Accelerometer Calibrations

[Hawaii596]

Basic question about calibrating accelerometers.  When calibrating accelerometers of various maximum g force ranges, are they calibrated at differing amounts of applied g force based on range, or are they calibrated at a standard amount of g force and a mV/g measured?  I am spec'ing hardware, and there are some portable vibration calibrators on the market that apparently just generate a standard amount of g force, and it seems that even for (as example) 5000 or 10000 g accelerometers, the calibration amplitude applied may only be 5 or 10 g's, then a mV/g is measured against spec, etc.

Any inputs?

[HalAC]

I performed calibrations for a year or two on accelerometers when I worked at Arnold AFB's PMEL. We used one specific amplitude for an accelerometer and would sweep through a set of frequencies to check it's response against a known standard accerlerometer.

Hope this helps.

[CalMachine]

Here at ENI Labs, we start at 1 g-rms @ Whatever the stated reference frequency of the accelerometer is (generally 100 Hz, but sometimes 159 Hz) and sweep up and down in frequency to provide a frequency response curve @ 1 g-rms.  Provided data points are in "deviation from nominal" form from the reference frequency sensitivity value found at the start.

[Hawaii596]

This is pretty much what I have found.  To keep this from being commercial, I am looking into one of the new accelerometer calibrators, and that seems to be the norm.  I also looked at a few USAF T.O. procedures, and they say the same.  The new type standards are pretty nice, and not too expensive. One model has a USB where you attache a thumb drive, and semi-automatically store each reading into a CSV file, then transfer the thumb drive when finished.

[briansalomon]

I got the Vibration Research Accelerometer calibration system last year and am just about ready to proceed with the accreditation process.

I like the system fine and have noticed the software allows me to configure quite a few parameters such as dwell time at the selected frequencies, input filter parameters and response time parameters that can affect the measurement quite a bit.

The manual is quite good but I am wondering if anyone knows of a good handbook or text book that will help me understand how to configure these parameters for best (truest) response to my reference accelerometer.

[lonewolf]

my equipment is older Bruel & Kjaer I use an 8305 for my standard mounter on a 4809 shaker, using the 2626 conditioning Amp on the 8305, I have an old 2706 Power Amp to drive the shaker.

my test subject mounts on top of the 8305, and I sweep thru the range that I plan to use the Test Subject to make sure my curve matches my Standard and the datasheet that comes with the test subject.

All I really mess with is the Piezo types, I have several other bruel&kjaer instruments also like the 2511,1405, and a bunch of Misc adapters and standard accelerometers.