency link
More bandwidth allows for faster data rates
More bands means more pilots can fly without interference (hundreds at least)
Widespread support (spektrum, frsky, eLRS, futaba, hitec, and hundreds more)
Small, compact antennas
Disadvantages
Shortest range, the higher the frequency, the lower penetration through free space and also through objects
Generally plug-and-play (iPhone style) other than expressLRS and openSky (and maybe others). Most of these systems are black-boxes
Noisy channels (everyone uses 2.4GHz)
900MHz
Advantages
Longer range, better penetration
Still good support (CRSF, eLRS, R9, etc)
Perfectly acceptable packet rates, even for freestyle/racing quads
Outside of the common 2.4GHz wi-fi link, so less noise floor (please don't fly over houses)
Disadvantages
Longer, heavier antennas
More sensitive to antenna placement (though this is the case with any long range)
Smaller bandwidth means a limit may be reached
433MHz
Very similar to 900MHz, however more range, bigger antennas, and less common support.
72MHz
A relic, antennas are so obstructively large that it is annoying to construct 1/4-wave antennas on groundstation or in the air. Easy to transmit on and low noise floor, and most transmitters require very little power due to the impressive penetration. May require customization for long range, however. Low data rates. |
H: For a typical dual controller + companion computer UAS, is it normal to have UART only from controller, and USB only from companion computer?
When I'm reading the extension capability of 3DR Solo, I realise that its breakout board interface has some peculiar limitation.
Here is the context of the entire 3DR Solo design: it is an open-source, fully extendable quad that doesn't stand out as a consumer flying camera, yet became popular within researchers and developers (it is still popular). And typically serve as the reference board for an open source UAV system.
It has 2 computers: a controller and a companion computer:
controller:
pixhawk 2.x
running ardupilot 4.x, on ChibiOS
low latency control
companion computer
freescale iMX.6
running a custom yocto poky linux distro
high latency, complex, high level processing
According to this doc, here are all the UART and CAN interfaces:
9. SER5 TX (DEBUG) UART5 TX output from Pixhawk™ 2.
10. SER2RT UART2 RTS output from Pixhawk™ 2 for flow control. Connect to device's CTS pin.
11. SER2Tx UART3 RX signal to Pixhawk™ 2. Connect to device's TX pin. Voltage is 3.3V.
24. SER5 RX (DEBUG) UART5 RX input to Pixhawk™ 2.
25. SER2CT UART2 CTS input to Pixhawk™ 2 for flow control. Connect to device's RTS pin.
26. SER2Rx UART3 TX signal from Pixhawk™ 2. Connect to device RX pin. Voltage is 3.3V.
12. CANH1 CAN bus high to the Pixhawk™ 2.
13. CANL1 CAN bus low to the Pixhawk™ 2.
Here are all the USB interfaces:
1. USB D- Negative differential data signal to iMX6 OTG USB port.
2. USB D+ Positive differential data signal to iMX6 OTG USB port.
In short, all UART & CAN are from controller, and all USB are from companion computer.
I wonder what the purpose of this design is (and other similar designs in UAV)? Assuming that if I need to add another low-latency robotic part, do I have to route it through pixhawk or USB?
AI: Short Answer
There are several different protocols that have different uses and different limitations. The ones mentioned that have a name like SER and description of UART are built into most controllers. Things like CAN and USB are a bit more modern and USB replaces SER on modern computers. These last two may or may not be built into a controller.
Long Answer
The SER is an RS-232-like protocol that uses one wire for Transmit and one wire for Receive. It is one of the simplest protocols and has been around since about 1960. It is intended to connect two and only two devices together (semi-permanently) and has a lower speed than most other protocols. There is some configuration required on both devices to enable communication.
The USB protocol came out in the late 1990's to overcome some of the limitations of SER. It still has two wires but they are both used at the same time for transmit and receive. The protocol self configures so the user does not need to make any setting changes. The speed is much higher and up to 127 devices can talk on the same bus. This has replaced the SER on all modern computers and is the standard for most tables, phones, etc.
CAN also came out in the 1990's but was used almost exclusively for automotive purposes. It is somewhat similar to USB, but does not require a host computer to oversee communications. It can work at a higher voltage and is a little less susceptible to noise than USB. It also can have many devices sharing one bus.
UAS Usage
A controller is limited to (usually) 1-3 serial ports. This means it can only talk to a few things. The SER is generally used for things like Telemetry / Data collection.
The CAN is used for add-on sensors like GPS and safety devices that keep the Drone flying / moving.
The USB is used to connect the controller to the companion computer for things like software updates, data transfer, etc.
Each one has a particular use that leverages the strengths of the protocol. |
H: FC powers RX only if USB is connected?
this is my first UAV build, so please don't bare with me :)
I use basically a S500 frame with 4S setup and a flightstack using a Holybro Kakute F7 1.5 FC, connected with a Holybro Tekko32 4in1 ESC board. Connected with the default 8pin bus cable and like the FrSky R-XSR RX wired up like the FC manual recommend (RX,TX and GND, 5V). Everything is working fine, if I connect LiPo and FC with USB.
