07-25-2012 10:19 AM
I tried to measure a IEPE sensor (simulator) ouput in parallel to the amplifier/signal conditioner. Looking with a scope I got about 12V bias with a nice sine 2Vpp.
So I setup a PXI 5922 with a PXI 5900 differential preamp to get 40Vpp inputrange.
I need to measure in DC mode. The output of the 5900 is clipping at about 11V . (CH+ to signal, CH- shortcut to GND) . Tried differend 5900.. all the same 😞
Another point: reading the input range is giving me 160V ... Oops bug?
NI-Scope 3.8.2 , LV2011
Another question: What is the output impedance of the 5900? 50Ohm? It is not noted in the Spec !!
Solved! Go to Solution.
07-27-2012 09:08 AM - edited 07-27-2012 09:10 AM
Some more information:
I updated the NI-Scope driver to the recent version.
and put together a short vi to show the effects:
Setup is a AWG generating a sine with about 1Vpp (in the vi Vss for the German Spitze-Spitze) feeding a IEPE (ICP) Simulator (coarsely applying an 11V offset) connected to an amplifier. The awg output and the Simulator output are captured with a PXI5922/PXI5900 combination.
If I monitor the 5900 input and output with a scope I get this picture . Yellow is the input measured in DC mode and green the output of the 5900 in AC (couldn't set the offset) but divide the red graph in the vi by 4 and you get the DC values.
All three cards tested so far show this effect.
Ether the input amplifier or the output amplifier of the 5900 is clipping.
And the range output of the NI-scope driver still gives me 160V range .....
Finally a picture of the setup.
07-27-2012 09:18 AM
I tested three different PXI systems , not only 3 different 5900 amplifier..
08-01-2012 10:52 AM
Seems to me that something is wrong in the Spec of the 5900 😞
I hook up a two channel generator that allow me to feed each input of one differential amplifier with two sine variable in phase and amplitude
Every thing is fine if you keep the input voltage range in the +-10V (20Vpp) range , no matter which phase.
But if the inputs where feeded with a 28Vpp sine with a 90 deg phase shift giving ~40Vpp Diff and +-10Vcm (common mode) the output of the 5900 is a triangle 😞
5900 was configured to DC and 1M input impedance , if I get my hands on a >2Channel scope I post a pic.
08-02-2012 05:30 AM - edited 08-02-2012 05:32 AM
Well, I think I found the solution for the clipping.
And it's nothing wrong with the specification of the 5900, however the definition of the common voltage range was missleading me.
While the 'common' definition of the common voltage in a diffential measurement is Vcm=(Vpos+Vneg)/2 the 'Common Voltage Range' isn't the possible range of the common voltage, instead the definition (by NI?) I found here states:
'The common-mode voltage range is defined as the maximum allowable voltage swing on each input with respect to the measurement system ground. Violating this constraint results not only in measurement error, but also in possible damage to components on the board.'
Something that should find its way into the Understanding Instrument Specifications and Data Acquisition Specifications -- a Glossary
So I can't measure a 12V to GND signal with my 5900 😞 the input unity gain amp is probably the one clipping
Leaving the 160V input range calculated by NI_SCOPE ..... CAR?
08-02-2012 08:24 AM
Henrik,
After giving it some thought, I agree with you that the specification should be clarified.
I suggest that the terminology be changed to be more like that used of op-amps: An absolute maximum voltage which may be applied to the inputs without damage, typically referred to ground (or, in the case of op-amps, to the power supplies). Additionally a maximum input voltage referred to to ground for which the device is operational and will perform within specs is important. Those two voltages may be the same or they may be different, depending on the nature of the input protection circuit. If the input characteristics are not symmetric about ground, then the limits shoudl be specified in each direction.
Lynn