#2, #4 Motors Shut Down During Power Up
GlassKNees
Posts: 181
My Elev-8 suffers from intermittent control loss, typically during takeoff, resulting in flipping over. After a recent crash like that, I broke two props. After replacing them and trying again, I noticed that the controller would reset when I throttle up close to take off speed. After several attempts, I gave up. Later, I tried troubleshooting - powered up partially and after a second or two, the #2, #4 motors shut down while #1, #3 spun just fine.
I have flown this craft a number of times - yesterday, I was hovering in my front yard - seemed okay.
Does anyone have any ideas what could be going on?
I have flown this craft a number of times - yesterday, I was hovering in my front yard - seemed okay.
Does anyone have any ideas what could be going on?
Comments
Are all of you BEC's in parallel into the controller board, or do you have only one,providing the 5 volts with the others disconnected?
The quad crashed through a tree which helped to break its fall, but it landed between two plastic barriers on the side of a road under construction. Hidden from view, my wife and I were unable to located it after two days of searching. Finally, a construction worker found and returned the quad to me. The weather was clear but warm during the days - highs in the mid 90's. The only apparent damage was a couple of broken props.
After replacing the props and recharging the batteries, I could not get the controller to work well - went into trouble mode - fiddled with the receiver connections and even tried another receiver - no joy. Thinking that the controller was damaged, either from the crash or exposure to the heat, I ordered a new board (hasn't arrived yet), Then I hit upon trying to reflash the firmware. After doing so, I experienced some strange behavior - like the ESC's all rapidly beeping. I resynched them and that seemed to work. I did fly it successfully several time, but each time I powered up from a cold start the aircraft either would not respond to the transmitter, or the board would go into trouble mode. After several tries - powering the aircraft on and off, it would respond and I could arm and launch the aircraft. But I restricted flights to low level hovers in my front yard.
Then, one a takeoff, the craft flipped over, breaking two props. I replaced them, and then discovered that the board would reset as I throttled to take off speed. I haven't flown since. Then, while trying to figure out what is going on, I powered up the aircraft and had the motors idle, and that's when the two motors cut out...
Now, I'm thinking that the Hoverfly board is indeed damaged. I won't know for sure until I receive and install the new board. The only other thing I can think of is that perhaps the ESC's are damaged? I'll try one thing at a time - replace the controller, then the ESC's if the trouble persists...
Let us know after you receive the new controller.
After the second crash, I tethered it to a table and could reproduce the issue without breaking props. Cutting off the shrinktube I could see evidence of hardware failure. Parallax sent me a replacement and I haven't had a failure since. I think it will be two years old this August and I've flown pretty regularly, several times a week, when the Ohio weather is cooperating.
One simple tool that I've found very useful over time is a servo tester. Like this one: http://www.robotshop.com/en/rc-servo-esc-tester.html I think I got mine from eBay for a few dollars. You can use a propeller board, arduino, teensy or other controller to drive the ESC, but it is so simple to plug in this guy and a Lipo to the ESC for a reality-check test.
-Russ
https://youtu.be/n94iZYldXzg
So, when I powered up a second time, all four motors responded, though #2, #4 tend to shut down at idle. I don't know if all are running at the same rpm higher (flight) throttle settings. I'm leery of flying it at this point. I think I can get it to take off and perhaps hover, but I worry that one or two motors will cut out and I'll lose control.
At this point, I'm thinking of replacing the #4 ESC and see if that makes any difference...
It's probably trying to yaw and you're holding it down, so two motors are stopping to help it yaw faster. Be careful, that kind of thing can go bad real quickly.
-Russ
Yeah, I recognize that I was taking a great risk holding down the aircraft with my hand while powering it up - I had nothing to tether it down and I was anxious to see if the new controller solved my problem. I won't do that again!
I have removed the bottom plate and removed the #4 ESC. I will replace it with a new spare and do a thorough visual inspection of all connections before I button it back up again. I will retest later today with it safely tethered.
Thanks for your input Russ!
Brian
I've had that happen periodically with the HoverFly. When it does, I abort the take-off, cycle the power and it seems to be ok on the next go around.
-Russ
When I applied power, the Hoverfly board started behaving like the one I replaced - would not respond to the transmitter, would to into trouble mode, etc. After several attempts at power cycling, I finally got the motors to arm. As I applied throttle, I had to apply left yaw to get all four motors going then returned to neutral. As I added more power, the aircraft wanted to rotate counterclockwise; I applied yaw control to conteract, then it wanted to roll, so I corrected and decided to abort the takeoff.
I brought the aircraft back inside and now the board goes into trouble mode. I tried another receiver but no change. Then I decided to try to plug an ESC directly into the receiver's throttle channel - I get a slow beep from the ESC no indication that the receiver is responding to the transmitter. I tried binding the receiver/transmitter, but no change. I tried plugging three ESC's one at a time with the same result - a slow beep/twitch from the motor.
Now I'm totally stumped.
Next, I hooked the power distribution harness back and and again test each motor, one at a time to confirm proper spin direction and confirm that each motor started at the same throttle setting. All was good. Next, I reflashed the Hoverfly board and connected it up to the ESCs and receiver channels, and here is where I made a discovery:
Back when I first built the aircraft, I labeled all of the receiver channel wires, but when I connected them, I had one cable (Rudder) oriented 180 degress from the other. That is, in order to correctly plug in Rudder, the plug was oriented opposite from others. Apparently I noticed this during the initial build and didn't bother to change it. But when I repaired the aircraft and reconnected, I plugged Rudder in like the others. Fortunately, power is the middle plug. But I hooked the signal to ground and ground to signal.... No wonder the Hoverfly board was complaining!!
