Teensy 4.0 Suddenly Stops Working?

Status
Not open for further replies.

Epartsman

Member
Hello,

So I had a few hours with my first Teensy board and then all of the sudden it seemed to stop working. I had just gotten about half way through the third tutorial (Hello World via serial to the monitor) on the website when I noticed I had no LED activity on the board.

I then remembered that you have to press the button to get the board to boot and load code. I pressed the button several times and nothing.

So I started looking at the board and found that the end of the board where the USB connects was extremely hot.

I unplugged the board and left the board cool down. I re-applied power and checked to see if I had 3.3V at the end of the board opposite the USB connector (per the enclosed pinout card). I did not. And, again, the USB end of the board became too hot to touch.

I also tried to reset the board back to 'factory' code, but there was no indication that the code was written or saved. The board remained inoperable. And hot.

A search of the forums seemed to indicate that this happens from time to time.

Can anyone offer other troubleshooting suggestions?

I really need to get this board back up and running to keep on schedule with my project.

Thank you in advance for your time,
 
Do you have any pictures of your board or code you were running when it happened? What connections you have to it?
 
I don't currently have a photo of the board, but it looks just like the board photos shown throughout the tutorials. There is no charring on the board to photograph.

If you still feel that a photo would be helpful, I'll gladly post one.

The last code that I ran on the board was from the tutorial page, about mid-way down the page the "Hello World" found here:

https://www.pjrc.com/teensy/tutorial3.html

Followed by attempts to reset to factory.. that link I will have to find again, I don't have it handy.

There were no external connections to the board other than the USB cable.
 
Correction..

I was preparing to run the code below, and had loaded it but had not yet made a connection to my I2C device. I was going to run this code with a logic analyzer connected to see that the I2C port was wiggling pins.. but never got that far due to noticing the LEDs were not illuminated..


Code:
#include "Wire.h"

void setup() {
  // put your setup code here, to run once:
  Serial.begin(115200);
  while(!Serial){}

  Serial.println("Start I2C Scan...");
  Serial.println("\r\n");

  byte count = 0;

  Wire.begin();

  for(byte i = 8; i < 120; i++)
  {
    Wire.beginTransmission(i);
    if(Wire.endTransmission()== 0)
    {
      Serial.print("Found I2C Device:  ");
      Serial.print(" (0x");
      Serial.print(i, HEX);
      Serial.println(")");
      count++;
      delay(1);
    }

    Serial.print("\r\n");
    Serial.println("Finished I2C scan ");
    Serial.print("Found  ");
    Serial.print(count, HEX);
    Serial.println(" Device(s).");
  }

}

void loop() {
  // put your main code here, to run repeatedly:

}
 
What you desribe, the hot spot near the USB connector, looks like the TLV75733P is overheating. This can be due to several issues, usually a simple shortcut somewhere.
Try to measure thr resistance between 3.3V pin and GND. It has to be quite high. If it is low, there is a short. Find it and eliminate it.

Are you sure you did not overclock the Teensy 4.0 without cooling?
Are you sure you did not connect more than 3.3V to any pin except the one labeled Vin? This Vin can take up to 5.5V. Many powerbanks and some powered USB hubs deliver 5.6V, which is definitely out-of-specs, but in China everything is different. Also, did you use a passive USB-C to USB adapter? This could lead to an overvoltage problem, killing the Teensy.
Are you sure there are no metal particles, carbon dust, gallium/liquid metal thermal conductor or solder beads hangin around on the Teensy 4.0 board?
Has this board been exposed to extreme temperatures or aggressive substances, causing the growth of whiskers?

Did you try to recover the board by holding the pushbutton for more than 15 seconds after applying power?
 
Last edited:
I did note that the hot spot was located in the corner of the board where the hard-wired jumper is for external / battery power (as noted on the card). I presume that this is where the regulator you mentioned is located.

- Measurement of the resistance between the 3.3V and GND pads located at the end of the board opposite the USB connector reads 6.1 ohms on my Fluke DMM. Is this the proper place to check the 3.3 V voltage?

- If I overclocked the part, I have no idea how I would have done that. This is my first-ever Teensy. I didn't even know you could overclock the board.

- There were no connections made to the board.

- Power was provided from my PC (and HP production model) and the connection was made using a generic USB Type A to Type B micro cable that I've used for many other development boards. I don't have any support for USB Type C connections.

- While my work area isn't surgically clean, it doesn't contain anything that I would expect to cause a short on the board. I've looked at the board under a microscope and I don't see any suspect shorts. I had our production folks inspect the board under x-ray and they found nothing odd.

- I had tried, and tried again before hitting reply, the 15 second push-button reset. No change in behavior.

- Unless the board was exposed during shipment to unusual environments, it has not been exposed after receipt.

Thanks for all the suggestions.
 
6.1Ohms is the DC impedance of copper traces on the board. So there is a short circuit, probably on the USB side area of the board, regarding you measured on the opposite side. Note: Vbat does only need to be connected to a RTC backup battery. It is not intended for running the board on battery. See documentation.

Please post high resolution photographs or scans (on a flatbed scanner) of the board, both sides.
 
Photos as requested. I hope the resolution is sufficient. The originals were too large to upload.

