0
$\begingroup$

I have a 2D vibration sensor that can only measure in the X and Y directions, and it is mounted on top of a motor that is fixed into place, as shown in a rough sketch below.

rough depiction of vibration sensor on motor

Would I be able to work out the vibrations in the Z direction based on a cross-correlation of just the vibration in the X and Y directions?

$\endgroup$
1
  • $\begingroup$ No -- X, Y, and Z are completely independent so long as they are all orthogonal. Unless you are dealing with a system where 'cumulative' properties apply. For instance, the cumulative mass participation(s) of an object fundamental modes in three axes. It seems we need more information. $\endgroup$ – Austin Prater Jul 16 '19 at 1:23
2
$\begingroup$

Think it through.

Put a penny on a table. Slide it away from you and back -- that's x. Now slide it left and right -- that's y. Now pick it up, without moving it forward, back, right or left -- that's z.

If you're sensing x and y, will those change if you move the penny straight up?

If your sensor senses x and y, will it see z in any way shape or form?

$\endgroup$
1
  • $\begingroup$ No, the sensor should not be able to work out the z-axis based on x and y-axes alone. Thanks for the confirmation. $\endgroup$ – plu Jul 13 '19 at 2:26
0
$\begingroup$

What vibrations are you interested in?

Are the vibrations somehow interlinked - what else beyond the question you placed to us do you know?

If you can constrain the question more, the situation may change. If you cannot, then @timwescott's answer is obviously the correct one.

$\endgroup$
2
  • $\begingroup$ I'm doing some verification on some previous paper publications, where one party claims that a cross-correlation of the X and Y-axis determines the rotor shaft speed, while another claims that simply the Z axis is enough. In that specific case, either the X or Z-axis would work; @TimWescott 's answer is enough to confirm that in principle, any cross-correlation of X and Y should have nothing to do with Z. $\endgroup$ – plu Jul 15 '19 at 19:27
  • $\begingroup$ Sounds like your comment would be well placed in the question itself. $\endgroup$ – akauppi Jul 16 '19 at 8:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.