Checking for Shaft Eccentricity with a Strobe

[rstock]

I have a shaft that I suspect may have an eccentricity problem, and I am wondering if there is a procedure for checking for eccentricity with a strobe light. It is a three piece shaft through a screw pump. We could use a dial indicator on the straight stubs at the ends, but the centre section with the screw can't be measured that way. Have you ever checked for eccentricity using a strobe? How did you go about it and were the results helpful? Thanks!



rstock

 

[GregLocock]

use a tacho pulse 1/rev and then use the freq delta on the strobe to give a slowly rotating view of the 1/rev shaft deflection.

yes, I have used a strobe for that sort of thing, main problem is that the deflections have to be quite large to be visible, so they only really work at low speeds.







Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[zappedagain]

I like GregLocock's idea, except I'd use 2 pulses/rev. That way as you vary the delay (rotation angle) you may be able to accentuate the ecdentricity. For example, if you think of the eccentricity as a cosine you'll want to strobe it at zero degrees (+1) and 180 degrees (-1) to observe it; if you strobe at 90 degrees (0) and 270 degrees (0) you won't see a thing.

 

[rob768]

What about a pulse at half speed? You see both maximum delfections. With one or two (or any multiple), you just see the shaft in the same position over and over again

 

[TPL]

I cannot see how more than one pulse will achieve anything - Greg L says it all when suggesting that you slightly de-tune the strobe frequency away from the running speed so that the shaft appears to move slowly allowing you to see the eccentricity BUT only if it is large enough to see.

If you are looking at less than 0.100inch eccentricity, then, you might want to consider using a proximity probe: look at the display on a scope to determine the value you are looking for with a good degree of accuracy - if you have more than one probe you can get a measure of rotor deflection shape.

 

[ivymike]

how about using the across-the-room shadow of the object as strobed in the manner GL suggests? (deflection magnified by trig)

 

[electricpete]

Interesting idea for the shadow. Two thoughts would be
1 - you need the right geometry to be able to do it.
2 - you would presumably need a strobe resembling a point source. A wide source like our handheld strobe would create a blurry edge of the shadow (draw the lines from each side of the light past the edge of object to see width of shadow edge).

I would definitely endorse slight detuning from 1x to obtain slowly moving image... rather than trying to strobe at twice per rev. I have looked at things strobed at 2x before and I see two images but my brain has a hard time separating them. Difficult to tell what you are really looking at (try it).

rob - strobing at half speed (once per two revolutions) - makes even less sense to me. You would see the same image as strobing at 1x speed, except the image is not refreshed as often. Maybe you are thinking of strobing at twice speed (twice per revolution).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[GregLocock]

Hmm, I like the idea. I imagine an LED torch would be bright enough to act as a point source in a dark room. Then lay a plank at an angle away from the shaft, and the shadow will dance up and down the plank, exaggerating the motion.

Interfacing an LED to the strobe's trigger circuit shouldn't be too hard, otherwise build a delta F trigger using an F to V and V to F chip (I think that's the cheapest).


Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[ivymike]

how about just masking the output from the strobe, with a slotted or pin-hole template, and using a reasonably well lit room?

 

[rob768]

Electricpete, you are right, my mistake of answering without thinking.

 

[MikeHalloran]

Greg's idea can be enhanced a bit by adding a simple convex lens between the shaft and the plank, creating a projection microscope. The motion will be reversed, and magnified as the lens gets closer to the shaft, or the plank gets farther away.



Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

ivymike, none of the strobes I've used has been especially bright in the first place, but it would be a simpler alternative.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.