Discussions

Hi everyone. I am a “newbee” to RC flying and I am building my first quadcopter.

I have two major problems. The first is that 3 of my 4 motors randomly spin up, or randomly beep, even when the APM2 is disarmed. The second is, that after re-installing APM Planner and Arducopter firmware, it won’t boot from battery; only if the USB is plugged in first.

Update: - Problem Solved.

I sent the board to Dany at CanadaDrones. He verified the board was bad and managed the RMA to 3DRobotics and shipping the new board back to me. Thanks, Dany, for all your help.

Once in a while, it is, actually, a hardware problem.   Grin.....

I have tried as many workarounds as I have been able to find on this forum and others, but none work for me. I have disassembled my quadcopter and spread the parts out on the bench to try and isolate the problem.

I have been working on this for 5 days and finally decided to RMA the APM2 board. However, I would greatly appreciate knowing if I did something wrong, or if there is a solution available or planned for Arducopter 2.6.

My system is as follows:

Transmitter:   Turnigy 9X modified with FrSky DHT transmitter and running ER9X

Receiver:       FrSky D8R-II Plus 8 Channel with telemetry (not using telemetry)

PDB:               jDrones Arducopter PDB Rev. D

ESC:               jDrones ESC 20

Motors:           jDrones AC2830-358 850KV

Battery:           Turnigy nano-tech 2.2 3S 2200mAh 35-70C

Connectors:  Battery=XT60, ESC=Deans, Motors=bullet (will solder once working)

Controller:     APM2 standalone GPS (remote board)

Frame:            X525 (not is use. Other parts are disassembled on bench)

APM Planner: Version 1.1.92 Mav 0.9 updated June 7

PC:                 Windows 7 current updates, Norton 360 current

These are my notes from recent testing:

First Tests

• I tried powering up the quadcopter again.
• It powered up OK, but I let it run a half throttle for a minute or so.
• I could hear a different frequency to the motors occasionally.
• Then Motor 3 stopped completely, then started again.
• Then Motor 4 slowed down and sped up again.
• I stopped the motors (Throttle Full Off).
• I disarmed the motors (Throttle Full Off, Yaw to Full Left).
• The APM2 showed it was disarmed. (flashing red LED “A”).
• Motor 3 started to beep slowly (once every 2 seconds). Motor 4 was quiet.
• Then Motor 4 started to spin up for 1/2 second and stopped. multiple times.
• I turned off the transmitter. Motor 4 kept trying to start.
• I unplugged the quadcopter.
• I tried again.
• Power the transmitter, throttle down.
• Plug in the quadcopter.
• Still disarmed, but Motor 4 spins up and Motor 3 beeps.
• Then Motor 4 stops and starts to beep.
• Stops beeping. Clicks a few times, tries to start. Spins up for 2 seconds and stops.
• Motor 3 stops beeping for a few seconds. Motor 4 starts beeping.
•  My Voltage alarm shows 11.2V 3.74, 3.73 3.78

Rebuild PDB Tests June 5

Symptoms:

• Power on the transmitter, throttle down
• Power up quadcopter
• It is disarmed
• Hear 123, 2 second pause, one long beep
• Motor 4 beeps once per 2 seconds, and randomly starts and stops
• Motor 3 occasionally twitches or beeps

Test 1 – Resolder PDB

• Dismantle quadcopter and spread out on the bench
• Resolder all connections on 3DRobotics PDB
• Connect power to PDB
• Result: same symptoms as before resoldering.
• Conclusion: 3DRobotics PDB solder joints are good

Test 2 – Swap ESCs

• Swap ESC #4 (bad) with ESC #1 (good)
  • Motor 4 still beeps and randomly spins up with ESC #1
  • Motor 3 randomly spins up and beeps (with ESC #3)
  • Motor 2 Randomly spins up and beeps (it had no ESC change)
  • Motor 1 OK (but it now has the ESC #4, which was bad before swap)
• Conclusion: Something is sending specific commands to the ESCs.

Test 3 – Disconnect APM2 from PDB and ESCs

• Unplug the 4-pin signal cable and 2-pin power cable from APM2
• Connect battery to PDB
• Result:
  • 123, 2-scond pause, all motors twitch and beep with short beep, 2-second pause, repeat; indicating no transmitter signal detected
  • No spin ups
• Conclusion: ESCs are not generating the errors on their own

Test 4 – Connect APM2 without Receiver

• Disconnect 8-wire cable and 2-wire power from input rail of APM2 to FrSKY receiver
• Plug in 4-pin signal cable and 2-pin power cable from PDB to output rail of APM2
• Power off transmitter
• Connect battery to PDB
• Result:
  • Tx and Rx are off
  • Red “A” and Blue “C” LEDs flash fast (while APM boots) then settle down to alternating about once per second.
  • After 2 minutes Motor 3 started to spin up for 1 second on its own, the for 4 seconds, pause 3 seconds.
  • Then motor 2 spun up.
  • Then motor 3 started beeping once per 2 seconds.
  • Unplug
• Conclusion: Problem is not the transmitter. Problem is not the receiver.

