Hi guys,


3DR arducopter, APM2.5 :

during ESC calibration i noticed one of the fourth ESC (motor4) an extreme overheating. Three possibilities : ESC problem, motor problem or PDB problem on output motor 4.

I excluded ESC and motor problem by swapping the "hot" ESC by a nother ESC : this other ESC was also getting hot. Then I used PDB output n°3 on motor 4 and everyhting was fine -> this excluldes a short on motor n°4.

So obviously I have a problem with my PDB on output 4, that provokes extreme overheating of the ESC. I measured with an ohmeter the resistor between the + and - of PDB output 4. It is the same as all other outputs (so no short at that level). Plus the fact that even if the ESC overheats, the motor 4 still turns fine !


So what could be a cause on that PDB that makes the output 4 , thus the ESC4 overheat, but still driving the motor to turn fine ??? A short on the PDB ? Did I solder something wrong on the PDB ?


Please help because I have exhausted all I could test. Should I disassemble the whole thing to get a closer look at the PDB ?  What  other tests could I do on the PDB to detect an eventual issue ?

I want to avoid having to replace a PDB, not only is it more dollars to dish out, but further it is a pain to assemble and solder.



Views: 1775

Replies to This Discussion

@ Tim & Luke :

By "Hot" I mean impossible to touch it , you burn yourself terribly. Even the yellow shrink tube around was starting to stink like burned plastic.

This happened just by connecting the Lipo battery, not having the motors running (and JP1 was off and I am using the 3DR power module to power APM).

Everything disappeard after I removed the Red&black wire between the PDB and APM outputs. Apparently others using 3DR quad kist have had similar issues (PDB design faults ?). I did not find the root cause anyway.

If it's getting that hot there could be a short. Doesn't matter if the motor is running or not because the BEC is a separate circuit (just happens to be combined with the ESC for convenience). Got a multimeter handy to troubleshoot?

The limiting factor is that the APM PM I-sense will put out more that 5V to the ADC with over 90A. The inline resistor is capable of the extra current (its rated 4W). You could try and scale the 5V output using a small voltage divider to scale back to 5V e.g.. R1 of 10K and R2 of 47K see http://www.raltron.com/cust/tools/voltage_divider.asp

Just remove JP1 and connect the BEC to the PWM inputs 5v and GND lines. This is if you are not using the APM PM.
Yes i am using 3DR ESCs. I have on the outputs only the 4 signal wires,nothing else connected. No more hot ESCs. However I started anothe thread for an even worse problem now : no more throttle control from these 4 signal wires and I am stuck totally: http://www.diydrones.com/group/arducopterusergroup/forum/topics/pro...


I'm using an external BEC (non-lineal) with my octo. I think it is a must.


You need JP1 removed, otherwise you have ~5.3 on the Outputs rail and approx 4.8V on the input. That will cause a current to flow from the ESCs as it tries to pull-up the 4.8V line to 5.3V.

You only need 5V power from the PDB to the outputs if you are connecting something that needs to be powered, like a servo, You should have the GND line connected though.

See http://code.google.com/p/arducopter/wiki/AC2_Current#APM_2.5:_Using...

we've been building our SteadiDrones now for a while, over a year and we've always simply connected ALL ESC 3pin (red, ground and signal) wire plugs directly to the APM, with jumper on to power RX via the input rail and NEVER had any problems with this. We use this on al our Quads, H6X and large EI8HT copters that have flown thouands of flights with no problems, I'm not sure how others are doing it, but this works very well for us.

Hello Tim, no i cut the red wire and kept only the black wire connected on the APM's output rail.

If your using the power module I think you leave the red and black wire in, just remove the little piece that goes over the jp1.


© 2018   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service