version without qwiic switch for UTokyo

[jerabaul29]

@tnn77 @T-Kodaira

• I now pushed a 2-ways-communication version of the buoy without the need to use the qwiic switch:

  • firmware: https://github.com/jerabaul29/OpenMetBuoy-v2021a/tree/main/legacy_firmware/firmware/steval_gps_waves_pcbtraces_2ways_noqwiicswitch
  • binary (so I know that this compiles on my machine :) ): https://github.com/jerabaul29/OpenMetBuoy-v2021a/blob/main/legacy_firmware/binaries/steval_gps_waves_pcbtraces_2ways_noqwiicswitch.bin

• if your branch / internal UTokyo OMB firmware is a bit different from mine, it should still be very easy to take in the changes I have introduced: the changes on this branch should be orthogonal to your changes. See the log of changes (may be extended when we test and debug, so this may not be everything in a few iterations):

https://github.com/jerabaul29/OpenMetBuoy-v2021a/commit/69d259b37d01aefb018d5decc8412b5ae5a3dcdb

• the pin used to power the IMU is defined in:

https://github.com/jerabaul29/OpenMetBuoy-v2021a/blob/f320d43214e4752da8bce15e7450fbcf4fe0184a/legacy_firmware/firmware/steval_gps_waves_pcbtraces_2ways_noqwiicswitch/params.h#L17

naturally feel free to edit this :) .

• the wiring will be:

  • SCL AGT <-> SCL IMU as usual
  • SDA AGT <-> SDA IMU as usual
  • GND AGT <-> GND IMU as usual
  • IMUPwr AGT <-> PWR 3.3V IMU

• to do tests, please remember to:

  • do a full functionality test; i.e., let the instrument work for a few cycles and check that everything looks good, there is no crash / reboot etc even after a few cycles of waves / gnss / sleep / iridium repeating
  • check the power consumption (in particular in wave mode and during sleep)

If you can report the results you get during tests (observations / power use / serial logs if relevant), that would be perfect :) .

The Readme for the corresponding firmware should be updated, let me know if you have any comments.

Let me know if you have any comments / feedback :) .

---

This is inspired by the discussions we had with @gauteh when implementing a similar thing for https://github.com/gauteh/omb-2024-nano ; thank you / credits @gauteh for the early tests! :)

[T-Kodaira]

@jerabaul29 Thank you so much for your prompt job! We will try the following two items as you suggest

• do a full functionality test; i.e., let the instrument work for a few cycles and check that everything looks good, there is no crash / reboot etc even after a few cycles of waves / gnss / sleep / iridium repeating
• check the power consumption (in particular in wave mode and during sleep)

[gauteh]

Hi Jean,

the qwiic power switch is hard to get. Do you have a way we can wire up the standard PCB to use this code?

-- gaute

[jerabaul29]

Hi @gauteh ,

Yes, this is the same challenge that @T-Kodaira and @tnn77 have. I think the simplest is to use the same method that you use in the omb-nano you are working on. This is the goal with the branch that I point to here. I have not had the time to test this myself yet - I will try to find some hardware and do some tests either tomorrow or next week. Is that early enough for you @gauteh ? :)

[jerabaul29]

@tnn77 @T-Kodaira this ends up now being the exact same code as #98 , and the other issue has progressed further in the testing etc. So let us share experiences and checks done there! :)

This means that for the testing ( i) test over a few days to check stability, ii) power checks), you should use the code from https://github.com/jerabaul29/OpenMetBuoy-v2021a/tree/main/legacy_firmware/firmware/steval_gps_waves_pcbtraces_2ways_noqwiicswitch , and you can use if you want the binary https://github.com/jerabaul29/OpenMetBuoy-v2021a/blob/main/legacy_firmware/binaries/steval_gps_waves_pcbtraces_2ways_noqwiicswitch_v2.bin .

@tnn77 @T-Kodaira if adapting to your own code, note the latest push I have made: you will need this too otherwise GPS does not work:

https://github.com/jerabaul29/OpenMetBuoy-v2021a/commit/0369dee4bf815722de8dcb1d922388ddb37baf6d .

We need to check if power use is effective or not as of now; If not, I will have to push some more updates to force relevant pins to float. this will be determined by the power use measurements during the tests :) .

[jerabaul29]

@tnn77 @T-Kodaira do not hesitate to let me know if anything is unclear! :)