I saw Robert at Drone Con #1 talking about the traditional heli and now I want one! I'd like to get into this a cheap as possible and still get pretty good quality gear. I do FPV with everything else I've got so I'll want to do that too. So what kind of trouble can I get into with this new branch of APM?
You need to be a member of diydrones to add comments!
Replies
Rob, sorry to resurrect an old thread, but this is exactly where I am right now and I've got a few lingering questions ...
The servos you recommend (TGY-306) - do you use them for both cyclic and tail rotor, or do you use a different servo for the tail?
Have you had any issues with these servos interfering with the GPS? I had one a while ago with an APM2.0 and the MediaTek GPS. GPS worked great until I added the servo to drive a camera, then it couldn't find more than 1 or 2 satellites. Took the servo off and we're good again, so I just went with a different servo type and never thought about again ... until now. Did I have a bad servo, or is it endemic to the breed? Is this the reason for putting the GPS near the tail or do the blades also interfere with the GPS signals?
Like Markus, I'm going with the HK450 Pro V2, but I'm also planning on converting it immediately to a belt-driven tail. I also plan to maintain this 450 as a development machine and general test platform.
I been reading your "Attractive Alternative" thread. This 450 build is a stepping stone to a full-on, purpose-built, long endurance, aerial photography platform capable of carrying a 3-axis gimbal and a NEX-class camera, etc. Any advice you can offer would be appreciated and it would be my pleasure if I can help with any of the testing for new ArduCopter releases.
Hi Brian, very interesting, I have had exactly the same problem with those servos interfering with the GPS! I really don't know why. I've made the situation "manageable" by keeping the GPS far back on the tail, and putting ferrites on the servo wires. But GPS performance on this thing is not good. This is only acceptable as a test bed.
I use the same servos for the tail as the cyclic. They seem like good servos, other than this problem.
I actually just crashed mine yesterday. I believe the tail belt failed in flight. The APM did a pretty good job of keeping the heli upright and airborne long enough to allow me to choose when/where to crash it. I stuck it high up in a tree. Luckily I was able to get it down by restarting the motor which knocked it out. I'm not sure if the first or second crash caused the most damange.
But, this is another example of the reason I prefer direct drive tails. I find the powertrain for the tail of helis to be the #1 most unreliable system.
Sorry to hear about the crash!
With the direct drive tail, I assume you use a second ESC. To control the tail motor, do you have custom hardware/microprocessor to start it and set the speed when the main motor starts, or are there functions in arducopter to accommodate a second ESC on a traditional heli?
Brian, yes, there is a second ESC, running in governor mode, and I have built in functions to Arducopter to control the ESC. The tail ESC is fed a fixed PWM signal from Ch7. The PWM is set by a parameter, but is otherwise unchangeable in-flight. The tail motor starts when the main motor starts.
I have thought long and hard about allowing a 3-position switch so that you can switch the main motor off while leaving the tail rotor running, for autorotation practice. But just haven't gotten around to it yet.
I re-did the testing of the S306G servos and I was mistaken in my initial conclusions. If the servo is connected to the APM, the GPS performance drops off immediately. It loses satellites and the HDOP goes through the roof. Disconnecting the servo causes GPS to recover almost immediately.
The interference seems unrelated to proximity and more related to the physical connection. Initially, I had been testing the servo using a separate power source and no physical connection, assuming that we were dealing with an RF issue. Guess not.
Brian, I believe it was released in 3.1, so you should have the option now.
Handling, there's nothing that immediately jumps out. I wouldn't do it for a 3D Acro machine, but for basic flying it seems fine.
That is very strange that GPS firmware helped cure interference?
That's interesting, Rob. Is this feature in general release, or are you working on this in a private branch? Are there any handling issues created by putting that much weight out at the end of the tail boom or can it all be managed within the standard PID's?
Also, on the GPS interference issue with the S306G servos ... since I was trying to decide on servos for my new heli build, i decided to test the servo again with the latest firmware on both the MTK GPS (v1.9) and the APM 2.0 (v3.1.2). With the new firmware, the servo didn't seem to any impact on the GPS performance at all. I guess that's good news. I'm planning on using the Pixhawk and external uBlox GPS on the heli anyway, so it's not a critical issue, but it is interesting that the new firmware seems to have made a difference.
Hi Rob and Brian,
These 306g servos are very strong and fast, I highly recommend it for 450 heli, even on the tail, but it consumes more power. I was using turnigy 113MG before, they were slow relative to 306G, but the intefrence was very little. After changing to 306G things started to get interesting.
On PX4 with telemetry, servos will not move at all if PX4 is still locked. After releasing the safe button all the servos start to jump up and down together, all of them, just like synchronized to some signal. There is no twitching without telemetry.
The most interesting part is, that if you connect an LED to the servo plug (I guess it's red and black wire), then move the servo real quick, it actually generates current and light the LED! There is some youtube video about it but I couldn't find it again.
Considering the very fast response of the servo (people often use 333Hz with this servo) I guess that the servo can pick up some inteference very easily, or it's very sensitive to noise from the RC signal.
If it's sensitive to the interference from outside, it should simply twitch when noice comes, but they are totally still when PX4 is still locked. So it seems it's sensitive to the RC output noise, which could be the wires picked up the interference, or even a wild guess, something inside PX4 picked up the noise. However I have the same twitching with APM, I really don't know what's the core issue of this twitching/interference. Perhaps we need a good oscilloscope to check the signals.
I also have heard the GPS issue, you said the GPS issue was better with ferrites, does that mean it's actually the noise in the servo wire that interfering the GPS through USART port?
Xin, I used to have a problem with servo noise like you're suggesting with other servos, back in the day when using Xbee radios. The problem had to do with the way the Xbee was designed, it doesn't have a ground plane and uses the electrical ground as the ground plane. Something like that. Anyway, it hasn't happened to me since I started using the 3DR radio.
I'm not entirely sure how, where or why the GPS interference is created from the servos. I don't have an Oscope. But through trial and error, the beads seemed to help.
I did confirm that the problem is indeed RF, and not through the power system, because I would get the interference even when the GPS was running on a PC that had no electrical connection to the heli. Amazing really.
Cool thread!
I have got quite unexpectedly an extra APM 2.5 and think about to use it for a cheap(!) 450 heli. What do you guys think about the HK-450TT PRO V2 Flybarless?
-
1
-
2
-
3
of 3 Next