• State Suggested Answer
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 122 views
  • Users 0 members are here
Options
Related

AD9834 Single Output

DLobb384
on Jan 18, 2021

Hi, 

I have a quick question about the AD9834 DDS. I have designed a board that is going to become an automated test rig for a few products. It will take out a lot of hard work when testing a board. For instance, plugging in a board and it measuring voltages at certain points, checking quiescent current etc. One of these test is to feed a small signal sine wave to a specified pin. The amplitude needs to be configurable so I have used the CN-0156 application note and used a DAC on the FSAdjust pin. This works well.

However, I think I completely misunderstood the pins IOUT and IOUTB. I thought they were essentially 2 seperate and configurable outputs. But when testing them I was getting odd outputs on both. Then, when I measured between the two pins, I got the intended sine wave.

This wouldn't ordinarily be a problem but I need to send the sine wave to one pin, and cant send both IOUT and IOUTB. Is there a way to get a sine wave with reference to GND on just one of IOUT / IOUTB?

Alternatively is there a similar AD - DDS that has adjustable amplitude with a single voltage output?

Thanks, 

Damien

 

  • 0 on Jan 19, 2021 4:15 AM 1 month ago

    You should check the reference schematics on CN0304 and EVAL-AD9834..

    AD9837 is somewhat similar and with voltage output - but lower frequency range.

  • 0 on Jan 19, 2021 9:03 AM 1 month ago in reply to languer

    Hi, thanks for the reply. 

    Yes I have been reading through those but I just want confirmation. In CN0304 all it says is that "There are two current outputs available on the board: IOUT and IOUTB. The IOUT is unfiltered, and the IOUTB is filtered". 

    I see the schematic at the start of CN0304. It looks to me that if the output of IOUTB goes into a op-amp+LPF you get a sine wave with reference to GND?

    Can you confirm that is the case?

    Secondly, can you confirm that the way I have my schematics would result in a sine wave between IOUT/IOUTB rather than either to GND?

  • 0 on Jan 19, 2021 5:29 PM 1 month ago in reply to DLobb384

    The IC outputs are complimentary - IOUT and IOUTB are identical in every aspect but their outputs are 180deg out of phase. In the reference note, CN0304, IOUTB is ran through an OPAMP for increasing the voltage range and re-referencing to GND and then through a filter; IOUT is fed directly without amplification or filtering.

    Your schematic looks ok, but the IOUT output will have a very small voltage swing and won't be filtered.

  • 0 on Jan 19, 2021 6:11 PM 1 month ago in reply to languer

    Thank you. 

    OK so I will need the op-amp to rereference to GND.

    The Small voltage swing isn't an issue, for my application I will need an amplitude range between 0.5V and (hopefully) 64mV. Depending on how noisy this becomes there maybe need to be further attenuation.

    As for the rereference to GND, I assume this is done via the two 2.4k Resistors going into the input of the opamp? 

    **EDIT**

    So I had a brainwave this morning. If the two outputs are complimentary, connecting one to GND should result in  the other giving a sine output. I just tried it, and it worked. 

    Whilst it works, is this something that should be done?

  • 0 on Jan 20, 2021 4:51 PM 1 month ago in reply to DLobb384

    You are correct; both outputs will be referenced to ground (there is only one supply rail) in complementary fashion. The OPAMP provides the buffering, gain, and "filtering" (filtering is really after OPAMP). Whatever you are driving you need to make sure it doesn't impact the current drive of the IOUT (you need to remember the output is a current waveform which gets converted to voltage via the 200-ohms).