Bricked my Teensy 3.2 on the first try

Status
Not open for further replies.

ac2ev

Member
First download to the board (just got it) and I bricked it. Went through the basics, shutting down computer and restarting, trying different cables, different ports, different PCs, out of circuit, holding the button then plugging it in. Plugged in other boards (Adafruit) I have they work fine.

Board is not seen when plugged in. No USB "Ding", nothing. No unknown drivers in device manager. No lights on board, no smoke, no heat.

Files and libraries are attached. I was attempting to use a I2C LCD display on the 2nd I2C port. I modified an exising LiquidCrystal_I2C (drfrobot?) I had found online and used before on an UNO. I modified this one to use i2c_t3 library added the ability to pass the scl and sda pin numbers, and change _I2C to _I2C_T3 (so I knew the difference) and that was it.

I have no idea if it worked or downloaded to the Teensy 3.2 as the board acts completely dead.

What to try next?

Miracle Max: With all dead, well, with all dead there's usually only one thing you can do.
Inigo Montoya: What's that?
Miracle Max: Go through his clothes and look for loose change.
 

Attachments

  • LiquidCrystal_I2C_T3.zip
    5.9 KB · Views: 52
Do you have everything but USB disconnected? Did you solder anything onto the board? Did you check for shorts. Do you have a multi-meter? if so might check while unpluged resistance from VIN to GND, also 3.3v to GND and VIN to 3.3v and see if they look reasonable. Also check when plugged into USB. What voltage are you seeing on VIN? How about on 3.3v.

It can help if you take pictures and post them of top and bottom as sometimes someone will see something. Things like solder bridges. Or we cut the VIN/VUSB etch on the bottom to make safer to run with external power, only to find out when you are on USB only there is no power.

Did you try loading a real simple program, like blink. Again may go through some of the steps you mentioned about maybe compile with verify and when the Teensy program is up try pushing the program button on your T3.2 and see if it programs.

I am assuming Windows as you mentioned Device Manager? Which version?
 
I modified this one to use i2c_t3 library added the ability to pass the scl and sda pin numbers.
What to try next?

Did you bridge two selection pads to make the LCD an i2c instead of SPI? If yes then carefully check if they did not use zero ohm resistors to select the SPI mode which means you have a short from V+ to GND when you connected the LCD to the Teensy

Similar to the selection pad here:
oled1.jpg
 
Last edited:
Do you have everything but USB disconnected? Did you solder anything onto the board? Did you check for shorts. Do you have a multi-meter? if so might check while unpluged resistance from VIN to GND, also 3.3v to GND and VIN to 3.3v and see if they look reasonable. Also check when plugged into USB. What voltage are you seeing on VIN? How about on 3.3v.

It can help if you take pictures and post them of top and bottom as sometimes someone will see something. Things like solder bridges. Or we cut the VIN/VUSB etch on the bottom to make safer to run with external power, only to find out when you are on USB only there is no power.

Did you try loading a real simple program, like blink. Again may go through some of the steps you mentioned about maybe compile with verify and when the Teensy program is up try pushing the program button on your T3.2 and see if it programs.

I am assuming Windows as you mentioned Device Manager? Which version?

I soldered to Pin 29,30,A14(DAC). Right now they are just wires, but will inspect further tonight and desolder them. I'll also check with a meter.
Can't load a program as board is "dead"
Windows 10 Device Manager
 
5v and 3.3v are well within spec. Program pin floats at 0.3v then goes to 0v when you press the reset button. Those wires aren't connected to anything at the moment.

IMG_1687.jpg
IMG_1688.jpg
 
Did you solder those header pins, or was the board purchased with the pins already soldered?

Looking at your photo, there might be some sort of material on the side of the MKL02 chip. Difficult to see what it is. Any chance that could be shorting the pins on that chip?

t3.jpg
 
I don't see anything obvious, maybe others will.

Again I am assuming Windows. Version 10? What version of Arduino do you have installed? What version of Teensyduino do you have installed?

Is this the first time you tried to program it or have you programmed it a few times?

I am assuming you have Arduino installed, with Teensyduino, and have T3.2 chosen as the current board type? And you have USB set to Serial.

Again I would try hold in program button and plug it in. Load real simple sketch like examples->basic->blink.

Click the verify button in Arduino and have it build. This should launch the Teensy tool.

