No access to STS1TX following power-up

[Spencer.Andrew]

Now and again, following power-up, we find the STS1TX in our product in a state where we cannot interface with it. The SPI and GPIO lines are outputting ~2V levels, so no commands can be sent.

Typically, cycling power may allow the chip to initialise correctly, whereafter it then works perfectly for the entire power session.

Power application is fast (<45µs to 3.3V) so the internal reset should be working fine. VBat has 1µF, 100nF, 10nF and 270pF decoupling capacitors. Assembly x-rays show no soldering problems.

The SDN pin is connected to ground, so we have not tried to toggle it (There is no spare microcontroller port pin, and we also did not expect that this might be necessary.)

This is a problem that we have encountered over a number of years, with chips coming from varying distributor sources. Generally users have just toggled the power, but we are needing to solve the problem. Any advise would be very much appreciated.

[MDorm.1]

I've seen similar problems, and its doing my head in too

(I've added a switched rail for the chip in one design, but its FAR from ideal)

Does anyone have insight in this ?

[ASpen]

We received the attached PDF on SDN management. I've only been able to test it in the last few days, and the results look good.

We had put aside some ST1TX chips which seemed to consistently malfunction following power-up. We found a spare microcontroller I/O port and fly-wired a 10K pull-up to Vcc on SDN, and connected up the port.. The port is obviously an input following power-up, so some time later (after other microcontroller setups, and definitely not as fast as the Fig7 timings on page 3 imply, we turn the port inot a high output for 50ms, then low, wait the Treset, and get on with things.

It seems to be working.

Since we are waiting a reasonable time after taking SDN low, I am not doing the Poll READY test, or the GPIO0 test, mentioned on page 4.

I hope that you get yours to work too!

[ASpen]

Further note : We did the mod on two STS1TX-equipped boards that we had set aside as being problematic. They had failures to access the STS1TX following power-up almost every single time. With the SDN control from the micro, we got 100% success with many power cycles on both boards.