I am setting up another Skylark and bought a TBS Lucid H7 Wing flightcontroller for this due to Matek H7 boards being either unavailable or terribly expensive at the moment. I am running Arduplane 4.7-Beta5 (because of the AUTOLAND mode). This plane is meant as a fun plane mostly flown in acro/manual mode.
ESC used is a Sequre 2670 AM32 ESC connected using DShot with telemetry running over a separate line to a serial port.
Physically, the battery is connected directly to the ESC. Power for the FC is taken from the ESC’s pads and routed with thinner wires. This keeps the high current path short and simple and prevents running power wiring back and forth.
I would prefer to use the ESC current and voltage measurements over the Lucid H7 builtin analog measurements. I am most interested in voltage, current only when trying a new prop or so.
Now, I have some issues with battery measurement. When connecting a 4S battery somewhat above storage (15.5V) both ESC and analog measurement under-report the voltage:
I measured the voltage with a multimeter on the ESC pads and the flightcontroller/ESC reports are both incorrect. I can also light up the 12V COB LED strips on the wings, which I would not be able to do if voltage was actually below 12V
When setting the BATT_VOLT_MULT parameter from it’s default of 11 to 21 (which seems a common number; many Matek F405-WING style FC’s seem to use it), at least the analog battery voltage measurement is correct. ESC is not. But it is also not correct in the AM32 configurator.
I have seen this behaviour before. Back then I flashed back to Arduplane 4.5 which did not resolve the issue. I managed to resolve it back then, but did not take notes in the heat of the battle and forgot what I did to resolve it. Maybe someone here knows?
Here is the current set of parameters. I did take a diff from another Skylark I have and used that to do part of the initial setup; plane has not flown yet.
I also recently installed a TBS Lucid H7 Wing FC in a fixed-wing aircraft (AtomRC Swordfish) using 4S, and had the same issue…reported voltage and current was incorrect.
Using a multimeter and performing a manual calibration through MP, the voltage is now very accurate and the current is close.
My BATT_VOLT_MULT ended up being 20.87567 (close to 21, as you stated), and BATT_AMP_PERVLT landed on 37.25927 (calibrated at 4A and near the default 40).
The BATT_VOLT_MULT of 11 in their manual may be a typo.
I’ll reach out to TBS to see what they say and report back.
Same here. Recently tried the TBS Lucid H7 wing for the first time.Flying Arduplane 4.7 beta5. The current seemed correct but voltage was about half of true voltage. I did a manual calibration and got a BATT_VOLT_MULT of 20.9. All has been working as expected since correcting the voltage calibration. Of note, I did not cross check against ESC telemetry. If ESC telemetry is also incorrect it seems this would argue against it being only a FC voltage multiplier error. I’m not home.to check the ESC telemetry voltage but will try to do so in the next couple of
That’s OK if it’s accurate which isn’t always the case and there is no way to scale it if it isn’t. You want to set the multiplier by measured battery voltage anyway so it doesn’t much matter what the default value is.
Because the firmware for the Lucid H7 Wing is using the Lucid H7 HWDEF with a few modifications the battery related parameters are the same. So if there is a hardware difference between these two in this regard then the HWDEF for the Wing needs to change.
Last time I checked their website and the PDF with all their Lucid controllers in it said 11.0. Seems they are quick to correct then.
Still weird that the ESC is providing approximately the same erroneous voltage. Coincidence? I did update the firmware without improvement.
The routing is
XT60+ —(thick wire, AWG12?)–> ESC+ pad —(AWG24 wire)–> FC BAT+
XT60- —(thick wire, AWG12?)–> ESC- pad —(AWG20 wire)–> FC BAT-
There are capacitors directly on the ESC and an extra one on the FC BAT pads.
That was the original plan (works fine on my other Skylark which uses a Tekko32 AM32), but the ESC also reports the incorrect voltage and that one cannot be adjusted.
So, analog voltage it is, which can be adjusted using the multiplier and offset. Current is less important for a ‘basher plane’.
I will send a mail to Sequre if I can find a support address. I thought it was the better class ESC but maybe I am wrong.
It would be nice if we could do Ax+B calibration on any monitor value in Ardupilot like we can with the analog voltage/current sensors.
On the other hand, it is not unreasonable to expect useable results from an external sensor.
