• Content Count

  • Joined

  • Last visited

  1. I've given the board a visual inspection and clean and it has made no difference. If I spray some freezer spray on IC11 the 26kHz peak disappears. Not sure what to do next other than mark channel 2 as U/S :-( Kevin
  2. Thanks for the reply. I've tried it with the inputs shorted and it makes no difference to the frequency or amplitude of the 13kHz harmonics. It does however prevent pick-up from the AWGs so unfortunately it looks like I have a hardware fault somewhere on channel 2.
  3. My trusty AD 1 seems to have developed a fault. If I connect the AD to the USB and run Waveforms and use the spectrum view I see some peaks at 13.02kHz, 26.04kHz, 39.06kHz on channel 2 that are not present on channel 1 and are not present on a different AD unit that I have. The units are both Rev C. Take a look at the attached screen capture. There is another peak at 52kHz but this is down in the noise. I have also noticed that if I generate a signal on the AWG2 with no external connections, say 50kHz, this then appears on the spectrum of the channel 2 input. If I do the same with AWG1, then nothing appears on either channel. Perhaps there is some connection between these 2 issues. Unfortunately I am looking at some small signals in the audio band and these peaks are getting in the way. I suspect that I may have somehow damaged channel 2 in the past and not realised. Does any one have any ideas where to look in the unit? I have some of the schematic from the documentation. Thanks, Kevin
  4. KevinS

    PmodDA3 Glitch

    Thanks for the response, a simple filter on the board would be ideal. Kevin
  5. KevinS

    PmodDA3 Glitch

    We have a couple of DA3 modules connected to a pair of Xilinx Spartan 3 FPGAs. We are generating an audio ferquency baseband signal with at 200 ksps. Looking at the buffered output from the module we periodically see a sub-1us 500uV pulse approximately every 100 samples when outputting a small signal sinewave. Offsetting our sinewave by different amounts can make the problem either better or worse. The datasheet for the AD5541 DAC talks about a 'Digital-to-analog glitch impulse' that seems to explain our unwanted pulse. We have crudely filtered out this pulse with an 18pF cap from the DAC output to GND. This does not seem to cause any problems at our relatively low baseband frequency. Is there an accepted method of dealing with these unwanted pulses? Would it be possible to add a position on the Pmod board for a filter to be added?