Problem with ADuM6401

Hello!

I have a problem with ADuM6401. I use this part for isolation AD7793. I have it on SPI bus with other parts (LCD, EEPROM, another AD7793 (non-isolated), DAC).


And now about my problem. When ADuM6401 is connected to bus, there is a problem with MISO wire. Sometimes data goes, sometimes not, and it is for all parts, which use MISO wire (EEPROM, non-isolated AD7793), other parts (LCD,...) which don't use MISO wire are OK. I looked at this by osciloscope and sometimes there is indeterminate state (not zero, not 3.3V).

I tried slow down communication speed, but the problem is still there. Than I tried connect two AD7793 on breadboard, one was isolated and the another wasn't. The problem was still there, but when i set right input filter for DC/DC converter, the problem was gone.
But when I get it back on the PCB with other parts, the problem is still there, and I have input filter, which is described in datasheet. I tried change its parametrs, but no change on result.
I use AVR MCU.

Do you have any ideas?

Thanks for your answers. 
And sorry for my english.

Parents
  • Hello MsCantrell,

    Thank you for your answer and sorry for my late response.

    Yes, I have three devices on MISO line (EEPROM, AD7793, ADuM6401+AD7793). I have there two AD7793. One is isolated with ADuM6401 and the another one is non-isolated. I have problem with master side bus segment of ADuM6401.

    To eliminate the problem with Slave select. I gave EEPROM and ADuM6401+AD7793 away. On the bus was only non-isolated AD7793, nothing else, and it worked fine.


    But when I put ADuM6401 back on the board (parallel to non-isolated AD7793), the problem was back. There was only AD7793 and ADuM6401 (with nothing on slave side) on the bus. I looked with an oscilloscope on MISO line and I could see that data went from AD7793 to MCU, but high level of logic state was 1V and not 3.3V.

    I interrupted MISO wire which is connected to VOD ADuM6401 and high level of logic state is on correct value 3.3V.
    It seems, that VOD takes high level of logic state on MISO wire down to 1V. And it is why another devices can not communicate with SPI bus.

    VDD1 is 3.07V and VISO is 3.3V, it seems be good.

    Thanks for your answers. 
    And sorry for my english.

Reply
  • Hello MsCantrell,

    Thank you for your answer and sorry for my late response.

    Yes, I have three devices on MISO line (EEPROM, AD7793, ADuM6401+AD7793). I have there two AD7793. One is isolated with ADuM6401 and the another one is non-isolated. I have problem with master side bus segment of ADuM6401.

    To eliminate the problem with Slave select. I gave EEPROM and ADuM6401+AD7793 away. On the bus was only non-isolated AD7793, nothing else, and it worked fine.


    But when I put ADuM6401 back on the board (parallel to non-isolated AD7793), the problem was back. There was only AD7793 and ADuM6401 (with nothing on slave side) on the bus. I looked with an oscilloscope on MISO line and I could see that data went from AD7793 to MCU, but high level of logic state was 1V and not 3.3V.

    I interrupted MISO wire which is connected to VOD ADuM6401 and high level of logic state is on correct value 3.3V.
    It seems, that VOD takes high level of logic state on MISO wire down to 1V. And it is why another devices can not communicate with SPI bus.

    VDD1 is 3.07V and VISO is 3.3V, it seems be good.

    Thanks for your answers. 
    And sorry for my english.

Children
No Data