Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
OLi

Sr. Member
Registered:
Posts: 1,918
Reply with quote  #1 
Say you would balance a reasonable hi speed gas-turbine, 6 or 8000 RPM having both Bently probes and bearing sensor and you only have a trim balancing to be done in 1 plane, you can naturally calculate them separately and see what it gives but if you got a stupid idea like me and you run a 2 plane solution since you have that instrument and what would it give, something completely not useful?
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
ivibr8

Sr. Member
Registered:
Posts: 163
Reply with quote  #2 
OLI - Not 100% sure what is the difference between:

"Calculate them separately...."   vs   "....run a 2 plane solution..."

Regards
Jim P
spciesla

Sr. Member
Registered:
Posts: 107
Reply with quote  #3 
I'm confused as well.  When 1 read 1-plane or 2-plane solution, I think that you are referring to the number of balance planes where weights are added (or subtracted).

However, based upon your post, I'm guessing that you are doing a single plane balance, but doing the calculations independently based upon the shaft relative (proximity probes) and casing measurements. Are you getting significant different results between the two sets of calculations, resulting in a "Crazy balance"?
OLi

Sr. Member
Registered:
Posts: 1,918
Reply with quote  #4 

Normally you do get a significant difference btw. calc's from absolute and relative probes for all kind of reasons. So using a 2 ch instrument that do a 2 plane calculation and yes you need to cheat one way or another. It would anyway be a quick and dirty way to get weights calculated for both the Bently and Accel data sort of but the result may be a mixup you don't want. Yes the alternative is to do it properly, measuring and performing 2 one plane measurements and calculations and that in my case either require some fiddling that usually go wrong or 2 set of instruments that is an alternative. Idea just fell in to my head but it may be a stupid idea. It is not that many numbers so you can just do the measurements and enter them in 2 instances of the simplest balancing software. It is likely better and then look at the differences and decide what compromise to accept.

Maybe even better is to have a setting in the instrument where you select to have 2 single plane calculations and not a 2-plane. If the problem do get frequent that may be the best way.

I have had cases where control room displays are uncompensated Bently data, we would like to use compensated data if we use Bently data for balancing and then have absolute bearing data as a safety to make sure we don't get way out.


__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
MachDiag

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 130
Reply with quote  #5 

I often setup a 2-plane balance job, with just a single correction plane (CSI's FastBal II).  Get my reference run and trial weight run data, then see what the FastBal II predicts the end results will be.   If the predicted results are favorable, I continue using the single correction plane and pretty much always have good results (as predicted, usually better).  If the predicted results are not within tolerance, I'll go back and add the second correction plane to my setup and make the additional trail weight run to solve for true 2-plane balancing.     

More info on how from the old forum >> https://www.maintenance.org/fileSendAction/fcType/0/fcOid/399590942964733324/filePointer/399590942964879882/fodoid/399590942964879880/Simplified_Field_Balancing.pdf  

Not sure if this helps what your trying to achieve.

OLi

Sr. Member
Registered:
Posts: 1,918
Reply with quote  #6 

Yes I saw something like that where you did setup and use 2 sensors and the instrument calculated a single plane and 2 plane sort of prediction from the first test run so it would then advice that you could get away with a single plane solution or if data indicate that you really need a 2-plane balancing. That would be nice to have. I think I have seen it described somewhere so someone have done it. It is not exactly the case I am thinking of here but another likely better good thing to improve.

Thinking some more, you are correct FastBal II would do exactly what I liked to do the same that you do, only difference is that I would like to have different sensor types but that would maybe be possible..... Thank you for the input, that would work.


__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.