A PCI data-acquisition board has 16 referenced single-ended channels. Here is the board: http://www.mccdaq.com/pdfs/manuals/PCI-DAS6034-35-36.pdf

I'm using all 16 channels with BNC cables. The two channels are used as output and the rest 14 channels as inputs. When I make single-ended measurements with all channels plugged, I get 50Hz noise and its harmonic in the signals. For low signals this causes a problem for data analysis.

Problem is: when all the 16 channels are plugged, the data from channel 1 has 50Hz noise, similarly other channels as well. But if I only plug channel 1 and plug off all the other channels there is no 50Hz noise.

All the input channels have isolated power supplies. I couldn't figure our where the noise can come from?

  • \$\begingroup\$ This needs more information on what you are plugging where, how and why, because just "plugging" and "plugging off" means nothing to me. Especially since these cards have no BNC connectors. Regardless, sounds a lot like common mode noise or ground loop currents. \$\endgroup\$ – Asmyldof Nov 10 '15 at 21:45
  • \$\begingroup\$ 14 channels are input signals from traducers with isolated power supplies. 2 channels are analog output. BNC cabels are used with RSE configuration. The noise is 50Hz and its harmonics. When only one single channel is connected there is noise. Can 50Hz noise come from a GND of an input signal's power supply even-though it is isolated form the mains? \$\endgroup\$ – user16307 Nov 10 '15 at 21:52
  • \$\begingroup\$ Is your AC 50Hz? Often times you can get crossover/interference from a power supply. \$\endgroup\$ – Daniel Nov 10 '15 at 22:08
  • \$\begingroup\$ yes 50Hz. do you mean one of the inputs' power supply causes the noise. The power supplies are isolated from earth. Can this still be the reason? \$\endgroup\$ – user16307 Nov 10 '15 at 22:12

Line noise can definitely be picked up in shielded cables; the longer they are, or if there are poor connections, the more easily it happens.

I do not know about this card, but is it possible that the software has not been set to make all of the RSE?

| improve this answer | |
  • \$\begingroup\$ it is set to RSE. but are you saying that one of the BNC might be picking up the line noise electromagnetically since earth is isolated? \$\endgroup\$ – user16307 Nov 10 '15 at 23:01
  • \$\begingroup\$ The isolation is a possible issue, but that wasn't what I was saying. I've had shielded cables pick up resistance because the connectors were dirty or the cable quality was otherwise poor. Even in a well-connected, good cable, if it is very long or extremely close to a source of EMI, some noise pickup can occur. As far as isolation, it might allow the grounds to float at different levels, so that noise can be induced from one cable to the other. I think that you will have to experiment with various permutations and look at the noise for each one. \$\endgroup\$ – Thagomizer Nov 11 '15 at 23:51
  • \$\begingroup\$ how can I eliminates if I cannot make the cables shorter? what if I connect the GND part of the BNC cables to earth ground? I was thinking if the in RSE configuration GND was already connected to the earth through the PC. \$\endgroup\$ – user16307 Nov 11 '15 at 23:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.