Another Broken T3.6?

Status
Not open for further replies.

WMXZ

Well-known member
Scenario
I have a T3.6 connected to an IMU and custom Audio board.
Running on battery (DC-DC converted to 6V on Vin and Vin-USB cut)

System is running for some time now (on demand power will be connected)
It is programmed for MTP-Disk (i.e seremu)

Now, 'suddenly', during field test yesterday, USB is not responsive anymore (no device manager action, no linux hid ports)

initially power consumption was same as always
after 15 s press of button power consumption dropped significantly (it seems only ADC board and IMU where on power)

So I measured some electrics (with the power supply that I used to develop all system)
Vin OK with 6 V,
V3.3 OK with 3.3 V
Prog OK with 3.3 V
Reset line 0.003 V (with and without USB connection) (LA confirms regular pulses)

It seems that K66 does not come out of reset.
So I assume K66 is broken.

before replacing whole system (T3.6 is soldered to adapter board) i would like to get a feeling how this could happen

Hardware was not touched (it is in a pressure housing, that is usually not opened)

However, I made a SW change:
Until yesterday, code had while(!Serial); in setup, requiring USB connection and Serial terminal running, before continuing with setup and loop.

So, the new idea was to let system start as soon power is applied, consequently I commented the while(!Serial); line , but I let all Serial.print statements, relaying on the hope when USB buffer is full, nothing will happen. At the end of test, terminal will be connected and system stopped, file downloaded, etc.

I did some tests (with dedicated battery) to calibrate IMU, downloaded files via MTP and provided system for field test.

While SW was working also after SW changes, I noted, however, some difficulties to connect Teensy to system, but power toggling always solved the connection problem.

Now system in permanently in Reset mode,
So real question is: is this a sign that K66 is 'broken'?
If yes, could it be done in user SW, (say by printing to USB without USB enumerated (no USB connected))?
or is it more likely that a physical stress (heat or shock in field test) disconnected MCU?
 
Last edited:
When using USB host - you don't want over 5.5V on VIN. Check the USB Host chip power switch specs. They are typically limited at 5.5V on the high side - because they are switching directly to USB devices - and they are just switches AFAIK - unless the 6V VIN is regulated down - the only Teensy regulation is to 3.3V - so VIN at 6V would be over USB spec to feed devices as well.

This was observed/posted sometime back after the T_3.6 shipped.
 
When using USB host - you don't want over 5.5V on VIN. Check the USB Host chip power switch specs. They are typically limited at 5.5V on the high side - because they are switching directly to USB devices - and they are just switches AFAIK - unless the 6V VIN is regulated down - the only Teensy regulation is to 3.3V - so VIN at 6V would be over USB spec to feed devices as well.

This was observed/posted sometime back after the T_3.6 shipped.

Not using USB host. Teensy is USB device (MTP not MSC protocol for data transfer)
AFAIK the regulator for USB2 (TPD3S014) is rated max 6 V, so it should not break down
 
Did you try a 15 Restore to take the code away?

Yes,
before this consumption was 'normal', after 15s press consumption was lower, indicating to me that program was not running anymore, obviously I cannot verify
did also look into power-up (0 to 3.3V needs <1 ms, which should be good)
 
Yes,
before this consumption was 'normal', after 15s press consumption was lower, indicating to me that program was not running anymore, obviously I cannot verify
did also look into power-up (0 to 3.3V needs <1 ms, which should be good)

… I see that was noted 'after 15 s press of button power consumption dropped significantly (it seems only ADC board and IMU where on power)'

Bummer … and puzzling …
 
Status
Not open for further replies.
Back
Top