How to Compassmot without short circuit?

Hello folks,

At the moment I am building my hexa with an APM 2.5.2 and the 3DR power module.

When I want to do the Compassmot-calibration I need to connect the APM via USB to my PC as well having the Lipo battery connected. As far as I know and can read out of the schematics the 5V USB power goes directly over a 500 mA fuse to the Vcc input rail of the APM. The 3DR power module feeds 5.37 volts to the power connector of the APM and these 5.37 go over the Schottky diode and another 500 mA fuse. The diode drops around 0.37 volts so that Vcc is 5 volts then.

When the Lipo is connected and also USB both voltages collide on the Vcc rail of the APM and result most likely in a current which destroys the Schottky diode or one of the 500 mA fuses if the current is higher. This solution protects the APM for sure but in this constellation it destroys components on the APM board.

I am thinking of not using the +5.37 V from the power module and use a separate BEC with 5.0 V on the input side of the APM to power the Vcc rail directly. I would then need to disconnect power when doing Compassmot.

Does anyone have any suggestions regarding my problem?

Kind regards and merry Xmas!

Sven

APM 2.5.2 as in RC-Timer APM 2.5.2? That would be not supported as this forum is for genuine 3DR hardware only!

Anyways, you don’t do compassmot via USB. You do it via telemetry radios because of the power conflict issue and also for safety reasons. You don’t want to have a copter spinning up the props to half or even full power inside your living quarters, let alone near to you or your computer.

Thanks for your answer Stefan!

Well, the original 3DR APM suffers from the same problem…

Does CLI work over telemetry radios? AFAIR this does not work over radio…?

What about the plan using a separate BEC for powering the APM via the input rail? Is this a good or a bad idea?

Greetings!

Sven

Yes, cli works. The trick is to press the reset button on the APM and then quickly click connect in console mode in Mission Planner.
I’m not sure about the BEC but if you don’t have servos, you wouldn’t need any additional power supply to the APM.
Still, the safety issue persists! You just don’t want to have a copter spinning up close to your pc and inside!

Okay, I will try connecting to the CLI by pressing the reset button before.

Maybe the instructions for Compassmot on the website need to be updated. It is reading the way to connect Lipo and USB at the same time which could result in a defective APM.

Many thanks for your help!

Sven

I don’t think there’s any issue with connecting the APM using the USB connection and the Lipo battery at the same time. I’ve done it plenty of times and my APMs are all fine.

I don’t think this forum is for 3dr genuine hardware only…I’ve never read that anywhere. I wish people would buy 3dr hardware because that money goes back to support the open source hardware and software that people are using but I don’t think it’s a rule of this forum.

viewtopic.php?f=49&t=4646 - Section 6
viewtopic.php?f=49&t=4676
viewtopic.php?f=52&t=4633

That depends on how high your USB voltage is. If your USB voltage is less than say 4.5 volts, there is a current flowing from the power module’s 5.3 volts (source) through the schottky diode and fuse 1 with 500 mA rating into your USB controller (sink).

It depends on how high this current is if your fuse, your diode or the PC’s USB port gets damaged or not. I would not recommend connecting both at the same time since the effects are unpredictable. The problematic point is that the schotttky diode is rated at 1 A current max, but the cooling surface of the layout is too small. That means in an overcurrent event the diode blows earlier than the fuse. Bummer if this happens in air…

Unfortunately this is broken by the design of the APM. It would be better to have a low-resistance switch (logic level MOSFETs in SOT-23 case) between the USB side and the power coming from the power module and switch between them in a smart way (higher voltage). The circuit could be made self-locking so that only one power source at the same time is connected. - Hopefully the design of the PX4 is or will be changed in that point!

In the meantime I decided to not use the protection circuit of the APM and power it from the analog input pins via a dedicated UBEC 5V/3A. The two +5.3 volts pins have been removed from the power module connector. This way I can still measure voltage and current but using another power source.

Anyway…if the teams needs a helping hand in designing safe to use electronics, please come back to me!

Kind regards!

Sven