IEPE force measurement scaling/sensitivity problem

[persyMG]

Helo - my problem are inconsitent force measurement results from two modal hammers:

Hammer 1 – PCB 086C03

range = +/- 2224 N

sensitivity = 2.458 mV/N

Hammer 2 – PCB 086C04

range = +/- 4448 N

sensitivity = 1.239 mV/N

for tracking down the problem I additionally use

Force sensor – PCB 208C02

range = +/- 444,8 N

sensitivity = 10,83 mV/N

When I measure impacts from both hammers (I hit the force sensor), the results from hammer 1 are as expected but from hammer 2 are about half of the expected force (See figures, white – hammer 1, red – hammer 2, green – force sensor).

I’m getting this problem when using DAQAssisant (code attached), huge LV application using DAQmx and small application written in LabWindows/CVI so it might be rather DAQmx problem than LV or CVI. In LV the hardware used is PXI 4496, in CVI its cDAQ 9230.

Please help - what am I missing or doing wrong?

(note: when I test hammers in NIMAX test panel as voltage signal the response from hammer 2 is half the voltage from hammer 1 – and this is as expected. With the same force applied, the lower sensitivity sensor should respond with lower level of signal. This suggests that both modal hammers are working OK and there is a problem with internal scaling from voltage to force values.)

[Henrik_Volkers]

Seems that the scaling information is not correctly applied, as you already suggest.

Unfortunately I can't open your vi... my LV installation is to old ..

[persyMG]

Here is a version saved for LV15 - is it OK?

+ a screenshot of setting tabs for Hammer 1 and 2.

I played around with settings but I can't get any idea of what may be wrong apart of wrong / no scaling applied by DAQmx. 😞

The question is - what is the basic reason of the problem?

If I understand - than I'll manage to fix it.... or find some workaround, and it must be some "general" solution that works properly in all cases - I can't just multiply the result by 2 😞

[Henrik_Volkers]

I would capture raw voltage signals and do the scaling by myself (in LV)

The DAQ scaling of the expected voltage range needs to done by 'by hand'  ... but shouldn't be a problem.

If you plan to use LabVIEW to measure and process the data   make yourself familar with the basic DAQmx vis... the express vis hide a lot ... just play around with the provided examples.

Impact testing...   do you have acess to the sound and vibration vis?

[persyMG]

I'm familiar with DAQmx - I use it a lot in LabView and recently in LabWindows/CVI so managing raw data is not a problem for me. I used DAQAssistant to make sure that there is no mistake in my own LV or CVI code as DAQAssistant creates it's own code which I assumed to be 100% correct.

If I decide to directly process raw data, there is no vi / CVI function for creating voltage IEPE channel - sholud I configure standard Voltage AI and than, using property nodes enable IEPE excitation?

Yes, I have Sound&Vib package.

[persyMG]

It took some time, but finally the problem is solved. It come out that it was a faulty hammer. It was quickly fixed by my local PCB dealer and now everything works as expected.

Graeat thanks to both NI and PCB supports for their suggestions and help.