TPS35-Q1: TPS35AA38AGADDFRQ1

Monica bagade

Part Number: TPS35-Q1

I am using TPS35AA38AGADDFRQ1 as watchdog reset IC in my design. Below is schematic.

Crst=100nf, trst= 500mS.

Twd= 6.66Sec.

I am expecting reset pulse of 500ms after every 6.66 sec in case software freeze. It is happening properly also as expected.

However, in one of board, IC hold reset pin continuously, and not in pulse form of (low 500ms, high 6 sec).

The IC doesn't have functionality of output latching.

I am not sure, if that is assembly issue or IC behavior.

It started working normally as expected after we removed, clean, reflash and reset.

Could you please let me know what could be the issue?

over 1 year ago

0 Jesse Villanueva over 1 year ago

TI__Mastermind 19010 points

Hi Monica,

You are correct, this device does not have latch functionality. The 2 conditions that will give a consistent LOW RESET output is VDD < Vit and MR = LOW. There is also a possibility that the GPIO that the RESET is connected to is sinking too much current preventing the RESET pin to go HIGH.

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

I have used pull up resistor of 10K, with Vin =3.3V. it is drawing less than 500uA as per suggested by datasheet. I hope this is correct?

0 Jesse Villanueva over 1 year ago in reply to Monica bagade

TI__Mastermind 19010 points

What else is connected to your ESP_RESET net?

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

One pull up resistor of 10k and 1uf capacitor

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

pull up Resistor of 10k and capacitor 1uf

0 Jesse Villanueva over 1 year ago in reply to Monica bagade

TI__Mastermind 19010 points

Hi Monica,

Looking at the schematic provided there should not be any issues. I do have some clarifying questions below.

Monica bagade said:
It started working normally as expected after we removed, clean, reflash and reset.

When you said "reflash" what was reflashed from this schematic?

The Toggle_Switch part, I am assuming it is a simple switch or push button?

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

Yes, it is a simple push button for manual reset purpose. Reflash is for MCU( Sometimes after resetting also WDI pin was full time low, that issue was not related to Watchdog IC, not applicable here ).

Does connecting capacitor at reset pin can make it latch permanently?. It is load capacitor of 1 uf.

0 Jesse Villanueva over 1 year ago in reply to Monica bagade

TI__Mastermind 19010 points

Hi Monica,

The 1uF cap is on the higher side, our typical recommendation is only .1uF at most, but 1uF should not cause an issue. It is possible that the GPIO of the MCU was not setup correctly. For instance, if the GPIO is not set to a high impedance input, it is possible that the GPIO will sink too much current, which will prevent the RESET to stay high.

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

Okay. But, when I am disconnecting watchdog reset from MCU Gpio, it is pulled high. It is getting pulled down from watchdog IC.

Does EMI/EMC can cause this kind of issue? some hight current wires are passing nearby due to which some transients are there on pins? But can it cause permanent Reset without damaging. Because IC is functioning normal after power recycle.

0 Jesse Villanueva over 1 year ago in reply to Monica bagade

TI__Mastermind 19010 points

When the MCU is disconnected to the watchdog and no input to WDI, does the reset pin toggle between high for the timeout duration then toggle low for the reset duration?

If the failure is reproducible, you can measure the impedance between RESET and GND. If the device is functioning properly, you should see the resistance toggle between low impedance, few hundred ohms, to high impedance when WDI has no input.

How much current are passing through the high current wires?

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

No, IC is holding reset pin continuously High, MCU is giving WDI properly. When watchdog IC's supply removed and connected again, it again became functional.

Around 60A current

0 Jesse Villanueva over 1 year ago in reply to Monica bagade

TI__Mastermind 19010 points

Hi Monica,

Monica bagade said:
IC is holding reset pin continuously High, MCU is giving WDI properly

This is an expected behavior.

For the 60A parallel trace, is that on the same layer? I will check with our device if that will be an issue.

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

Sorry for the confusion. I meant to say IC is holding reset pin continuously LOW (not high), MCU is giving WDI properly.

0 Masoud Beheshti61 over 1 year ago in reply to Monica bagade

TI__Mastermind 18071 points

Monica - our team will respond in 2 days, Sorry for the delays

0 Jesse Villanueva over 1 year ago in reply to Masoud Beheshti61

TI__Mastermind 19010 points

Hi Monica,

Do you happen to have a way to add a damping resistor between the ESP_RESET output and the MCU input?

Jesse

0 Monica bagade over 1 year ago in reply to Jesse Villanueva

Prodigy 240 points

yes, possible. What value is preferred?

0 Jesse Villanueva over 1 year ago in reply to Monica bagade

TI__Mastermind 19010 points

A 10K should be fine. If it is a sinking current issue, meaning that the GPIO may not be setup correctly, you will see that the ESP_RESET will pull up to a faction of the pull-up voltage depending.

Jesse

Power management

Power management forum

TPS35-Q1: TPS35AA38AGADDFRQ1