ADS7953: the data of ADS7953 is sometimes incorrect , please help ...

51622465stm32F405RG-ADS79XX-MannulMode.rar

Edmond,

Welcome to the TI E2E forum.

To help us get to a root cause, could you please provide the following details:

1. Could you clarify what you mean by “some data is not correct”? Are you occasionally getting bad data or are some of the bits incorrect? Could you share a log from the DebugPrint for 20 captures?
2. What is the sampling rate during this experiment? Does the problem go away at lower sampling rates?
3. What is the input range and ADC_HS1 voltage? Is the resistor divider on the REFP required? Ideally this should be replaced by a reference IC?
4. Could you please share a screenshot of the SPI interface signals (probed near the host controller) during a data capture frame?

 Thank you.

Regards,

Sandeep

Hi, SanDeep,

Thankful for your concern,

I updated my answers for your questions, please review it and if having more questions, please help to let me know them in early time,

and I would like to describe our schematic, we tried to read the data of each channels of 8 ADS7953 through SPI line,

currently, when trying to experiment : some channel connect to 3 V or 0 V directly, 

I anticipated to read 0xfff or 0x000, but I checked only 0xfff and strange values like 0x1fff so on.... I can't check 0x000, 

after that, I checked the communication line of SPI, and found some problem : MISO line is strange, the waveform is not normal like the following image,

please help to check it,

ref) we control cs pin of each ADS7953 with 2 gpios like NSS pin and each ADC_SEL pin with OR gate, 

thankful for your support in advance,

1. the data can be presented in 0 ~ 4095(12 bit resolution), when checking the data, I got strange data like 

if using below,

change state STIMSTATE_RAMPUP -> STIMSTATE_STIMULATING, 5
ADS7953 CH0 data is LS1: 0x089f
ADS7953 CH0 data is LS2: 0x0899
ADS7953 CH1 data is LS2: 0x0898
ADS7953 CH0 data is LS1: 0x08a1
ADS7953 CH0 data is LS2: 0x0899
ADS7953 CH1 data is LS2: 0x1895
change state invalid state -> STIMSTATE_STIMULATING, 6
ADS7953 CH0 data is LS1: 0x08a1
ADS7953 CH0 data is LS2: 0x0897
ADS7953 CH1 data is LS2: 0x1896
ADS7953 CH0 data is LS1: 0x089e
ADS7953 CH0 data is LS2: 0x0897
ADS7953 CH1 data is LS2: 0x1896
ADS7953 CH0 data is LS1: 0x089f
ADS7953 CH0 data is LS2: 0x0898
ADS7953 CH1 data is LS2: 0x1895
change state invalid state -> STIMSTATE_STIMULATING, 6
ADS7953 CH0 data is LS1: 0x08a0
ADS7953 CH0 data is LS2: 0x0899
ADS7953 CH1 data is LS2: 0x1895

I used the code as follows, 

memset(SPI2_RX0, 0, sizeof(SPI2_RX0));
SPI2_LS1_ManulMode(spi2_cmd0,SPI2_RX0);
Hal_DebugPrint("ADS7953 CH0 data is ");
Hal_DebugPrint("LS1: 0x%02x%02x ",SPI2_RX0[1],SPI2_RX0[0]);
Hal_DebugPrint("\r\n");

memset(SPI2_RX0, 0, sizeof(SPI2_RX0));
SPI2_LS2_ManulMode(spi2_cmd0,SPI2_RX0);
Hal_DebugPrint("ADS7953 CH0 data is ");
Hal_DebugPrint("LS2: 0x%02x%02x ",SPI2_RX0[1],SPI2_RX0[0]);
Hal_DebugPrint("\r\n");

memset(SPI2_RX1, 0, sizeof(SPI2_RX1));
SPI2_LS2_ManulMode(spi2_cmd1,SPI2_RX1);
Hal_DebugPrint("ADS7953 CH1 data is ");
Hal_DebugPrint("LS2: 0x%02x%02x ",SPI2_RX1[1],SPI2_RX1[0]);
Hal_DebugPrint("\r\n");

sometimes, the print sentence show the value over 4096(oxfff). like 0x3f83....so on...

