http://www.mt-online.com/compo...nt.html?directory=90

Check out the 3rd picture, the caption says "A test for the Westfalia: balancing a nickel while running; that's precision."

Enjoy!

Laugh if you want, but I assure you that if they can stand a nickel on their centrifuges while running, then they have little to worry about.

Westfalia was turned off for the photo, had dealings with there centrifuges before and that photo is impossible.

This may be a repeat question, but does anyone have an idea how low vibration has to be in order to balance a nickel on edge?

A manager told me he balanced a nickel on a machine. It is a vertical 1800rpm motor vibrating 0.35ips horizontal and 0.12 ips axial/vertical (I imagine both the horizontal and vertical vibration would be relevant.) I didn't see it with my own eyes. My own attempt to balance a nickel there was unsuccessful.

I have done this once before...
I used some "adhesive" as the couplant between the transducer (nickel) and centrifuge

I like Pete's story. It's real world experiences along with science and technology that get my attention.

Artsy stuff like balancing a penny or hey Rocky, wanta see me pull a rabbit from my hat don't take a backseat to proven science. Oh, I know some things are believed when seen; dating a lady magician once - we were driving down the road, she grabbed my leg and I turned into a hotel.

Please; no comments on vacancy - no vacancy ;-P

quote:

Originally posted by electricpete:
This may be a repeat question, but does anyone have an idea how low vibration has to be in order to balance a nickel on edge? ../

I was thinking of trying this out with our shaker (adjustable frequency / amplitude). Any preference on frequency and units?

Just thinking out loud, perhaps set a constant frequency and then increase the amplitude until the nickel is disturbed, then change the frequency and repeat. I could generate a stability curve.

If I had to guess I'd say that the acceleration/g level is more relevant to nickel balance test than the velcoity/ips level. If you do end up tabulating results results vs frequency, that would answer the question which is more relevant.

1800 would be an interesting frequency from my perspective.

Now what are the specs for the levelness and surface roughness of the table? (just kidding).

I believe what you will find is that the most important parameter is frequency as it applies to displacement. The 'nickel' test was very popular when the fastest running machine in the plant was 900 rpm. You can probably balance a nickel on a 30,000 rpm spindle drive with unacceptable vibration levels in velocity or better yet unacceptable levels in acceleration.

Bear in mind that when the nickle test was in extensive use, there was no acceleration measurement as the standard sensor was a moving coild velocity pickup (IRD 544). There were also no spectra or waveforms as the instrument of choice was the IRD 350 (and similar). When you said "vibration" about the only thing people thought of was "it must be out of balance" since that's all most folks knew about vibration.

Still, the nickel test is probably a very good indication of balance condition.

I used to work for Westfalia as a field engineer and was "taught" the nickle test when I started. It does work somewhat, but there are too many variables for it to be of much value other then the gee wiz factor.

A centrifuge like that usually runs in the 1 mm/sec range and if running rough would be 2-3 mm/sec. The nickle at best lets you know if it is really bad, but an experienced person can easily tell that just by hand. It's been a number of years since I did that, but I seem to recall it was in the 3-4 mm/sec range at the centifuge frequencies when the nickle would fall. I would be interested in seeing the results of the shaker test.

What happens when the nickel falls off and into the coupling guard?

Results: Balancing a Nickel on a Shaker

Short Answer: About 1 g (0-pk)

Long Answer: See attached power point slides. I used a Gilchrist 4000 Portable Vibration Calibrator (Shaker). I started with the nickel standing on the shaker head at a low amplitude with a frequency of 30 Hz. I slowly moved up the amplitude. At around 0.6 g's I could easily see visible motion of the nickel. I was able to get the amplitude up to approximately 0.9 g's before it came unstable. Note that this is around 20 mils pk-pk.

I repeated the exercise at 60 Hz and 100 Hz and got maximum stable amplitudes at approximately 1.1 g's in both cases.

The table on page 1 shows the vibration amplitudes in accleration, velocity, and displacement.

One big difference (besides using a shaker instead of an operating machine) was setting up the nickel prior to the commencement of the vibration. It is definitely harder to "balance" the nickel on a shaking surface.

The second big difference is that the vibration only has a vertical component. I suspect that the threshold for horizontal vibration would be much lower.

This message has been edited. Last edited by: Steve Ciesla,

Nickel_Shaker_Test.ppt (3,215 KB, 24 downloads)

Bill, we used to do that on our toy grinders when training balancing, falling down on the balancing wheel it makes a dent in a metal inner roof in a conference room but not a hole at 1 SEK coin and 3000 RPM. We stopped after that and yes, insurance would have covered but we were lucky. I know of a fan manufacturer still using this method for final factory testing at the machine end that do not vibrate... then with a 10 SEK coin despite some arguing, still doing it. Steve, I guess 1g+ should be expected, that´s when it should get airborne so your shaker is about correct. Next target must be to obtain release speed and get it into orbit Olov

Great experiement Steve. Thanks for sharing that.

Interesting as Oli said when we exceed 1g's peak downward acceleration the coin should lose contact. But I would have never guessed the results would be so tightly clustered around 1 g.

Similar test for horizontal vibration I imagine would be quite a challenge.

This message has been edited. Last edited by: electricpete,

I wonder how phase would affect the horizontal tests? I would guess that you could hit spots where it would withstand some pretty high stuff and others where it would go flying.