MicroMod Teensy Custom Carrier - USB Comms

Status
Not open for further replies.
Looked at the gerbers. Can't see any reason this wouldn't work.

I'm not familiar with AMS1117-3V3. Quick search turned up a datasheet which says it wants a 22uF tantalum capacitor for stability (on page 4 under "Stability"). PCB layout looks like 4 ceramic capacitors. Maybe this is one of the types of LDOs which needs capacitor ESR between a certain range for stability? If all else has failed, I'd remove those 4 ceramic capacitors and try 1 tantalum.

Or maybe try a LM1117 chip instead?
 
Thanks for the help Paul. I tried a tantalum instead and still no change. I put a scope on the 3V3 while it was attached to a pc and got this, which doesn't look particularly worrisome:

DS0000.jpg

I don't know! Nothing's attached to the board now, I desoldered the other chips. Just the regulator and a few caps. I've also had a good look under the microscope.
 
Sorry I know you have probably done all of these things, but if it were me some of the things I might try include:

a) Have you moved the MMOD into some other board like ATP and make sure it is still working recently...

b) again double check that the board is receiving 3.3v and GND (I know you have done so), and the USB VBUS maybe... Also to help verify no shorts, that the voltages are fine...

c) would probably ring out RESET/BOOT pins, to make sure reset is not held low which goes to ON/OFF on teensy.

d) Try different VR on off chance that maybe the power on ramp up has some issue? Mine right now is rigged with a Pololu DC/DC converter...
And I mean rigged:
screenshot.jpg
I keep meaning to test it some more, to see if other things work like the 5v VR... But been distracted..
 
Hello Kurt,

a) Yes multiple times.
b) I've checked around the board and the pins on the M.2. USB to the M.2 isn't shorted.
c) Yes, those are fine.
d) I've tried a separate power supply and applied 3V3 both before and after 5V is applied.

My USB traces must be terrible, which I can't believe. Compared to my TSynth project and Sparkfun's own ATP board, they look ideal. Short, about 90ohm (I think) and unbroken ground plane underneath. Even using wires between connector and PCB usually works.


Cut down to the basics and four more boards to go...

20211103_114100.jpg
 
I tried to reproduce this problem by building one of the breakout boards I made with only the 3.3V regulator (MCP1826) and its capacitors, the USB connector, and pushbutton.

micromod.jpg

I uploaded code a few dozen times, some using Upload from Arduino, other times by pressing the pushbutton. So far I have not managed to get a single NXP ROM failure.
 
I'm getting the same problem with the TeensyMM on the ATP board now. When plugging in, the Serial USB is not present in Arduino IDE and a simple Blink program fails to run. Pressing Boot on the ATP presents the Bootloader in Arduino IDE but uploading code usually fails at the erase stage. This is on two different PCs running Win 7 and 11. Intermittently, it works fine.

Could it be the MicroMod device, which was working fine from July when I received it? I'll order another and see if that's the problem.

When it manages to get past erase, I'm getting a message that "Automatic upload has been disabled, because the board returned to bootloader mode..."
 
Last edited:
Could it be the MicroMod device,

It's possible something went wrong with the MicroMod, if it's failing the same way on the ATP carrier.

My understanding from everything that was said before msg #32 was this MicroMod was working in the Sparkfun carriers but mysteriously failing on this custom carrier board. On msg #28 Kurt asked "Have you moved the MMOD into some other board like ATP and make sure it is still working recently..." and the answer on msg #29 was "Yes multiple times."

If it's actually failing the same way on all carrier boards, of course that's a pretty strong sign the MicroMod has a problem. This particular issue described would be most likely explained by damage to the flash memory chip on the bottom, or perhaps damage in this area on the top side.

mm.jpg
 
It started failing on the ATP board yesterday and today. The MM isn't visibly damaged, I've looked at both sides under a stereo microscope.

Just now, I've managed to get it programmed by using the ATP and working momentarily on another one of my PCBs (not the one I cut down in #29). It reads from the SD card and I'm getting Serial messages and then the USB fails - this is a big improvement, my PCB actually works! Plugging out and in, there's no USB port once more.

I'll order another, see what happens and report back. It's a screwy problem, please don't spend any more time on it.
 
For T_MM Beta PJRC provided a single processor unit here. It moved a dozen or two times between a PJRC breakout, ATP, Machine learning, and at least once to a Display carrier : Never an obvious issue.
> Even bought a longer screw and added double nuts (at the right clamp height) to replace the tiny screw without dropping it for all the changes it did.

Using a NEW 'just unwrapped' production board some weeks back on a new ATP ( still not found the original ATP carrier ) it was connecting funny/intermittently as posted at the time. Swapping it with another NEW production unit and carrier ALL came to working reliably as far as tested. Not sure what was behind that. Seemed the cable at hand was pushing power before seated and maybe dropping out when carrier end was final connection - then it didn't seem that it was the cable.

But if not the cable then it seems it was the M2 seating on the carrier. It got set aside for new boards and not scope examined as it was working then.

Having a second T_MM processor board is a good thing for sanity testing, luckily the rare Beta board was good.
 
Hello again. :D I've received a new Teensy MicroMod and it works on both ATP board and my PCB. The other one failed in an odd way, refusing to work on all my PCBs immediately and then after a couple of weeks of swapping between boards, the ATP board also. Again thanks for the time and help. I'll attempt to investigate the problem with the MicroMod board. Could I have fried something due to static? Could the MOSFETS attached to PMIC be damaged?
 
Hello, I'm getting around to having a look at my non-functioning TeensyMM. There isn't much to easily change really other than the MOSFETs connected to the PMIC. I've looked at both the crystal oscillators and their producing 32.76KHz and 24MHz as expected. I've got a wire attached to the PMIC test pad, what can I expect at power on? PMIC drives high to 1.1V and then stays at this level? I'm measuring 3.2V on it.

Other than that, I can't really go much further. USB still doesn't appear when plugged in, no LEDs and no amount of reseting changes this. I'm testing with the ATP carrier board. Thanks for any suggestions.
 
Last edited:
Hello, I know people have been tearing their hair out and broken the F5 key on a couple of laptops waiting for an update on this. I've got my dodgy TeensyMM working again. I replaced the two PMIC MOSFETs because, well what else can you replace? That didn't make any difference, so I went over both sides blasting it with heat from the hot air gun because, well what else can you do other than wishing really hard?

I think doing the top of the PCB, either the big cap (C21?) or the bootloader chip cured the problem. It now shows up as normal in Arduino and can be programmed.

16402-SparkFun_MicroMod_Teensy_Processor-01.jpg
 
Status
Not open for further replies.
Back
Top