TMP392: After POR, OUTA/B

Hi Hiroshi-San,

Thank you for your post.

Does the customer only see this behavior when starting up with the DC Power supply, and never with the onboard power? Are these tests being performed at room temperature? 

Can you share a bit more of the schematic, specifically I would like to see what OUTA and OUTB are connected to, and what the pullup values are. 

Also, if possible, can they probe the SETB pin during one of these events?

Best Regards,
Brandon Fisher

Hi Brandon-san,

Thank you for your reply.

LMR23630 is used as the on-board power supply.
In this case, both OUTA and B frequently output low pulses (only after POR).

As a result of removing "L1", "R40", and "C27" and using an external power supply to isolate the problem, only OUTB output a low pulse.

・ 8 x TMP392
・ 8 x OUT A / B is wired OR connection & 10k ohm Pull up

Therefore, which TMP392 is outputting the pulse? It's difficult to judge.
Also, since it is not possible to monitor 8 SET Bs at the same time, the customer has started to modify the circuit.

The results will be posted at a later date.

Best regards,

Hiroshi

Thank you Hiroshi-San,

I would suggest checking the tolerance values of the used resistors (1% at room and 100ppm/C over the temperature range) and then taking a measurement of them with an Ohmmeter or digital multi-meter to verify they are compliant. 

This type of behavior is possible if the resistor values are not being properly decoded at POR. This can happen if the set resistors are outside the 1% tolerance, or if there is significant board noise at startup (please see this thread for reference). 

Best Regards,
Brandon Fisher

Hi Brandon-san,

The result of the customer's test using EVM is attached.

TMP392 EVM .xlsx

It seems that the noise caused by the power supply is the cause, but customers feel that the noise immunity of this product is weak.

However, I think it is a very problem that it does not work properly in the environment of the attached file [Test 2].

If a stronger noise filter is inserted, the probability of starting failure will decrease, but it seems that it will occur depending on the environment even with the configuration of [Test 3]. I don't know when and when the output goes low due to noise, not just at startup.

[Test 1]-[Test 6] All of them should start normally, but why does it malfunction due to noise?
Or how to use it? Please let me know if you have any countermeasures.

Best  regards,

Hiroshi

Hi Hiroshi-San,

Thank you for sharing the updated data. I am checking with our design team to see if they have any other possible causes. At first glance, it does seem to me like the TMP392 is having trouble decoding the resistor values at startup when under their power supply conditions.

I will update you here as soon as I have some feedback from our designers on this. 

Best Regards,
Brandon Fisher

Brandon-san,

Thank you for your support.

We look forward to hearing back from you.

Best Regards,

Hiroshi