One Component on Teensy 4.1 gets very hot

[RobotMan]

Hi there,

My teensy 4.1 used to work perfectly but, just now, when I plugged it in to the computer and tried to upload code I got the following error:

No Teensy boards were found on any USB ports of your computer.
Please press the PROGRAM MODE BUTTON on your Teensy to upload your sketch.

I tried pushing the button, restarting my computer, and using a different usb cable. Nothing helped. I then noticed that one of the components on the board gets very hot very quickly when it is plugged in. Does that mean that the board is ruined beyond repair? I don't know which component is which. I am including a picture of which component is getting hot.

[PaulS]

You may want to check out this thread.
Most probably something [bad solderjoint?] is shorting the 3V3 output.

Paul

[RobotMan]

Thank you so much for the quick response. I did not solder anything on the teensy board, and it worked well when I first bought it. How would I find a bad solderjoint? Is it possible to fix it or do I need a new board?

[PaulS]

First I would do a visual inspection using an optical magnifier. Check for loose/missing/shifted/misplaced components. Also check for tiny solderballs/metal particles anywhere that may short signal lines.
Don't forget to check the bottom side as well.
Then using a voltmeter, measure the actual voltage between the 3V3 pin [marked "3V" on the PCB] and GND pin.

Paul

[PaulS]

Is it perhaps possible to send a close-up photo of your board?

Paul

[RobotMan]

[PaulS]

The little wire in the red circle is not supposed to be there:



Paul

[PaulS]

Is the top of the package of the chip in the yellow circle nicely flat? It looks like a little bulge is sticking out. If so, then the regulator might be killed.



Paul

[RobotMan]

Thank you for noticing that. I removed that wire and tried again. However, I am still having the same problem. The same component is getting hot and the computer cannot find the teensy. However, I am not noticing any bulge in component. Do you think it might be dead anyway? Is there a way to fix it?

[PaulS]

Before concluding that the regulator and/or board is killed, you could do 2 more tests:
1. with the USB cable connected measure the voltage between the 3V3 pin [marked "3V" on the PCB] and GND pin. If it's not around 3V3, there is some short between these pins.
2. without the USB cable or other power supply connected, measure the resistance between the 3V3 and GND pin. My Teensy 4.1 measures around 8.3KΩ.

Paul

[PaulS]

Oops, sorry, I mixed up Teensy 3.2 and Teensy 4.1. The chip that is running hot is the MKL02Z32VFG4, the so-called Bootloader chip.
That chip should not get hot as far as I know.
Please forget the what I said and asked above.
Not sure what to do next, perhaps replace the chip? But it would be sensible to get to the cause of running hot first.

Paul

[RobotMan]

Would it make sense for me to mail the chip back to the factory then?

[PaulS]

The problem is that little wire that you removed from the board. That could have been around the board anywhere and may have shorted out some other lines with a failing board as result. Normally this is not covered by warranty. And as you stated in your first post, the board used to work perfectly...
It happened to me long time ago with a non-Teensy board as well - also a little piece of wire/solderblob lying around the board. I just wrote the board off as my own error.

Mailing the chip/board back to PJRC for repair is not economical I assume.

Paul

[RobotMan]

If it is helpful, I just noticed a second component that is getting hot. It is circled on the picture

[PaulS]

That's the DMG2305UX, a P-channel enhancement mode MOSFET that is setup to function as a low voltage drop diode. It carries the 5V USB voltage to the 3V3 regulator. That FET should not get hot at all.
Hmm, I'm afraid the board is end-of-life...

Paul

[PaulStoffregen]

Very unusual for either of those 2 parts to get hot.

Usually when things go wrong, the PTC fuse (small white part near the USB connector and 3V pin) gets hot. The voltage regulator (6 pin part just above the center pin of the USB host pins) also tends to get hot if something is shorting the 3.3V power. Those 2 parts are right next to the 2 parts you indicated as heating up....

[MarkT]

I'm afraid the board is end-of-life...

