Vibrating reed frequency meter- or alternative?

[FreddyZ]

I have a fan of diameter 4 ft or so, the outer panel of which is vigorously vibrating, but only in some few installation configurations. I would like to know the frequency of the vibrations that are being observed in order to help devise a solution. I recall seeing little boxes years ago at another job, in which reside various reeds which are tuned to a particular frequencies. Holding the box on the part in question will then excite some of the reeds at their natural frequency, which is labelled for your convenience. I cannot find such a thing for sale now with a search of the internet. Maybe there's a "better" electronic means to map the frequencies of your vibrating equipment? We do have a Balmac 200 accelerometer which reads out mils of motion, g's of acceleration, and velocity in inches per second, but it seems to be designed more for shaft and bearing analysis, with a massive sensor head [which itself kills half the vibration] and limited motion detection range on the order of 200 mils. We have more like 20mm [750 mils] of motion at, I would guess by eyeball observation, something on the order of 8-20 Hz, with a slower superimposed frequency as well.

What's a good way to get a handle on the frequency and amplitude of this vibrating panel? I tried a stroboscope but the lighting isn't conducive to that process, and the scope's adjustment is rather coarse, and so it skips past the desired flashing frequency.

So far the most viable solution to the excess vibration is to add mass to the outside edges of the panel, moving its natural frequency away from the operating frequency. Part of the problem is that the fan is mounted to a structure that readily transmits vibrations from one fan to its twin and vice versa. This was readily observed by inducing vibration by hand, with the fans off. Shaking one at its natural frequency had both of them bouncing very nicely in no time.
-Freddy

 

[btrueblood]

The little boxes are (were) called reed vibrometers...but yeah, good luck finding one today. There are electronic versions of the things, which use a small accelerometer, but not sure if any of them do more than the unit you have (i.e. do some kind of Fourier analysis).

A smart phone (Android) has a tuner app that might pick up a resonant frequency. I tried to find an app that would use the internal accelerometer and do FFT on it, but didn't see one in a minute or two of looking...

 

[Strong]

Here is one source if you want to play with a reed vibrometer:

http://www.mcmaster.com/#reed-tachometers/=oy48qp

Walt

 

[FreddyZ]

Strong -

Yes, thanks, that is what I had in mind.

I have no clever phone so apps are not going to help. Unless I get a 200/mo boost to cover that sort of thing.

 

[electricpete]

Maybe there's a "better" electronic means to map the frequencies of your vibrating equipment? We do have a Balmac 200 accelerometer which reads out mils of motion, g's of acceleration, and velocity in inches per second, but it seems to be designed more for shaft and bearing analysis, with a massive sensor head [which itself kills half the vibration] and limited motion detection range on the order of 200 mils. We have more like 20mm [750 mils] of motion at, I would guess by eyeball observation, something on the order of 8-20 Hz, with a slower superimposed frequency as well.

I assume your plate is vibrating highest at the center and much less around the edges. Attach your measurement device at the edges and it will disturb the system being measured much less. There are a variety of ways to mount an accelerometer - prep the surface and epoxy on a quick-connect… or use a magnet (massive, so put it at the edge),... or use a stinger. Since you are not looking for very high frequency content any of these should be adequate.

There are plenty of smartphone apps. but can be a bit of a challenge because it needs to be rigidly mounted to the vibrating thing. If you want to tighten it on with a C-clamp, you'd probably get a good measurement but might end up with a few more cracks.

So far the most viable solution to the excess vibration is to add mass to the outside edges of the panel, moving its natural frequency away from the operating frequency.

For mass addition, I'm guessing it would be most effective at the center of the panel (lowering resonant frequency if that's your aim). In general (not 100%), you will have better luck with stiffening to increase resonant frequency than adding mass to reduce. Something like an angle iron welded across the panel.

Part of the problem is that the fan is mounted to a structure that readily transmits vibrations from one fan to its twin and vice versa.

Are these mounted on vibration isolator springs?


=====================================
(2B)+(2B)' ?

