Problem Statement: Need help to isolate "over-range" errors which occur intermittently during vibe measurement
*Objective: Characterize HALT chamber vibration range - Measure acceleration PSD on vibration table of HALT chamber and compare to setpoint
*Instrumentation:
-Accelerometers: 5 single axis, PCB Piezo, +/- 500g range, 10mV/g sensitivity, all attached to table (next to each other) with bees wax
-DAQ: 8-channel Wavebook 512 by IOTech; eZ-Analyst software (both > 12yrs old)
-Configuration: Freq range 10-5000Hz, selected correct accel input range, sensitivity
*Observation:
-At 5Grms - all 5 accels read within 0.3Grms of each other and setpoint
-At 10Grms - all 5 accels read within 0.9Grms of each other and setpoint
-At 15Grms - start observing "over-range" errors on few channels
-At 20Grms - all accels have "over-range" error
-Swapped some accels; error reproduced intermittently/at higher vibe level; Also occurred on 3-axis accel
-Changed accel location to different area of table. Again mixed results - Some channels over-range and sometimes dont.

Would like ideas on how to isolate issue to instrumentation or software. Note: IOtech was bought by National Instruments but do not have any development/support on Wavebook.

I'd try fixing your signal leadss to the table with more care.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Incidentally I'd be a bit leary of beeswax for a test rig, especially when you aren't happy with the signals. yes, i know it has the frequency response.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Did this setup EVER work? You're supposedly using a ±500g accel with ostensibly ±5V output feeding an ADC set to what gain?

TTFN
FAQ731-376: Eng-Tips.com Forum Policies

In my opinion, it's not a good news. This certainly indicates that your accelerometers are soon out of order. Actually, the lifetime of a PCB piezo is around 6 years (from my own estimation ;) and eachtime, I got the same type of fault before accelerometer fails: over-range when level is higher than around 10g...

I have both a WaveBook and ZonicBook. The 8-channel Wavebook does not have anti-alias filters, so this may contribute to overload issue with out of frequency range high values. I had to add a separate 8-channel module with ICP and anti-alias filters.

Walt