Unfortunatly, the RX isn't powered, if I use only the LiPo. It's LEDs are just off and also the FC has only the green LED on, but not the blue one flashing.
I double checked the manuals and the web but have no idea whats the problem. There is a B+ cable within the bus to share full Bat power from ESC to FC but why is the 5V down? And why is there a different beep sequence?
AI: A quick look at the kakute shows that it definitely has a regulator on board, so it should be able to make 5V from the provided vbat. I would suspect a bad cable from ESC to FC, which can be confirmed with a continuity check on each line.
Regulator issue
Compare the 5v pad where the rx is connected (and try some other 5v pads, to rule out one particular bad trace). Use a multimeter and compare against the battery negative terminal.
Vbat issue
Does betaflight show a battery voltage when connected to the PC and also plugged into battery? This confirms your supply to the fc is good.
Ground issue
Find a ground on the flight controller, and temporarily jumper it to the main negative battery terminal. Note, it is not recommended that you fly like this (ground loops), but it rules out a potential grounding issue. |
H: Loose drone propeller mount, flying off at high throttle
I have seen this post here that asked the same question, however, the user forgot to use one component (collar). I have similar components and even when I mount everything correctly, the mount is loose. I bought a kit for a quadcopter and 3 of the 4 propeller mounts are loose.
Like you know how the propeller should turn when the physical motor turns. well the props also turn by themselves on the surface of the motor since it is not tight there. I have tightened the nut of the propeller (cone shape thing) with some pliers pretty tightly, and that stops the propeller nut from flying off, but it doesn't stop the entire mount from flying off.
How do I tighten this? I tried to tape the motor stick thingy that the props sit on top of but it peels off as i slide the propeller mount on it. I have also added some images if needed and in the first one, you can see the how it sort of slides vertically on that motor stick thingy. Thanks!
AI: The piece with the threads on it should probably fit over the motor shaft (motor stick thingy) all the way down to the motor body. When tightened down the slots in the threaded piece should close up and grip the shaft tightly.
You can put a screwdriver (or similar) through the hole in the conical top piece (the nut) and use that to tighten it down.
Fit Check
In the bottom picture tighten this down as much as possible without it being on the motor shaft. The diameter of the hole should be less than the diameter of the shaft. If it isn't, then these mounts will never work on your motors. |
H: Why do people say flying FPV is harder than LOS?
I've heard people say that FPV is harder than LOS and I'm curious as to why- it's easier for me to stay oriented when I can see the aircraft's perspective, especially flying acro quads.
AI: Because flying FPV you have a limited view of the UAS position relative to yours and because your field of view is limited by the characteristics of the sensor (mainly the camera) that you have on board.
Moreover flying FPV you may get a misleading perception of the distance and a limited situational awareness on the obstacles surrounding your Aerial System that may lead in extreme cases to the loss of the UAS because you loose the radio coverage, you hit something or you run out of battery.
This is why in most of the Countries the rules dictates that you must fly LOS unless you have a proper clearece from the National Authority.
Hope this answers. All the best. |
H: Influence of propeller blade shape and tip shape on its behaviour?
Newbie here, still learning how to fly and having to buy new propellers. Of the various properties to be chosen among, there are two that have me a bit baffled:
symmetrical vs. undercambered
and
swept-tipped vs. straight-tipped
Looking at pictures of historical aeroplane propellers (my model is a Piper Cub, the original of which was first flown in 1938), there only seems to be the symmetrical+straight variant:
So when did undercambered, swept-tipped propellers come up, what were designers trying to achieve with them, and what are their advantages and disadvantages?
Here's an example of a prop sold by a popular Hongkong-based brand:
As another example, the model I'm learning with came with an APC 7x5E prop, which has these two properties even more pronounced:
(source: netdna-ssl.com)
(Full disclosure: this question evolved from that one over on Aviation.SE where I was told I'm off-topic and lacking focus and to go here instead. Please let me know if either is still a problem!)
AI: The main thing the designers are trying to achieve is reducing tip losses by making the chord of the tip of the prop small relative to the rest of the blade. Tip losses are due to air moving around the tip from the high pressure area behind the blade to the low pressure area on the front.
Swept tips theoretically delay transonic effects as the tip approaches the speed of sound. Air speeds up as it flows around the curves of an aerofoil, and so shockwaves start to form before the whole blade reaches Mach 1. This is about 30,000rpm for a 7" prop, so well about the RPM limits for glass-reinforced nylon prop. However carbon 'race' props can spin that fast. I guess that shaped developed for high-performance props have just been copied in cheaper props for marketing reasons.
The bulge (wide chord at about half span) in the last prop is because it's designed for an electric motor and doesn't have to deal with the forces that occur when an IC engine fires. The thick part of the blade is extra area away from the tip (reducing losses because more of the prop is now 'not the tip'). A prop for an IC engine would need a massive thick root to support the extra weight, but electric props don't need to be as strong, and saving weight is important.
