Pixhawk and 3DR Power Module: keep your eyes open

[Marco Robustini]

Hi all!
With my small "kamikaze hexa" i'm doing some tests with the new Pixhawk V2.4 with 3DR Power Module, powered with a LiPo battery 4s 4A 45C (new), I set the battery failsafe to14.1V.
Takeoff, after three minutes "beep beep beep" (yellow rgb led), battery failsafe engaged...... what?
I'm going to read the logs and I see an high LiPo voltage drop that reaches 10V.
I thnk "impossible", i replace the LiPo pack with another, same thing, three different battery and same problem.
I speak with Randy that obviously tells me that the LiPo pack could be bad ... three out of three? Naaaa...
I decide to install a Futaba 7008 rx (with the internal telemetry, 100% much reliable for this), where it can monitor the LiPo voltage in real time.
I takeoff again and when the "Pixhawk builtin voltage failsafe" begins to bite I see on Futaba a voltage of about 15.7V and on my phone (via telemetry) 12V ... damn!
Then I choose to remove the 3DR Power Module, power Pixhawk with a dedicated BEC 5.3V and create a voltage divider with two precision resistors calculated for the 6s LiPo: 25.2V input, R1 = 10K (in series with the +), R2 = 1.5K (between + and - after R1), 3.3V output (to Pixhawk) with full charged LiPo 6s.
Having provided a 6-pin connector parts i create the cable to inject the reference LiPo voltage in Pixhawk, i follow the right value of voltage divider via software (some like 7.67) and voila, the LiPo voltage in the Planner is the same of my Futaba telemetry, even after 10 minutes of flight.
Now, definitely my Power Module (new) is defective, so if there such a thing were to happen to you too careful to set actions like RTL on the "battery failsafe", first check the operating voltage during the flight corresponds to what in theory should be, the electronic components can also fail.
What interests me is to have the LiPo voltage in the Planner, the current does not interest me, and in this way I can even fly with 6s, while the Power Module bears a maximum of 4.
If you want to create you then the voltage divider values ​​are those written above, the correct parameters for Pixhawk is this one:

BATT_VOLT_PIN,2
BATT_CURR_PIN,-1
FS_BATT_VOLTAGE,14.1  <- for 4s
BATT_VOLT_MULT,7.67

Bests, Marco

[Marco Robustini]

PS: Craig, when you think to send me another Power Module (tested, please)? ;-)

Merry Christmas!

Marco

[Marco Robustini]

The "Pixhawk" power pinout (six pin) is this one:

VCC / VCC / V / I / GND / GND

Marco


On Tuesday, December 24, 2013 6:54:05 PM UTC+1, Marco Robustini wrote:

[Craig Elder]

Marco,

It sounds like the multipliers that are programmed for the power module are not correct.

and the settings your are recommending are not all correct either

BATT_VOLT_PIN,2

Battery Voltage sensing pin (BATT_VOLT_PIN)

Setting this to 0 ~ 13 will enable battery current sensing on pins A0 ~ A13. For the 3DR power brick on APM2.5 it should be set to 13. On the PX4 it should be set to 100. On the Pixhawk powered from the PM connector it should be set to 2.

BATT_CURR_PIN,-1

Battery Current sensing pin (BATT_CURR_PIN)

Setting this to 0 ~ 13 will enable battery current sensing on pins A0 ~ A13. For the 3DR power brick on APM2.5 it should be set to 12. On the PX4 it should be set to 101. On the Pixhawk powered from the PM connector it should be set to 3.

FS_BATT_VOLTAGE,14.1  <- for 4s
BATT_VOLT_MULT,7.67

http://copter.ardupilot.com/wiki/arducopter-parameters/#Voltage_Multiplier_BATT_VOLT_MULT

Voltage Multiplier (BATT_VOLT_MULT)

Note: This parameter is for advanced users

Used to convert the voltage of the voltage sensing pin (BATT_VOLT_PIN) to the actual battery’s voltage (pin_voltage * VOLT_MULT). For the 3DR Power brick on APM2 or Pixhawk, this should be set to 10.1. For the Pixhawk with the 3DR 4in1 ESC this should be 12.02. For the PX4 using the PX4IO power supply this should be set to 1.

--
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.

[Marco Robustini]

I know Craig, I explained bad, sorry, I wanted to say is that: my final posted settings are fine for my "home made resistive divider" that I'm using now, not for the Power Module.

Marco

[Robert Lefebvre]

Hi Marco,

I have observed this too on my PM's.  I don't know if yours is defective, or if something is going on.  I wonder if the shunt resistor for the current sensor is dropping the voltage or what.  I know that doesn't make any sense, but something is going on.  I'm flying a little F450 quad with a giant 4S5000 battery that I use for me helicopters, and I can see the voltage drop to 12V at full throttle.  I haven't looked into it much since I don't typically use the PM much.

Rob

--

[Craig Elder]

There is some variation on the power modules and on the ADC in the APM.  You can still fine tune each of the multipliers to work better for a specific PM

[Marco Robustini]

Thanks Craig but ... how?
Excuse my ignorance but I do not know how to do... you mean inside the code??

Marco

[Robert Lefebvre]

It's in the parameters Marco.  The Voltage vdiv ratio.  Normally 10, but 10.1 or 10.2 is typically right.   But I don't think this is the problem.  I've tuned it for correct voltage being shown while not under load.  The problem is the amount the voltage drops while it is under load.  The PM does not seem to be measuring properly.  I dunno. I haven't dove into it, but this has always seemed not right to me.

[Marco Robustini]

No, the problem is not the voltage vdiv ratio, because i play a lot with this without good result, this "phoney voltage drop" is another story.
Why with my home made partitor resistor the voltage in the Planner is perfect and this drop is absent?
However i stay with the "two resistors" system, my Power Module back in the box and i wait the 6s version, when available.

Cheers, Marco

[Hampus Lundin]

I have the exact same problem as Marco Robustini had and I can confirm that his PM wasn't faulty. I returned my PM to the seller and got a new one, the exakt same problem with this phony voltage drop during load still persists. 

[Paul Riseborough]

We will be getting ground offsets between the different parts of our system (battery, PM, flight controller, etc) and are not using balanced signals for our analogue inputs, so there is plenty of scope for the types of offsets being reported.