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.

AMC1301: Connecting outputs of multiple AMC1301 to a single-ended (not differential) multi-channel DAQ

Part Number: AMC1301

Hi,

I'm developing an application where I have 30 boards each one containing one AMC1301 that I would like to connect to a NI PCIe-6323 card. This is a 32-channel Analog Data Acquisition card. The VOUTP and VOUTN outputs of the AMC's both vary in regard to GND (from VDD2 power supply). So basically my analog reference is floating in regards to GND. If I were to connect the VOUTN outputs of these 30 AMC1301's together and connect them to a floating reference pin (called AI SENSE on that card), would this work?

For those of you not familiar with NI DAQ cards, the analog inputs can be connected as shown in the following picture:

If the VOUTP and VOUTN are really "floating" in regard to GND, than I could connect all of my AMC1301 outputs like shown in the column "Floating Signal Source" and pick either the RSE or NRSE configurations, preferably the latter. See this paper for more info:

www.ni.com/.../

But if there not floating, and instead the VOUTP and VOUTN outputs are referenced to VDD2-GND, connecting the VOUTN pins together, as a virtual AI GND, would not work, because the VOUTP outputs will not maintain their voltage differentials in regard to this virtual reference, as it would if the output were a truly floating signal source, like an array of batteries or thermocouples.

Please let me know if my assumption is correct or not.
Will I be able to connect the VOUTN lines of my 30 AMC1301's together, or not?

Thank you in advance for your help.

Habil Silva
CONTROLAR
Portugal 

 

  • Hi Habil,

    Welcome to our e2e Forum!

    The outputs of the AMC1301 are referenced to GND2 which would typically be the same ground reference as the AIGND in the figures you provided. The only way they would be 'floating' is if you had 30 independent power supplies (i.e. batteries) powering up each VDD2. The outputs of the AMC1301 are true differential outputs so, assuming the inputs are not the same signal, tying all 30 VOUTN's together would most likely give measurement errors. If I read the datasheet for the NI-DAQ card correctly, there are 16 differential inputs or 32 single ended inputs so you could just use the VOUTP outputs and leave the VOUTNs floating.
  • Hi Tom,

    Thank you for your answer.

    We could actually make all of the VDD2 supplies floating by using an isolated DC/DC converter but I'd suspect that all of the differential outputs, by having their VOUTN pins tied together, would cause some sort of measuring error.

    Regarding your suggestion of using only VOUTP and use GND as reference, we noticed during testing that both VOUTP an VOUTN vary in regards to GND as the output increases. This means that without knowing the value of VOUTN, whatever is measured on VOUTP would be of no use:

    It's exactly as you said: the outputs of the AMC1301 are true differential outputs.
    To solve this problem we're going to add a differential amplifier to the outputs of the AMC1301 and have the amplifier's output referenced to a known GND so that the DAQ card can be used as 32x single ended channels with AI GND as reference, instead of AI SENSE.

    That would solve the issue, right?

    Best regards,
    Habil

     

  • Hi Habil,

    Take a look at this TechNote: www.ti.com/.../sbaa229.pdf . You can certainly convert the differential output of the AMC1301 to a single ended signal and simplify the connections to the NI-DAQ card.
  • Thank you very much Tom!
    This has solved my issue.

    Best regards,
    Habil
  • Great! Thanks for letting us know!