Finally, the last prop would be a difficult shape to make from wood, where you want the grain of the wood to align with the tension.
A note about under-camber. This doesn't really do anything directly, but it's a result of picking a thin aerofoil with lots of camber. If you play with an aerofoil plotter you can see that at 2% camber and 12% thickness, the bottom is approximately flat. If you increase the camber or decrease the thickness, an under camber appears. Here again the difference is mostly because electric props don't need to be as strong as ICE props. The camber is there because the prop is optimised to work in one direction (as planes don't fly backwards) and thinner props are lighter and have less drag. |
H: Aileron and elevator axes are swapped on the remote
According to all resources I've found, this is not a remote mode issue - no modes have horizontal elevator and vertical aileron axes. My current setting is Mode 2.
Is it safe to swap the cables running from the aileron and elevator sockets on the helicopter?
(Probably just a result of me having plugged them into the wrong sockets after replacing the motherboard, but I don't want to brick another one)
AI: In essentially all cases I can think of, you should have no problem with switching the servo/signal connections around on your RC RX. In fact, they're usually designed with that in mind.
What likely happened is that you misidentified the physical channel order of the servo/signal channels and have them plugged in the wrong order. This is quite common, as two of the most popular default channel orders are AETR and EATR. In fact, the only difference between them is the order of the aileron (A) and elevator (E) channels! |
H: DJI Mavic Mini Specs
There are some specifications I was wondering about the DJI Mavic Mini.
First of all, is there a minimum or maximum temperature the drone can operate at?
What is the capacity and C rating on the battery?
Does the Mini support ActiveTrack?
What is the maximum range of the drone, either straight up or in one direction, taking into account battery life?
What is the maximum payload of the drone (how much weight can it carry)?
AI: First of all, is there a minimum or maximum temperature the drone can operate at?
0° to 40°C (32° to 104°F)
What is the capacity and C rating on the battery?
Capacity is 2400 mAh; C rating not given, as the battery is matched to the drone - but they list the charging power as 24W and the discharge rate is often higher than the charge rate.
Does the Mini support ActiveTrack?
Apparently not; from the user manual:
Mavic Mini supports QuickShots Intelligent Flight Mode. QuickShots shooting modes include Dronie, Rocket, Circle, and Helix.
No mention of ActiveTrack, you would expect they would advertise the feature if present.
What is the maximum range of the drone, either straight up or in one direction, taking into account battery life?
You'll need to do some maths depending on what exact flight profile you want, but DJI give a few helpful numbers:
Flight endurance: 30 minutes at 14 kph
Controller range: 2.4 GHz: 2000 m (MIC/CE), 5.8 GHz: 500 m (CE)
Max Ascent Speed: 4 m/s (S Mode)
Max Descent Speed :3 m/s (S Mode)
Max Speed: 13 m/s (S Mode)
What is the maximum payload of the drone (how much weight can it carry)?
Take-off weight is 249 g; I doubt DJI warrant the aircraft for any payload.
Sources:
Mavic Mini - Specifications - DJI (Accessed 2021-04-04)
Mavic Mini User Manual v1.2 (Accessed 2021-04-04) |
H: DJI Mavic Mini - excess arm vibration
I recently noticed while flying my Mavic Mini that its front right arm vibrates loudly while descending. Especially when I start to descend, the arm visibly vibrates and makes a low-pitched sound as opposed to the normal high-pitched buzzing. Is this an issue?
I checked the screws holding the props on - they're tight enough, and the props are only slightly chipped (much of the silver stripes on the end is still there), and that goes for all of the props, not just this one.
Could someone tell me what's going on / how to fix this? I'm concerned the drone will fall out of the sky if something breaks loose.
AI: So I found the problem. One of the props on that arm only was warped, causing irregular airflow and vibrating the arm. I changed out that prop and the issue disappeared. |
H: Is it safe/good to store a Mavic Mini 2 LiPo battery at ~18% for an unplanned next flight?
They say it is best to store at ~60% but I personally don't want to spend CC for charging it first and then discharging my battery.
Note: someone said, I can just charge the battery and unplug the battery individually one by one when it reaches 60% since the charging hub will charge the battery sequentially one by one, but it is safe tho?
AI: 18% should be safe for a few weeks.
Lipos self-discharge at 1-2% per month, and 15% is about as low as you want to go, so you don't have much safety margin. If you were to leave it for 6 months or a year, it could discharge to the point where permanent damage occurs.
60% is recommended because it's low enough to avoid "storing while fully charged" and allows for a year or two of self-discharge, and can be topped up quite quickly. It's not an exact number. A 30% charge could be safely stored for 6 months, and possibly a year, depending exactly how much it self-discharges.
I've no idea if the battery management chip in the DJI batteries draws any power, but it might. In which case the self-discharge rate might be quite a bit higher.
