This is a new problem for me but I am sure the answer lies in the physics of batteries and that one of you can explain it to me.

I have a new X6 build using Pixhawk I have Taranis telemetry working. I have had a few test flights to see how things were going and everything looked to be as it should be. 

Then I was ready for the hover test and flight duration test. I have noted on the short flights that the hover current was about 35A which agreed pretty well with eCalc. When I tried the longer duration hover test with a fully charged battery (6 x 4.2v, 25.2v) I had a low voltage warning at just under 8 minutes.

I have attached the log file of voltage and current. It shows the current at 35A (average) .

The voltage started at 25v + and stepped down to 23.8v (6 x 3.99v). It is then pretty linear decay down to 21v (6 x 3.5v). The battery warning is set here and it told me to land. Which I did (then didn't kill the throttle and it fell over and I have a CRASH badge of honor in my graphs).

The final voltages I checked with no load were at 22.9v (3.82, 3.83, 3.82, 3.86, 3.77 ,3.87).

The battery is a Turnigy 12000mAh Graphene 15C.

Lastly, the total current graph shows just under 5000mAh which makes sense (8 mins, 35A, (8x35)/60=4.67). The telemetry was showing about 60% left on battery charge. 

The AUW is 6.5KGs, Tarot 5008 - 340KV motors. 

1) Does this look right to anyone?

2) Could the Attopilot current measure be out by a a lot and I'm pulling a lot of current? Although eCalc is not far off. Each motor lifting 1.1Kg looks like 7A on the motor curve. 

3) Is this battery a piece of crap? 

Views: 157

Attachments:

Reply to This

Replies to This Discussion

The Ardupilot current and voltage sensors are not the best and the firmware even less so, voltage just has a simple filter on it to smooth down the value shown a bit.

By my calculations (https://play.google.com/store/apps/details?id=appinventor.ai_graham...):

80% of 12Ah is 9.6Ah which, at 35A should give you 16m27s of flight time. 80% consumed should read around 3.5V per cell (which you've set) so either the voltage being sensed is wrong or the battery does not have 9.6Ah to give or it is unable to hold it's voltage at the 2.9C load that it's under. Is it very cold where you are?

The only way to make sure that the low voltage point is being read properly is to independently measure it somehow. I use a Tx with telemetry for an independent voltage reading. But even a DMM strapped temporarily can give some indication.

I am pretty sure my problem is one of calibration of the voltage sensors. I use Attopilot and I will attempt to tune this ti the right voltages. The thrust calculation for the motor and prop combination I have should not exceed 5A at the AUW I have so something doesn't add up. 

I am also not happy with the unbalanced battery that comes backout of my plane so I will see what I can do with that as well.

Thanks for your input and nice app.

PS It's still 85 degrees in Texas.

I have mixed feeling about Turnigy batteries, but most have done pretty good for me. I have noticed telemetry showing underload voltages being considerably less than noload voltages. What is C rating of battery? While not a number considered real trustworthy, a higher C rating should result in less voltage sag under load. 35amps is a decent load at hover.

As noted it's a 15C battery and I am allegedly running at just under 3C. I have re-calibrated the voltage sensor and also ordered a Craft & Theory telemetry voltage sensor as a live second opinion. I really only care about the voltage and flight time. The current I can calculate form those. The Attopilt current measurement does not fill me with confidence. TO be fair, it is as likely to be the A-to_D in the STM32 that is inaccurate rather than the attopilt. 

When the wind drops to reasonable I can follow the current calibration suggested above.

Reply to Discussion

RSS

© 2017   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service