NI 9234 with analog quasi-static DC voltage

[AWUoB]

Hello,

I have a NI 9234 (4ch +/-5V 24-bit IEPE) module attached to a cRIO chassis. This module is ideally suited for accelerometers and microphones where the DC voltage is always changing (ie; measures rate of change).

I also have a NI 9237 (4ch 24-bit full bridge analog input module) attached to the same cRIO. This module is ideally suited for measuring varying voltages from strain gages (quasi-static and dynamic loads).

The attached graph shows the two channels collected synchronously, but as you can see the loadcell (red trace) goes down (as it should) but then drifts back up to zero of its own accord when in fact it should remain low just like the Strain Gage beam does. After all, the two sensors are physically coupled together.

Q1: Could this have something to do with internal AC/DC coupling of the 9234 module?

Q2: Is it actually possible to collect "quasi-static" DC voltages using a NI 9234 module?

Any information as to why this happens or if there is a way to correct it would be appreciated.

Regards,

Andreas

[AJost]

To Q1:

I think the answer will be: depends on the time scale and your settings, but I don't think so. May I ask, what is the time scale of your graph? I gues the scale of the y-axis is volts?

In AC mode, all frequencies below 0.5Hz will be filtered. You can set the mode inside your project by right clicking on the Module -> properties

To Q2:

Sure. The 9234 is able to work in DC mode, thus should be no problem.

To share my thoughts what might be happing:

Could it be, that Loadcell is measuring differential and the strain gage total?

I realized, that when you differentiate the strain gage curve/integrate the loadcell curve, both curves should very closely follow each other.

You should be able to find this kind of information in the manual of both devices.

Andreas Jost

Applications Engineer

National Instruments

[AWUoB]

Thanks for the info (another Andreas),

The ticks along the X-axis are 2 second intervals. Collection time was 15 seconds.

I've read the AC & DC coupling tech note on the NI website which explains coupling and I don't see this as having anything to do with my question.

The load cell is a +/-20kN piezo-crystal type and both it and it's bespoke amplifier appears to be single ended (BNC). The NI 9234 is (as you say) has four differential inputs. The strain gage beam has a full Wheatstone bridge, two gages on the top and two on the bottom and has a differential output (+/- 5V full-scale) which is fed feed into one of the channels on the NI 9237.

It appears to me that the internal circuitry of the 9234 is designed really only for rapidly changing quantities like that of microphones and accelerometers, hence the nice smooth roll off back to zero even with force still applied!

I retested the same setup yesterday but this time I fed the load cell’s output into a separate laptop running dedicated proprietary software and got a more closely resembling force trace to that of the SG-beam trace, so at least I have a solution. Y-axis in in N and is negative as it’s in tension.

regards,

Andreas

[AJost · Accepted solution]

The AC/DC coupling has to do a lot with your question. In AC mode static voltages will be fitered out and the 9234 will indeed only measure changing voltages. In DC mode, the voltage goes directly to the AD converter and you can detect also static voltages.

A minute of googling gave me the answer, that piezo electric load cell can only measure dynamic changes, as any charge will leak through the lowest resistance path and the signal will go to zero after a while. I guess that the other device you were using has a larger internal resistance (which is relatively small on the 9234), so it takes some more time until it leaks, but it also did on the picture you attached of your second test.

You can find more for example here under the headline

"WHY ONLY DYNAMIC FORCE CAN BE MEASURED WITH PIEZOELECTRIC FORCE SENSORS"

http://www.pcb.com/techsupport/tech_force

Andreas Jost

Applications Engineer

National Instruments