DIY Teensy 3.6 MK02 problem

Status
Not open for further replies.

Koka

New member
Dear community members, first of all I’d like to thank all of you for your discussions. They are of incredible value for those who follow.

Yesterday I assembled my first DIY teensy board (T3.6). After 20 hours I’m still awake, trying to understand why it fails. I read everything available on the forums.

Since MK02 chips did not arrive on time, I removed two bootloader chips from Teensies (3.2) I had lying around, carefully, with a lot of flux and hot air. Then I added one of them to my custom board.

I can measure stable 3.3v on the input of MK02, but the PROG pad seems to be floating instead of being pulled high. When I push the program button, the pad is pulled low, but when I release the button, it continues to float. It is floating in the range of 200mV.

RESET pins (PTA4, PTA13) are pulled high, and when I push the program button, they continue to stay high instead of being pulled low.

Maybe someone has had a similar problem, I would be extremely grateful for any suggestions.

1. All solder joints seem to be smooth and clean, inspected through a microscope, no offsets, no bridges;

2. 3.3v is provided nicely to both, MK66 and MK02;

3. I re-inspected schematics and layout several times in KiCad. No missing traces, no copper dirt. Everything is clean. I followed up and inspected my board with a multimeter. It corresponds to the layout;

4. Crystal is placed according to recommendations on the PJRC website;

Thanks in advance!

Follow up:

My mac sees the MK66. When I go to system report, I can see a “Kinetis Bootloader” in the USB section, but Arduino (1.8.10) sees nothing and therefore I cannot program the board.

MK02 seems to fail to respond to a “program” call and so it never sets the MK66 into programming mode.

I desoldered the MK02 chip and soldered another one, just to make sure it was not a damaged chip. The behavior is the same. RESET pad is NOT pulled high by default, which it is, on Teensy 3.2 boards I examined.

https://www.dropbox.com/s/l8tu79wkr8n338a/Screenshot%202020-03-24%20at%2016.36.42.png?dl=0
https://www.dropbox.com/s/h2tsiixsp409po8/Screenshot%202020-03-24%20at%2016.39.37.png?dl=0
 
Last edited:
A quick update:

I used an oscilloscope instead of a multimeter to do the measurements.

It turns out that the bootloader chip speaks with the MK66 chip. I did a cross-comparison between my board and original teensy 3.2, and all signals coming from the MK02 seem to be identical.

The MK66 is seen by my mac. System report displays “Kinetis Bootloader”, but it is not seen by Teensyduino (1.52 beta).

When I push the program button, I see activity on the USB line for a couple of seconds, but I cannot make my board display in the “port” menu in Arduino IDE, so that I can program it for the first time.

Please help if you can!
Thanks in advance,
Koka
 
PROBLEM SOLVED!

For someone who might stumble upon this thread, the problem was incompatibility of a bootloader chip that I removed from original Teensy 3.2s.

Bootloader chips for DIY teensies sold by PJRC store can identify microcontrollers and work with teensies 3.x, but the ones that come with factory-made boards are programmed to work with those particular models ONLY. I.e., if you desolder a bootloader chip from T3.2, it won’t work with your custom T3.6 board. In order for it to work, you need to grab one from a T3.6 board.

We did so and it worked. We will soon get a plenty of bootloaders from PJRC and do a small production run.
 
Thanks for this thread. It’s encouraging. I’m taking a similar path to a 3.2 device, and am only at board design phase.
 
Good luck David,

one thing that I learnt in this process is that if you follow the reference schematics, if you follow the additional tips on the PJRC website (check the bootloader page), if you do an ok layout and most importantly, if you do a good soldering job, then there is no reason why it should not work.

Make sure you order an SMD stencil, buy some good quality paste, bake it in a good oven, etc.

In my case everything was right, except that I rushed forward and did not wait for the right bootloader chips. If I had done that, it would be a success with the first try.
 
Here's the Bootloader section of my schematic. @Kola (or others), does this seem right for a Teensy 3.2
Bootloader.jpg?
 
No, that does not look correct. Here's that part of the Teensy 3.2 schematic.

schematic32.png

The Reset signal is supposed to connect to pins 4 and 13. But on your schematic, it's connected to pins 4 and 5.

Pin 7 is supposed to connect to PTA3 on the MK20. On you schematic, it's net is labeled "PTA7_PTA0". Maybe that really does connect to PTA3, but looking only at the name seems to suggest it does not.

I could go on, but I think you can get the idea to re-check all your connections. Hopefully this will help save you from a non-working PCB build.
 
Oh my goodness!
I had built a very-confused Eagle part.
I rebuilt it (with the tool, from NXP) and here's the result:
Bootloader Schematic.jpg
If you would, please have another look.

I do have question about the ***Reset*** in yellow and the Prog
  • Is "PROG" to allow another pin connection to bring PTB2 to ground through software?
  • Is "***Reset***" to allow another physical switch to be added for resetting?
 
Last edited:
One more post to goose this to the top. Any possibility of a quick review of the boolloader schematic?
My EE consultant has no experience with this type of part.
 
Self-Corrected again.
I think this is correct.
Bootloader Schema.jpg
I hope I'm on to it now.
@Paul, perhaps you can delete my previous posts in this thread to avoid anyone using them.
 
Status
Not open for further replies.
Back
Top