ADS1259: wrong CONFIG0 data

Andrey Basov1

Prodigy 70 points

Part Number: ADS1259

Hello.

Project details:

STM32L4 MCU

ADS1259 using Internal oscillator

SPI seems to be working. I am able to read registers but CONFIG0 contains 0x25(00100101) instead of expected 10XX0101.

If i write 00000100 to CONFIG0 i read back 00100100.

All other registers contains valid data.

Varying SPI baud rate in range from 625kbit/s to 5mbit/s does not affect problem.

Received data corresponds to DOUT waveform on oscilloscope.

HAL_Delay(1000);//wait 1000ms
HAL_GPIO_WritePin(START_GPIO_Port, START_Pin, GPIO_PIN_RESET);//START pin - low
HAL_GPIO_WritePin(RST_GPIO_Port, RST_Pin, GPIO_PIN_SET);//RESET pin - high
HAL_GPIO_WritePin(SPI1_CS_GPIO_Port, SPI1_CS_Pin, GPIO_PIN_SET);//CS pin - high
HAL_Delay(100);//wait 100ms
HAL_GPIO_WritePin(SPI1_CS_GPIO_Port, SPI1_CS_Pin, GPIO_PIN_RESET);//CS pin - low
HAL_Delay(100);//wait 100ms

TxBuf[0]=0x11;
HAL_SPI_Transmit(&hspi1, (uint8_t*)TxBuf, 1, 5000);//transmit SDATAC command
TxBuf[0]=0x40+0x00;
TxBuf[1]=0x00;
TxBuf[2]=0b00000100;
HAL_SPI_Transmit(&hspi1, (uint8_t*)TxBuf, 3, 5000);//write 1 byte at address 0x00
TxBuf[0]=0x20+0x00;
TxBuf[1]=0x08;
HAL_SPI_Transmit(&hspi1, (uint8_t*)TxBuf, 2, 5000);//RREG 9 bytes starting from 0x00 
HAL_SPI_Receive(&hspi1, (uint8_t*)RxBuf, 9, 5000);//read 9 registers

Thank you in advance.

over 3 years ago

0 Bryan Lizon86 over 3 years ago

TI__Mastermind 35111 points

Hi Andrey,

Are you able to read any registers back correctly?

You mentioned an oscilloscope capture in your initial thread, can you post that here?

I will look into this some more in the meantime.

-Bryan

0 Andrey Basov1 over 3 years ago in reply to Bryan Lizon86

Prodigy 70 points

Hello Bryan,

Yes, i can read registers back.

Here is a screenshot with results of code posted above:

Oscilloscope capture of DOUT transmitting CONFIG0(it was captured with different code without writing to registers, read by MCU as 100101):

0 Bryan Lizon86 over 3 years ago in reply to Andrey Basov1

TI__Mastermind 35111 points

Hi Andrey,

It is interesting that the only bit that seems to be off is bit 7 in CONFIG0, which should be a 1 but is also read-only so the actual value really doesn't matter.

Is there any way the initial timing could be off such that the first bit is somehow missed after the RREG command? Do you toggle CS at all? Is there any other traffic on the SPI lines (intentional or not) that could be messing with the digital communication? Any other transactions occurring in your MCU such that the first bit is missed or the SCLKs are initially misaligned?

-Bryan

0 Andrey Basov1 over 3 years ago in reply to Bryan Lizon86

Prodigy 70 points

Hello Bryan,

ADS1259 is only device on this SPI interface. I will try to investigate SPI with logic analyzer soon. CS is controlled by MCU and set to low during communication.

I tried to read CONFIG1 at first and then CONFIG0 with same results - CONFIG0 is corrupted .

I can start conversions and monitor DRDY with MCU and oscilloscope. It works fine and change period if I modify data rate bits.

