Accelerometer Sizing

[dhwilliams]

I have designed a machine for my employer which has a radial arm 70mm long which oscillates through 180 degrees taking about 5 seconds to go from left to right and back. The "payload" on the end of the arm can be seen to vibrate (probably at about 15Hz) and too much vibration can cause operational problems. At the moment I am using an M5 Socket Head Screw of length 30mm standing on its head (8.7mm)on the "payload" platform - which is kept horizontal throughout its travel - to pass or fail the vibration level. If the screw falls over it's failed! I want to replace this with an accelerometer and some electronics to be able to quantify the vibration in each machine but can't get my head round the mathematics required to select the range of the accelerometer. I'm looking at dual-axis devices and don't know whether to go for 1.2g (as in gravity!), 10g or what. Can anyone give me a ball-park figure based on my "falling screw" test?

Thanks,
Dave

 

[tomirvine]

The G level depends on the frequency and displacement.

For example,

1 inch displacement at 15 Hz = 23 G

0.1 inch displacement at 15 Hz = 2.3 G

The displacements in the above examples are zero-to-peak.

Your machine probably has other excitation frequencies besides 15 Hz. So you need to add some margin to your G level prediction.

Tom Irvine

http://www.vibrationdata.com

 

[dhwilliams]

The displacement is difficult to measure because the payload platform is moving in a semi-circle but I guess it could be up to .05 inch so, by your examples, a 1.2 G device might be a reasonable first attempt. I've requested a sample ADXL213AE accelerometer from Analog Devices Inc which should arrive here in England in about a week so I'll give it a try.

Thanks, Tom, for your prompt reply.

Dave