TPS65219: TPS6521908 PMIC Timing Reversed Issue

susie yuan

Prodigy 40 points

Part Number: TPS65219
Other Parts Discussed in Thread: AM2431, TPS22965, TPS54320

Tool/software:

I have some questions about PMIC timing. PMIC part number is TPS6521908, MPU is AM2431.

Please assist in analyzing the causes and assessing the risks.

Thank you.

1.Power On Sequencing Issue

The sequencing between Buck1 0.85V and MCU OSC0-XO is reversed, We don’t know why it happened. What effect will the reversed timing cause?

Test point is MCU OSC0 XO output pin.

Item	Measured Value	Criterion
Time between Buck1 0.85V and MCU OSC0 XI/XO	-3ms	3ms

The test waveform is as follows, and required timing sequence is used for comparison.

2.Power Down Sequencing Issue

1)The sequencing between Buck1 0.85V and MCU OSC0-XO is reversed, We don’t know why it happened. What effect will the reversed timing cause?

Test point of MCU XI/XO is MCU OSC0 XO output pin.

2)The sequencing between GPO2 and VSYS is reversed, We don’t know why it happened. What effect will the reversed timing cause?

Test point of VSYS is PMIC Power input 6/20/22 pin.

Item	Measured Value	Criterion
Time between NRSTOUT and Buck3 1.1V	0.16ms	>=0ms
Time between Buck3 1.1V and MCU XI/XO	8.5ms	0ms
Time between MCU XI/XO and Buck1 0.85V	-8.4ms	10ms
Time between Buck1 0.85V and Buck2 1.8V	0	>=0ms
Time between Buck2 1.8V and LDO3 1.8V	0	>=0ms
Time between LDO3 1.8V and GPO2	0.04ms	>=0ms
Time between GPO2 and VSYS	-1.1ms	10ms

The test waveform is as follows, and required timing sequence is used for comparison.

Here is the schematic diagram of PMIC section as below. Please help check. Thank you.

Thanks.

over 1 year ago

0 Matt Sunna over 1 year ago

TI__Guru* 81035 points

Hi Susie,

The PMIC timing looks correct, but I agree that there are some differences when looking at the PMIC and MCU timing together with regards to the oscillator startup and shut-down and the VSYS turning off before GPO2. In the block diagram of the application report on this topic (https://www.ti.com/lit/an/slvafk3/slvafk3.pdf ), GPO2 should be controlling the enable pin of the power switch for VSYS, but in your scope capture, it appears that VSYS is somehow being turned off without GPO2 going low. Do you know why this may be happening independently in your system?

Are your connections between the PMIC and MCU the same as the block diagram in the guide?

Best regards,

Matt

0 susie yuan over 1 year ago

Prodigy 40 points

Thank you for replying to me. I design refer to the document . (https://www.ti.com/lit/an/slvafk3/slvafk3.pdf )

1).You said "GPO2 should be controlling the enable pin of the power switch for VSYS" ,I don't understand.

You means the VSYS(3.3V) is the load switch TPS22965 output? It isn't the load switch TPS22965 input power "VSYS"?

I am wondering which Test point VSYS is correct. Please tell me. Thank you.

2).You means the the oscillator startup and shut-down is sequencing correct. The MCU AM2431 can work well. Is it right ?

Currently our MCU can work well and run the program normally.

Look forward to replying to me.

0 Matt Sunna over 1 year ago in reply to susie yuan

TI__Guru* 81035 points

Hi Susie,

So there is something wrong since the VSYS for your system (input to the power switch) is turning off before the GPO2 pin switches off the power switch. So the PMIC power is being cut off before the shutdown sequence is completed. How is VSYS being turned off?

Regarding the oscillator startup, I found this previous thread that helps explain: https://e2e.ti.com/support/processors-group/processors/f/processors-forum/1337400/faq-am623-mcu_osc0_xi-starts-timing-slot-is-not-correct

Regards,

Matt

0 susie yuan over 1 year ago in reply to Matt Sunna

Prodigy 40 points

Hi，Matt

I can test that oscillator startup shortly after 1V8A power on. I think it's correct timing.