Yes, it's safe to remove a battery from the charger before it's finished. |
H: Quiet, dockable drone for patrolling and remote communication
Never had a drone before but now I want one which needs to meet this very specific set of requirements:
Dockable. I want to setup a drone dock on the roof of a building, where the drone will reside and recharge. It should be able to automatically return to the dock on command or low battery. The dock shall be weatherproof;
With dual audio (to talk to anyone in the vicinity) and camera for live video streaming;
Reasonably quiet. Both to avoid drawing attention and to enable efficient audio communication;
With powerful LED flood/searchlights;
Weatherproof, sturdy and reliable.
Typical use-case scenarios:
Remote patrolling/inspection of a property. Range from the dock is less than 100m
Remote communication with visitors
Aerial photography
All price ranges considered.
Is there anything like that on the market as a turnkey solution? If I was to build one myself, which specific components/brands should I be looking at?
AI: There are several options available for Self-Docking drones. Skydio, Icaros and Nightingale have some for security markets. There are some targeted at agriculture as well. However these may turn out to be very expensive (some number with 4 zeros behind it).
Most available drones have an RTL (return to launch) feature triggered by the user or a low battery condition.
Two way audio is probably not a feature on any expect Security type drones. You could add this, but the extra weight (speaker, mic, amp) would need to be accounted for in the payload capability.
The noise is also a consideration since most drones from 10 meters away are upwards of 70-80dB which is similar to traffic noise next to a highway. Some smaller units or hexacopters may be quieter.
It might be worth it to get a consumer grade unit and explore the possibility of customizing to fit your needs.
A note about batteries:
Higher power batteries give longer run time up to a point. Then weight becomes an issue.
The general recommendation is to have them at some reasonable temperature before use to prolong life. 'Reasonable' is anybodies guess.
They are generally discharged much faster than they charge. A fully depleted battery may take upwards of one hour to fully charge. |
H: How close can a drone be to a wind turbine in Germany?
I operate a rather professional fixed wing drone which requires pre-flight planning via an app and doesn't support direct control besides instant flight abortion and "return to landing zone" command.
Due to the fixed wings the covering of an area of interest includes U-turns at the end of each pass. These U-turns make things a bit complicated when it comes to flight planning, since they "collide" with a wind turbine. I was wondering, how much distance I have to keep. My location is Germany.
Flying above the obstacle is no option, since the drone is only registered in the open category, hence the maximum flight level is 120 m above ground.
AI: very intersting question
I am in your situation as well flying LOS of course: in the EU assuming you are operating an A3 Open Category UAS, according to EASA, we should keep the drone (fixed wing or rotary doesen't matter) 150 mt away from build up areas.
The D-flight site will provide additional limitation for the area you are operating. Please be sure to register your flight before flying. |
H: Lipo battery with an ESC and a BLDC motor
I'm not sure if I can connect four brushless dc motors to my Lipo battery. I don't know how to calculate it. The brushless motor I have is the Turnigy L2210C-1200 brushless motor (150w) and the Lipo battery I have is the Turnigy 5000mAh 2S1P 20C Hardcase Pack (ROAR APPROVED). I should also probably say that the ESC I have is Turnigy Plush-32 30A (2~4S) Brushless Speed Controller w/BEC (Rev1.1.0). Also how long can I use the brushless motors? I would like to make a flying car.
AI: The motor part number (Turnigy L2210C-1200) tells you two things: The 2210 is the physical size and the 1200 is the Kv rating. The Kv rating tells you the RPM (revolutions per minute) per volt applied.
The battery part number (Turnigy 5000mAh 2S1P 20C) tells you things about the battery. It is 2 cells in series and 1 parallel (2S1P). The 1P is a standard configuration. The 2 cells have a (nominal) voltage of 3.6v each, so a total of 7.2 volts. The 5000mAh says you can pull 5 amps for one hour. The 20C says you can drain it 20 times faster (at 100A) but only for a short time. This is useful for takeoffs and high speed maneuvers.
The motor specification has a minimum voltage of 7.2 volts and a maximum current of 15.8A. The ESC part number (Turnigy Plush-32 30A) is well above these requirements so everything should work together nicely.
Run Time
How long you can run the unit is not really a question that we can answer. It all depends on the drone that these things are going on. The size, weight and type of drone that you are making will determine the run time. Other factors like speed and the environment will affect the amount of time as well.
Depending on the above factors you can probably expect between 10 and 30 minutes of use. One of the stated motors running full speed would last about 20 minutes or less. Run times are exceeding difficult to calculate and actual usage will usually be less than the calculation.
With more information about the drone someone may be able to give you a better answer. |
H: Can't disable SAFE or change SAFE mode
I was gifted one of these (with souped-up motor and ESC) and was delighted when I discovered it could be bound to my Spektrum DXS transmitter. However, the receiver is stuck in beginner mode. I've been unable disable SAFE or change it to expert mode.
The plane holds the control surfaces at an angle when I hold the plane sideways, limits the throw and prevents me from applying any down elevator, indicating to me that it's in beginner mode (although this is my first plane with SAFE so I'm not certain). The SAFE only activates after I first increase the throttle - prior to that it behaves fully manually. This is extremely annoying as I consider myself a competent pilot and wish to be in full control of the aircraft.
