# Balancing Calculations @ Diffrent Speed

Dear All,

We have  Turbine Generator Set 10 MW Capacity.  Rated RPM : 3000

We analyse the Unbalance in  Generator Rotor.

During no trail weight we could not  Cross the  1400 RPM

But After adding Weight  of 160 gms. @180  deg. on Generator Front side we are able to cross the 2130  rpm Critical Speed. And could run  up to  the 2850  RPM.

But only capture the readings  @  up to 2500 RPM.

Now base on these 2 sets of readings  what should be the Calculated  Mass and angle  to bring down  the vibration levels @  3000 RPM.

Attached sheet  showing  readings  for  both trials @ different Speed

Readings are taken directly  from  Bentley Panel.

Vipul

Sr.Vibration Specialist

#### Attachments

Files (1)
Original Post

Do you only have the ability to tabulate amplitude/phase or do you have the equipment to generate polar plots?

()

Why wouldn't you balance at 1300 RPM at first? I would give correction of around 290 g / 75° (combine X and Y calculations). This should let you go to 3000 RPM where you can make another - final balancing.

()

Dear John & Becar

Thanks for  sparing time

We do not have any polar plot right now.

At present we have removed the generator rotor  for inspection ,but could not found any significant abnormalities except  two to three loose wedges.

()

My understanding is that wedges are on the stator. If that is the case, then there would be no affect on rotor unbalance. Apparently there was no visual indication of significant rotor deflection, such as a rotor-stator rub or bearing faults.

Your Trial Weight reduced vibrations, but Becar 's calculated Correction Weight was 130 grams heavier and 105 degrees different than the Trail Weight. I can't say whether Becar's solution is good or not, but I would be careful about reaching a low speed single plane balance solution for a rotor that operates above first critical speed (balance resonance). You might get lucky if the unbalance was mostly located near the rotor end where balance weights can be installed. I would also be careful about using proximity probe vibrations for balancing, unless you know that run-out (electrical and mechanical) is negligible or is slow-roll vector compensated.

Walt

()

Walt, my calculated correction angle is exactly 75° different from the trial weight angle (not 105° as you mentioned).

()

Becar,

You are using relative Correction Weight angle position instead of absolute position, and that at least falls in same quadrant as the Trial Weight!

Walt

()

Walt, I understand your point of view now. We were thinking the same way but expressing different. At 1300 RPM the trial weight didn't do much balancing improvement, so the calculated relative 75° correction angle should be reasonable.

()