Why it sorta worked sometimes is a mystery to me. But I think that also explains the extreme yaw problems! At any rate, I've tested (without props) and all motors are spinning smoothly. I've attached the propellers and now I'm just waiting for good weather conditions to test fly it.
1st thing - congratulations on getting going! Ah, sorry about your ESC/Motor test fail. I know I've done this testing myself and I wanted to find out why your testing didn't work. I'm pretty sure it is because that at zero throttle your radio/receiver is outputting a pulse width wider than acceptable for the ESC to consider as zero. When the ESC powers up and samples the input pulse width, it decides whether the throttle (input) is a valid zero, a valid scaling level or neither. If neither, it thinks the throttle level is too high to be zero and not high enough to rescale, so emits the "high throttle" error beeps you got.
I'm sure every brand has its own definitions for input levels and each particular device has its own variation. I did a little bit of testing on my ESC's (Parallax) and radio/receiver (Hitech Aurora 9). Using a scope to measure the pulse widths I found that 1.11ms was a valid zero, but that 1.32ms was not. The nominal is 1ms. I found that the lowest input considered a valid re-scale was 1.52ms. Nominal is 2ms. I also found that my radio (as configured @ +/- 100%) put out 1.12ms @ zero throttle and 1.88ms @ full throttle.
So my radio @ zero throttle was putting out a pulse width high enough to trigger a high throttle warning by the ESC at startup. Yours probably does too. If I change the end point adjustment to the throttle channel on my radio to bring the pulse width down (-132%), the testing method works fine.
The reason your ESC doesn't complain when connected to the HoverFly controller is that it is probably conditioning your throttle inputs.
Have fun flying and good luck,
-Russ
So I tried to power up the aircraft again, and ran into the familiar Trouble mode problem. I tried binding the transmitter and receiver, reflashing the board - nothing seemed to help. Then, I noticed at power up there is no led indication from the receiver that it's getting power. I tried another receiver but there was no change...
Now I'm beginning to suspect a problem with the cables from the Hoverfly board to the receiver... I found a source online and ordered some replacements.
I"m using an Spektrum AR 610 receiver. I'm wondering if it requires power from all channels that are plugged into it, or if it only needs power from any one.
On the rx (Spektrum AR610) the ground and +5v ports are all tied together, so as long as power is provided on one of the ports the rest will share it as well as the rx.
How are you powering the hoverfly?
I cut the power/ground wires on the rudder channel, leaving only the signal lead; the other four channels have all leads. It was working for oawhile....The Hoverfly is getting power from all four BEC connections. When I first built it, I was concerned about whether I needed to cut power/gnd from three of the leads as shown in a YouTube video, but the guy was building the earlier model. I was assured by Parallax support that I didn't need to cut any BEC wires on this version.
But on my APM build, there's a video where a guy was insisting on disconnecting power/gnd on all BEC's save one; ditto for the receiver channels, and so I've done it on that aircraft. I can't get it to respond to the transmitter either, but I've put that project on hold until I can get the telemetry link working so I can debug using Mission Planner.
I've taken it out for a test flight - it seems to be flight worthy. Here's a pic:
Here's a video of a flight at a local driving range: https://youtube.com/watch?v=ckw2yu9YWHQ
I watched both of your videos.
Pretty educational.
I have a Quadcopter, but it's not elev8
It's an inexpensive toy that I'm playing with before I attempt an elev8 build.
I have same orientation trouble as you show in your flight video.
With your setup, can you release the control and have the quad just hover in place?
and
Do you need to compensate for any very small changes in wind gusts.
I can only fly my quad on dead calm evenings.
even the smallest breeze is difficult to compensate for.
Thanks for starting this message thread.
gg
To answer your questions, no the Elev-8 does not have a Stabilize mode. You fly it pretty much like a small cheap quad, so I highly recommend you keep it near and low while you practice. I tend to fly on calm days but I've found that it's pretty stable in breezes.
I have built another Elev-8, and put an APM 2.6 controller/autopilot on it, but I've had a lot of glitches to overcome. I thought I had everything working until I tried to take it out and discovered that it rotated counterclockwise in Acro mode and I could not compensate with yaw control. When I put it in Stabilize mode, the #1 motor would not spin. I took the aircraft apart several times, checking, rechecking the wiring - everything checked out okay. Motor tests passes and all four would spin when I calibrated the ESCs but when I power cycled the aircraft, #1 failed to work.... I finally concluded that the APM module is defective, and I bought one from 3D Robotics, which I've received, but have not yet tried loading the firmware and configuring, as I cannibalized the ship for spare parts to fix my other. Replacement parts (boom, motor mounts) will arrive in a few days - then they get here, I will put the ship back together and see what happens.
With the APM, I will have Stabilize, Alt Hold, Loiter and RTL (Return To Launch) modes, which will be great for a novice pilot! I will post my progress in my other thread.
Good luck!
Brian
I saw an APM power module in the pic. If you're using that none of the BEC power wires should be connected to anything. The power module provides power to APM and the APM will provide 5 volts for your receiver. No BECs needed.
If the jumper is set to ESC power, and you're using the APM power module I'm not really sure what would happen, but it has to be bad.
The receiver is powered from the APM, you can connect all of the wires between the APM and Receiver, unless you want to save weight then all you need is one red, one black and all of the whites connected.
It's great you have two frames, and two flight controllers. Copter heaven... Keep on buildin!
*If you remove the APM cover, be sure to check the foam block that covers the altimeter is in the right place.
Brian
L3G4200D gyro plugged into Jason Dories QuadXV2 program. Jason did all of the hard work like tuning the PID's for the
ELEV-8 chassis. And that Gyro has slightly different output than the Hoverfly gyro.
My other Elev-8 is also flight worthy!