I tried moving all of the switches on my transmitter but none changed the mode. I thought possibly the previous owner had a transmitter with more channels and bound the SAFE control to a channel which no longer exists, so I tried these instructions to bind it to channel 5 on my transmitter but it didn't work.
It's possible that this is linked to an issue where the motor only works if I re-calibrate the ESC every time I turn it on.
My receiver has the part number eflr310013 and is described in this manual for an E-Flite Apprentice.
AI: I found a solution the next day.
I noticed that while holding down the panic button (A button on the transmitter, channel 6), I appeared to have full control of the controls. This led me to realise that the channel was operating in the wrong direction and so the plane was always in panic mode - this also explains the problem with the ESC. I reversed the direction of the button on my transmitter and it now functions correctly. |
H: What battery/motor setup will replace this 1980s vintage glow engine?
I have a sport model (RCM Terrier, balsa/plywood construction covered with Monokote or Econokote stick/shrink film) that I built in 1984, and it has somehow come through multiple moves apparently intact. I have a radio for it that's still FCC and AMA legal, as far as I can find out (though I'm certain the transmitter and receiver batteries will need replacing), but back in the day, power generally came via burning methanol, often with a small amount of nitromethane added, along with oil to lubricate the small two-stroke engines; electric models were heavy and performed poorly with their nickel-cadmium batteries and "can" motors.
In this case, my old engine is an O.S. Max .25 (appr. 4 cc), plain bearing non-Schneurle, good for around 14,000 rpm with a 9dx6p (inches) Master Airscrew reinforced nylon propeller. The model has a span of 60 inches (~193 cm) and weighs about four pounds (1.8 kg) with radio, engine, and a full fuel tank sufficient for 20-25 minutes of flight; when I flew it regularly it was mildly aerobatic, capable of loops (inside and outside), spins, snaps and aileron rolls, etc. -- but a little underpowered for the style of aerobatics that was common thirty-five years ago.
I'd love to restore this model to flying condition, but glow fuel is harder to come by than it was then, and even when current was recognized as messy (unburned oil would coat the model after a flight), noisy (even with a muffler on the engine), and I'd have to replace the tank and all the plumbing, as well as the engine, after this much idle time (hasn't been flown since about 1986).
I'm aware that most modern R/C is electric, using lithium batteries and brushless motors. What size/type motor, battery, and controller setup would I need to get similar performance from this model to what it had thirty-five years ago?
AI: I would suggest two answers:
Answer 1: The pragmatic approach:
A good rule of thumb for estimating power needs for RC applications is to use 100 watts per pound for sport flying, 150 for aggressive aerobatics, 200 watts per pound for 3D flying.
In our case with a 1,8 kg weight, that sound like about 4 pounds, I would suggest for sport flying at least a 400 watts system (Output power). To be more precise a 0.24 glow is providing about the equivalent of a 650 Watts brushless system.
I thrust Hacker brusless motor and I would suggest:
Engine: Hacker A30 10XL, 650 watts, Propeller 14×7 Battery 3S 2500mA, ESC 80 A (this systew will drain 50 Amps at full power, 7200 RPM at 80% efficiency). Please note that to get about 400 Watts output power you need a 650 Watts system due to the losses of the various components (see the graph for details).
If you want 'more power' , I would suggest Engine: Hacker A40 10S, 900 watts, Propeller 14×7 Battery 4S 3000mA, ESC 80 A (this systew will drain 50 Amps at full power).
Answer 2: the Aerospace engeneering approach:
Analyse the aerodinamical configuration of your aircraft using XFLR5 (https://www.xflr5.tech/xflr5.htm), then make your math to calculate the required power for level fligh, climb and turn at desired G. Choose your Engine-propeller-ESC system using one of the simulators available online, I suggest this one: https://rcplanes.online/calc_motor.htm or this one: https://ecalc.ch/motorcalc.php?usahacker .
Please verify your system for SAFETY, appropriate configuration, Center of Gravity (it will change consitently so posiition the battery accordingly), electrical connections, battery charge, Max current for each throttle setting (measure it !) etc. Please test your modified configuration on the ground to get familiar with it. Take care of your LiPo batteries using proper SAFETY, storing, charging procedures and install an onboard battery monitor checker to monitor the health of the battery and land with a proper battery reservoir. More info here: https://hackermotorusa.com/resources/rc-brushless-motor-application-guide/. Enjoy and fly safe. |
H: Can an extra 300s glide upside down and backward?
I had a crasy idea; here it is. Let's say I'm flying an 80" Extra 300s with a .6 nitro motor in it. I then go into a dive straight down with full throttle util my plane is 30 ft. off the ground. I then pull up, and go straight up until I reach a complete stall so the plane is not moving at all. Then, instead of letting the nose of the plane fall to point down and regain speed, I cut the motor, then I move the right controll stick, on the remote, to the top so as the elevator flippers point toward the bottom of the craft. Then (here's the crasy part) could I, by careful maneuvers with the control surfaces, cause the plane to start a reverse momentum and start gliding down backward for at least a few seconds?
