ADC32RF42: spi mode 3 setting not responding

J P

Part Number: ADC32RF42

Good afternoon,

We are trying to set our ADC into 3-wire mode but I only see data on the 4th wire. No matter if we write 0x00 0x10 0x01 (3 wire mode) or 0x00 0x10 0x00 (4 wire mode). I see this on multiple ADCs. Do we need to do anything after the 3 wire mode set? Soft reset? something else? Any ideas on why this is not working? We write and read from other registers to confirm ADC SPI functionality, and you can see from the diagrams that the response is on the 4th wire.

1st register write: reset and 3 wire mode

register write: value written 0x1F

register read: reads value of 0x1F on 4th wire

over 5 years ago

0 Vijayendra Varma Siddamsetty over 4 years ago

TI__Genius 10865 points

Hi Julia,

I see the same on my ADC32RF42EVM. Setting the register 0x10 to 0x01 is not switching SPI mode from 4-wire to 3-wire.

We have been using 4 wire mode on EVM and 3 wire mode might not have been verified. I'm checking with the design team regarding this issue. Please give me until Monday to get back.

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

Thank you for the quick reply. I will await your confirmation on Monday.

Julia

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

SDO output might not gated in 3-wire mode (I'm still waiting for design team to confirm this). On the EVM, as it is configured for 4-wire mode, SDIO is forced to low from the FTDI (USB IC) during SPI read. So it's not easy to verify this on EVM.

Are you sure that FPGA is not forcing SDIO pin when SDIO is supposed to be input? Can you probe the pin direclty?

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

I confirmed with the FPGA engineer that the line is not being held low.

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

I'm trying to test this on EVM. But it needs some hardware changes. I will get back ASAP before Friday.

Is it possible for you probe SDIO line directly with an oscilloscope?

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

Good morning,

I can probe the SDIO line and never see a data value on an ADC response. I see the line toggle when writting to the ADC

0 J P over 4 years ago in reply to J P

Intellectual 370 points

We did a test with an older board from a few years ago and a new board that was just built a few months ago. We are wondering if there are differences in the die.

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

Are you seeing that 3 wire mode is working on an old device but not new one?

I will double check on device versions. But I highly doubt anything related to SPI has changed over versions.

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

I can take a screenshot of the output in our protocol analyzer if that helps.

Yes an old device from years ago acts differently than the one using a new TI chip

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

It will be helpful if you can probe SDIO in 3-wire mode in both boards with oscilloscope and send scope screenshots.

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

rev b 3 wire

rev b 4 wire

rev c 3 wire

rev c 4 wire

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

Can you please read register 0x04 on both old EVMs? Please let me know the read value. With this we can find out if the Device version is different.

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

Good morning,

The Rev B board of ours has a readout of 0x80

The Rev C board of ours has a readout of 0x50

Thank you!

Julia

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

Device on RevB board (with readout: 0x80) is the latest and supported version.

Device on RevC board (with readout: 0x50) is very old version. How and when did you receive this device?

Is it possible to possible for you to read and share the markings on both the devices?

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

Device on RevB board (with readout: 0x80) az32rf42 ti 714 ztf2 g4

Device on RevC board (with readout: 0x50) xaz32rf4x x2651c ti 5b4 zje5 g4

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

Thanks for the quick reply.

I don't think device with readout: 0x50 is supported by the datasheet. I will double check with the design team and let you know about this by Thursday.

Can you also let me know how and when did you receive this device?

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

Thank you for your quick reply. I will get back to you with an answer. We look forward to hearing from you on Thursday.

Julia

0 J P over 4 years ago in reply to J P

Intellectual 370 points

We received the parts from an Assembly House in Colorado about 1 month ago.

Thank you!

Julia

0 Vijayendra Varma Siddamsetty over 4 years ago in reply to J P

TI__Genius 10865 points

Hi Julia,

I just got confirmation that device marking starting with 'xaz' is pre_RTM version and cannot be supported.

We recommend that replace device with post-RTM version device (marking should start with 'az').

Regards,

Vijay

0 J P over 4 years ago in reply to Vijayendra Varma Siddamsetty

Intellectual 370 points

Thank you very much for the confirmation!! We will replace accordingly.

Have a great day,

Julia

Data converters

Data converters forum

ADC32RF42: spi mode 3 setting not responding