New Teensy 4.0 Suddenly Not Recognized

Thundercat

Well-known member
Hi, I use a load of Teensys, but this one's a new on for me.

I attached the board and attempted to FuseWrite it. The board was recognized normally, the program ran, but no output to the serial monitor. This has happened a bunch of times, and I usually just run it again, because I'm OCD about this kind of thing, and I have doubts that it works correctly unless there is confirmation on the serial monitor. Usually it works after a few attempts.

Not this time.

After about 4-5 attempts, the Teensy is no longer recognized at all by Teensyduino. It does not even show up in the device list whatsoever.

I tried rebooting my computer, using a different USB port, changing the cable, holding the button down on the teensy prior to inserting the cable as detailed in a help thread, holding the button down for 20 seconds to hopefully reset it - nothing.

When I plug the teensy in, the red light on the Teensy flashes twice upon inserting the cable. It also keeps flashing 2 times every 20-30 seconds. I'm sure I haven't damaged it!

Oh, also, the teensy loader shows the message FuseWrite.ino.hex (unreadable).

Any ideas how I can get this little guy up and running again?

Thanks for your support,

Mike
 
Try the 15 sec restore. Hold the pushbutton for 13 to 17 seconds. At 13 sec, you'll see the red LED blink briefly to indicate you're at the beginning of the 4 sec window where releasing the button causes full memory wipe and restore of a known-good LED blink program.

You should see the red LED turn on bright while the erase is in progress. Erase and restore is slow, almost 1 minute on Teensy 4.0 and longer on Teensy 4.1. Be patient and allow it to complete. When finished, the red LED turns off and you should see the orange LED (pin 13) slowly blinking. Your PC should detect Teensy as RawHID (not Serial, so no COM port) in the Tools > Ports menu. Upload should work again.

This erase and restore only works if Teensy is in normal unsecured mode, not yet locked into secure mode. If you ran the fuse write sketch to program your key, it should still be unlocked. Only the 3rd program locks secure mode.
 
Hi,

I am having the same problem with the Teensy 4.1. I followed the steps mentioned before and restarted my computer just to be sure, and the Arduino IDE recognized the Teensy as a RawHID. However, when I try to upload a program, it says that it uploads it, but I can see that it disconnects with the Teensy just before finishing the process. The serial monitor does not work, and I am guessing the program does not fully upload since the blinking program is not working anymore but neither is the one I uploaded. I already uninstalled and installed the Arduino IDE and the Teensyduino. I have two Teensy 4.1 and thought that maybe one was faulty but both are having the same issue in two different computers. This issue started happening yesterday and I've been working on my Teensy everyday and it has been working fine. Any other suggestions as to what to do to solve the issue?

Thanks for the help,
LVargas
 
Try the 15 sec restore. Hold the pushbutton for 13 to 17 seconds. At 13 sec, you'll see the red LED blink briefly to indicate you're at the beginning of the 4 sec window where releasing the button causes full memory wipe and restore of a known-good LED blink program.

You should see the red LED turn on bright while the erase is in progress. Erase and restore is slow, almost 1 minute on Teensy 4.0 and longer on Teensy 4.1. Be patient and allow it to complete. When finished, the red LED turns off and you should see the orange LED (pin 13) slowly blinking. Your PC should detect Teensy as RawHID (not Serial, so no COM port) in the Tools > Ports menu. Upload should work again.

This erase and restore only works if Teensy is in normal unsecured mode, not yet locked into secure mode. If you ran the fuse write sketch to program your key, it should still be unlocked. Only the 3rd program locks secure mode.
Thank you Paul.

Wish I could say it worked.

If I hold down the program button for the requisite 15 - 20 seconds, it either never stops the flashing that I mentioned in the first post, or it does not flash at all - even after the 15-20 seconds. Holding for longer does not change anything.

These Teensys are so doggone hardy, I can't believe it's dead or unreachable somehow.

Any other suggestions?

Thanks,

Mike
 
OK I'm getting extremely frustrated.

Not one, not two, BUT THREE brand new Teensys are all doing the same thing! I am loathe to call out quality control, because issues are my fault 99.9% of the time.

But I'm having brand new Teensy 4.0's work for a short while - being recognized and allowing flash - and then stopping being recognized no matter what I do, no matter what USB cable I use, no matter if I try to hold the button for 15 seconds, etc. And strange flashing behavior on the red lights.

