ADS1247 for ECG

Hi Luis, thank you for your reply. I thought there is something more simple solution since I'm overly frustrated with that ADS1x9x chip. I simply cannot get it to work, it is like a bad dream for me. Thanks anyway.

Hi JustGreg,

I remember you had some issues with the RESV1 pin, but I had kind of thought you got that straightened out.  Can you possibly post a screen capture from that setup and let us have a look at the way you are implementing the SPIn interface?  Are you trying to work this through the C2000 processor?  I've noticed a few other posts regarding the TMS320F28335 so I am a little curious if that is the controller you are using here.

Hi Tom,

 

thank you for dropping by. Well, I had problems with that design, but since there were quite a few bugs on that PCB board I abandoned it. Instead, I've soldered a new chip on a Schmartboard with all the necessary circuitry, for testing purposes. I am interfacing the chip with an MSP430F2274 mcu with a proven SPI code. The mcu outputs the correct signals but the ASIC doesn't want to answer correctly - 0x00 or $60 is all I receive when I send read CONFIG1. Here is the schematic I am using. I checked the circuit for shorts, checked all pins, the power supply circuit voltages, all seems to be fine. Double checked the RESV1 pin. No idea what's the problem here.

 

Here is the SPI code for reset, read&write reg operations:

void ADS1298_PowerUp(void){

P4OUT &=~PWDN;
P4OUT &=~RST;

SPI_CS_ADS_LO();
P4OUT |=START;

//CLOCKSEL to DVDD -> internal clock 2.048 MHz
// tclk=488 ns

P4OUT |=PWDN;

P4OUT |=RST;

__delay_cycles(1000000); // MCU clock @1 MHz, wait 1s for tpor

P4OUT &=~RST;
  __delay_cycles(2); //trst=2 tclk, equals ~1 MCLK cycle but changed to 2 for safety reasons
P4OUT |=RST;
__delay_cycles(10); //18 tclk, equals ~9 MCLK cycles but changed to 10

  ADS1298_SendCommand(ADS1298_SDATAC);

}//ADS1298_PowerUp

 

unsigned char ADS1298_RREG(unsigned char address){

unsigned char status;
unsigned char data;

data=0x20|((address&0x1F));

SPI_CS_ADS_LO();

status=SPI_Write(data); //address

status=SPI_Write(0); //bytes to read - 1

status=SPI_Write(0xAA); //dummy byte

SPI_CS_ADS_HI();

return status;

}//ADS1298_RREG

 

void ADS1298_WREG(unsigned char address, unsigned char value){

unsigned char status;

status=0x40|((address&0x1F));

SPI_CS_ADS_LO();

SPI_Write(status);

SPI_Write(0);

SPI_Write(value);

SPI_CS_ADS_HI();

}//ADS1298_WREG

 

void ADS1298_SendCommand(unsigned char value){

SPI_CS_ADS_LO();
SPI_Write(value);
SPI_CS_ADS_HI();

}//ADS1298_SendCommand

 

Hi Greg,

Sorry for the slow reply here - its been a busy couple weeks.  I don't recall ever seeing a screen shot from you with the actual SPI communications setup you use.  If there is a phase inversion or large delays in the transmission to the ADS1298 from your host processor, you could get corrupted data.  The code snippet is great, but without know exactly what all the function calls do, it's a little difficult to try and put a finger on the root cause of your problem.  Can you capture the SPI transfer from that old board and send it along or perhaps sketch it out on the back of a (virtual) napkin and send it along to us? 

Hi Tom,

 

thank you for your response. Tried again the whole thing with a MSP430F2013 in a LaunchPad. My SPI setup is: MCLK ~ 1.2 MHz (MSP430 default), clock source SMCLK (=MCLK), all other settings are default (CPHA=0, CPOL=0). Here are some pictures: SCLK/CS , SCLK/SIMO , SCLK/SOMI. The main loop is:

 

BoardInit();
SPI_Init();
ADS1298_PowerUp();
P1OUT |=LEDR;



ADS1298_WREG(ADS1298_CONFIG3, 0xC0);
ADS1298_WREG(ADS1298_CONFIG1, 0x86);
ADS1298_WREG(ADS1298_CONFIG2, 0x00);


while(1){

status=ADS1298_RREG(ADS1298_CONFIG1);     //3 bytes: 0x21 0x00 0xdummy

__delay_cycles(1000);

}

}//main

 

  There is no answer from ADS1298, status always returns zero. Stepping into the RREG function, all the three spi write returns zero.  The START pin is low, RST, PWDN is high.

 

I went through the board with a conductivity tester. There are no shorts, all connected pins are following this schematic. The AVDD/AVSS/DVDD levels are also fine. DRDY is high and not changing. Seems like the SPI interface is not working correctly in my ADS1298. I really don't know what's wrong here.

 

Greg

Hi Greg,

I can't see your file postings due to some IT and firewall issues.  Could you possibly zip them all up and post them here?

Tom,

 

here you are. Thanks for taking the time.

Hi Greg,

Attached is a file that I used to capture the device ID data from the ADS1298 using the MSP430F2013.  In this code I've sent two 8-bit words which provide SDATAC and STOP commands followed by a 24-bit RREG command.  Try it out and let me know what you get as a result.

Hi Tom,

 

thank you very much for your efforts. I'll able to try this on Tuesday. Have a nice Easter ham till then.

Hi Tom,

 

I managed to solve the problem. Found a bug on my PCBs:   the soldermask below the chip was unusually thick and some pins did not touch the pads. With two hypodermic needles and a conductivity tester I was able to detect that the pwdn was not connected firmly to its pad.  I removed the chip, sandpapered the bump off and soldered it back. From that point onwards all seemed to work fine. Tomorrow I'll check my code since I used yours. Thank you for your input - it really helped me to focus on the hardware alone.

 

Great! 

Glad to hear you found the problem.  I think you should go out and enjoy a good 'health drink' to celebrate!