ads1248 problem

Thnx for your reply.  We are using ads1248EVM as stand alone module. It is been interfaced to ST72561 MCU through SPI interface.
We have send following data to configure some registers of ADS1248.

Data : 0x42, 0x01, 0x30, 0x78
            0x4A, 0x01, 0x0E, 0x01
with START   pin high

Foloowing are the jumper settings :

1) We have configured EVM board to run at 5V logic (Digital and Analog supply). (But labelled pin on the EVM-board display  3.3 - digital and +/- 2.5V Analog supply)
2)  Download link  http://rapidshare.com/files/260474529/ADS1248_EVM_Settings.JPG.html for our custom settings

Following are the observation after data is sent
1) DRDY : On scope we have found that DRDY after power ON goes to high for period 376msec (approx) then it goes low for around 10usec then again goes high for the 376 msec. after that it constantly low.
2) VREF pin showed 0 V
3) SPI is configured for 125KHZ baud rate in mode 2 (CLK has 0 phase i.e CPHA = 0 and polarity is invert i.e CPOL = 1)
4) /CS signal is been toggled after every command frame of SPI
5) DOUT  signal is following /CS signal

Shweta,

I'm currently looking through your setup and it looks ok.
However, I'm using DVDD=3.3V because my SPI interface is
expecting 3.3V digital signals and if I run DVDD=5V, the
signals may not drive high enough. Also, you need to make
sure that AVSS is connected to AGND (and to DGND). I've
used J10 to connect to those points as well.

I've put together this together and I can read and write
to the part with these supplies. I've changed the data rate
and seen the /DRDY pulses change period.

For a good first test, I would send a command to read the
registers to see if that was correct. It's always a good
first test to see if the communications are correct and the
results should be known and repeatable. The values read back
should be the reset values given in the datasheet.
(They should be in hex: 01 00 x0 00

If the register readback is correct, then try writing
something to the register to see that the part takes the
value correctly. Then, you can readback the changes.

If you change the data rate as I had done (that would be the
first four bits of SYS0 or address 03h), you can also monitor
the change of the /DRDY line. In the default case, the /DRDY
should be showing a narrow pulse every 200ms as a default.
If the data rate is changed, you can see the pulse period
get shorter.

Here are a few more comments to some of your notes:

1. Note that the setup that I have outlined did not have the
+5V or -5V supply. This is required for the REF5020 and the
OPA277 buffer. In this case you can use the internal reference,
but you'll need to make sure this is turned on in the ADS1248
and sent to an external pin to read it.

2. I think the SPI communication setup you describe is wrong.
The data is setup on the rising edge of an SCLK and sampled
on the falling edge. In this case, the base of the clock
should be zero. Therefor CPOL=0 and the CPHA=1. With the
clock dwelling high, the tCSSC spec may be violated as shown
in Figure 1 and Table 1 of the ADS1248 datasheet.

Thats all I have - Let me know if you have any other problems.

Joseph Wu

Thank you so much for your reply. I have followed ur suggestion. Now my EVM board (in stand alone mode) is working on 3.3V interfaced to ST72561 MCU through SPI. Following are my observations.

Observations:

1) DRDY interrupts are coming after 200 msec for 5SPS data rate(default)

2) I can read conversion result using RDATA command

 However i am facing another problem

Problems

1) Unable to write and read configuration registers (Commands WREG, RREG)

Then i have assembled ADS1248 IC on seperate board. And tried to interface ADC to my MCU ST72561. In this case i can successfully read and write registers.  After that i checked the same EVM board on MMB3-PDK kit. It is working on PDK setup.  can write configuration registers successfully.

Can you please tell me which signal i am missing in my interface due to which I am unable to access the Configuration registers?

According to datasheet this condition occurs only in shutdown mode. When START pin is low then only i cannot write registers. But in my case i pulled START signal high permenently.  Then also it is in shutdown mode

Regards.

Shweta

 Following are SPI Settings in my system:

1) CPOL = 0, CPHA = 1

2) Command Byte :  0x20, 00, oxFF ( 0x20 - Read MUX0 register.,  0x00 - No of bytes to be read -1 , 0xFF - NOP to get the data on DOUT)

There are three ways to interface ADS1248

Case 1)  Interfacing our MCU ST72561  to EVM board alone

Set up is available at following path           

http://rapidshare.com/files/267196635/Doc2.doc

Observation :

1) ADC is not responding to above command byte for RREG command. (No data on DOUT line. It is following /CS signal)

2) On DRDY event  RDATA command ( 0x12, 0xFF, 0xFF, 0xFF )  is issued by MCU. In that case ADC is responding with some data on DOUT line. It looks like ADC goes into SLEEP mode.

3) DRDY interrupts the MCU after 200 msec      

Case2 :  Interfacing our MCU ST72561 to  ADS1248 IC mounted on PCB

 Setup will be available at following path

http://rapidshare.com/files/267197098/Doc1.doc

Observation:

1) For the above command byte   (0x20, 0, 0xFF) , ADC responds to MCU with 0x01 data

2) If i change the data rate by writing configuration register, the DRDY interval changes accordingly.

Case 3 :  ADS1248 EVM board interfaced to MMB3 motherboard

Observation:

1) Made settings through ADCpro software and checked the performance of ADC. It is functioning properly.