In fact I had another one like this a month ago! But I just assumed it was a one-off.

This isn't cheap to keep replacing them.

I'm going to attach the log from the last one that failed as I was simply trying to flash it with the Lock sketch. Note it had ALREADY successfully flashed all 3 security sketches, so I was just testing the flashing to make sure it was working before loading the main sketch. But it not only failed to flash, but now it too is not recognized AT ALL when I plug it in!

I would be SO HAPPY to be wrong or have screwed things up. But I buy a LOT of Teensys, and now I'm starting to lose money badly here as they are not working reliably for me now, in PCBs that had already been tested as reliable with them. These are commercial PCB's I've developed over years and have worked great with Teensys.

Any insights would be much appreciated. Again, I just want to solve the problem, not point fingers, but I've had a lot of experience with these now for the last couple years and this borders on insanity.

Oh, and very worrying, I can hear a funny sort of scratching sound when the Teensy is plugged in, like something is frying or something. But the unit is not hot to the touch.

And I'm a good solderer, so it's not like I've got messy blobs of solder everywhere (or ANYWHERE!).

Thanks for any insights.

Mike
 

Attachments

  • log.txt
    11 KB · Views: 48
[photo deleted]

A pic of the poor defunct little bugger

Which was BRAND NEWLY INSTALLED!!!

Note the upside down "L" = that's because I was able to successfully flash it with all 3 security sketches. So the unit absolutely was working just beautifully as expected. Until suddenly now it's not. For no reason I can ascertain.
 
Last edited by a moderator:
The ONLY other possibility I can think of, besides pointing the finger at the Teensy, is that the commercial PCB I'm using has some kind of internal short that I'm not aware of that is somehow frying the Teensys. But again, I've done a ton of PCBs and used them with the Teensys without issue.

I'm at a total loss. Any support is welcome

Mike
 
Thank you. The design works. I have units running on this PCB no problem.

The PCB is also an nth generation derivative from previously working designs.

So I have perfectly working units on this same run of PCBs.

I’ve never had a teensy just “go bad” unless I abused it somehow. And I used same processes with these boards as previously successfully working boards.

Of course there’s always possibility for human error or something I am missing; I do not discount that at all, which is why I am posting here, to get any wisdom from smart ppl like you guys.

In any case I’m now at 4 defunct units. Something is seriously wrong.

“Cable is bad!”

Same cable works fine on previous units.

“Bad soldering or assembly!”

4 in a row? There’s only a few solder joints I do; rest are commercially manufactured. And I’ve been building gear for 20 years. I have OCD and I meticulously clean and check each board after assembly too.

So I just don’t know. I’m hemorrhaging money on these now. And they are for a product I sell.

Again such odd behavior. They work for a few minutes and then die. Pointing to something shorting out.

Oh and nothing to do with the code either. Same code works flawlessly on other units. Same entire process of manufacture works flawlessly on other units, and has for a couple of years.

I’d love to find my idiot mistake here. Or discover just WHAT is going on so I can fix it.

I suppose I can do another run of PCBs and hope it was some internal fault or short in the previous run that would be corrected by a new batch. I did once get a batch of PCBs made that were missing an internal trace. But no shorts or intermittent issues (so far).

But this seems an expensive and haphazard way to troubleshoot.

Thanks for any other insights,

Mike
 
Are all the failures happening with locked Teensys?

Given nobody here has seen any code, schematics, etc. of anything you're using, the chances of anyone having any insight into the problem is very low.
 
Are all the failures happening with locked Teensys?
Good insight.

Yes, all the hardware is lockable Teensys.

However, some failed prior to even going past first of 3 locking procedures. In other words they weren’t even yet locked.

Thank you.
 
So I checked out that thread that @PaulS shared; it seems there is something similar happening for me, but with the Teensy 4.0s. And not all of them. At least the older ones didn't have an issue.

I've posted a question there and linked to this thread. If there's something going on with the Teensy 4.0s it would be good to get a handle on it too.

Thank you everyone.

Mike
 
Just FYI I'm now at 5 "defunct" Teensys and main boards...wish I wouldn't have already soldered them into the main boards.

Obviously there's something going on here.

Per Paul's suggestion I tried a different USB cable, and port, but no joy.

The last one that "died" or stopped responding, I plugged it in, and it was recognized as a USB port. It had already been flash fused locked with all 3 sketches.

