Can't upload to previously working Teensy 3.6

Status
Not open for further replies.

rbushnell

Member
Hello!

I'm using a Teensy 3.6 to read some analog and digital values, and have run into an odd problem. When a particular external sensor has been uninstalled/reinstalled, the Teensy has started to display analog read values that are significantly noisier than usual. This persists until the Teensy is power-cycled, at which point it resumes normal operation. However, after this event has occurred, new code cannot be uploaded to the Teensy - the bootloader (via Arduino IDE) asks that the user press the program button, which has no effect.

I've had the same symptoms occur with multiple devices. It seems pretty obvious that this is due to something involving the external sensor.

What's baffling me, though, is that the sensor in question is isolated from the Teensy. It uses an isolated power supply and communicates via an optically-isolated interface that's incapable of over-voltaging the line.

My question, then - does anyone know what would produce those symptoms in the Teensy? I'm hoping to identify the potential failure modes in the Teensy, and to work backwards from there.

Please let me know if you need any additional details - I'm limited in what I can provide, but I'll do my best. I'm uploading using an unpowered USB cable, while the system provides the Teensy with its own 5 V supply.

Thank you!
 
Something sounds odd electronically with cable or wiring affecting the Teensy ?

When healthy - a button push takes the Teensy to Program Mode and shows up on USB for coding instructions.

Have you installed the released TeensyDuino 1.42? It has an updated interface shown on the Tools / Ports menu item where a Teensy specific section will show a connected Teensy when present in Bootloader or other USB state when it is running.

I put this post together some time back when my computer or something was having some issues at times and these steps worked to get back to normal operation.
forum.pjrc.com teensy-1st-setup-problem
 
I think I'm back on TeensyDuino 1.40. It's certainly possible that there's an intermittent fault in the cable, but I've been using the same one - for the initial programming of each Teensy, and then later for the failed uploads.

Thank you for the suggestions! I'll update TeensyDuino and see if I can try it out over the weekend.
 
Also note if you push the reset button for exactly 15 seconds, the teensy bootloader completely erases the program on the Teensy 3.5/3.6. I recall the time is actually something like 13-17 seconds, but the idea is it wants to guard against short or very long presses. At present, I don't think the 3.2 bootloader has this feature. I'm not sure about the LC.
 
As I note in that linked post … good cables do go bad … cost me a couple hours one night. 'Assuming' the cable was 'still' fine I was on a remote machine and went to a lot of trouble reinstalling stuff and other things … then tried another cable and it just worked.

With the new TD 1.42 the Tools menu Ports item will clearly show if there is a Teensy detected - if not with 'running USB' then it will appear as 'Bootloader' with a button push. If that is not there then the USB port is offline/confused or the cable is bad … if the Teensy is functional.
 
I was directly connected to the computer; no USB hub involved. I'll give the v1.42 software a try and post results, but it may not be until next week.

Thanks, all!
 
When you get back to it - if this doesn't show success - perhaps a picture of the Teensy and notes on the 5V supply. If possible remove the 5V supply and jumper the VUSB back to the 5V Teensy pin with a Powered data programming cable?
 
When you get back to it - if this doesn't show success - perhaps a picture of the Teensy and notes on the 5V supply. If possible remove the 5V supply and jumper the VUSB back to the 5V Teensy pin with a Powered data programming cable?

I can't post a photo right now, but there weren't any observable abnormalities under a 10x lens. I'm using the outer rows of pins, and the inner pins A10, A11, Gnd, and Reset. The 5 V supply is a TDK-Lambda CCE ... 2405, I think. ( https://us.tdk-lambda.com/ftp/Specs/cc-e.pdf ).

I'll try a powered cable on the bench as soon as I have access to the Teensy again. Thanks!
 
I was able to check things out last week and think that everything's okay now. It looks like the Teensy's were only unprogrammable when they were in-system after a particular situation occurred, for reasons that I don't completely understand. Once removed from the system, the Teensy's could be programmed normally. Apparently the periodic heavy rain in that area caused a non-destructive short between two otherwise isolated parts of the system, which eventually led to the Teensy's behavior.

Thank you all for your help!
 
Status
Not open for further replies.
Back
Top