What are the common reasons for Teensy 4.1 flash memory chip failure?

L

lincolnwang

Member
Hi,

We have been using Teensy 4.1 for over 2 years for a robotic project without major issues. But recently (last 4 months) we have started seeing more Teensy 4.1 having flash memory chip failure (4 quick blinks in red LED) after running for a short amount of time. The problem will get fixed by swapping to another Teensy. So I would like to know what are the common reasons for this failure so we can further investigate our system.

Thanks

Lincoln
 
This is the posted data for DIY bootloader chip ...
1772764368956.png
 
lincolnwang said:
We have been using Teensy 4.1 for over 2 years for a robotic project without major issues. But recently (last 4 months) we have started seeing more Teensy 4.1 having flash memory chip failure (4 quick blinks in red LED) after running for a short amount of time. The problem will get fixed by swapping to another Teensy. So I would like to know what are the common reasons for this failure so we can further investigate our system.
Click to expand...

I don't know of any "common" reasons for Teensy 4.1 flash failure. Are you using the flash for data storage? Is the Teensy 4.1 subject to any shock loads?
 
defragster said:
This is the posted data for DIY bootloader chip ...
View attachment 38933
Click to expand...
Thanks for the details. What do you think could be the reason for the increasing number of units having this issue? Also, I don’t think we see this issue recurring in the same system once we swap the Teensy.

In our system:
  • We use the Teensy to send PWM signals to motor drivers and read encoder signals from the motors.
  • All signals are strictly 3.3V logic.
  • We have TVS diodes under the motor connectors.

Do you recommend soldering a new 24 MHz oscillator onto the bad Teensy for further investigation?
 
joepasquariello said:
I don't know of any "common" reasons for Teensy 4.1 flash failure. Are you using the flash for data storage? Is the Teensy 4.1 subject to any shock loads?
Click to expand...
Thanks for the reply. As @defragster and @jmarsh point out, it is actually ARM JTAG or SWD failure.

In our system:
  • We use the Teensy to send PWM signals to motor drivers and read encoder signals from the motors.
  • All signals are strictly 3.3V logic.
  • We have TVS diodes under the motor connectors.

Do you recommend soldering a new 24 MHz oscillator onto the bad Teensy for further investigation?
 
Hey guys,

I reflowed the oscillators on three Teensy boards that previously showed the ARM JTAG failure, without replacing any parts, and they all started working again. I am now able to upload firmware, power cycle each board about 20 times, and control the motors without any issues.

Based on this, can I assume this might be related to a PCB assembly issue in the recent Teensy batches? I haven't seen such a high failure rate over the past two years.
 
Metal cans like crystals/oscillators absorb radiated IR heat less well than epoxy components, so I suspect the reflow profile wasn't quite right, leaving some un-flowed solder paste under them.
 
I had my first out of bag failure a couple of weeks ago. Crystal not working. Inspecting under a microscope did not show any obvious solder issues or contamination around the crystal.

I used flux and reflowed the solder with a soldering iron and then cleaned the board and that fixed it same as you.

I think MarkT is correct that their reflow soldering profile is/was not quite right on the new mfg line. Hopefully that is resolved on newer builds.
 
In the shipment that I received, there are a number of slightly different batch numbers . I don't have the bag from the exact failed unit, but all the batch numbers appear to be around 179878, 17881, 17882
 
You must log in or register to reply here.
Back
Top