Loop Impedance Meaning Data Logger Troubles

[katiencbabe]

Hi!
I am trying to measure the voltage through a "DaqPro", basically a data logger. I've used a voltmeter to get a reading, but then when I use the DaqPro it's totally off. So my question is--what does loop impedance mean, especially if it reads:

(Input Channel Measure: 0 – 50.000 mV)

Range 0 to 50mV
Resolution 3µV
Accuracy ±0.5% of reading
Loop impedance - 50Ω

Please let me know if you understand!

 

[FredGarvin]

The DAQ stuff I have seen usually state a channel input impedance. I wonder if this is the same thing. It essentially means that the signal is seeing the DAQ card input channel as a 50 ohm resistor.

The first thing I would do is double check your calibration of the channel in question. Are all of the measurements off by the same amount? It would also help if you specified the type of measurement you are trying to accomplish and any signal conditioning there may be.

 

[katiencbabe]

The input impedance is actually 200k ohm, and the load is 50.
( http://www.fouriersystems.com/products/8-channel/data_logger.php http://www.fouriersystems.com/products/8-channel/sensors.php )

The measurements usually aren't off by a standard ratio. Sometimes, the data logger reads a static number (+11.596 mV) while the Fluke will be reading the same source at somewhere around 17-18 mV (this is with a resistor near the daqPro's end). The reason I know it's not the DaqPro, and more so my cruddy attempt at electrical connections, is because I have tested a 9 V and 1.5 V battery on the daqPro's reader and it works.

I was thinking it may occur because there's too much 'noise', so I shortened the electrical cord which didn't really help. The Daqpro has already been calibrated, as has this large mixer. I am simply trying to examine how the voltage is affected by the substance in the mixer after a few hours (what should happen is that the voltage stays the same and then jumps a certain percent higher).

 

[berkeman]

(Input Channel Measure: 0 – 50.000 mV)

Range 0 to 50mV
Resolution 3µV
Accuracy ±0.5% of reading
Loop impedance - 50Ω

That's not what the specs say at the link you provided. The loop impedance applies to the 0-20mA input mode, not the voltage input mode. The spec says:

0-50mV input
Range: 0-50mV
Resolution: 1µV
Accuracy: ±0.5%of reading
Input impedance: 200kΩ

How are you obtaining the readings? Are you just looking at them on the Datalogger display, or are you connecting the logger to a PC or laptop? The thoughts about noise are a very real possibility, especially if you are using a laptop. Laptop power supplies generate a lot of common-mode noise at their switching frequency, and this can play havoc with low-level analog measurements. Have you tried placing a small capacitor across the differential sense lines of the datalogger? How long are the wires from the logger to the sense points?

 

[katiencbabe]

I am not understanding the difference between loop impedance and input impedance, nor do I even understand their meanings. I just copy and pasted from the site:
Input impedance = 200k Ohm; Loop impedance = 50 Ohm
So they should be what the specs are at the site's I provided. If not then maybe I'm looking at the wrong place. But no matter, I don't see the significance or effect these values can have on a reading.

I am both obtaining the readings straight from the DaqPro as well as connecting the logger to a laptop and both display the same value. I shortened the line significantly recently and it seemed to have little effect, however the logger's values are static as much. I will try the small capacitor thing, but what should that accomplish? The wires were originally about 8' long and have now been reduced to a little over 3'.

Thanks!

 

[chroot]

An ideal voltage measuring device connects to your circuit without disturbing it. Ideally, no current will flow through your voltmeter, and your circuit will behave exactly the same with and without the voltmeter connected.

However, real voltmeters do affect your circuit. All real voltmeters really do draw a little current from your circuit. You can model the voltmeter as a simple resistance. If you apply 1 V to your voltmeter, and it draws 1 mA, then it has an input impedance of 1 kilohm.

Your Fluke voltmeter probably has an input impedance on the order of 10 megohms, while your DAQ system apparently has an input impedance of only 200 kilohms, 50 times less.

It is entirely possible that the DAQ system is loading your circuit by adding additional resistance. This forms a voltage divider with other resistance in your circuit. The result? Your DAQ reads a voltage that is smaller than you'd expect.

Could you perhaps use a buffer (unity-gain opamp) to prevent your DAQ system from loading your circuit?

- Warren

 

[chroot]

BTW, what kind of signals are you trying to sense? DC? AC?

Berkeman's suggestion of using a small capacitor across your sense lines would serve to smooth out any high-frequency noise present across them.

- Warren

 

[berkeman]

As I said, I believe that the "loop impedance" specification applies only to the current input mode of the logger, and certainly not to the voltage input mode.

So when you disconnect the logger from the laptop and read the logger display, do you get the same numbers? The small capacitor is meant to reduce noise on the differential measurement. Start with about 0.1uF, and then try about 1uF to see if there is any effect.

Beyond that, all I can suggest is to use a good-quality differential probe and oscilloscope to try to trace down the differences. Alternately, make a large resistor divider (like divide by 1000), and test the DVM and logger in parallel as you use a power supply to step through small voltages at the output of the divider.


EDIT -- Great point by chroot. I missed the input impedance difference.