Hi all,

I got two diferent results in phase from two balance software. I would wanna know why? where should i place the weigth?

very thanks,
Fabián

WRONG_OR_NOT.doc (328 Kb, 86 downloads)

My calculations match your first program, i.e. the correction weight should be 14.7 @ 245 degrees.

Steve

This relates a little to another discussion on calling the first angle 0. If we look at the phase shift from the originial weight we have the following.

215-50 = 165 degrees and
245-50 = 195 degrees.

These angles are complementary. That is one is measured against rotation from the 50 degree mark and one is measured with rotation from the original weight at 50 degrees.

It looks like an sign convention.

I think Bill is on target, since 165 + 195 = 360 degrees.
I was balancing a gas turbine many years ago. The engineer from the Swiss/German manufacturer was looking over my shoulders and then telling the plant manager that I was wrong. I found out that he took my phase values that were based on reverse rotation (CSI stuff) and entered them into his balance program as phase in direction of rotation. I guess the moral of the story is if you define your point of reference and the direction you are going, then everyone will know where you are!

Walt

Good point by Bill to spot that it is a matter of convention and the angles are the same if we measure in opposite directions.

But I think it would be an unusual sign convention to measure weight angle with-rotation instead of against-rotation as implied in the second solution, especially since the second software identifies "lag" convention for phase. If you tried solve a balance problem graphically using a common angle measurement for vibration phases (lag angle) and weights (measured with rotation), your graphical solution would be garbage (i.e. it would be incorrect). Does anyone use that convention?

I would tend to say the first one is correct and the second one is based on a convention that is out in left field (unless there are some conventions I don't know about)

I've seen programs that would allow the user to set their own conventions. I think the balance program should set out a convention (the one I use, of course - actually this is the most common -- lag all the way) and stick with it. The program should be clear on the input and the output as to what the angles mean.

A couple of years ago (ok add a zero +) I had to write a manual for a balance program that I wrote. I included, in the manual only, all type of phase conventions, strobe light (protractor on shaft, protracter on stator, protracter in either case with or against rotation, shaft triggers, using a strobe like a shaft trigger to align a mark on the stator to a mark on the rotor - I don't rememeber if there were others, but one had to make sure the user would enter the correct data.)

Hi all !

Getting different results on phase angle for the position of the correction weights is not a rare occurence.

I recall using an old IRD computor program that wanted corrections applied at a certain angle. And another, Dr Wook's Excell program, calling for the same weight but at another angle got me very puzzled.

I then remembered that for the IRD program, every reading of phase angles, be it for trial weight placement or final correction, ALL READINGS had to be read from the same end always.

In other words, suppose that all our readings while doing a in situ job were taken at the drive end, if it said 10oz at 270 degrees for the NDE, we had to remember that it was to be placed physicaly at 90 degrees when facing the NDE.

I'm not sure if I'm explaining this right.
Yet, Dr. Wook's Excell program confirmed the above.

In conclusion, the phase angles must always be referenced to the one end, either the DE or the NDE throughout the balancing procedure with whatever program or instrument one uses.

I may be wrong but...

MarkoLeo

You should reference the weights with or against rotation. The end (of the same shaft) does not matter, this way.

Also, rotation has to be part of the reference. There is no absolute phase reference for balancing. Rotation determines how the force from the weight is phased into the system.

Old Ird Program Says 215. I believe this one.

That's strange. Maybe there are more conventions than I am aware of.

I think most of us are familiar with the graphical solution:
CW = - O * {TW / [(O+T)-O]}
Where CW, O, O+T, TW are the following vectors:
CW, TW = trial and correction weights
O= Original vib
O+T = Vibration during trial weight run.

As far as I know, that graphical solution method only works in two cases:
1 - Vibration phase expressed as lag angle and weight position measured against rotation (common).
2 - Vibration phase expressed as lead angle and weight position measured with rotation.

It wouldn't work with a convention implied by the 2nd solution that gives angle 215.

Attached is graphical solution which gives correction weight at angle 245.

Is there any graphical solution that produces 215? If not, why would one choose a convention that cannot be interpretted graphically?

SinglePlaneBalGraphGood.xls (78 Kb, 54 downloads)

As culprit of the "confusion" I have followed this discussion with great interest (feedback on the single plane balancing part).

1) Yes, the calculation comes from IRD, my documentation comes from the times of the stroboscope.
2) Actually it is a graphical method, converted to analytical. The math from Pete
3) Bill hit it right on the nail, the correction angle takes the trial weight position as reference.
4) For calculation purpose the direction is Counter Clock Wise (it makes the math easier) also for the place of the trial weight.

