Hello wizards and experts,
  I have a problem with a new ESC I have installed on my quad.  The ESC is a Holybro Tekko32 F4 Metal 65A... This is replacing a  Tekko32 F3 Metal 65A that caught fire one day when plugging in a battery, but I digress.

After the fire that ruined the original Tekko32 ESC, I replaced it with a newer F4 model, soldered the motors and it seemed to be working ok on the bench but one of the motors sounded different than the other three so I decided to replace all 4 motors.  So I now have the FC and ESC installed in the drone and the motors are soldered to the ESC.  When testing the setup on the bench with the battery plugged in, I spin the motors from ~0-50% throttle for ~10 seconds (with no errors shown) then all motors stop spinning and the voltage indicated in betaflight drops from ~15V to ~7V.  

The ESC firmware is 32.8 and I've tried betaflight 4.2.10 and 4.3 (RC3) Firmware for the Kakute F7 FC.  The motors are Xing E Pro 2306 2450KV. 

I've triple checked the solder pads on the ESC and I haven't bridged any connections that I can see, no solder balls (I mask the circuitry with masking tape before soldering).  Maybe I damaged the ESC with too much heat?  This is the second set of motors I've soldered to the board...

Any troubleshooting tips?  

Thanks in advance!

1. Im not sure the voltage shown is "wrong". Afterall, all motors spin.

2. Sorry to hear you had to spend so much $$$ to get back to flying.

3. Before connecting battery, always check each motor pad has no continuity to + and gnd. (check when powered off!!)

4. Regardless if BLHeli_32 or BLHeli_S, all motors should twitch at 1020, and spin smoothly at 1060. Emf voltage is proportional to RPM, so sensorless ESC sucks at very low rpm.

5. The "bad" motor. Check resistance between each phase. Should be iden tical. 90% likely one wire is loose.

If you upload photos, maybe one of us can spot if something is wrong.

Drones is funny.

All the soldering must be perfect. No frayed wire whiskers, or blobs or bridges.

But, when it comes to Betaflight setting, you can put random iterm, anti-grav, filter, etc.. and it still flies.

IIRC, there were some bugs in 32.8. You can try updating to the latest testcode, or downgrading to 32.7 and see if that fixes your issue.
https://github.com/bitdump/BLHeli/releases

(17-Feb-2022, 09:40 PM)romangpro Wrote: 1. Im not sure the voltage shown is "wrong". Afterall, all motors spin.

2. Sorry to hear you had to spend so much $$$ to get back to flying.

3. Before connecting battery, always check each motor pad has no continuity to + and gnd. (check when powered off!!)

4. Regardless if BLHeli_32 or BLHeli_S, all motors should twitch at 1020, and spin smoothly at 1060. Emf voltage is proportional to RPM, so sensorless ESC sucks at very low rpm.

5. The "bad" motor. Check resistance between each phase. Should be iden tical. 90% likely one wire is loose.

Thanks for responding, romangpro.  Can you please elaborate on 3?  
   On the ESC, there are three pads for each motor, do I check there is no voltage between any combination of the three?

(17-Feb-2022, 11:10 PM)cpintheloop Wrote: Thanks for responding, romangpro.  Can you please elaborate on 3?  
   On the ESC, there are three pads for each motor, do I check there is no voltage between any combination of the three?

With your multimeter in continuity mode, verify there is no continuity between any of the motor pads and Vbat or ground to make sure none of the FETs are damaged.

There are 2 basic multimeter tests to do when wiring up and soldering:

1. Powered off. Continuity.
Check adjacent FC pads you soldered to to make sure no bridging.
Check ESC 3 motor pads to Vbat and Gnd.
Worst is continuity Vbat to Gnd. Big smoke.

2. Powered on. Voltage
Black probe on Gnd...
Red on 5V pad.
Red on 3.3V pad.
Red on vtx power pad.
Can also check input and output to voltage regulators.

Powered off. Motor soldered on.. each motor pad beeps continuity to another. Thats normal.

Powered off. Sbus or Rx pad beeps for 0.01sec continuity to gnd. Thats normal.

Thanks for everyone's help.  I finally found the problem to be the smoke stopper between the battery and the ESC.  Spinning all four motors to about 30% throttle was enough to trigger its safety cutoff, I guess (the buzzer would make a sound and that was my clue). I didn't realize you weren't supposed to use one when testing the motors.  After removing the smoke stopper from the equation, everything works great!