ADS131A04: application problem

Hi Reed,

Can you please show me the raw data from the ADC and also what your input voltage on ADC's input is? let me know you configuration for the ADC in terms of gain and data rate. Did you see the jump only happening in the beginning of channel switch or it happened continuously? a timing plot for digital signals will be helpful as well. Thanks.

Best regards,

Dale

When a stable voltage is input, the ADC conversion value of the normal read-back is stable, regardless of the value. When the physical impact, the value will instantly produce a large fluctuation (more than 5 times the normal fluctuation value); The spi clock cycle is 80ns, and the sampling time is selected at 20ns after the falling edge of the clock.

The configuration is as follows:

always @(posedge clk)

if(clkv2) case(rCnt2[4:0])

0:  tReg[15:0]<=16'h655; //unlock

1:  tReg[15:0]<=16'h4b78; //inner reference,4v

2:  tReg[15:0]<=16'h4c3c; //no crc

3:  tReg[15:0]<=16'h4d88;   //sclk,sclk<25M,iclk<12M,fmod<4.25M

4:  tReg[15:0]<={12'h4e4,vset2[3:0]}; //0~15:4096~32

//4:  tReg[15:0]<=16'h4e40; //0~15:4096~32

5:  tReg[15:0]<=16'h4f0f; //all channels on

6:  tReg[15:0]<=16'h5101; //channel 1 gain: 2

7:  tReg[15:0]<=16'h5201; //channel 2 gain: 2

8:  tReg[15:0]<=16'h5301; //channel 3 gain: 2

9:  tReg[15:0]<=16'h5401; //channel 4 gain: 2

10:  tReg[15:0]<=16'h0033; //wake up

default:  tReg[15:0]<={4'h2,rCnt2[3:0],8'h0};

endcase

else if(clkv3)

tReg[15:0]<={tReg[14:0],1'b0};

Hi Reed,

Thank you for providing information. Can you please clarify what the physical impact is? Also, can you please provide the raw data from the ADC?

Best regards,

Dale

When external force knocks, PCB board will vibrate. ADC data also changes at this time.

Hi Reed,

Thank you for your explanation. When the physical is impacting and the pcb board is vibrating, did you monitor the analog signal on the ADC's input pin and the digital signals of SPI including CSn, SCLK, DIN and DOUT?

Regards,

Dale

Hi Dale,

Thanks!

Using an oscilloscope measurement will encounter a problem: the physical impact of the probe, such a movement itself will affect the measured waveform. 

Hi Reed,

I understood, however it will be hard to identify whether the input signal has already been affected or not. We have to figure out a solution to monitor input and output. Is it possible to put the oscilloscope with the ADC board or system together when the "physical impact" is happening?

Best regards,

Dale