Now I like to have the teensy tool visible on my screen and then try pressing the program button on new teensy and see if the Teensy app does anything. Hopefully it will show now programming... And then when it completes hopefully your system will go into the device and create a Serial port object for this teensy...

But again I am hoping that the hardware is not somehow damaged.
 
@PaulStoffregen
Not a short just a misleading image. Not to say I couldn't accidentally short a connection but I've had professional training and have been soldering for over 20 years.
IMG_1692.jpg

@KurtE
Arduino is 1.8.12
Teensduino is 1.5.1

The fact that none of my computers indicate a device is connected when I plug in the Teensy leads me to believe the device is dead. I'll have to spin up a linux machine and see what happens there.

On the Window10 machine I tried a USB hub and Nirsoft USB Deview, nothing shows up.

I did take a really close look at the main chip and I don't like this pin. Main chip last pin on lower right (not header)
IMG_1694.jpg

I touched it up and re-examined it and it looks fine. I don't see any indication of damaged traces or shorted pins. FWIW, I've had my fair share of board damage when I worked in a lab. We had a control loop go unstable and had a MOSFET blow/burn a hole through a 6-layer board. I rebuilt it and got it working again, though.


I guess at this point, barring a miracle from Linux. I'll have to buy another board. Shame, though it died on the 1st try. I've got a Teensy 2.0 that's still going strong after much abuse in projects.
 
Tried connecting to a couple machines at work with a couple different cables. Nothing.

Decided to buy a new board, cable, and header pins from Paul. Hopefully it's not the code.
 
The Program pin at 0.3V when not pressing the button sounds like hardware damage. It's normally supposed to be 3.3V, due to an internal pullup from the MKL02 chip.

When Program is low, the MKL02 pulls Reset low. So you might try looking at the voltage on the tiny Reset test pad on the bottom side. If it's also low, that's probably a good sign the MKL02 is working, but something external is dragging Program down. If Reset is high while Program is low, or if Reset is low but non-zero voltage and frequency measurement or an oscilloscope shows pulsing (which happens if the MK20 is rapidly watchdog rebooting), that's a sign the MKL02 isn't working.

I can't see anything obviously wrong in these photos. Hopefully this info about the Program->Reset behavior helps, if you want to keep trying to find the cause of trouble.
 
Well that settles it. There's the Reset pin to GND 1v/div at 50uS. My scope is only 20MHz but it was free and works pretty well.
Now the question is how did this happen? it's only been connected to USB so far. I work on a static mat. Perplexing to say the least.

Is there any possible way the code (in original post) has the chip stuck in a loop that I can't get it out of? Someone care to give me a sanity check?

I did order a new one so that should be here soon so that I can continue this project.

IMG_1696.jpg
 
Is there any possible way the code (in original post) has the chip stuck in a loop that I can't get it out of?

No, not really. If you look at the schematic, Program doesn't connect to the main chip at all. It only connects to the MKL02 chip. If Program is stuck low, something had gone wrong with the MKL02 chip which should be pulling it up to 3.3V, or if the MKL02's pullup is working something must be physically holding that signal low.

This one is quite a mystery. Very difficult to say what's really going on here.
 
No, not really. If you look at the schematic, Program doesn't connect to the main chip at all. It only connects to the MKL02 chip. If Program is stuck low, something had gone wrong with the MKL02 chip which should be pulling it up to 3.3V, or if the MKL02's pullup is working something must be physically holding that signal low.

This one is quite a mystery. Very difficult to say what's really going on here.

If you want I'll mail it back to you and you can investigate it. It's been a while since I dug this deep and I'd spend more time reading technical manuals than at the bench.
 
If you want I'll mail it back to you and you can investigate it. It's been a while since I dug this deep and I'd spend more time reading technical manuals than at the bench.

Well this is an interesting turn of events. The Teensy is not going back. When I got home I decided I'd de-solder the header pins and the 3 wires. I also figured I'd give it one last chance and touched up the solder joints on the MKL02 chip and I gave it a nice bath in Isopropyl followed by a a heat gun sauna. I plugged the Teensy into the PC and nothing. So I unplugged it. Wait what was that sound? Could it be?

Fired up the Blink.ino and Teensy loader and plugged it back in. It's alive.

So success is attributed to one of the following:
1. Touching up the MKL02 chip
2. Cleaning it in Isopropyl followed by a heat treatment (Taking to the Spa)
3. Threatening to send it back home

Personally I think it was #3 but who knows. IMG_1700.jpg
 
Status
Not open for further replies.
Back
Top