Bug with 3DR power module voltage and current reading

[David Dewey]

I've seen quite a few posts about getting incorrect voltage and current readings on APM 2.5 using the 3DR power module, with no straight answers on what causes the problem or how to fix it. Anyway, this issue never happened to me until recently when I upgraded to APM 2.6, and now I'm stuck with a wildly incorrect voltage and current indication in my Mission Planner and on my OSD display. Furthermore, I assume compassmot won't work correctly with a garbage current reading, and the whole reason I upgraded to APM 2.6 in the first place was to try and improve my magnetic field interference thanks to its external compass.

Currently I'm getting a voltage reading of 0.3 volts below actual value (as measured by multimeter), and a current reading of several times the actual idle value (as recalled from my previous APM 2.5 installation, disarmed current was less than 1 amp, now it's reading about 3 amps on average).

I've checked the settings and they seem to all be correct (although interestingly the "3DR Power Module" option defaults the "Amperes per volt" parameter to 18, whereas the parameter list description claims the default is 17; this inconsistency itself should be fixed in mission planner, although either way it doesn't seem to make much of a difference in the incorrect readings I'm getting):

Any help would be appreciated, as I've spent a lot of time upgrading to this APM 2.6 and don't want to have to revert back to APM 2.5

Thanks,

David

[Robert Lefebvre]

Hmmm.. interesting.  I'd never used the PM before, but just started using it on my Nerf Quad.  I have also found the voltage readings to be weird.  I'm running 4S, and I like to fly until the hovering voltage is 14V, which is 3.5V/cell under load.  A nice place to stop.  Typically after a rest, the battery is at 3.7V/cell, and takes an 80% charge.  I've never had an issue with this using my own voltage dividers.

But with the PM on this quad, I find that if I land at 14V, I still have TONS of power left in the battery?

--
You received this message because you are subscribed to the Google Groups "drones-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to drones-discus...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Randy Mackay]

     If you're using an APM2.6 with the external GPS+compass module you may find you can get away without having to do the compassmot.  I've been able to skip the compassmot if I put the GPS+compass on a 5cm stalk so it's about 3cm above where the APM was.

     I get accurate voltage readings by setting the VOLT_DIVIDER parameter to 10.1 (MP sets it to 10.0).  You should be able to measure the battery's voltage using an external voltmeter and then fill that voltage into the "1. Measured battery voltage" field but it looks like there is a bug in the mission planner screen.  You cannot edit that field unless, strangely, you go into the full parameters list and then modify VOLT_DIVIDER manually once.  After that you can change teh field back on the Battery Monitor page.  I've raised an MP issue here.

     We should probably ask michael to add some similar fields to allow the current multiplier (AMP_PER_VOLT) to be calibrated as well.   ..and I probably need to make a video showing people how to do the calibration.

     By the way, I've moved the battery monitoring to an AP_BattMonitor library so that the code will be shared between plane, rover and copter.  That will change most of the parameter names so they all start with BATT_.  This will mean some more work for MichaelO unfortunately and a short period of time where the mission planner screen won't work with the trunk code.

-Randy

---

From: Robert Lefebvre <robert....@gmail.com>
To: drones-discuss <drones-...@googlegroups.com>
Sent: Thursday, October 3, 2013 12:01 AM
Subject: Re: [drones-discuss] Bug with 3DR power module voltage and current reading

[David Dewey]

Thank you Randy! I was finally able to get very accurate voltage readings by setting VOLT_DIVIDER to 10.2. When MP set it to 10, it was off by 0.3 volts, and setting it to 10.1 (the value specified in the parameter description) resulted in a reading off by 0.15 volts or so.

Not sure if my ampere readings are correct now, since I don't have an easy way to measure the actual current, but it seems closer to what I expected at least.

I did the compassmot, and got interference at only 1% of magnetic field (down from 12% before upgrading to APM 2.6). Nice!

My GPS isn't even on a stalk, just on a 1-cm thick piece of foam.

Anyway, there should definitely be a more reliable way to calibrate the power module, because I'm sure this is something that will greatly trip up new users when their voltage and current readings are way off and they can't figure out why.

Regards,

David