Data Acquisition Device for 0-0.4V signal

[katieh412]

Hey everyone!

My name is Katie and I'm currently working at Queen's University to communicate an analog output signal with a range of 0-0.4 Volts to a computer so that I can track the voltage over time in LabVIEW.  I'm currently using the National Instruments USB-6009 Data Acquisition device to do so, but unfortunately I'm experiencing quite a bit of noise in the signal.  Do you know of a comparable data acquisition device that would be able to take in a signal of this magnitude with better accuracy?

Thank you very much for your help in advance!

[crossrulz]

Change your max and min voltage settings of your DAQ task to be -1V to 1V.  That will lower the range of the DAQ to the min range, which should help your resolution.

You could also use an op-amp circuit to increase your signal's voltage range.  If you do a gain of 2, your signal would then be 0V to 0.8V, still within the -1V to +1V range of the DAQ.

[katieh412]

Thank you so much for these suggestions and for getting back to me so quickly! Formerly, my signal input range within the DAQ task assistant was -10 to 10 V, which would definitely increase the noise! The specs say the voltage accuracy in this larger range is ±7.73 mV, but when it's set to -1 to 1 V the accuracy should increase to ±1.53 mV.  Unfortunately after I changed the range, the accuracy didn't seem to improve.  Attached is a picture of the altered settings, did I do this right?

Thank you in advance for your help! 

[cbutcher]

Yes - that looks right. A quick question - you say that the previous range gave a noise specification of +-~8mV and the new ~1.5mV. Would you say that the noise you measured (previously or now, since you said it didn't make a big/any improvement) was of approximately that (~8mV) magnitude?

If the noise is much larger than +-8mV, then it's probably not unreasonable to conclude that the noise source might be external to the DAQ device. Is it possible that the wires are for example unshielded and in an electrically noisy environment, or similar?

I have a bunch of torque measurements next to a pair of large AC motors - if I could get results with only +-8mV of noise I'd be ecstatic! 🙂 (although in fairness, the range of measurements is larger so the increase in noise is only very frustrating at low torques).

[katieh412]

Hmmm so for one of my measurements LabVIEW output numbers of 0.348542 to 0.396996, which corresponds to a difference of 0.048454 volts.  This definitely seems higher than the minimum reported accuracy of ±7.73 mV... Basically my system is composed of the pH meter, a handmade adapter by our electrical technician that inserts itself into a headphone jack on the pH meter and outputs two red and black wires, and those wires are attached to two terminals in the USB-6009.  The USB-6009 is just set up as usual, with the cord it comes with attaching the device to the computer via USB.  Does any of that sound like it could be causing noise?  I can send pictures of the system if you think that would be helpful.

Thank you again for all of your help!

[lagelian]

Hi katieh412, have you also taken a look at our Field Wiring and Noise Considerations guide? It has some tips on reducing noise for DAQ systems.

http://www.ni.com/white-paper/3344/en/

Angela L. 

National Instruments

[katieh412]

I'm looking at trying these methods today or tomorrow, I'll let you know how they go!

Thanks!
Katie

[katieh412]

Hello everyone!

Thank you again for all of your help- it's greatly appreciated!  I tried attaching a short wire to the two terminals, and still received noise of a magnitude of approximately ±7.6505 mV.  For the maximum voltage range (-10 V to +10 V), the factory stated tolerance is ±7.73 mV, which is close to the value I obtained.  However, the settings were changed to the minimum voltage range (-1 V to +1 V), so the tolerance should've been ±1.53 mV.  Do you know of a way that I could obtain the accuracy of the minimum voltage range?

There were no additional boards attached to the DAQ, so this isn't the problem!  I was working with a data acquisition specialist, and he stated that the noise may be coming directly from my pH probe - do you think this is a possibility?

Looking forward, I'm planning on installing an amplification circuit to increase the magnitude of the input voltage and thus hopefully improve the filtering and the accuracy.  Do you have any further suggestions?

Thank you very much for all of your help!
Katie

[Henrik_Volkers]

You have a USB-6009 and a pH-Probe....  usually these probes need an amplifier or a high input impedance (GOhm) reading device. The 6009 has an input impedance of 144kOhm , far too low for a pH-probe (migth get damaged!)

If so, an amplifier is needed.

or

https://www.atlas-scientific.com/product_pages/circuits/ezo_ph.html

Noise:

If speed isn't a question, you can read 1000 sample with 1kSPS and take the mean of that value. (Try that with a short cut input)