AI: The plane would be extremely unstable in reverse - like trying to throw a dart with the fins first.
It's not really the same thing, but some indoor models with reverse thrust do manage to fly backwards for short distances at low speed.
Another example is RC helicopters flying backwards at high speed, despite a vertical tail - this is only possible because the tail gyro can react very quickly, and the tail rotor works in either direction. It wasn't possible before heading-hold gyros were available.
However for the sort of model you're talking about, you'd not only need superhuman reflexes, but the loads on the control surfaces could overpower the servos. |
H: Question about Mars Ingenuity helicopter's wide-chord blades
I marveled at the two flight videos today on YouTube by Perseverance helicopter flights during Ingenuity mission.
Got me thinking.
Why aren’t the helicopter blades wider? I would like to think that in order to compensate for Mars’ thinner air.
Doesn’t increasing the area of wing surfaces equal bigger lift?
Video 1, Video 2
AI: Longer blades are more efficient, but it's only practical to a point. Blades have to withstand high RPM, be balanced, and be light. Also the linear velocity of the tip of the blade increases a ton if u have a long prop. At the same RPM, every rotation the tip travels 2pi*r meters, so more r means more distance in the same time.
This may not seem like an issue, however propellers really don't work very efficiently close to or above the speed of sound. Having the tips go supersonic can lead to compression and other issues. This is why there haven't been any successful supersonic propeller planes. The speed of sound is even lower on mars.
Here is some reading on that if you are interested, that explains it very well:
https://mars.nasa.gov/mars2020/participate/sounds/
https://www.highskyflying.com/can-propeller-planes-go-supersonic/ |
H: How to calculate how much lift to be generated for my helicopter to takeoff?
I am currently making a model helicopter. I wanted to know how fast must my blades spin so that it can takeoff. I guess, for that I would need to calculate lift produced right? Can someone help me find what rpm must the blades spin at, so that it can takeoff? I amn't sure about the parameters required to calculate this, so I'll just give some of the important parameters (atleast, the ones I feel might be needed) :
No. of Rotors = 1
Surface Area covered by blades = $400 cm^2$
Mass of helicopter = 1.5 kg
If any more parameters are needed, please let me know. Thank you in advance.
AI: Using the equations from this question
$$L = C_l \cdot A \cdot 0.5 \cdot r \cdot V^2$$
Where:
$L$ is lift force
$C_l$ is the lift coefficient
$A$ is the blade area (0.04 square meters)
$r$ is the air density (approximately 1 kg/m^3)
$V$ is the speed of the blades at about 70% of the radius
Taking the speed at $0.7r$ is only an approximation, as we should really integrate the lift over the length of the blade, to account for the tips moving faster than the root.
The required lift is 15N. (to lift 1.5kg) The maximum lift coefficient is 1.3 for a symmetrical aerofoil, according to this question.
That gives $15 = 1.3 \cdot 0.04 \cdot .5 \cdot 1 \cdot V^2$, or $V = 24m/s$
You don't specify the dimensions of your blades, only the area, so I'm going to guess that you have a 100cm diameter and 4cm chord to make the numbers easier. That is similar to a '500' size RC helicopter (which is about right for the 1.5kg weight too).
0.7 of the radius is 0.35m, and a circle of that radius has a circumference of 2.2m
24m/s divided by 2.2m is 11 revolutions per second, or 660rpm.
That seems low (normal head speed for a T-Rex 500 is 2500-3000rpm) but we're calculating the absolute lowest speed to hover, with the blades right on the edge of a stall, while this sort of heli is designed for aggressive aerobatics.
This approach also ignores losses at the blade tips and the non-lifting section near the hub and ignores the effect of downdraft on the body of the heli. |
H: Is there a simulator in which you can test your own helicopter model?
I am currently trying to make a model helicopter. I have calculated lift generated by the blades and upward acceleration of the helicopter. However, I amn't very sure if my helicopter will fly or not. Is there a simulator in which you can test your own helicopter model? (It would be very nice of you if you could suggest a simulator that is compatible with a Mac).
AI: RC Desk Pilot is a simple, free, open source RC simulator that has an aircraft editor where you can create your own models, or modify an existing model.
If you really want to get into this, X-Plane is a popular open source simulator for full-size aircraft. It's super detailed. Just looking though the overview of the plane design tool is quite daunting! |
H: DJI spreading wings 900 propeller size
I need to know the size (length and chord) of the UAV spreading wings s900 propeller. It should be 1552/1552R and the only info I was able to find was this size: 15×5.2inch (see here). I found that the 15 inches is diameter, so the length is half of that. However, 5.2 is a pitch, and I cannot understand how do I get chord size from that.
P.S. I don't own the drone. I only need to know the size, so I can't measure it myself.
AI: Here's a picture of the blades. I rotated it and measured it with an image editing program. The lower blade is 918 pixels long and 215 pixels wide at the root, 76 pixels wide at the tip.