20200309_141903A.jpg20200309_141841A.jpg
 
I can not see any obvious short circuit. On pin 18 and 19 is visible the board had been contacted, also on the button-side edge the 3.3V and GND. edit: After inspecting the photos, I see that the 5V has been contacted as well.
My guess (!) is, the CPU has been fried with a signal higher than 3.3V on pins 18/19 of the Teensy 4.0. This would lead to latch in the output driver circuit, causing a proper short between supply rails with low impedance. The destruction is very likely to be permanent.

An other guess is, the 3.3V rail has been overloaded by drawing too much current to supply an external circuit. This could lead to thermal desctruction of the TLV75733P or the DMG2305UX, causing short to ground. This damage could be rapaired.
 
Last edited:
- Power was provided from my PC (and HP production model) and the connection was made using a generic USB Type A to Type B micro cable that I've used for many other development boards.

Sorry, only to understand: Do you mean you plug in usb cable from PC AND provided power from HP production mode (whatever that is)?
 
@radias:

Yes, I had attached clips to 18, 19 and GND in anticipation of connecting the logic analyzer at those points.

At the point of failure, there was nothing attached to those clips.

I also probed many of these pin locations you note with the DMM in an attempt to figure out what was going on with the board.

Again, no external circuit.

Looks like I got a dud of a board. Unfortunately.

A disappointing introduction to Teensy for me.


@WMXZ: The USB connection (and the only electrical connection) to the board was via a generic USB A to Micro-B cable that I use with other development kits. The source of the power was not a stand-alone supply, or other mysterious source, but it was a standard mass-produced desktop PC. Basically saying that the source of the power to the board was the bog-standard desktop PC USB port. I hope this helps your understanding.
 
No.. it always has a reason.
And it is always wrong handling.
You connected the wrong pins, accidentally, that's all.
A board does never die when you handle it correctly.
 
Just a couple quick things
1. did you hold the button down for 15 sec while powered and if so did you see the red light come on? If you did, did you release the button at that point so the chip could reflash itself to" factory" config.
2. Didn't see any headers attached to your board, did you have anything connected to the board at the time it died. Just want to be sure. You didn't have any other power connected did you?
 
It's likely, that a 5V i2c device was connected.
Why connect a logic analyzer to i2c when there is no other device.
 
@Frank:

I take issue with your absoluteness...

Agreed, there's always a reason.

Perhaps, statistically, troubled boards tend to have 'wrong handling' as the root of their malfunction. But, I disagree that it is _always_ wrong handling. Parts fail. It happens.

Other posts (here and elsewhere) indicate that occasionally there are Teensy boards that have had issues out of the box. Statistically, it happens.

My experience as an engineer is that one can never say never.. there are times when the peculiar and unexpected happens and we may never be able to discover what that event have been.


@Others:
This being my first ever exposure to Teensy I realized that it could be 'operator error'. I came here seeking help with why my particular Teensy appeared to go belly up.

I had hoped that it was something simple that I was missing and that I could get back on track quickly.

The helpfulness of others here confirmed my own trouble-shooting efforts, and, gave me a couple of additional things to check.

I'm thankful for that assistance and confirmation.

At this point it appears that the board is dead and I'm out the cost of the part and shipping. Disappointing, but, so be it.
 
@ mjs:

1 - I tried the 15 second button press. There was no LED activity when I did the button press.

2 - I had intended to connect a logic analyzer to confirm that I was wiggling the pins on proper port. It was not clear to me which port would be active. There was no other power connected to the board. The board was only powered by the USB connection to my bog-standard HP desktop PC.
 
@Epartsman

What concerns me is that the board was working when you originally hooked it up. I have about 6 of the t4's and I managed to fry one myself by putting 5v into one of the pins.

Did want to make sure you followed the reset procedure.


EDIT: what do you mean by wiggling the pins? You mean the clips?
 
I've tried the reset procedure several times. I held the button for 15-16 seconds.. as I read that holding it longer would not cause the reset to take place. I must have tried it at least two dozen times by now.

I don't even have 5V in any of my target circuits. The closest 5V to all of this is the USB connection.

I'm simply perplexed.
 
I've tried the reset procedure several times. I held the button for 15-16 seconds.. as I read that holding it longer would not cause the reset to take place. I must have tried it at least two dozen times by now.

I don't even have 5V in any of my target circuits. The closest 5V to all of this is the USB connection.

I'm simply perplexed.

Ok had to make sure.
 
Sorry,

By wiggling, I meant electrically 'wiggling' the pins.

When I ran my code, I expected to see the pins changing states.. my next step was to run the code, and use the logic analyzer to see what the pins were doing.

Sorry, wiggling pins is 'technical slang' that my colleagues and I have adopted.
 
OK - since maybe you have nothing to loose. One of the things I did was put 5v directly on the VIN pin and GND to gnd between the teensy and power supply. Tried a 15 sec reset. Sometimes it worked sometimes it didn't. Let it sit and then try to plug into the PC but with a different usb cable.

Oh - is it still getting hot when you plug it in with a USB cable?
 
I can try what you suggested with a different power source. I'm not sure if it will make a difference.

Yes, still gets very hot when plugged into USB.
 
Status
Not open for further replies.
Back
Top