• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 46 subscribers
  • Views 789 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DAC8775: DAC8875 Calibration for different ranges

Lasto Roma
Intellectual 600 points
Part Number: DAC8775
Other Parts Discussed in Thread: REF5050, ISO7141CC


Hello,

I'm using DAC8875 and would greatly appreciate some help with the following issues:

  • Issue 1. The output voltage measured for the following configurations (channels A and B) are always saturated at 4.96 V (in both calibrated and un-calibrated modes, uni/bipolar ranges).

reset config: ['0x2', '0x1e12']
channel selection: ['0x3', '0x0020'],['0x3', '0x0040']
channel configuration: ['0x4', '0xd000']
channel data: ['0x5', '0xffff']
BBC selection: ['0x6', '0x0001'],['0x6', '0x0002']
BBC configuration: ['0x7', '0x061f']
calibration enable: ['0x8', '0x0001']
calibration gain: ['0x9', '0x81eb']
calibration offset: ['0xa', '0x0026']
dac status: ['0xb', '0x1781']

  • Issue 2. The status register indicates a fault with channel A, but not channel B. Both channeled are connected only to DMM. The fault is not cleared on software reset. Any suggestions what might be the cause of the fault? Both channels A and B are generating identical output measurements.

  • Issue 3. I'm trying to get a better understanding of the calibration steps of the DAC8775. The calibration of the unipolar modes 0x00 (5V) and 0x01 (10V) based on the datasheet, and as illustrated by following example works fine. How do we calculate calibration data to bipolar ranges? Do we need to keep separate calibration parameters for bipolar ranges?

Unipolar Calibration (V_requested = Cal_Gain x V_measured + Cal_Offset)
Cal_Offset = 0.002279831V (0x1D)
Cal_Gain = 1.007500601V/V (0x81EB)

Thanks for your help.

  • 0
    TI__Expert 5820 points

    Hi Lasto,

    Issue 1: Assuming these are written to the device correctly they look okay. One thing I would recommend is configuring and enabling the buck-boost before configuring and enabling the DAC outputs. What are the outputs connected to? It looks like you have clamped Vpos to 15V and Vneg to -15V based on your register values. One thing that could be happening is the voltage output is being current limited due to a short or low resistive load. Can you provide a schematic and also measure Vpos and Vneg after enabling the buck-boost? I would also recommend first ensuring the output is as expected before writing to the calibration registers.

    Issue 2: The fault could be for several reasons. The only voltage fault condition is caused when the short circuit current limit is on Vout.

    Issue 3: The calibration should work the same with the bipolar ranges. It is still just changing the overall slope and vertical offset of the transfer function. The only difference is that negative full scale will be at 0x0000 and positive full-scale will be at 0xffff with '0V' being at mid-scale 0x8000. When changing any ranges the gain and offset error can change due to switching in different gain setting resistors so for each range you may need separate calibration values.

    Thanks,

    Garrett

  • 0 in reply to Garrett Satterfield
    Intellectual 600 points

    Thanks very much for your help.

    Issues 1 and 3

    I'm using the EVM board and simply reading using DMM. I think Issue 1 and 3 are related to deviations in reference voltage. I measured Vref at the REFIN TP and it is 4.97V, which likely explains the readings. I tested with both the internal and REF5050 references and I'm continuously getting readings indicating deviations from 5V.  For now I adjusted the calibration to use VRef=4.97V.

    Could this be related to jumper settings on the EVM or my wiring to the test points for SPI (I'll be adding an ISO7141CC Isolator for the SPI interface, but I'm seeing the same Vref deviation when using SM-USB-DIG only).

    Issue 2

    The channel fault status entry seems to be a transient fault that is being latched. For some reason the fault status is not cleared via software reset or writing to the status register 0xB; it only clears after hardware reset.

    [Yes. the posted data are readback from the registers via the test points (double checked with the SM-USB-DIG and LabView app of the EVM). I checked the Vpos is set to 15V and Vneg to -15V.]

  • 0 in reply to Lasto Roma
    TI__Expert 5820 points
    Hi Lasto,

    The internal reference voltage is specified as max/min from 4.99V - 5.01V. The only way I could see it being out of spec if it is loaded but this should not be possible on the EVM. Are you using a DMM with enough accuracy to measure the reference? I don't think this could be caused by jumper settings.

    That is correct these are sticky to indicate that a fault has occurred as described in section 8.3.8 of the datasheet. These should clear on software reset or by resetting them in the status register. Have you tried this with the EVM software? Maybe the fault is still occurring so the get are latched immediately after you set them.

    Thanks,
    Garrett
  • 0 in reply to Garrett Satterfield
    Intellectual 600 points
    Ok. I'll use other DMM to check and double check the reset again with the EVM software.

    Thanks for your help.