2. I tried to catch up ADC value every 1 second, the period is 1 second,

3. ADC_HS1 is 3.3V, we used the divider resistor to set the voltage as 3.0V, what do you mean as to "Ideally this should be replaced by a reference IC?" ?

4. I found the problem in  MISO waveform, please refer the image : I checked sclk and MISO line simultaneously, and I have a question,

Should ADS7953 use Mode 2 of SPI communication ? which mode should use ? and how about baud rate ? I am using 2.625 Mbits/s. is it right ?

and I am using the data to select each channels with array as follows, is it right ?

uint8_t spi2_rst[2] = {0x00,0x40};   //reset

uint8_t spi2_cmd0[2] = {0x00,0x10}; //CH00

uint8_t spi2_cmd1[2] = {0x80,0x10}; //CH01
uint8_t spi2_cmd2[2] = {0x00,0x11}; //CH02
uint8_t spi2_cmd3[2] = {0x80,0x11}; //CH03

uint8_t spi2_cmd4[2] = {0x00,0x12}; //CH04
uint8_t spi2_cmd5[2] = {0x80,0x12}; //CH05
uint8_t spi2_cmd6[2] = {0x00,0x13}; //CH06

uint8_t spi2_cmd7[2] = {0x80,0x13}; //CH07
uint8_t spi2_cmd8[2] = {0x00,0x14}; //CH08
uint8_t spi2_cmd9[2] = {0x80,0x14}; //CH09

uint8_t spi2_cmd10[2] = {0x00,0x15}; //CH010
uint8_t spi2_cmd11[2] = {0x80,0x15}; //CH011
uint8_t spi2_cmd12[2] = {0x00,0x16}; //CH012

uint8_t spi2_cmd13[2] = {0x80,0x16}; //CH013
uint8_t spi2_cmd14[2] = {0x00,0x17}; //CH014
uint8_t spi2_cmd15[2] = {0x80,0x17}; //CH015

and as to the setting of SPI communication, please refer below,

hspi2.Instance = SPI2;
hspi2.Init.Mode = SPI_MODE_MASTER;
hspi2.Init.Direction = SPI_DIRECTION_2LINES;
hspi2.Init.DataSize = SPI_DATASIZE_16BIT;
hspi2.Init.CLKPolarity = SPI_POLARITY_HIGH;//SPI_POLARITY_LOW
hspi2.Init.CLKPhase = SPI_PHASE_1EDGE;
hspi2.Init.NSS = SPI_NSS_SOFT;
hspi2.Init.BaudRatePrescaler = SPI_BAUDRATEPRESCALER_16;
hspi2.Init.FirstBit = SPI_FIRSTBIT_MSB;
hspi2.Init.TIMode = SPI_TIMODE_DISABLE;
hspi2.Init.CRCCalculation = SPI_CRCCALCULATION_DISABLE;
hspi2.Init.CRCPolynomial = 10;

thanks.

Hello Edmond,

As described in Figure 50 of the ADS7953 datasheet, the channel requested over SDI in the n-th frame is selected by the internal mux in the (n+1)th frame and the conversion data is driven on SDO (MISO) on the (n+2)th frame.

This can be seen in the latest log that you have shared, where the channel number in your print string and that in the data have a lag of two frames. If you want to see this behavior even more clearly, please connect CH0=0V, CH1=1.5V and CH2=3V and repeat the same experiment.

As shown in Figure 2 of the datasheet, the data is launched by the ADS7953 at the falling edge of each clock and should ideally be captured by the host at the rising edge of the next clock. But due to trace delays or buffers in the MISO path, the time at which the data arrives at the host input pin may be different. A zoomed in screenshot of SCLK and MISO at the host pins should show us a clear picture of the interface timings.

Regards,

Sandeep

Hi, SanDeep,

firstly, thankful for assertive support earnestly,

as requested, I attached 2 image related with MISO waveform,

one image show overall waveform including issue : MISO (yellow line) waveform is not normal,

the other show the zoomed signal, 

if you can find the cause of issue, please help to let me know the method to solve it with detailed information,

and currently, I can get the exact data from ADS7953, if connecting to 0 V, got 0x000 data, if 3.0V, got 0xfff data, if 1.5V, 0x832 so on...

and it includes the number of channel like 0x1000 : this means that the 1 channel is 0x000.

thankful for your support in advance,

thanks.