Custom 3.2 Board - Partially working - Bootloading process?

Status
Not open for further replies.

Davidelvig

Well-known member
New board based on Teensy 3.2 and MKL02 boot loader from PJRC

When running Teensyduino (and no Teensy Loader) , connecting the new board, and pressing reset, I see this in the Tools/Port menu
Teensy32-Reset-NoLoader1.jpg
This is identical to the behavior on an off-the-shelf Teensy 3.2 purchased form PJRC

After running Teensy Loader, I see this with my custom board
Teensy32-Reset-AfterLoader1.jpg
- no useful Serial ports

In comparison, the off-the-shelf Teensy 3.2 board after Teensy Loader runs shows the usual
Teensy32-Reset-AfterLoader1.jpg

My sense is that I have a lot working here.
What's next to test?

Thanks
Dave
 
One of my consulting EE's has done some testing, comparing my custom board (DB Board) to an off-the-shelf Teensy.

There's a fundamental difference in behavior between a Teensy board and the DB board. On Teensy, after a reset (but prior to loading a sketch), the 16MHz crystal is oscillating. After loading a sketch, it is still oscillating (or maybe more accurately, it's oscillating again).

On DB, The 16MHz crystal is oscillating after a reset, but prior to loading a sketch (same as on Teensy). However, during the sketch load process, the 16MHz crystal stops oscillating and doesn't start again, at which point the board is unresponsive. Pressing the reset button will start the crystal again.

I believe the crystal will only oscillate if software configures the chip correctly to enable it. I don't know what this means, but it seems like it could be significant in understanding where things are getting stuck.

Any ideas as to why this board is failing from this crystal behavior, or the previous post?

Thanks in advance!
 
Last edited:
Did you connect 3.3V power to the processor's VBAT pin?

Check the troubleshooting info at the bottom of the MKL02 page.

https://www.pjrc.com/store/ic_mkl02.html

When the processor runs the bootloader but not compiled programs, it's usually VBAT missing. When the compiled program tries to initialize the RTC, it crashes if there is no VBAT.
 
Wow, I think that's it.
From the Bootloader page, this Troubleshooting link appears to be the one you refer to.

When I short VBAT to 3V3, the reboot completes, and the sketch runs... and the Serial port appears.

I know I read that post too as I was designing. I'll review where I went wrong, and continue testing.


Thanks, Paul!!!
 
Just to follow-up... the problem was indeed that VBAT was not connected to 3v3 on startup.
Successful board after that jumper.
 
Status
Not open for further replies.
Back
Top