Hi !
I have an issue with my TAS3251. Working on my new 2 boards configuration, suddenly on one board, all board's leds went down. This was surprising as there was only a low volume and PS is 20V (voltage on low side for experiments ; but workbench was a bit of a mess, tight wires... maybe my fault).
Schematics is very similar to TI TAS3251, TPA3251, TPA3255 EVM boards.
The 12 rail is NOK: it starts a about only 3V at powering and then slowly drops to 1V in about 10s. I quickly unplug the PS. TAS3251 FAULT Led is ON (which may be related to the fact that the 12V is NOK):
- 15V is OK
- If I put my finger on the TAS351, it gets hot,
- If I open the jumper after the LM2940 12V regulator, the 12V is OK,
- the 3.3V section looks OK, as my microcontroller get the associated led blink as long as there is sufficient voltage on the 3.3 section (powered from the 12V rail ; about 10s)
I fear an issue in the TAS3251 GVDD circuit with a shortcut to GND.
Did something similar happened to some of you ? Up to now, I had found that this chip was fairly resilient...
Do you have troubleshooting ideas I could experiment, before trying to change the TAS3251?
I could unsolder the 3.3V regulator to decouple the 3.3V section to make sure.
Are there meaningful controls I can do on the unpowered board with a multimeter (resistance/load measurements, continuity checks)?
Help very welcomed !
Best regards,
JMF
I have an issue with my TAS3251. Working on my new 2 boards configuration, suddenly on one board, all board's leds went down. This was surprising as there was only a low volume and PS is 20V (voltage on low side for experiments ; but workbench was a bit of a mess, tight wires... maybe my fault).
Schematics is very similar to TI TAS3251, TPA3251, TPA3255 EVM boards.
The 12 rail is NOK: it starts a about only 3V at powering and then slowly drops to 1V in about 10s. I quickly unplug the PS. TAS3251 FAULT Led is ON (which may be related to the fact that the 12V is NOK):
- 15V is OK
- If I put my finger on the TAS351, it gets hot,
- If I open the jumper after the LM2940 12V regulator, the 12V is OK,
- the 3.3V section looks OK, as my microcontroller get the associated led blink as long as there is sufficient voltage on the 3.3 section (powered from the 12V rail ; about 10s)
I fear an issue in the TAS3251 GVDD circuit with a shortcut to GND.
Did something similar happened to some of you ? Up to now, I had found that this chip was fairly resilient...
Do you have troubleshooting ideas I could experiment, before trying to change the TAS3251?
I could unsolder the 3.3V regulator to decouple the 3.3V section to make sure.
Are there meaningful controls I can do on the unpowered board with a multimeter (resistance/load measurements, continuity checks)?
Help very welcomed !
Best regards,
JMF
Humm tried some continuity checks with the multimeter, board unpowered.
It seems that the multimeter does not ring (anymore) between the 12V rail and GND. But it rings between the 3.3V rail and GND. Could it mean that the 3.3V rail could be the culprit ?
First in the circuit: 3.3V regulator (AMS1117-3.3) and 100uF 6.3V Electrolytic Capacitor.
What are the failure modes of those type of components ? Some particular type of compoment could "leak" current to the ground, showing then a decreasing voltage (in about 10s)?
JMF
It seems that the multimeter does not ring (anymore) between the 12V rail and GND. But it rings between the 3.3V rail and GND. Could it mean that the 3.3V rail could be the culprit ?
First in the circuit: 3.3V regulator (AMS1117-3.3) and 100uF 6.3V Electrolytic Capacitor.
What are the failure modes of those type of components ? Some particular type of compoment could "leak" current to the ground, showing then a decreasing voltage (in about 10s)?
JMF
Tonight I unsoldered the 3.3V reg and the 100 uF cap just behind. I still measure 1.5R resitance between the 3.3V part of the circuit and the ground.
I have seen the uC alive, so should not be the bad component. Low probability that it is the EEPROM...
Unfortunatly, the TAS3251 still seems to be the component to replace...
No more ideas about tests to do to confirm the culprit.
JMF
I have seen the uC alive, so should not be the bad component. Low probability that it is the EEPROM...
Unfortunatly, the TAS3251 still seems to be the component to replace...
No more ideas about tests to do to confirm the culprit.
JMF