          We want to know that since the ADS1248 EVM board is working properly when interfaced to MMB3 mother board but the same ADS1248 EVM board when interfaced to our ST7 MCU gives us some problem to read the ADS1248 IC therefor our question are we missing some kind of hardware/ Software interlock in case 1 ? Awaiting for reply.

Thnx n Regards,

 Shweta

Shweta,

First of all, there is no hardware/software interlock on the EVM, it is designed so that the part can be used by itself.

We have looked at the interfacing diagram and seen no problems with the way you have it connected. However, would recommend measuring the supply voltages at the pins of the ADS1248 to make sure that the part is getting the proper supplies. I would also make sure that the proper signal is going to the control lines as well for START and RESET.

It appears that you have case 2 and case 3 working. So using case 2, look at the SCLK, DIN, DOUT and CS lines with an oscilloscope, and make sure that you can duplicate the signals for case 1. When using the EVM standalone, there is certainly a chance to pick up more noise. I have had cases where the communications are so noisy that the part sees the noise as extra SCLKs and either commands are sent that are unintended or the data output advances bits unexpectedly.

Also in case 1 is the data that you read out correct? If so, what is the exact data? If the data is correct, then the communications may not be a problem, but the part may be in reset or powerdown because of another bad signal.

Joseph Wu

Thnx for yout concern.  Observing the SPI communication in case 3 , we came know that after pressing the reset button provided on MMB3 board , the first command it sends is 0x00  three times i.e WAKEUP command according  datasheet. 

We followed the same sequence of transmitting WAKEUP command thrice in case 1 , we were able to write and read registers of ADS1248.

We dont know  why we need to send WAKEUP command thrice? 

Does the EVM board goes into SLEEP mode after power ON?  This question arises because this is not true in case 2.

Thanx n regards,

  Shweta

The ADS1248 does not go into SLEEP mode after the power ON.

We've gone back to look at the firmware in the ADS1248EVM and the original intent was not to issue the WAKEUP command, but it is a read data that is associated with an interrupt service request for the MMB3. Normally this would be FFFFFFh used to read it out, but it was coded with the 000000h as an oversight.

Joseph Wu

Thnx for your reply. I have some doubt regarding your reply

1) As you said 0x000000 is send to read the data. But according to datasheet of ADS1248 , we need to send RDATA command  0x12 / 0x13 followed by  0xFFFFFF (NOP command) to read the conversion data.  But in case 3 I have not seen such kind of waveform. It sends only 0x000000 after DRDY interrupt.

2) Then I removed the statement which sends 0x0000000 on interrupt from my firmware   i.e in case 1,  but still it is working fine.  In my firmware I am writing configuration registers after DRDY interrupt occurs. That means ADS1248 does not go into SLEEP mode.

3) Again i have checked my previous firmware version where i write  \ read  configuration registers of ADS1248 before DRDY interrupt occurs with the delay of 20 msec after pulling START signal high.

 Following are my observations

  In case 1 :

                  Again i am unable to read and write configuration register.

3) In case 2 :

                 I can successfully read and write register of ADS1248

      We were able to read \ write registers in EVM board only after DRDY interrupt occurs. Whereas  in case 2  i.e  ADS1248 mounted on separate PCB ,  we can read / write registers irrespective of DRDY interrupt,  Can you please tell me why there is difference between behavior of ADS1248 in EVM board and ADS1248 mounted on PCB ?

Thnx n Regards,

 Shweta

I had problems with comm until I noticed that my ADS1248 was in constant Reset due to DVDD not being tied properly. I'm am using the ADS1248EVM board as standalone and noticed that the data sheet incorrectly labels J11.8 +3.3V however J11.9 is +3.3V. I'm looking at Users Guide SBAU142A - Apro; 2009- Revised July 2009. 

Thanx for your post. I will check it out and let you know.

Right now i am using ADS1248 EVM along with MMb3 PDK kit to chek ADC performance for RTD application.  I have some problem regarding RTD wire compensation. 

In data sheet of ADS1248, the application note for ADS1248 in hardware compensated RTD measurement is given. 

For my set up  I refer the application note of ADC AD7711 from analog devices (Fig 20.3) to avoid the external components. 

As according to  AD7711 (Analog Device ADC) Fig 20.3 application note, we  don’t need any external hardware compensation as ADC is working in common mode configuration so it will not introduce any error.

With reference to that , I have simulate the RTD configuration with ADS1248EVM  along with MMB3 – PDK kit to check the performance of ADS1248 for 3 wire compensation, but I did not find such results.  I have interfaced 100 ohm resistor from calibrator as RTD. My set up information and observation will be find on following link.

 http://rapidshare.com/files/270802280/ADC1248_in_RTD_application.doc.html

 To compensate the wire drop I have to subtract the drop across AINN wrt to GND from the analog input reading (i.e AINP and AINN) to get actual drop across RTD. But according to data sheet of ADS1248, I cannot get the reading of AINN wrt GND, as there is no provision to select MUX channel for AIN and GND in mux0 register.

If I connect one of the channel from 8 analog channels to AGND then only I can read the drop across AINN and AGND, but there is no free channel available in my application as I am using all 8 channels.

 In datasheet Hardware compensation technique is available. But is there any other way (using software) to do 3 wire compensation in RTD? Is there any application note available for that? 

Thnx n regards,

  Shweta