4-in1 ESC / FC current read

ArduCopter Copter 4.6
1

Hi, analyzing MP log i see an issue on logged value of Bat current…

my setup:

-Fc Matek H743 Slim V4 (4.6.3/MatekH743-bdshot)

-ESC 4en1 Sequre Blueson A2 AM32 65A 6S ( AM32_SEQURE_4IN1_F421_2.20)

-Li-ion 6sp1 4000mAh.

why the current report by esc has a wrong value?

why the current report by ADC cur pin of fc has a wrong value too?

Vbat seems to be well reported by esc and fc.

i may be set wrong paramerter value ..?…?

all wires are connected between ECS and FC

image
image1347×670 231 KB

image
image1526×890 136 KB

@ 0 rpm(before arming) , ecs report 18A (same for all ECSs), should be 0A

@ 0 rpm (before arming), FC report 4A, should be 0.2A as measured

i try to ajust with BATT_AMP_OFFSET and BATT_AMP_PERVLT but i definitely return to default parameter because this issue cannot be solved with a simple ratio and offset (seems to be not a linear correction…)

Could you please help to understand what’s wrong in my setup?

unable to add zipped BIN log file here .. over 8Mb…

thank you

have a nice day

/olivier

2

Hi Olivier,

Are your parameters BATT_CURR_PIN: 11 and BATT_AMP_PERVLT: 40?

It’s possible you may need to do a manual calibration for your current sensing with a multimeter and adjusting in MP.

3

Hi Mike thank you for this reply!

Yes i fallow manual calibration guide and Matek recommendation too, so i already set:

BATT_CURR_PIN to 11 and BATT_AMP_PERVLT to 40

Both parameter handle the read ADC mV input pin (“CUR” pin on the matek FC) and computed Vbat Current used by Ardupilot and display in MP. mV is comming from build in power CMS resistor (shunt) on ESC.

i understand that at low current the calculated current has a very poor accuracy. this could explain why 4A display for 0.2A real current?.? huge poor accuracy!

Am’i wrong?

So i don”t understand what’s wrong… hardware issue?

About Esc1/2/3/4.Current value coming from esc Telemetry ?

i read that telemetry value (including current for each ecs) could comes from bidirectional Dshot Motor wire M1/2/3/4 (pad S1 to S4 on matek fc) OR via command RX via the UART (SERIAL5 = UART8 for Matek Fc)?

it’s not very clear for me!… and if the current report wrong value by esc, it will definitily an ecs issue?

note: i ‘ve tried with same ECS hardware but brand new part => same result ! may be i’ve to change Sequre supplier by an other…

param list attached.

param_18_06_2026.param (25.9 KB)

thank you again for your help…

Have a nice day

/olivier

4

This is common.

I would re-calibrate again while pulling more amps (e.g., 5A, measured on multimeter) and enter that as shown below.

This will change your BATT_AMP_PERVLT to a value different than 40.

1
1703×648 77 KB

If the Sequre sensor won’t cooperate and you end up looking for another option, inexpensive ACS758-based sensors like the one below have worked well for me.

1 Like
5

It’s uncommon to have Current in the telemetry data with 4-in-1 ESC’s. A very few do.

1 Like
6

@dkemxr @nspector
Dave and Mike hi,

Thanks for your feedback and useful advice.

We fly drones with enormous complexity, PID, filtering, etc., but we failed with read an accurate battery current at full range. :slight_smile: @SEQURE @MATEK

I’m going to integrate a Matek I2C-INA-BM board.

Do you have any experience with this board using ArduPilot?
Accuracy at low current, up to 50A? Coulomb metry? Is the flight time calculation ultra-reliable?

thank you again for the time you spend for my issues!

Have a nice day!!!

/olivier

7

Those Matek I2C-INA-BM sensors are very accurate, I use them on my solar rovers.
https://www.mateksys.com/?portfolio=i2c-ina-bm

2 Likes
8

I don’t have experience with that particular Matek power monitor, but it looks very good. Matek says 2% current sensing accuracy.

The data sheet for the TI INA228 can be found here.

I’ve used Mauch current sensors and can recommend them as well.

1 Like