• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 46 subscribers
  • Views 783 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ADS1248 IDAC Does not Turn on

John Merrill
Intellectual 420 points
Other Parts Discussed in Thread: MSP430F5232, ADS1248, MSP430F2618

I am using an ADS1248 in conjunction with an MSP430F5232 to perform 3 measurements.  DVDD = 3.3V.  AVDD = 5V.   The Start and Reset lines are pulled High by the MSP.  I measure Vref = 2.048V. I appear to have good SPI communications between the MSP and the ADS because I can read back the expected value of the internal temperature diodes. All points of connection appear to Ohm out correctly (i.e. no unexpected shorts or opens).   

However, when I attempt to connect IDAC 1 to IECX1 to provide a 500uA bias current to a restive bridge, i get no current through my bridge.  IECX1 appears to have no output.  The same is true trying to use IDAC 2.

Setting the registers of the ADS to the following:

IDAC0 = 0x04  500uA current

IDAC1 = 0x8C  Connect IDAC1 to IECX1,  Disconnect IDAC2

MUX0 = 0x1A   measurement 1

VBIAS = 0x00 = Disable

MUX1 = 0x30    Internal ref on

SYS0 = 0x30 (PGA = 8; 5 Hz sample rate)

We are experiencing this issue on every prototype board we have worked on thus far.  This prototype board was developed based on a Proof of Concept PCB that we designed and were able to successfully test the ADS1248 with an MSP430F2618. The Proof of concept PCB was also RIGID FR-4, where the prototype PCB is a RIGID-FLEX design.  Schematically they are the same.  The layouts may differ but as i stated before i checked all the lines and they check out.

Is there a difference between the 2618 and 5232 that could account for this issue?

Is there something in my register settings that are incorrect?

Are there any other items to examine that could be causing this issue?

Is there something else i can try or did i receive a batch of bad parts?

  • 0
    TI__Guru* 97455 points
    John,


    The first thing people miss for operating the IDAC is that the internal reference should be turned on. However, you note that you've turned on the internal reference so just check the REFOUT pin to make sure that the reference voltage is coming out without problems.

    I am not familiar enough with the different MSP430 devices, but I would doubt that they would be different enough to cause this problem. Besides, if you can read and write to the device, that should be enough to show the operation.

    I also don't see anything in the register settings to cause problems and it's extremely unlikely that you would get bad parts or damage parts enough that it works for everything except IDAC settings (each of these devices are tested before leaving the factory).

    To start, I'd try to shunt the IDAC current to some other channel to check the current. If you have a free input pin, you should be able to shunt the current there to check it. Are you measuring the bridge measurement externally? Can you disconnect your bridge just to see the current by itself? Have you taken an unpopulated board to check to see if IEXC1 isn't accidentally grounded? If you're able to read you're supply current, you may be able to see the IDAC current when you change the value.

    Check these things and let me know if it helped. If you have a schematic, and are able to share it, that might be helpful also.


    Joseph Wu
  • 0 in reply to Joseph Wu
    Intellectual 420 points

    We went through the list of suggestions.  There are no hardware issues that are causing comm issues.  We took another look at our MSP code.  Turns out the SCLK was set to about 4MHz which is double the maximum frequency (2MHz) based on the datasheet.  We slowed down the SCLK to 250kHz and the IDACs turned on and we got the readings we expected.  However now we cannot set the SCLK to 1 or 2 MHz and maintain communication.  Is there something else we need to look at in the ADS1248 settings that could be preventing the faster comm?

  • 0 in reply to John Merrill
    TI__Guru* 97455 points
    John,


    There shouldn't be anything that would disrupt communications between 1 and 2 MHz inherent to the device. If you're having problems with the communications, I would look it over with an oscilloscope. Check, SCLK, DIN, DOUT, and /CS.

    What exactly is happening? Are you losing the ability write to the device? read from the device? Are the problems associated with reading/writing to the registers? or reading the data?

    One issue might be extra noise on the SCLK line. If there is noise, then the DIN/DOUT might be accidentally advanced while reading from or writing to the device.


    Joseph Wu
  • 0 in reply to Joseph Wu
    Intellectual 420 points
    After a closer look it came down to timing on the MSP. The time between Power up of the MSP & ADC1248, Releasing the ADC Reset line, setting up the SPI bus, and beginning communication violated the 0.6msec delay required by ADC when we attempted to use a SPI clk at 1MHz and above, but not when we set up the SPI clock for 250kHz. By adding a delay we were able to achieve communication at 2MHz.

    However in both cases the the MSP was set to Fosc = 8MHz, so the timing between start up and begining communication should have been constant. The only difference in code was the SCLK frequency. Does the time to set up the SPI bus change with the SCLK speed or does the MSP start attempting communications faster with the increased SCLK frequency? My guess is the latter as the number of operations/sec could not have increased with the clock fixed at 8MHz in the micro.
  • 0 in reply to John Merrill
    TI__Guru* 97455 points

    John,


    Sorry about the late reply, but many of us have been on break for the holidays.

    When it comes to the SPI bus setup time for the MSP430, that's a question for the MSP430 engineers. It's likely that they require some sort of setup that is related to the oscillator clock. For the ADS1248, it has a setup time that is based on the given oscillator clock. Once the device detects the power supplies coming up, there is a count of 2^16 oscillator clocks before the device starts operation. This time is used for the power supplies, reference, and any analog circuitry to settle, and to load the default configuration into the digital section of the device.

    If you have any other questions, let me know.


    Joseph Wu