Teensy 4.1 likely dead - any suggestions

M

mind2023

Member
I fear I've likely damaged my Teensy 4.1 and I'm wondering if there are any steps you'd recommend to diagnose/repair it, or if I should just buy a replacement?

Here are some details:

I'm using a Teensy 4.1 (https://www.pjrc.com/store/teensy41.html) in the implementation of the MicroDexed-Touch (MDT) project (https://codeberg.org/positionhigh/MicroDexed-touch). As described on the project page, the Teensy is also providing power to an external PCM5102 (https://www.amazon.com/dp/B07Q9K5MT8?psc=1&ref=ppx_yo2ov_dt_b_product_details), an external 128 Mbit SPI Flash memory (https://www.amazon.com/dp/B0BD7G5HRG?psc=1&ref=ppx_yo2ov_dt_b_product_details), and a HiLetgo ILI9341 display (https://www.amazon.com/dp/B073R7BH1B?psc=1&ref=ppx_yo2ov_dt_b_product_details). Also, the MDT PCB board wires a USB socket to the Teensy's USB Host connection, which powers an Alesis VI25 MIDI keyboard (https://www.alesis.com/products/legacy/vi25) from this USB connector.

I downloaded the MDT source code from its project page and compiled it using Arduino IDE 2.0.3 with the Teensy V 1.57.2 board manager installed, on a Windows 10 PC. At various stages of development, I've had this project implemented on a breadboard and on the MDT-specified PCB, and working as expected in both cases. The Teensy was sometimes powered via USB from the computer, and sometimes via USB from an Anker PowerCore III Sense 10k (https://www.anker.com/products/a1231-1?variant=37438359961750) via the Anker's USB-A port. I operated the MDT for a total of probably 2 hours total over the course of a month, without any issues.

Recently, I installed the project into its case. I used it for about 15 minutes without trouble. At this point, the audio output went dead, the screen went blank and it would not respond to inputs from the two encoders. I powered it down and rebooted. The Teensy LED did not come on. Both the PCM5102 board and the Flash board had one brief red blink. The screen was white, as when it first failed. I rebooted many times, with the same result each time. When connected to the computer, the IDE failed to detect the device. I pressed the reset button for over 15 seconds, with no change in behavior.

I measured Vin at 4.9V, but the 3.3V pins on the Teensy only measured about 0.4V. I tested all adjacent solder connections for the project and didn't find any shorts.

One question I have is if it is likely that powering the project from the Anker portable charger could have damaged the Teensy, maybe by providing too much current? Or is it more likely that something else went wrong in the project software or in my building of the hardware?

Thanks for any inisghts!
 
The 3V3 regulator on the Teensy is short-circuit protected so it should not fail easily.
Can you isolate [or take out] the Teensy from the rest of the circuit/PCB and then test it again?

Paul
 
Looking at the schematic of the MDT board, the only (external) connections on the Teensy 3V3 output are the PCM5102 board and the SPI Flash.
Can you take those boards out and check whether the 3V3 returned to normal?

Paul
 
I just had a chance to pull the Teensy off the PCB and test it by itself on a breadboard. Unfortunately, there wasn't any change in behavior. The 3.3V pins still measure about 0.4V.
 
Hmm, that is too bad.
I suppose you already did a close visual examination of the whole PCB?
Is the 3V3 regulator [U4] or any other part getting hot?

Location of U4:
T41, U4.png

Paul
 
Here are a few photos. I notice that the U2 component has a bit of a cracked, rough appearance. I don't know if this is normal or not.
teensy-01.jpg
teensy-02.jpg
 
Yeah, U2 looks a bit weird [it's the bootloader chip]. Don't know what to think about it...
But I also noticed that the regulator housing misses a little chip in the bottom-right corner:

T41, U4-2.png

Paul
 
That does look like a chip on the regulator. I wonder if it's melted rather than a chip? I could have gotten the soldering iron too close to the regulator when I soldered the pins for the USB Host? I wish I had a better magnifier.

So at this point, I'm thinking I should buy a replacement and try again, unless you think there is anything else I could try?
 
Oh, and I just noticed your first reply from today. I did look the board over and didn't notice anything odd offhand. And nothing gets noticeably warm when its plugged in.
 
Did you write that you (occasionally) pushed +5V into the USB host connection? With nothing else providing Teensy power? That’s quite likely a good way to kill the
https://www.ti.com/product/TPD3S014 chip. And more… That 5V Host USB is a power out, not a power in option.