But I test the voltage of oscillator MCU-OSC0 XO is low. I don't find the spec of oscillator startup voltage.

I want to know the voltage is 0.9V. Is it within the normal range? Thank you.

0 Matt Sunna over 1 year ago in reply to susie yuan

TI__Guru* 81035 points

Hi Susie,

I can reassign your thread to the processor team since I am only able to answer questions about the PMIC.

Regards,

Matt

0 Nick Saulnier over 1 year ago in reply to Matt Sunna

TI__Guru** 104760 points

Hello,

I have assigned your thread. Please note that this is a holiday week in India. Feel free to ping the thread the week of Nov 4 if you have not received a response.

Regards,

Nick

0 Kallikuppa Sreenivasa over 1 year ago in reply to Nick Saulnier

TI__Guru**** 200294 points

Hi Susie,

Please refer below E2E

(36) AM6442: 25MHz Crystal logic level for HFOSC - Processors forum - Processors - TI E2E support forums

Regards,

Sreenivasa

0 susie yuan over 1 year ago in reply to Matt Sunna

Prodigy 40 points

Hello,

It's still about the problem that the power-off sequence VSYS 3.3V is reversed.

I use DC-DC TPS54320 to produce VSYS 3.3V to input PMIC.

Then I change the Resistor value of Vstart and Vstop. I change Vstop threshold from 6.98V to 3.9V ，but It also fall fast before Vgpo2. The figure is as below. Do you have any ideas to change the fall timing of 3.3V.

Thank you.

0 Kallikuppa Sreenivasa over 1 year ago in reply to susie yuan

TI__Guru**** 200294 points

Hi Susie,

I do not understand how this query is related to oscillator amplitude.

If this is a different issue, please start a new thread and add more details.

Regards,

Sreenivasa

0 susie yuan over 1 year ago in reply to Kallikuppa Sreenivasa

Prodigy 40 points

No,this issue is about "TPS6521908 PMIC Timing Reversed Issue".

It's still about the problem that the power-off sequence VSYS 3.3V is reversed.

I use DC-DC TPS54320 to produce VSYS 3.3V to input PMIC.

Then I change the Resistor value of Vstart and Vstop. I change Vstop threshold from 6.98V to 3.9V ，but It also fall fast before Vgpo2. Do you have any ideas to change the fall timing of 3.3V.

Thank you.

0 susie yuan over 1 year ago in reply to Matt Sunna

Prodigy 40 points

No,this issue is about "TPS6521908 PMIC Timing Reversed Issue".

It's still about the problem that the power-off sequence VSYS 3.3V is reversed.

I use DC-DC TPS54320 to produce VSYS 3.3V to input PMIC.

Then I change the Resistor value of Vstart and Vstop. I change Vstop threshold from 6.98V to 3.9V ，but It also fall fast before Vgpo2. Do you have any ideas to change the fall timing of 3.3V.

Thank you.

0 Kallikuppa Sreenivasa over 1 year ago in reply to susie yuan

TI__Guru**** 200294 points

Hi Susie,

Thank you.

susie yuan said:
Then I change the Resistor value of Vstart and Vstop. I change Vstop threshold from 6.98V to 3.9V ，but It also fall fast before Vgpo2. Do you have any ideas to change the fall timing of 3.3V.

Looks like this is the question regarding TPS54320

Can you confirm this for me to assign to an expert.

here is what the PMIC expert mentioned before

Matt Sunna said:
So there is something wrong since the VSYS for your system (input to the power switch) is turning off before the GPO2 pin switches off the power switch. So the PMIC power is being cut off before the shutdown sequence is completed. How is VSYS being turned off?

Regards,

Sreenivasa

0 Ashwani Goel 11 months ago in reply to Kallikuppa Sreenivasa

TI__Mastermind 27440 points

Hi,

As there is no response for long. I am closing the thread. Feel free to ping, if you want to continue discussion.

Regards

Ashwani

Power management

Power management forum

TPS65219: TPS6521908 PMIC Timing Reversed Issue