TPS70933EVM-110: TPS70933EVM-110

Jan Placek

Prodigy 40 points

Part Number: TPS70933EVM-110

Hello

I have a question about how to proceed with the following problem with part 1174210 (TPS70933)

The claim number for invoice 60089874 with item 595-TPS70933QDBVRQ1 is s / o 689005. (Mouser Electronics INc)

Description of a fault:

Device switch to fail state (Critical Error 93), if the temperature of product is near to -30°C.

Working temperature of device is -30°C/+50°C

The LDO N1 doesn’t generate a correct reference voltage p3V3_REF.

Failure – at the temperature near to -30°C the output voltage of N1 is about 3.15V instead of 3.3V. The voltage tolerances are 3.3V+/-1%. The output voltage is out of tolerance.
It causes that the ADC Selftest in MCU fails.

3 Conclusion
the defect part 1174210 (TPS70933) on position N1 was detected.

Part details:
Internal part number: 1174210
Manufacturer: Texas Instruments
Manufacture part number: TPS70933
3x on PWRBoard (N1, N2, N13)

Please, whot will be nex step?

over 4 years ago

0 Nick Butts over 4 years ago

TI__Guru 50025 points

Hi Jan,

To evaluate this failure I will need more information. Can you please provide the following information:

Input voltage
Load current
Scope shots showing the failure during steady-state operation. Please include Vin, Vout, Iout

If you are interested in a return, you can start the process here:

https://www.ti.com/support-quality/additional-information/customer-returns.html

Note that for returns you will inevitably need much of the information I have requested from you as well.

Regards,

Nick

0 Jan Placek over 4 years ago in reply to Nick Butts

Prodigy 40 points

HI Nick

I am sending an answer to your questions

0 Nick Butts over 4 years ago in reply to Jan Placek

TI__Guru 50025 points

Hi Jan,

It looks to me that you are operating the device within specification and the device output should be within specification as well. Please give me a couple business days to do some testing here.

Regards,

Nick

0 Grzegorz Bartosz over 3 years ago in reply to Nick Butts

Prodigy 120 points

Hi Nick,

Could you provide any new to this case?

Thank you

Best Regards

Grzegorz

0 Jan Placek over 3 years ago in reply to Grzegorz Bartosz

Prodigy 40 points

Hello

I do not understand the question?

what do you think new ...

I described the mistake and we have noK pieces with us.

Please allow us to send it for your analysis. You will certainly confirm our assumptions.

0 Grzegorz Bartosz over 3 years ago in reply to Jan Placek

Prodigy 120 points

Hello Jan,

The question is for Mr Nick, not to you and it is chasing message.

Have a nice day

Best Regards

Grzegorz

0 Nick Butts over 3 years ago in reply to Grzegorz Bartosz

TI__Guru 50025 points

Hello,

I am very sorry for the delay. I tested on the EVM down to -30C and I did not see the deviation that you are seeing. I measured VOUT = 3.31V at room temperature and VOUT = 3.29V at -30C.

I do not see an obvious reason for what you are seeing. Is there any way you can send us a sample board so we can verify the behavior?

Regards,

Nick

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hello

Our claim is based on 29 nOK pieces that have the same problem.

Yes, I can only send you a defective part. Will that be enough for you? The part was desoldered from the board by our technician. How many pieces do you need to send?

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hello

Our claim is based on 29 nOK pieces that have the same problem.

Yes, I can only send you a defective part. Will that be enough for you? The part was desoldered from the board by our technician. How many pieces do you need to send?

0 Nick Butts over 3 years ago in reply to Jan Placek

TI__Guru 50025 points

Hello,

If you can send 5-10 pieces that should be enough.

Please send them to the following address (and please include the ATTN):

251 S Williams Blvd

Tucson, AZ 85711

ATTN Nicholas Butts

Regards,

Nick

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hello

Ok ... I can prepare max 4 nok pcs of component

I'm preparing to send it today ...

0 Nick Butts over 3 years ago in reply to Jan Placek

TI__Guru 50025 points

Hello,

4 will be enough. Thank you.

Regards,

Nick

0 Nick Butts over 3 years ago in reply to Nick Butts

TI__Guru 50025 points

Hi Jan,

I have received the parts. Please allow 2-3 business days to evaluate them.

Regards,

Nick

0 Nick Butts over 3 years ago in reply to Jan Placek

TI__Guru 50025 points

Hi Jan,

I tested all 4 parts with Vin = 5V and Iout = 5mA and I did not find that any of them are out of specification. See the table below. I did a final test on the part labeled 7425 with Iout = 100mA (because this causes additional load regulation deviation) at -30C and I measured Vout = 3.277V. This is again within specification.

Package label	Vout @ 25C (V)	Vout @ -30C (V)
7236	3.306	3.282
7205	3.309	3.296
7770	3.310	3.302
7425	3.300	3.289

I think the way forward is to troubleshoot your system because I do not think the LDOs themselves are causing the inaccuracy. Can you tell me the setup you used to take these measurements?

Can you also tell me what else the LDO is supplying?

Regards,

Nick

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hi Nick

I have discussed your conclusions with two techniques and these are the conclusions.

image from oscilloscope, output voltage from component N1.

The component is defective at start-up when the signal is not "immediately" 3,3V but has a lower voltage level (from the image - 3,19V) and after time the voltage goes up to 3,3V.
The time extension in which voltage goes from lower voltage to a correct voltage correlates with the temperature. With a lower temperature longer the output voltage stays on the lower voltage level (from the image - 3,19V). Because of this time extension, it's necessary to use an oscilloscope and measure immediately after start-up.

The current output 5 mA can cause an internal warm-up in the component and if measurement is not performed immediately after start-up, the measurement can be misleading

0 Nick Butts over 3 years ago in reply to Jan Placek

TI__Guru 50025 points

Hi Jan,

How would you like to proceed? Is this delay in the startup tolerable?

Regards,

Nick

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hi Nick

I have any feedback to your question..

image of output voltage

If the temperature in the component falls below a certain temperature, the output voltage is low. approx. 3.13V (voltage from the image). This bad voltage is at the output as long as the temperature in the component is below this certain temperature. If the temperature in the component ABOVES this critical temperature due to the self-heating of the component, the output voltage jumps to the correct value, 3.3V.

P.S.: The first image from the oscilloscope was just to show that the output of the component has two voltage levels. In the appendix there is another picture from the oscilloscope, where it can be seen that even if the time axis is two seconds, the output is still 3.13V.

So answer to your question Is this delay in the startup tolerable?
No, the this is not tolerable

0 Nick Butts over 3 years ago in reply to Jan Placek

TI__Guru 50025 points

Hi Jan,

In that case would you like to start the return process?

Regards,

Nick

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hi Nick

no .... the process is already claimed (

CPR211057182

CPR211057183

and the pieces were sent based on a discussion on the forum.

We are waiting for an opinion from TI, including sending an 8D reprot. Can you speed up our complaint process, please?

0 Nick Butts over 3 years ago in reply to Jan Placek

TI__Guru 50025 points

Hi Jan,

I am looking into this. I will update you when I have more information.

Regards,

Nick

0 Jan Placek over 3 years ago in reply to Nick Butts

Prodigy 40 points

Hi Nick

It will be great. Thanks a lot.

Power management

Power management forum

TPS70933EVM-110: TPS70933EVM-110