I saw threads with very similar symptoms(0x25 in CONFIG0) but cant find the answer:

https://e2e.ti.com/support/data-converters/f/73/p/653360/2424850#2424850

https://e2e.ti.com/support/data-converters/f/73/p/827561/3063460

Here is my schematic:

0 Bryan Lizon86 over 3 years ago in reply to Andrey Basov1

TI__Mastermind 35111 points

Hi Andrey,

Thanks for the additional details. I would be interested to see what the logic analyzer shows.

In one of the posts you linked to, the issue was resolved as the customer had noise between his ground planes that was causing issues (I don't think there was a resolution for the other post). Fortunately, your schematic implies one solid ground plane so it's unlikely that this is causing the issue, unless there are grounding issues between the ADC and MCU.

I will keep looking into this issue offline and see if I can come up with any other things to try.

-Bryan

0 Andrey Basov1 over 3 years ago in reply to Bryan Lizon86

Prodigy 70 points

Hello Bryan,

I recorded SPI and signal lines during initialization:

larger images:

First DRDY falling edge occurs 100ms after START goes high:

Capture file(Saleae Logic):

ADS1259_init_0x25.zip

0 Bryan Lizon86 over 3 years ago in reply to Andrey Basov1

TI__Mastermind 35111 points

Thanks Andrey,

Everything looks good on the Saleae.

I am following up offline with our back-end teams to see if there is any issue with this bit. I will follow up with you after I hear back from them.

In the meantime, can you confirm that you are able to read data correctly? E.g. you can put a 1V signal in and get the appropriate code at the output? I just want to make sure that everything else is working properly and that the only issue is this single bit.

Thanks!

-Bryan

0 Andrey Basov1 over 3 years ago in reply to Bryan Lizon86

Prodigy 70 points

Hello Bryan,

I have made several unsuccessful attempts to read the data(continuous and pulse mode). So I decided to fix CONFIG0 readback problem first.

I will also try to capture ADC data reading.

0 Bryan Lizon86 over 3 years ago in reply to Andrey Basov1

TI__Mastermind 35111 points

Thanks Andrey, it appears that the ADC is reading back CONFIG0 correctly other than bit 7. So if you are able to read data successfully then at least we have isolated the potential issue. If not, then something else might be wrong that might be worth solving first, and that also might illuminate the cause of the original issue.

Have you tried using other ADS1259 devices? Do they all share the same issue(s)?

-Bryan

0 Bryan Lizon86 over 3 years ago in reply to Bryan Lizon86

TI__Mastermind 35111 points

HI Andrey,

Also, if you can share the topside markings on your device(s) that would help us check the lot data

-Bryan

0 Andrey Basov1 over 3 years ago in reply to Bryan Lizon86

Prodigy 70 points

Hello Bryan,

I ordered additional ICs but haven't received them yet.

IC marking:

0 Bryan Lizon86 over 3 years ago in reply to Andrey Basov1

TI__Mastermind 35111 points

Thanks Andrey, we will look into it

-Bryan

0 Bryan Lizon86 over 3 years ago in reply to Bryan Lizon86

TI__Mastermind 35111 points

Hi Andrey,

I checked internally with our test team, and they confirmed that bit 7 in the CONFIG0 register should be a 0 for the ADS1259-Q1. The 1 only applies to the ADS1259 (non -Q1 version), but was not updated when the ADS1259-Q1 released. You can in fact see that the majority of the ADS1259-Q1 datasheet is just linked back to the ADS1259.

We will update this accordingly in a future datasheet revision of the ADS1259-Q1. Thanks for bringing this issue to our attention.

If you have further questions about the ADS1259, please feel free to start a new thread and I can assist you there. This way it will be easier for engineers who have a similar problem in the future to be able to find an answer.

-Bryan

0 Andrey Basov1 over 3 years ago in reply to Bryan Lizon86

Prodigy 70 points

Hello Bryan,

Thank you for your help!

0 Bryan Lizon86 over 3 years ago in reply to Andrey Basov1

TI__Mastermind 35111 points

No problem, thanks for your patience.

-Bryan

Data converters

Data converters forum

ADS1259: wrong CONFIG0 data