It was visible for about 3 minutes, then stopped showing up in the port monitor of Arduino. Subsequent disconnecting/reconnecting did not cause it to show up again, ever. I'm beside myself trying to figure this out. I've got products to ship using the boards and they are just not working as they did before.

Again, these Teensys have not been abused in any way shape or form! Brand new Teensys...

When I get a little more time, probably next week, I'll unsolder one of the Teensys from the main boards they are in, and test them out of the circuit as a first step. But I stress this same main board they are soldered into, has worked successfully without issue, no design change, with other Teensys in past.
 
In the interest of discovery, I am posting the schematic for the unit in question. I won't keep it up too long, but hopefully it can help figure this out.

Thanks,

Mike


[schematic deleted]
 
Last edited by a moderator:
Could you explain how the power supply works? Is all the board's power coming from the USB cable connected to Teensy? Or is external power connected (maybe not shown on this schematic)?
 
Could you explain how the power supply works? Is all the board's power coming from the USB cable connected to Teensy? Or is external power connected (maybe not shown on this schematic)?
Thank you Paul. No external PSU. Power is being taken directly from the VIN pin to the regulator, which is onboard the PCB; it's 2A as I recall.

Only reason for the regulator is the unit has 4 RGB buttons with 3 LEDs each, so 12 LEDs, which seemed a bit too much to expect the onboard Teensy regulator to handle (12 * 20mA = 240mA; I think Teensy doesn't supply much more than this IIRC).

Thank you!

Mike
 
I think he meant what is supplying VIN - we can see the Teensy's VIN is connected to the regulator's VIN, but what's producing the 5V?
 
I reread your messages again and this one worries me also:
Oh, and very worrying, I can hear a funny sort of scratching sound when the Teensy is plugged in, like something is frying or something.
Did you ever measure the USB voltage with an oscilloscope? [assuming the unit is indeed powered over USB]
I wonder if there is some high voltage, high frequency noise present on the USB +5V.

Paul
 
Did you change over to a different programming PC/laptop recently?
Could the USB host port be power-limited and act up at its max USB power?

Paul
 
I reread your messages again and this one worries me also:

Did you ever measure the USB voltage with an oscilloscope? [assuming the unit is indeed powered over USB]
I wonder if there is some high voltage, high frequency noise present on the USB +5V.

Paul
My last scope died awhile back and I haven't needed it...well perhaps until now.

I only heard this on 1 of the 5 defunct units. It was very strange.

Units have been plugged directly into a Mac M3, not using a hub, with different cables tried.

Last unit that died, I plugged it in only, watching the serial port input on the Arduino. It showed up for about 2 minutes, then suddenly disappeared, and subsequent reconnecting did not revive it at all. Holding the button for 15-17 seconds yields no lights, nothing. They are for all intents and purposes, dead.

Yet next to the dead ones are perfectly working models with the same exact PCB board, and other Teensy 4.0 lockable.

These are all the lockable ones. I have not yet tried one without locking it, as I had premade a batch of about 6 boards with Teensys soldered on for units I sell. I didn't think to test the Teensys prior to soldering as I had not encountered anything like this prior. But I will for future boards, test them prior to soldering.

I will do everything I can on my end to follow this through; my business depends on it so I will not be slack. Next step I can take is to desolder a couple of the defunct Teensyes and check them out of circuit. It's hard to get them out even though I have a Hakko desoldering unit, but I think I can do it.

Thanks everyone, and I'm sorry to present this. I'd be more than glad for this to 100% something I'm doing wrong; I just want to solve it, and I'm more than grateful for each of your expertise and time. I'm loathe to post the PCB board pix as this is something I've developed and sold and spent thousands of hours developing, but I can even do that if it would help.

Hoewever, again, I have working boards - same batch same exact PCB - with Teensy 4.0s installed and working/flashing reliably. It must be either a fault in the commercial PCBs, a run of a few "bad" Teensys, or some error I've made but haven't caught yet...

I'll keep posting the troubleshooting steps I'm taking, and the results I'm getting, if you can keep giving me ideas I'd be beyond grateful.

Thanks kindly,

Mike
 
Did you change over to a different programming PC/laptop recently?
Could the USB host port be power-limited and act up at its max USB power?

Paul
Thank you Paul. Yes it is a relatively new laptop (since beginning of this year). How could I check what you are asking?

But wouldn't the regulator I'm using, well, regulate the input? I'm using it literally out of the data sheet example to make sure I am using it correctly...
 
Back
Top