Oscillations in the battery voltage value

Julio_Diez · April 20, 2023, 11:55am

Hello all,

I am building a Hexacopter with a 6S battery of 13000 mAh. The flight controller is Holybro Durandal with a power module Holybro PM07.

I have seen in all the logs that I have a lot of oscillations in the voltage value of the battery. In the following plot, it can be seen the battery voltage and the compensated one. As you can see the battery sag is compensated but in both, there are high oscillations in the value. This leads to the activation of the battery failsafe just after the takeoff.

Does anyone know why this is happening and how can I solve it?

Thank you.

amilcarlucas · April 20, 2023, 12:56pm

Use shorter and thicker wires in all power connections
Add electrolic capacitors as near as possible to the ESCs
Get a better/newer battery

Allister · April 20, 2023, 2:54pm

Also make sure that all your connections are good and clean. Voltage jumping around like that is not good.

Might be worth checking that the readings are accurate. If the oscillations on the voltage are that bad, I wonder if you’d be able to hear it in the motors? Is the current also jumping around the same way? If the current is stable, then you may have a bad connection on the power monitor or a faulty unit.

Julio_Diez · April 20, 2023, 4:21pm

The current is also jumping around. I will check all the connections in case there are any bad ones.
I will check also with a multimeter that the reading from the power monitor is correct.

It was a windy day so I couldn’t hear the motors properly but they didn’t make any strange noise as far as I know.

Thank you very much for your replies!

Allister · April 20, 2023, 6:02pm

Might be time to try a different battery.

xfacta · April 20, 2023, 9:12pm

We need to see the .bin log file to know more.
It could be the voltage and current variations are caused by motor outputs oscillating or something like that.

Julio_Diez · April 21, 2023, 9:41am

Thank you for your reply. Here you have the log file.

I was trying to do some test flights before enabling the harmonic notch filter. But I wanted to solve the battery issue before moving on with the tuning.

xfacta · April 21, 2023, 1:46pm

As suspected the voltage variations align with the wildly oscillating motor outputs, in particular it seems

motor outputs 1, 3 and 6 go down= voltage goes up
motor outputs 1, 3 and 6 go up= voltage goes down

So those motors and ESCs might be drawing more current for some reason.

I’ve picked one of the cleanest sections here so it’s a bit more obvious

The voltage problem will sort itself out when you fix the vibrations that are causing the motor output oscillations.

X axis vibes are good, but Y axis is going to bad, and Z axis is very bad and will likely cause a fly-away.
Do not fly again until you are certain vibrations are improved.
Look for wiring or something affecting the flight controller, such as pulled too tight or hanging around loose.
The Z axis vibrations are usually prop wash.
Motor balance and prop balance can contribute to all of those too.
Can you supply a photo of the flight controller mounting and surrounds?

Also please provide a few more details, like what props, motors and ESCs you have, and the frame too.

Set these before the next test flight:

Upgrade to latest stable firmware

ARMING_CHECK,1
ATC_RAT_PIT_FLTD,13
ATC_RAT_PIT_FLTT,13
ATC_RAT_RLL_FLTD,13
ATC_RAT_RLL_FLTT,13
BATT_FS_CRT_ACT,1
BATT_FS_LOW_ACT,2 or 3
INS_ACCEL_FILTER,10
INS_HNTCH_ENABLE,1  // set this then refresh params to see the rest
INS_HNTCH_MODE,1
INS_HNTCH_REF,0.11
INS_HNTCH_FREQ,50
INS_HNTCH_BW,25
INS_HNTCH_FM_RAT,0.7
INS_LOG_BAT_MASK,1
INS_LOG_BAT_OPT,4

Start in Stabilise mode for take off and some small pitch/roll movements. If everything is going OK then use AltHold mode and do some more gentle pitch and roll. Use Loiter in the same flight only if stability is very good.
Let’s see that .bin log file.

Julio_Diez · April 21, 2023, 2:38pm

Thank you Shawn for the analysis. I am not really experienced, so I don’t know what can be causing the Z vibrations. I will check that all the motors and propellers are properly balanced.

The drone is composed of:

18’‘x5.5’’ propellers
Motor: T-Motor U7 KV420
ESC: Hobbywing XRotor 40A
Frame: Tarot X6
Battery: Lumenier 13000mAh (20C - Burst 40C)

Here is a picture of the flight controller mounted.

I have just changed the parameters. I will update you with the new .bin log file when I can fly (it is really windy these days).

Thanks all

dkemxr · April 21, 2023, 3:00pm

1st Remove the zip tie around the FC for the GPS cable. Strain relieve it to the frame with slack to the Flight Controller and any other cables that might be pulling or constraining the FC. It looks like some pretty thin CF for those plates.

xfacta · April 21, 2023, 9:07pm

Everything is about vibrations!

With that GPS cable, secure it to the frame, not the Flight Controller.
The wiring must have some flexibility where it plugs into the FC and not be pulled tight, but not so loose it is flapping around in the wind.
Where all those red/white (servo?) wires are connected at the end, they need a small amount of length in a straight line before they are bunched up and strapped down, so they are not transferring vibrations to the FC.

The FC should be mounted on some sort of double sided anti-vibration foam.

Also go over any wiring and parts that are hidden in that photo - everything must be secured so it doesnt vibrate.

Julio_Diez · May 2, 2023, 4:33pm

Hello again! After two weeks of not being able to fly due to bad weather, today I went to do the test flights you suggested. I changed all the parameters you mentioned and also tie all the cables that were loose in the frame.

I flew first in stabilize mode and had no problems while flying, really stable. But when I was about to land I put some throttle to put it back in the air because it was to land in a muddy place and it flipped and fall in its back. You can have here the log of this one.

Then, I check that everything was in order after the crash and I flew in alt-hold mode without any issues and then in loiter as everything was working properly. Here is the log of this flight.

I have checked the logs and I can see that I still have high Z vibrations. Also, the battery voltage oscillations are still there. I do not know what else I can do to improve it.

Thanks for your help.

Julio_Diez · May 4, 2023, 2:24pm

Also, I have the parameter MOT_HOVER_LEARN to 2, so it learns and saves. The parameter MOT_THST_HOVER is now 0.135, but in the documentation, it says that it has to be between 0.2 and 0.8. In the picture, you can see how ThH moves towards ThO (approximately 0.135).

Is there any problem with this value of MOT_THST_HOVER or should I change it to 0.2 to be in the range defined in the documentation?

dkemxr · May 4, 2023, 3:31pm

No, leave it at whatever has been learned. I don’t know where you read it has to between 0.2 and 0.8. It can’t possible be 0.8 it wouldn’t have enough thrust to leave the ground. And below 0.2 is very common. maybe you are referring to Expo.

Julio_Diez · May 4, 2023, 3:44pm

Here in the documentation, it says: “The value should be between 0.2 and 0.8.” That is why I was a bit surprised by my value.

https://ardupilot.org/copter/docs/ac_throttlemid.html

dkemxr · May 4, 2023, 3:49pm

I suppose it’s there for the same reason that a hover throttle value below 0.125 won’t be learned or there are Nanny messages in Mission Planner warning not to use a perfectly acceptable value for a particular parameter.
Hasn’t caught up with the times…