One of the options I’ve considered is using an appropriately sized electrolytic capacitor - with enough capacity to run the flight controller and it’s accessories (gps, telem radio, etc.) for a battery swap.
I was thinking of maybe installing such a capacitor to the Power-2 port on a carrier board.
I don’t know if this is possible - or what the possible pit-falls might be. One possibility might be the flight controller going haywire if the battery is isn’t swapped fast enough and the current from the capacitor begins to fade.
Before I begin experimenting, I thought I’d look to the community for comments. Thanks!
Time ago I tried that supplying the FC with a power bank through the USB connector during swap. It worked, although the remaining battery capacity was not initialized, and eventually was exhausted (I had to make it five times that of the battery for five batteries).
So the FC was missing recognizing the battery swap. This happened long time ago; I don’t know what happens now with recent FC firmware versions.
Mission Planner now has an option to reset the battery statistics. So as long as you’ve got a telemetry connection - you can set your failsafes properly for your battery size.
I haven’t actually tested this - but I’m confident it will work.
I’m just trying to figure out the best option for keeping the FC running. I’d like to also keep my other electronics running too - so I’m probably going to make some sort of connection to the Vcc downstream from the current sensor. That should keep everything alive. I may have to install a safety switch - because connecting even a small power source to the Vcc makes it possible for the motors to spin.
Some quick numbers with the farad+diode+resistor solution: if you want to supply 0.5A during 30s with a voltage drop of 1V:
C=0.5*30/1F=15F.
You can find 15F/5.5V 3x2x1cm 50g (not too bad).
Charging: 5 times constant a flight time of 5’:
R=300/(5*15)=4ohm (say 5W), so the rush current would be 1.25A (not too good, but probably the C internal resistor would limit it more).
All above if I am not too wrong, and pending experimentation (it could work). A battery instead would be an ugly solution.
I would prefer the power bank. MP reset is a workaround (manual intervention). The FC alone should do the job.
Thank you for the analysis. This level of electronics is not my strong suit. I appreciate your help.
While I agree on principle that the flight controller alone should do the job - that doesn’t appear how it works on the DJI equipment and software I’ve tested. The resume function relies on the ground control program.
With the DJI equipment I’ve tested, after landing for a battery change, the ground control program uploads the remaining waypoints of the mission before takeoff and resuming the mission. The aircraft retains no information other than the drones configuration settings.
This seems a bit in-elegant - but it works well.
Since none of the common ArduPilot ground control software has this capability - keeping the flight controller active is the only option for now.
There is another consideration. There may be other electronics on the drone that might benefit from retaining power. One such item is the FoxTech Map-02 camera I use on a copter. This camera has a peculiar requirement when powered up - and it would be much easier if it stayed powered up after landing.