If each blade is 7" long (allowing for the hub), 1" is 918/7 = 131 pixels.
That makes the root chord 215/131 = 1.6"
And the tip chord 76/131 = 0.6"
The taper doesn't appear to be straight, but I'd guess that only reduces the area by a few percent. |
H: Does the Mavic Mini 2 drone support the DJI FPV controller and goggles?
The drone: Official Website | Where I'm buying it from
I am looking for a drone for a class at school, but we need one under 250 grams, and that supports First Person View, so that we can use VR goggles to fly it. This one has the video quality to support it, and you can connect your phone to the controller and stream live video, so does that mean we could connect VR googles to the phone and fly in first person? I don't think it works with the actual DJI first person controllers and headset, but I am not 100% sure.
Any help would be appreciated, thank you in advance!
AI: As @strepto said, the Mavic Mini 2 is not supported by DJIs FPV system, they are only promoting their (very expensive) DJI FPV drone for FPV flying these days.
There are however various hacks to use 3rd party VR goggles with the Mini 2, you can read more about it in this thread on the DJI forum: https://forum.dji.com/thread-228895-1-1.html |
H: How to use real DJI controller in simulator
I've got a DJI Mini 2 and I'm aware their official application (DJI Simulator) does not support it.
Anyway, I wonder if I can use the real Mini 2 controller in any third-part simulator.
I'm more interested in an Android app that simulate the camera view of a drone (like in real flying) rather than a desktop application (in any case I run Linux).
I tried a couple of apps that claims to support USB controller but they didn't recognized mine (for example Freerider).
AI: The transmitter needs to appear as a USB joystick to the computer. This is often displayed as 'HID' (human input device). At least, that's for desktops, I don't know if Android is the same.
I've no idea if the DJI transmitter can do this.
If not, you can buy a USB dummy RC transmitter. These used to be about \$20, but I've just checked and they seem to be around \$60 now. It could be cheaper to buy a basic real transmitter like a FlySky FS-i6
All the flight simulators I've seen have a choice of 'camera' positions. They can show you the view from the drone or the view from the ground (and often a 'chase plane' view). However they are all for race drones with fixed cameras (and LOS flying for fixed wings and helis). I don't think there are any that attempt to model the various GPS modes available on a DJI drone, or the camera gimbal. |
H: Using BMP280 as an altimeter for CanSat
Situation:
Me and my student team want to use a BMP280 air pressure sensor connected to an Arduino Nano to measure the altitude of a CanSat.
A CanSat is a coca cola can-shaped and can-sized aerial vehicle, which
can be released from a sounding-rocket or a plane-/quadcopter-type
drone, packed with scientific instruments to do predefined
measurements/operations.
In our case, this CanSat is dropped from 300m hight and has to perform an operation as close to 70m altitude as possible.
Questions:
How precisely (practically) can the BMP280 measure the altitude (given physical noise and internal tolerances of the chipset)
Where is the ideal spot to drill the measurement inlet(s) into our airframe, when the airframe is shaped like a can and is falling vertically without any drift to the 'sides', and where do we have to position the sensor in relation to the position(s) of this/these inlets?
We would really appreciate the oppinions and practical knowledge of experienced people and enthusiasts :)
AI: Sensitivity
Barometers are quite sensitive, however "air pressure" is very much not consistent. Generally, you can expect about one meter accuracy, maybe a bit less, from a barometer, once it has been given time to settle. This is important to consider if your CanSat is falling quickly, especially if you have small pressure regulation holes
The primary workaround on traditional quadcopters/fixed wing aircraft is to rely heavily on multiple integrations of the accelerometer, and use the barometer as a heavily averaged baseline. This requires more sensors, but you effectively use the barometer to rule out drift in the accelerometer's readings. This is exceptionally complex if you can't predict the orientation of the can (a multirotor will be able to do this math with only one axis), but a vector sum of all axes should get you a close enough approximation
Airflow
As for airflow - you don't want direct airflow to the barometer, don't put it out in the wind. Common practice is to use a permeable foam covering the barometer - similar to how a microphone is wrapped in a very light foam to block pressure waves that will cause interference.
For the can - a similar setup might be ideal. Don't drill a single hole in a single side. You can end up with either high or low pressure depending on the orientation. I would recommend a set of holes on each axis, blocked with foam. This effectively gives you a vector sum, including slight vacuum and slight overpressure.
GPS
GPS provides altitude information. I'm not sure on the accuracy of it, but you can measure pretty consistently at 10hz, which is probably fast enough if you can trust the readings and don't have to average them
Of course, orientation and being in a metal can are issues with GPS |
H: What is the most invisible obstacles to avoid and notice when flying a drone?
Now I am understanding one thing: there are two kinds of obstacles, in addition to the visible obstacles, such as trees, there are also invisible obstacles that will affect the flight of the drone.
What invisible obstacles do you think need the most attention and avoid? And best practice advice on avoiding it and improving the safety of drone operations.
This is a meaningful question because the visible ones will be noticed and avoided easiler, but the invisible ones need to be discussed and made more clearly understood.