 

[IRstuff]

"Unless I get a 200/mo boost to cover that sort of thing"

Virgin is only $30/mo, but you'd buy the phone outright:

https://www2.virginmobileusa.com/phones/catalogPurchase.do?quantity=1&sku=885909634019

http://www.virginmobileusa.com/shop/cell-phones/samsung-galaxy-victory-4G-LTE-phone/features/

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[Tmoose]

tunable reed tachometer less than 30 US bucks.

http://www.amazon.com/Briggs-Stratton-19200-Tachometer/dp/B0044AV5VK

Unwind / extend the wire until the end, brightly painted like an angler fish's lure, becomes a blur.
Best for ~constant frequency phenomenon. The body must be in contact with the vibrating thing. The wire is round, so responds nicely to vibration in any direction but "along" the wire.

 

[FreddyZ]

This is a 50" panel fan with more or less an "X" strut between motor/ bearings and the panel/ orifice [struts are steel rectangular tube perhaps undersized at about 19x50mm]. If mounted by the panel edges, say to a building, all is well, no excess vibrations from this type fan. Other sites using nearly identical mounting bracketry do not have objectionable vibration.

However, in this application, at this site, observations reveal that the motor and the prop/ shaft tend to be quite still or have very little motion [maybe 3-4mm at the motor], while the orifice panel shakes vigorously at the outer 4 corners- movement on the order of 15-20mm or more. The primary motion at the corner of the panel is along the axis of the fan, as if the panel is twisting or perhaps moving axially as a more rigid assembly. Evidently, then, the fans' struts are serving as the spring, with the panel as the vibrating mass.

Mounting: each fan is secured to an "L" shaped steel tube [about 3" diameter and 3ft x 6ft legs] by a pair of brackets at the struts near the center of the fan, maybe 10" apart. The fan to building steel "L" tube is mounted to, at this site, a wooden post about 6x8" and perhaps 20 ft between floor and rafters. The fan pairs- one on either side of the post, blowing in the same direction- are secured to the post right about the center of the post's free span- the most flexible and bouncy spot of the post. With fans off, one can induce a harmonic frequency vibration to one fan by hand [jiggle or bounce the fan, in plain English], and the twin fan on the other side of the post immediately responds in similar vibrations. One can feel the wave go back and forth from one fan to the twin. Surely that conduction of vibration contributes to the problem.

At the site, changing props removed about half the vibration from some fans, but not others. Adding 2-3kg mass to each of the lower corners of the panel removed about half the vibration in every case, and resulted in operation deemed acceptable by the site manager.

Back home, we can get similar but not quite as vigorous vibration on our test fan, using a prop removed from one of the worst shakers at the site, and we find that adding mass to the panel corners [about 5kg per fan] effectively reduces the vibration- to an amplitude more on the order of 2-3mm. We don't have a similar 20 ft span of wooden post to try mounting the test fan to- we have only the steel post of a pallet rack, which has totally different vibration transfer characteristics of course. But, we are able to get SOME similar vibration of one fan's orifice panel, on the order of 10mm amplitude. Part of the struggle is that, without an *identical* mounting configuration, we can't be sure that any solutions we devise will behave similarly at the site.

Changing or stiffening the struts for a few fans at one site is unacceptable to management. Bracing the building's post is not acceptable to the site manager. To help reduce vibration transfer through the building's post, it was suggested to flip the "L" tube, so that the building post end of the L tube is above fan center rather than below, moving the closest post contact point up about 4-5 ft up, which would place the load and contact much closer to the rafters, where the post should be orders of magnitude stiffer. This suggestion was not warmly welcomed. Nor was the suggestion to erect a single similar wooden post for testing under more nearly identical mounting conditions.

At this time, the favored solution is to secure a sizable plate to each of the lower corners of the panel. Adding a stiffening member to the panel is difficult unless we can give up on tying it structurally to the fan's struts, and make do with this added member just stiffening the panel alone. It was suggested to weld the corners of the panel- they have a lip about 30mm bent at 90 degrees to the panel of course. This has not been tried as yet.

 

[Tmoose]

can you share pictures of a similarly constructed fan?

Have you determined the frequency(s) yet?
The Balmac's lower frequency range is probably a couple of Hz.

 

[MikeHalloran]

If you can find or make one tuned to the correct frequency, maybe you can bolt a damper to each free corner of the orifice plates. Something like this might do:

http://en.wikipedia.org/wiki/Stockbridge_damper

Mike Halloran
Pembroke Pines, FL, USA