PCF2131 SPI Interface issues

[ctmaker]

Hello,

I am interfacing PCF2131 with an MCU utilizing the SPI interface. On a brand new RTC, I find I can read the time and the time increments correctly.  As soon as I write to register 0x02 (Control_3) to enable battery switch-over, the RTC stops responding. Every register I read back is read as zero. Even a power cycle or writing default values to all control registers does not make a difference. How do I "unfreeze" the device?

Thank you.

[Ryan_Le]

Hi @ctmaker ,

Have you find out the root cause of this issue?

I have a similar problem with this PCF2131 and would know if you were able to fix it?

Thank you

[ctmaker]

@Ryan_Le , no, never found a solution, destroyed more than a dozen units. I gave up on the chip and redesigned using a different chip. This chip is a dud. Who would have thought RTCs would be so difficult.

BTW, NXP needs to change the 2FA timeout to 30 minutes or 1 hour. It takes 6-7 tries before I get the code in time. Or, they can upgrade the servers to make them more functional?

[markus-k]

Same here, had to switch to another chip. The PCF2131 simply does not work correctly and I don’t understand how NXP can sell such devices. I verified voltages and voltage ramps multiple times, no chance. Now using a PCF85263a, which I’ve been using successfully in previous designs.

[ctmaker]

Hello @TomasVaverka ,

Please see attached schematic. utilizing PCF2131. VDD is set to 3.30V, confirmed with measurements. BBS pin is floating. VBAT was tested set to both 0V (no backup cell) and 3.20V (backup cell in holder).

At startup, register 0x07 and 0x08 are read (sec and min reigsters). They worked initially when I had not tried to write to any register. I tried to write 0x20 to register 0x02 and ever since I did that, the RTC stopped responding with anything other than 0 for any register read. I tried this on two units I have and both are now non-responsive. I can try to write other values if you can advise anything that may recover the RTC.

Best Regards,

Jigney.

[TomasVaverka]

Hello Jigney,

Could you try to remove the coin cell battery and force VBAT to GND?

Then, without installing the coin cell again, power up the part (without VBAT) and try the same register write (0x20 to register 02h). Does it still fail after this write?

Best regards,

Tomas

[ctmaker]

Hi @TomasVaverka ,

I tried the following:

Disconnected main supply (3.3V), remove coin cell, ensure VBAT is at same potential as circuit GND and then just try reading 0x07 with no coin-cell. I got nothing back. I then tried to write 0x20 to 0x02 and still nothing back by read 0x07. It is almost like the IC is dead.

Does the IC have any EEPROM or flash that would save the settings and would persist over power cycles without backup power?

I had only two samples and unfortunately, they are both stuck in this state after write. I'll get some more units from my CM and try just reading again. 

Best Regards,

Jigney.

[markus-k]

I have pretty much the same issue, but on an I2C bus. The RTC just stops responding (no ACK) randomly after removing Vdd. It only starts responsing again after removing Vdd and Vbat and then powering it back up. The selected power management mode is direct switching with no low battery detection (but it happens in any mode, except disabled with the supercap removed).

I have attached a screenshot of my schematic. The 3V0 rail is backed by a 1F supercap.

Did you come to any solution? @TomasVaverka 

[TomasVaverka]

Hello Jigney,

I have not heard about such an issue before. In order to further investigate it, I need to know:

1. What is the value being written to Control_3 register?
2. Are there more register writes happening? If yes, please list them all with values.
3. What is the voltage measured at VDD, VBAT and BBS pins just before the RTC freezes?
4. Can you share the schematic?

BR, Tomas

[ctmaker]

and @TomasVaverka , please see attached SPI trace as well, where I am trying to read register 0x07. One would think that if I disconnect from all power, the IC should reset to defaults, but it is puzzling that once I write, I am unable to read, ever. Don't know if there is something else that I may be missing.

[ctmaker]

Hello Tomas,

Thank you for the response. I am out of the office for a week. I will measure and provide you with a response next week on Friday.

Thank you again for the help.

Best Regards.