TPS65910's PWR_HOLD setting

Jiansheng Liu

Other Parts Discussed in Thread: TPS65910, AM3715

It is described PWR_HOLD pin of TPS65910 to be tided to VIO in user guide of TPS65910 with OMAP. I am using TP65910 for AM3715CBP. Currently I have PWR_HOLD of TPS65910 connected to VIO directly. However when we implement software controlled power off (by using DEVICE_OFF bit), it was found that the power was not fully turn off, instead it just was turn off for short time and then back to on state, like a reset event, not really power off event. I wonder it was caused by PWR_HOLD status. Should we tide this pin to VIO directly, or through a pull up resistor? Please advise.

over 13 years ago

0 Jiansheng Liu over 13 years ago

Intellectual 310 points

Just checked from the board and found that VIO can't return back to zero after DEV_OFF bit was set, but remain at about 0.8 to 1V for about 3.5mS, and then return back to 1.8V, because PWR_HOLD pin is tided with it directly, it trigger another power-on process automatically. So in the end, the power -off process was not sucessful and end up at power -on state.

Is there anything to be taken cared at this point? Does a pull up resistor on PWR_HOLD make difference? Please some one advise!

Jiansheng

0 Fabian Stel over 13 years ago in reply to Jiansheng Liu

TI__Intellectual 2090 points

Regarding, PWRHOLD has an internal pull down and it should bring back VDDIO to 0. Which is not the case here...
The best way to connect it in your case will be to connect it to NRESPWON. Meaning that PWRHOLD will be maintain automatically at the end of POWERUP seq.

0 Jiansheng Liu over 13 years ago in reply to Fabian Stel

Intellectual 310 points

If we pull PWRHOLD to NRESPWON, it means that any of reset event from AM3715 will trigger a PMIC power off sequence, that is not what we looking for. I will find out the root cause of that VIO can't return to zero, whis was caused by a off board signal, and has nothing to do with the PMIC. So I will keep PWRHOLD pull up to VIO. The difference is that the pull up resistor will be changed to 100K to ensure the voltage on PWRON pin will drop enough to maintain the power off status.

thanks for the help.

0 Fabian Stel over 13 years ago in reply to Jiansheng Liu

TI__Intellectual 2090 points

What was the initial VIO pull up value?
By doing this did you fix your issue?

0 Jiansheng Liu over 13 years ago in reply to Fabian Stel

Intellectual 310 points

it used to be 10K

Power management

Power management forum

TPS65910's PWR_HOLD setting