Test 5 – Swap ESC 1 & 4 Back To Normal

• Swap ESC 1 & 4 back to their original positions (matched to same motor #)
• Results:
  • 3 rapid quiet beeps, long beep, 5 second pause,
  • Motor 4 starts to spin up and click, random spin ups
  • Motor 3 beep-short + 2 second pause, repeat, Longe, beep, goes quiet 3 secons, spins up 2 seconds, random after that
  • Motors 1 & 2 seem OK and quiet
  • After 4 minutes, Motor 3 gave a 123 tone, 3 rapid beeps, spun up for 3-4 seconds, random rapid beeps, clicks
• Conclusion: ESC 1 is OK, ESC2 probably OK. APM (or noise on PDB) is sending random commands to Motor 3 and probably 4 as well.

Test 6 – Swap the PDB for a different PDB

• Remove the 3DRobotics PDB from the loop and replace it with a newly soldered jDrones Rev. D PDB that came with the ESC/Motor/Propeller/PDB kit.
• Solder up the jDrones Arducopter PDB Rev D. using Deans connectors between PDB and ESC
• Remove bullet connectors from ESCs and solder Deans Connectors (M)
• Completely disassemble X25 frame and remove motors and wires from arms.
• Assemble PDB with ESCs and Motors on the bench in the Open without APM2
• Connect Lipo
• Results:
  • Got 123, then 3 beeps, then a beep from each motor every 2 seconds indicating no transmitter signal detected.
  • The 3 beeps indicates LiPo 3 cell, buti I programmed all 4 ESCs as Nickel, so it should not even do that test.
• Conclusion: At least one ESC has lost it’s programming

Test 7 – Test New jDrones PDB with APM Planner

• Plug in APM2 Outputs to jDrones PDB 4-wire signal and 2-wire power
• Do NOT plug battery
• All ESCs and Motors and signal cables are plugged into PDB, but not powered
• Start APM Planner in Flight Data
• Connect USB to power APM2
  • APM2 “A” red LED and “C” Blue flashing , green is solid
  • Receiver is flashing (indicating the transmitter is off)
• Click “Connect”
  • Tx flashes slow, then very rapid, then medium
  • APM Monitor says “Disarmed” “GPS: No Fix”
  • VSI gauage is moving between -1 and +1
• Turn on the Transmitter
  • Receiver LED is solid (indicating talking to transmitter)
  • Rx now flashing, but not as fast as Tx
• Click on Configuration in APM Planner
• Click on “Radio Calibration”
• Click on “Calibrate Radio” button
• Moved sticks and switches 123 and AIL and Trainer to set CH 1-5 and 7
• Set Flight Modes
  • Stabalize
  • Alt-Hold
  • Loiter
  • RTL
  • LAND
  • RTL
• Set Level
• Set Declination
• Set “X” Config
• Conclusion: Everything seemed to work correctly in the APM Planner
                        It seemed to transmit to the APM2

Test 8 – Manually Program the ESCs & Calibrate

• Because Test 6 indicated at least one ESC was not programmed correctly, I will program all 4 ESC manually
  • Brake: OFF
  • Battery Type: NICKEL (Changed from default of LiPo)
  • Cutoff Mode: SOFT
  • Cutoff Threshold: LOW (Changed from the default of Medium)
  • Start Mode: NORMAL
  • Timing: MEDIUM (Changed from the default of LOW)
• At the same time, right after programming, I did a Manual ESC Calibration on EACH motor
• Conclusion: It seemed to work fine. Each ESC could spin up its motor after the calibration.