Paul

To me "end of life" means "no longer manufactured". I'd just say the boards fried!

[ To "end of life" in the industry means to set a schedule for the phasing out of a particular device's
production. Often seen as "That part's been EOL'd". Before that, if you are lucky, you'll see "no longer
recommended for new designs" slapped over datasheets and search results. ]

[PaulS]

I know EoL well, I should have said end-of-its-life... :-)

[ptitneo]

Very unusual for either of those 2 parts to get hot.

Usually when things go wrong, the PTC fuse (small white part near the USB connector and 3V pin) gets hot. The voltage regulator (6 pin part just above the center pin of the USB host pins) also tends to get hot if something is shorting the 3.3V power. Those 2 parts are right next to the 2 parts you indicated as heating up....

Hi there, I believe I'm in this situation and I don't now how it happened / what caused it other than indeed running my circuit (off a battery via 5v pin) a bit longer than usual. The result is the Teensy still works but USB Host is dead. I'm at peace with replacing the Teensy but I'd rather not fry another so any suggestion as what could cause these issues (other than a 3v3 short) would be helpful!

Thanks

[ptitneo]

(PS in my case the 3v3 output reads correctly and resistance between 3v3 and gnd is a little above 8k, but the USB Host 5v output is flat)

[defragster]

Have used USB Host here direct and with powered hub - never on batteries and have not seen this on T_4.0 or T_4.1.

There is a chip onboard that switches the power pin on the USB Host and feeds directly from the connected VIN power (through whatever is in line with that). Perhaps that chip got something done to it from the battery supplied voltage? Until the USB Host code is activated that switch stays off driven by a specific pin on the MCU.

That chip IIRC is the limiting factor on T_4.1 input voltage as it cannot handle something over the 5.5V listed on the card. T_4.0 does not have this chip IIRC correctly that limits startup current from Host.

[ptitneo]

Thank you defragster, that's interesting. In my case this time there is no powered hub, it's direct. Could it be that the draw was too high for too long and that chip failed?

[defragster]

Thank you defragster, that's interesting. In my case this time there is no powered hub, it's direct. Could it be that the draw was too high for too long and that chip failed?

Not sure about the chip and what could stress it beyond over voltage. AFAIK it is a switch that controls current flow on startup to conform to USB spec and prevent VIN from a surge that would brownout the Teensy host device. After that it seems the switch is open - though in normal usage here current is limited by Host USB cable to the Teensy to ~500 mA - perhaps device connected was passing more current from the battery and caused damage?

Devices using lots of current here (multiple devices or HDD's) have been through powered hub given the Teensy itself was always USB powered and current limited.

Once that switch is open there is one powered hub here that will backfeed power and keep the Teensy running even when its USB power is switched off. This may or may not violate the USB spec - but it works and suggests that once the switch is open it can feed current both ways. That isn't meant to suggest a solution there - just observed behavior.

Speculation (unfounded/dangerous?): if the GND is common and the battery voltage is constrained to 5V that power could go direct to the USB Host +5V wire and not using that pin on the Teensy? If that battery voltage is not 5V constrained, then that could be the trouble.

[jmarsh]

When you say "USB Host", are you actually talking about the HOST port on the teensy (an external port that connects to the 5 pins on the top of board, that are not shown present in your photos), or the PCB mounted micro usb port usually used to upload code onto the Teensy?

As Defragster says, the host port has software controlled power that is only turned on when the port is initialized in code.

[ptitneo]

Yes i'm talking about the host port indeed. My problem is that it stopped outputting 5v and a chip is getting very warm. There was no change to the software when that happened (and it was working as expected before). The power source is a USB power bank that outputs 5v and feeds into the 5v input pin of the teensy. The 5v bridge on the teensy's micro usb was also cut off (but that micro USB wasn't plugged at the time of the failure).

Edit: I see the confusion, these aren't my photos I just piggybacked this thread because my issue seemed similar at least at first...