Apart from that, note i had a series of Teensy41 die because i had a too high capacitive load on the 3V3 output pin, plus a low resistive load on the 5v input. When the 5v disappears, suddenly, say when unplugging the Teensy micro USB, then with near zero volts left on the 5v line, the LDO regulator dies due to reverse current from it 3v3 output to its 5v input. The (new 8 pin) LDO has no adequate protection against backcurrent… That’s why i now always add a diode from 3v3 pin (a) to 5v pin (k) on designs that use Teensy41 modules.
 
For the USB host, it was the other way around. I powered my MIDI keyboard from the USB host on the Teensy.

That's great info about protecting from potential backcurrent. What does "(a)" and "(k)" refer to?
 
PaulS said:
Yeah, U2 looks a bit weirdl
Click to expand...

I can confirm that is the normal appearance for U2 (the new chip we had to start using last year due to shortages on the original U2 part). Yeah, it does look a little weird.


mind2023 said:
That does look like a chip on the regulator. I wonder if it's melted rather than a chip? I could have gotten the soldering iron too close to the regulator when I soldered the pins for the USB Host? I wish I had a better magnifier.

So at this point, I'm thinking I should buy a replacement and try again, unless you think there is anything else I could try?
Click to expand...

I'd probably just replace the whole board.

But if you want to try salvage by replacing U4, just order anything from PJRC (even just 1 header pins) and in the "additional instructions" field on the checkout page, ask for an extra NCV8186AMN330TAG voltage regulator. This part is out of stock everywhere, but I have some leftovers loose in an anti-static bag. Would be happy to send you a couple at no extra charge.
 
Or just for diagnostic sake, without any other power, I'd try applying 3.3V from a current limited power supply. If you get the same 0.4V, then desolder U4. If you continue to see 0.4V without U4, that's a pretty sure sign some other chip is internally damaged and replacing U4 probably won't help. But if the 0.4V "short" goes away with removal of U4, odds are much better you'll manage to salvage the board.

Don't expect Teensy 4.1 to boot up with just 3.3V external power. The main chip doesn't start up unless SNVS power is applied first. The main idea of external 3.3V is to check whether the 0.4V short goes away with removal of U4 (or any other part you try to replace).
 
PaulStoffregen said:
Or just for diagnostic sake, without any other power, I'd try applying 3.3V from a current limited power supply.
Click to expand...

I don't have a nice benchtop power supply, so I'd have to breadboard some 3.3V supply. Also, I don't really have the right tools to remove the U4 chip. But a buddy of mine does. So the next time I go over there, I'll try to run the diagnostic, just for curiosity's sake. It might be a few weeks before I get the chance.

By the way, since you mention a current limited power supply ... Did you happen to see my comment that I was powering this project from an Anker PowerCore III Sense 10k, and I was wondering if that could cause any trouble? Do you think there is any possible issue in powering the Teensy from this battery pack?
 
It is kind of off topic but since it is the same project, and there is some response going on, i try my luck. After the change of the voltage regulator and other changes ? during corona peak, we got reports that sd card access is only working after flashing the teensy and running our application without powerdown from usb / pc power supply. after a reboot, from same power source or other, access to sd card seems broken. Since i have not found similar issues in pjrc forum, is this somehow a known issue ? we use only default 5v in, from teensy micro usb.
 
mind2023 said:
I was powering this project from an Anker PowerCore III Sense 10k, and I was wondering if that could cause any trouble? Do you think there is any possible issue in powering the Teensy from this battery pack?
Click to expand...

I don't know anything about that product. Quick Google search landed here:

https://www.anker.com/products/a1231-1

Looks like it's just 5V USB power. Should work fine. Or at least shouldn't damage anything, but those USB battery packs do have a reputation for automatically turning off when they don't see a large load like a phone trying to recharge.

That page does have some specs.

USB-C Input (PD): DC 5V=3A, 9V=2A, 15V=1.2A

USB-C Output (PD): DC 5V=3A, 9V=2A, 15V=1.2A

USB Output : DC 5V=2.4A

Total Power Output: 18W
Click to expand...

The USB-C output is capable of higher than 5 volts, but my understanding is that would only happen if plugged into another USB-C device with power delivery negotiation.

I'm guessing "USB Output" means the USB A port. Looks like it's only capable of 5 volts. Seems like it ought to work fine.
 
mind2023 said:
For the USB host, it was the other way around. I powered my MIDI keyboard from the USB host on the Teensy.

That's great info about protecting from potential backcurrent. What does "(a)" and "(k)" refer to?
Click to expand...

anode and kathode/cathode.
 
You must log in or register to reply here.
Back
Top