Test 9 – Test all 4 motors with APM2 and Transmitter and jDrones PDB

• Turn on transmitter
• Plug all 4 ESCs and motors into PDB
• Connect LiPo
• Results:
  • Got 123, pause, 2 beeps and quiet while APM2 is booting
  • Motor 3 pause 10 seconds then spin up and low beeps, 10 beeps or so,then quiet for 10 seconds then spin up again
• Unplug battery
• The Wiki for ESC Calibration now says on the last line “(You may have to do a final automatic calibration)

Test 10 – Automatic Calibration as per Wiki

• Put throttle high
• Connect battery
• Results:
  • 123 beep, pause, beep, pause, beep, pause, etc.
• Disconnect power
• Immediately reconnect power
• Immediately move throttle low
• Results:
  • Motor 4 spins up without warning
• Disconnect power

Test 11 – Repeat Automatic Calibration

• Put throttle high
• Connect battery
• Results:
  • 123 beep, pause, beep, pause, beep, pause, etc.
• Disconnect power
• Immediately reconnect power
• Wait for 2 beeps (it happens within 2 seconds)
• Move throttle low
• Results:
  • Motor 4 goes crazy, spinning, stopping, stuttering
  • Motor 3 goes crazy too.
• Disconnect power
• Conclusion: The APM is sending bad commands to the ESCs

Test 12 – Do a CLI Setup Motor Test

• LiPo is disconnect
• Connect USB
• Start APM Planner
• Enter Terminal
• Results: still spews garbage along with some intelligent words
• Hit Enter 3 times
• Get Arducopter V2.5.5] prompt
• Type “Setup”
• Type “Motors”
• Results:
• Motor 4 is the first to start and stop correctly,
• Then Motor 2
• Then Motor 3
• Then Motor 1
• Then Motor 4, and Motor 3 spins up at same time
• Then Motor 2
• Then Motor 3 but Motor 2 spins up at same time

=======================

The next attempt was to start from scratch, but this resulted in not being able to boot without the USB port connected. The motors still spin up on their own, but only after about 5 minutes of sitting, powered with the battery and the APM still not booted.

I reset the APM2 to factory, erased the EEPROM, uninstalled APM Planner, deleted the entire folder structure that was on C:, reinstalled APM Planner 1.1.92 from
http://code.google.com/p/ardupilot-mega/downloads/list MissionPlanner64-1.1.92.msi Mission Planner Installer (64-bit).

It had two executables for APM Planner; one for Mav 0.9 and one for Mav 1.0

From what I could find, Mav 1.0 is faster and smaller, so I chose that one.

I started APM Planner 1.1.92 Mav 1.0 and loaded the firmware for Arducopter V2.5.5 Quad

But when I tried to connect to the APM2 board, it said it detected a Mav 0.9 heartbeat and this Planner was for Mav 1.0.

I don’t know how to upgrade an AP (I assume they mean the APM board firmware) or what to upgrade it with.

So I closed APM Planner Mav 1.0 and started APM Planner Mav 0.9 (which I have used since).

Note: I did NOT reload the firmware from the Mav 0.9 firmware, but instead, tried to connect to the APM2 via USB and it worked, so I carried on.

I ran through the entire configuration process.

When I got to the point of doing the all-at-once ESC calibration, it only worked on 3 of the 4 ESCs. ESC 3 did not work.

I disconnected ESC 3 and checked that it was still programmed using a Turnigy ESC Programming card. It was.

So I manually configured the limits on ESC 3 using the receiver channel 3 and the transmitter, as per the Wiki.

I put it back into the PDB and it still did not work.

So I ran the all-at-once ESC calibration again.

It worked. All 4 motors spin up together.

I unplugged the battery and shut off the transmitter.

This is where the different symptoms started.

The APM would not boot when I reconnected the battery.

The blue “C” LED flashed once, then all “A”, “B”, and “C” LEDs were dark.

The green LED was solid. The Tx and Rx flashed when I turned on the transmitter, and continued to flash even if I turned off the transmitter (that last part is probably normal).

I have read that others have had similar bootloader problems for months, but that it was supposed to have been fixed.

A workaround on Youtube says to power the APM2 from the USB port first, and when booted, connect the battery and disconnect the USB.

This worked for me yesterday. and there were no strange spin-ups of the motors.

The only “difference” was the boot took about 15 seconds (with LEDs dark) before they flashed and indicated “disarmed”.

It also “seemed” to take longer to arm (about 8 seconds of holding the stick to the bottom right) before it would arm (solid red “A” LED).

I could arm and disarm and the motors would all spin up normally when armed, and would not work when disarmed. Great but...

This workaround won’t work for me in the field, since I don’t have a laptop.

Another workaround proposed was to power the Output Rail with the BEC on a spare ESC and another battery.

This flat-out does not work for me. The board powers up, the green light is solid, the Tx and Rx flash, but “A”, and “C” stay dark. Resets don’t work.

So the USB power up first on the bench “seemed” to work. But while I was documenting this, and the APM2 was disabled, but powered, Motor 3 spun up on its own.

Since then other motors have also randomly spun up while disabled.

I sent the APM2 board in for replacement, since I am convinced it is a hardware problem.

Was there anything else I could have done?

Thanks…..Wayne

Quadcopter APM2 Test Board.jpg