unbalacing without phase probe 3

[molossi74]

Does someone know a method or mathematica formula to balance a shaft without using a phase probe ...but just trial masses?
Many thanks !
Molossi74

 

[GregLocock]

yes, it is possible, I had to do it at uni.

Sorry, I cannot explain how to do it, it sucks. Nobody would do it that way in the real world. If I really had to, I'd use trial and error.



Cheers

Greg Locock

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

 

[TPL]

There is a method known as 3 point balancing (or similar). Briefly, keeping the same speed throughout, you measure the 1X amplitude with no weights. Fix a 'suitably sized' trial weight at 1 location and measure the 1X amplitude. Stop the machine and move the weight 120 degrees and remeasure the resulting amplitude. Move the weight in another two 120 degree steps and measure the 1X vibration at each step.

You then use a little geometry to determine the amplitude and size of the required correction weight. Do a google search for details.

However, as mentioned, this is a really poor way of approaching a machine balance. Setting up a tacho/phase reference pulse makes things so much easier: you can reduce the number of runs required to obtain a balance and with a minimal understanding of rotor dynamics you can get a reasonably good estimate of where the correction weight should go. This is important if you are working on a badly balanced machine - a trial and erro approach could result in placement of the trial weight on the existing heavy spot making a bad situation even worse.

It is easy to make errors when balancing - by using 1X amplitude and phase obtained from a tacho/phase reference, you can work backwards to figure out what is going wrong and try to correct things.

 

[Tmoose]

Sometimes called "4 run balance" because it takes 4 test runs to get enough data points. I'd start and stop the device a few times to ensure the data is repeatable. You still need a filtered 1X signal. What instrument are you using that prevents getting a phase signal?

A method was posted by William Foiles here in the last year.

http://maintenanceforums.com/eve/forums/a/frm/f/3751089011

http://www.spintelligentlabs.com/SL-docs/4runtutorial.pdf

 

[sms]

"However, as mentioned, this is a really poor way of approaching a machine balance."

For cooling tower fans, or fin fans, the 4 run method works really well, and is still often used, as it reduces the influence of other operating cells in the tower.

-The future's so bright I gotta wear shades!

 

[dgallagher]

Victor Wok's book Machinery Vibration: Balancing, explains 4 run balancing without phase:

http://www.amazon.com/exec/obidos/ASIN/0071348611/reliabilitydirec

Dave G.

http://www.ReliabilityDirectStore.com

 

[Tmoose]

Even When balancing a cooling tower with phase, if there is enough blade/hub axial runout the single plane balance solution will make NO sense (since the unbalance is 2 plane or couple). Usually not a problem (maybe 1 in 100), but, just in case you suspect you may be impaled on the horns of a dilemma .

 

[GregLocock]

The other thing that annoyed me about it (other than the geometrical construction we had to deduce to get a solution) is that you have no real way of telling whether your trial mass is big enough, without doing three runs. Or is too big, which takes two.

Cheers

Greg Locock

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

 

[molossi74]

thanks to everyone

 

[WCFoiles]

Last year in the 2006 IFToMM I presented a paper that extended the methodology of balancing without phase (amplitude only) to using different size trial weights, multi-planes and/or speeds (not very practical because of Number of runs required), numerical solution method with multiple weights (3 or more), and an analytic solution (for the 3 trial runs).

This can be found in the proceedings of that meeting.

Regards,

Bill