PF3000 fails to start

[tetsuomaeda]

Hi,

We are using PF3000 as the power source to iMX.6 CPU.
In the high temperature environment, PF3000 sometimes fails to start.
At the first power ON, PF3000 always start normally.
We turn off the system, then we start it again, PF3000 sometimes fails to start.

Attached is the output behavior of PF3000.

Upper photo shows the behavior of the normal operation and the second photo shows the behavior when PF3000 fails to start.

When PF3000 start normally, VSNVS goes up to around 3V, but when it fails to start, VSNVS stays around 1V and  does not move up.

We are struggling to solve the problem.

I'd appreciate your advice.

Regards,

Tetsuo Maeda

[wenhuachen]

Hi,

One question, if the PF3000 failed to start, except the VSNVS is 1V, is there any output on the other regulators? If not, you can check if VCOREDIG is 1.3/1.5V and VCORE is 2.775V during  PF3000 failed start. The phenomenon VSNVS is 1V, VCOREDIG and VCORE output normally and no outputs on the other regulators means the OTP is not loaded. You can check VDDOTP(pin), it should be keep connecting to ground (0V) all the time.

And one more thing, check the totally loads on VSNVS, make sure it is not greater than 35uA during OFF mode/1mA during normally mode.

Thanks.

[tetsuomaeda]

Wenhua-san,

Thank you for the advice.

My colleague is studying it and we would get back to you in a few days.

Regards,

Tetsuo Maeda

[b36835]

HI Tetsuo,

Is the board made of i.MX 6UL + PF3000? What is about the test ambient high temperature?

PF3000 VSNVS can provide 1mA current with 3.0V output voltage, its limit is as below:

Symbol        Parameter                                   Min. Typ. Max.    Unit

ISNVSLIM   Current limit, VIN > VTH1             1100 – 6750       μA

1) Need to check the schematic of PF3000 VSNVS connectiion, check any loadings to it, includes any pulling up connections, may its rush current at start-up moment need to be captured.

2) Supply an external 3.0V or 3.3V to replace the VSNVS output, check whether it can be improved. 

3) Supply an external 3.0V or 3.3V to LICELL (pin #36) to check whether it is improved in this scenario.

Thanks,

Sean Liu

PMIC team

[tetsuomaeda]

Zhi-san,

Thank you for the advice.

We have checked the total load current and confirmed that it is within the specifications.

Initially the total load capacitor of VSNVS was more than 1uF. We have reduced it to around 0.1uF and there have been considerable improvement. 

We would like to check other advice and would be able to respond in a few days.

Regards,

Tesuo Maeda

[tetsuomaeda]

Hi,

Considering your advice, we have tried the following.
1. According to the reference design, main input of PF3000 becomes ON after VSNVS reach around 2.8V. So VSNVS has to go up to 3V before other outputs become active.
2. We have supplied external 3.0V to VSNVS, then the system works properly. So far we have not found side effects (problems).

We have two questions.
a. Our system doesn't use a battery cell (no back-up). Do we need to use VSNVS? As far as we have checked the behavior of the system, we have found no issues when we used an external 3V source instead of VSNVS output.
b. We think we need to clarify the reason why VSNVS stays around 1V at 80 degC environment. We would appreciate information regarding this phenomena.

We would like to use an external 3V source instead of VSNVS to solve the problem.

We would appreciate your comment about this solution.

Regards,

Tetsuo Maeda

[art]

Q. Our system doesn't use a battery cell (no back-up). Do we need to use VSNVS?

A. Do you mean the VSNVS output of PMIC? If so, the answer is: no, it is not strictly necessarily. For that case, you can simply connect together the VDD_SNVS_IN and VDD_HIGH_IN power rails on the processor side (this is the recommended solution for no backup case).

Q. We think we need to clarify the reason why VSNVS stays around 1V at 80 degC environment.

A. Actually, the maximum ambient operating temperature for the Consumer grade PF3000 parts is specified as +85C. So, in some near-limit conditions the part can be overheated locally (i.e. due to internal power dissipation during operation) and, thus, exceed its maximum specified operating temperature. So, do you use the Consumer, Automotive or Industrial grade parts? Please specify the exact PF3000 part number to make me able to define that.

Have a great day,
Artur

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[tetsuomaeda]

HI,

Very sorry for my long silence.
Thank you very much for the usage of VSNVS.
We have decided not to use internal VSNVS LDO.
We have done the test production and there seems no problems.

We are using "industrial grade" devices.

I think that the temperature of the chip does not exceed the specified value.

The parts code is "MC34PF3000A4EP".

Thanks again for your advice.

Regards,

Tetsuo