It does not matter where the trial weight is placed, the correction mass will always be the same.

quote:

In other words, suppose that all our readings while doing a in situ job were taken at the drive end, if it said 10oz at 270 degrees for the NDE, we had to remember that it was to be placed physicaly at 90 degrees when facing the NDE.

When you are using a stroboscope, you will have this "mirror" efect. With a tachometer pulse, the phase reference would be the same for all readings.

Based on above points, what are the suggestions to improve the program (nomenclature etc..) I am also thinking about some graphical stuff to put in.

I wasn't thinking about strobes. You're right - I can see where other conventions may have been appropriate in the strobe days.

If you have strobe triggered from vib and move an unbalance in the direction of rotation, then the trigger pulse would occur earlier in time, and a reference point on the rotor would appear to move opposite the direction of rotation. That would require a different approach or convention than I'm used to.

ElPete,

In your equations TW/([O+T]-O), you calculate the angle by subtracting the angle of T from the angle of TW. Just add the angles (from the graph) of TW and and T instead. This is still graphical. Algebraically, this would be the same as using the complement of the denominator (We already assume that TW is written in lead angles. The calculation of T is in lead angles; we convert to lag by making the angle negative - a complex conjugate.)

My suggestion for the program would be to make everything lag angles and be clear about it.

Depending upon how one uses strobes (see preceding comments about marks and protractors on shaft and on stator), one can get either a lead or lag angle.

Something similar that I've seen is to plot all lag angles as negative and lead angles as positive (I suppose vice versa works too.). Then the graphical method works.

As I've said in my earlier post :

quote:

I then remembered that for the IRD program, every reading of phase angles, be it for trial weight placement or final correction, ALL READINGS had to be read from the same end always.

In other words, suppose that all our readings while doing a in situ job were taken at the drive end, if it said 10oz at 270 degrees for the NDE, we had to remember that it was to be placed physicaly at 90 degrees when facing the NDE.
I'm not sure if I'm explaining this right.

I realise that I've added some confusion amongths some of you.

OK, I'll try once more : In the above quote, I was referring to the good old strobe light system where a mark was put on the shaft and, placed on the housing with its center part cut out to clear the shaft, a typical polar diagram calculation graph sheet was taped on. All of the readings were taken from that end.

It is when one places either the trial or the final weight at the opposite end that it should be installed as if were viewed from the end where the readings were taken. Hence, the simple calculation as stated.

Gosh... not sure if I'm explaining this right...

In this case, there are'nt any fuss as to the direction of rotation, phase lag or none, since all readings were taken from the same end.

Hummmm...

MarkoLeo

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

When I studied, the discipline "mechanical vibrations" was feared because it was confusing

We had limited equipment, a strobe light, some test apparatus with small electric motors, frequency analysis (documentation in a foreign language), and the prof himself was a "monster" who did not hold himself at: a mass spring system vibrating at..

Instead of that he was coming with: A vertical turbine in a hydroplant suffered.. at ... speed
The Schaum Collection Books with zillions of examples, obviously did not help here.

Balancing practicum was done with a borrowed analog instrument, and if that was not available, you still had to balance it anyway, because the assembly was on springs. Criteria of balancing: the Nickel Balancing Standard

When I started working, for a long time I did not come close to vibrations until two years ago.
My confusion started when the equipment was hooked up (balancing class) , no strobe, no visual effects and the analyzer dictates what to do.

For those who never tried to tune a stroboscope to read the reference numbers, this discussion is confusing.

quote:

Originally posted by fabuzon:
Hi all,

I got two diferent results in phase from two balance software. I would wanna know why? where should i place the weigth?

very thanks,
Fabián

Fabian, que programas estas usando para hacer el balanceo?

From everything I saw you needed to move the TW 165 degrees with rotation which gets me to 245 degrees. I think the second program moves the weight against rotation and ends up at 214 which is incorrect. This is simply based on drawing it out. All my balance programs also agree with this.

RENESIS estos programas los descargue por internet, pues pense en disminuir el tiempo de ejecución del balanceo ya que el método grafico vectorial hasta ahora usado me toma mucho tiempo.

Saludos.
Fabián Buzón

Hola Fabian.
Gracias por tu respuesta.
Podrias guiarme por donde puedo descargarlos para hacer un método gráfico que requiero para mi tesis.

Muchas Gracias