Noise & Ringing/Mismatched Impedances

[mjs12]

I'm currently using a DAQCard-AI-16XE-50 and 50 pin AI connector block and trying to diagnose some noise (possibly ringing) that I'm seeing in my analog inputs.

My program sends a signal out from GPCTR1_OUT to a photovoltaic relay (a PVU414) and returns to DGND (there is a 470 ohm resistor in this circuit to limit the cuurent flow). The GPCTR1_OUT signal also goes to PFI0_TRIG1 in order to trigger the data acquistion.

The counter signal is typically on for 100ms and the overall acquistion may be from 500ms to 2000ms. Once the counter signal stops, I see large oscillations on my analog inputs (+- 0.25V). If I leave the counter on for the duration of the acq. I don't get any noise, so it is specifically once the signal from the
counter is stoppped and relay opens.

I'm measuring in differential mode. The output impedance of my pressure transducers is 50 ohms. I have tried using one & two bias resistors according to the E-Series User Manual and nothing has made a difference. Currently I have the negative side of my signal also going to AIGND.

I have also added a voltage divider (just two resistors) to the transducer power supply and am sending a 4.2V signal into one of the channels. Strangely, I'm not seeing any noise on this channel. This also has the negative side (which is the same signal as mentioned above, just the common side of the transducer power supply) going to AIGND.

Also, when I short between the +ve and -ve of a chennel I don't get any noise.

So there seems to be a specific interaction between when the counter and hence relay is switched off and the noise that is being induced on the analog input.

The cabling is shielded, 1.5m long and there is reasonable physical separation.

Ca
n anyone suggest some more steps that I can try to get to the bottom of this problem??

Thanks,
Matthew.

[mjs12]

It turns out the sensor I'm using for that particular measurement actually has an output impedance of 4.5k Ohm, which is much higher than the 50 ohm output impedance of the other sensors that I'm using.

Also, to rule out the cabling as the culprit, I used two resistors at the sensor end of the cable to mimic the ouput of the sensor. When it was setup like this, the noise/ringing was gone.

So with the sensor in place I get noise/ringing but when sending a reference voltage down the cable to the daqboard I'm not getting the noise/ringing.

So what is happening with the interaction between the sensor and board that is causing the noise/ringing?

Thanks,
Matthew.

[Sean_C.]

Hello Matthew,

Thank you for contacting National Instruments.

Are your relay and your transducers connected in anyway? Or are they completely separated? I find it strange that the counter output has such an effect on the analog inputs. Does this noise appear on all the AI channels, or just the pressure transducer?

Some things to try are:
Acquire one AI channel at a time. Is this signal noisy?
Try acquiring all AI Channels without any counter output.
Try looking at your analog inputs in Measurement and Automation Explorer (MAX) Test Panels. Do these signals show noise?
If the signals are still noisy through all of this, make sure that the input voltage is not noisy. Measure these signals on an oscilloscope and compare them the the acquired signa
ls. Is the noise coming from the sensors, or is does it appear after being read in to the DAQCard?

Let me know the results of these tests... I hope this helps.

Sean C.
National Instruments
Applications Engineer

[Jamie Fraser]

Hi,

It sounds as if you are switching your signal on/off with the relay and measuring the signal you are truncating/passing with an analogue input. Could it be that the signal you are switching is inductive and you are getting back-emf spikes when you kill the current with your relay? Use a scope to check.

Regards

Jamie