AI: Overhead wires are probably the biggest issue. You really have to watch out for the poles and use them to ensure you're high enough to avoid the wires because you often won't see the wires until it's too late. You should be aware of any wires before you fly, and make sure you're aware of where they are at all times during the flight ("maintain situational awareness" in pilot-speak).
The one people talk about a lot is 'scraggle', or thin branches and twigs with no leaves. It's really only a problem if you're flying very close to trees and bushes. If you're trying to fly under a low branch, or through a gap in some trees, it's worth inspecting it first.
The other class of 'invisible' obstacle are those that are behind you and out of sight. These are the ones that usually catch beginners. Don't fly backwards or sideways unless you're absolutely sure it's empty space. It's safest to keep moving forwards, as the wind could be carrying you backwards. |
H: Will different brushless motor sizes work with the same flight controller and esc
I recently built a 5-inch quad using:
Diatone Mamba F405 Mk2 F40 Mk2 3-6S 30x30 Stack/Combo (F4 FC / 40A 4in1 ESC)
Motors: iFlight Xing-E PRO 2207 1800Kv Motor with DShot 600 and 4S battery cell. I didn't notice any problems with the build when I did a test flight.
I am now trying to build a larger drone (850mm or 950mm X-Class). I was wondering if the same flight controller/esc combo would work for motors:
Cobra CM-2820/16 Multirotor Motor, Kv=740 (link)
and
iFlight XING X4214 400KV Motor (link)
The cobra motor specifies that the pwm frequency is 8khz, but I didn't see any details on pwm frequency for the other brushless motors. So now I'm not sure whether pwm frequency is a parameter I need to consider when choosing an FC/ESC for a motor. Or is the 8khz a standard frequency that all FC/ESCs support? Usually I just make sure the ESC supports the maximum amp from the motor and propeller combination I plan to use, and that I have the correct LiPo battery cell count.
AI: That should be fine, brushless ESCs can normally run a very wide variety of brushless motors.
If they stutter or otherwise don't spin up reliably, you may need to adjust the timing settings in the ESC firmware. This controls when the motor winding are energised as they pass the magnets. Lower Kv motors can need a lower setting.
Here's an article on BL Heli settings (the ESC firmware) |
H: How to calculate lift for multiple rotors?
So I know that $L = \frac{1}{2}ρv^2AC_L$, but if I want to find the lift for a three-blade helicopter, do I just multiply by 3 or what? Let's say I have a 700mm rotor at 61mm wide, would it be:
$L = 1/2\cdot1.225\cdot73^2\cdot0.0434 = 141.6$ newtons. If I have 3 blades would it simply be: 141.6 x 3 = 424 newtons?
Side question, what sort of motor would I need for a 700mm 3 prop setup, you can ignore this if you want.
AI: That equation is for wings, where the whole wing is moving at the same airspeed. Rotor blades are spinning, so their airspeed changes along their length. To use an equation like that, you need to calculate the speed for any position along the blade and integrate for the whole blade. For multiple blades, you're right, you just multiply by the number of blades.
However, it's generally more useful to consider the momentum flow through a rotor disk. The area of the disk is known, and the blade pitch gives the maximum air speed through the disk. Every second a mass of air is accelerated from stationary to that speed. Force = mass * acceleration, so you can work out the force. You can also work out the kinetic energy input ever second, which is the power.
Some of the equations are here but this was an option in the third year of an aerospace engineering course, so it builds on stuff from the earlier years and I don't remember all of it!
The topic you should look for is actuator disk theory |
H: (Beginner) I have a Radiolink Mini OSD with a not so clear schematic. How do I know if it can handle the input voltage needed to power my VTX?
My VTX and FPV camera can both handle my battery voltage (6V-36V input for the VTX and 4V-25V input for the camera). My mini OSD, unfortunately, only has an input voltage of 4v-6v and an operating voltage of 5V. I am in too remote of a place to replace the mini OSD or VTX, and was not sure if it would be able to handle the 6V input needed to power the VTX.
The schematic shows that the VTX port can only handle 5V input, but the pin on the PCB is marked as V instead of 5V. Similarly, the marking on the 5V pin that connects to the flight controller is correctly marked as 5V, both on the PCB and the schematic. To my understanding, V refers to the input voltage range. Does this mean I can input 6V?
AI: You don't need to power the camera and VTX from the OSD, it's just easier when everything runs at 5v because you can use a single 3-wire cable from one thing to another.
Normally, power comes from the battery to the power distribution board (or 4-in-1 ESC board), which also provide regulated 12v and 5v power. You supply 5v power to the flight controller, which passes it on to your OSD.
It seems like you can use a 3-wire cable to pass that 5v power on to your camera, but you won't be able to do the same for the VTX as needs a minimum of 6v.
Instead you'll have to split the 3-wire cable from the VTX. Take just the signal wire to the OSD, and the power and ground wires to the 12v supply. I wouldn't connect these directly to the battery as they might pick up interference from the ESCs. The 12v power should be cleaner. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.