Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
RustyCas

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #1 
Related to the balance quality spec post, do any of you know how the actual residual imbalance left in a rotor is determined?

Am I correct in assuming that it’s simply the product of the “required weight” calculated on the final balance check, multiplied by the radius of weight placement? For example, after attaching my final balance weight (or last material removal), if the software calls for the addition of 5 grams and I placed the final weight at a radius of 20 in, is the “residual imbalance” 100 gm-in ?

To me that’s the implication, but it doesn’t account for the required angular position of the weight relative to the position of my final balance weight. Or does that matter?

__________________
"The trend is your friend"
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #2 
Quote:
Originally Posted by RustyCas
Related to the balance quality spec post, do any of you know how the actual residual imbalance left in a rotor is determined?

Am I correct in assuming that it’s simply the product of the “required weight” calculated on the final balance check, multiplied by the radius of weight placement? For example, after attaching my final balance weight (or last material removal), if the software calls for the addition of 5 grams and I placed the final weight at a radius of 20 in, is the “residual imbalance” 100 gm-in ?

To me that’s the implication, but it doesn’t account for the required angular position of the weight relative to the position of my final balance weight. Or does that matter?


I’m a bit confused by your question but this could be a matter of semantics.  If you place 5 grams at a radius of 20 inches, "yes" what you have added is 100 gm-in.  But in theory you are adding the 100 gm-in to offset the rotor residual imbalance.  If you nail the shot, i. e., the vibration is extremely low, then that supports that the rotor residual imbalance was close to 100 gm-in and opposite the angular position of your shot.  The vector summation of the rotor residual imbalance and your shot essentially are cancelling each other.

Now, there is a part of most API specifications that address the "Residual Unbalance Test", part of the shop balancing procedure.  In this test, while the rotor is in the balancing machine, a known weight is added at six positions and the result are plotted.  This, in the words of API, "leaves no doubt as to the amount and location of the assembled rotor's residual unbalance."  I believe the procedure is discussed at length in API RP 684.
OLi

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #3 
If your last run with your IRL added weights result in a new weight calculated to be less than the required weight according to the ISO class defined or other definition on the spec or drawing all is good. According to German rule of thumb you should for safety go to half down to the next better class but that may be normally seen as overdoing it here but I have seen it in Chinese belt and coupling parts so they may be trained like that sometimes.
I see no requirement of having the residual weight in any specific angle according to ISO but in special cases for special rotors like gas turbine stacked rotors it is sometimes of great value to have residual weight at opposites angels on each disc if control can be kept up to stacking the rotor. In some gas turbines it may increase the process total yield after test run of complete machine significant, like 20% or more.

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

Sr. Member
Registered:
Posts: 670
Reply with quote  #4 
I had the understanding that residual unbalance was you "final prediction" multiplied by the correction radius, if you added 43 grams at 20in (your final correction), check with another run, this time the trim correction is 5 grams, your residual unbalance 100g-in.
OLi

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #5 
I think that what you measure your acceptance limit in does not matter, weight on specific radius, gm-in, gm-mm, distance center of gravity to center of rotation in some distance measurement. It is the same unbalance defined if numbers convert correctly as they should. The system I currently use state 3 various numbers as above to aim for and it print them on the certificate for the selected setup rotor according to selected ISO class, in my case that is metric. It may be convenient to aim for the same definition as stated in the drawing that often is the same as some standard ISO class like 6.3 but for what designer only knows not always. If there are any such numbers to be found on the drawing that is...... IF there are nothing stated and nothing can be retrieved I use class 6.3 unless it is some some special rotor or the requirement is "as good as you possibly can".
To make it complicated in the case of a 2-plane balancing and a not axially symmetric rotor the acceptance according to the ISO standard formally do depend on the distribution of rotor radius so the "heavy" end have a higher acceptance level and the lighter end have a tighter spec....... So why make it easy? There are some logic for that though. I rarely encounter such rotors but some do it all the time.

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

Avatar / Picture

Admin
Registered:
Posts: 1,810
Reply with quote  #6 
So on the part that kicked off this discussion (313 lbs, 900 rpm service speed, G6.3 required) our final additional required corrections for the 2 pieces we balanced on the IRD-B50 were 1 gm on the first piece and 5 gm on the other, at a radius of 22.371 inches.  If I'm calculating correctly, that's a final grade of G 0.4 and G 2.5.  An actual grade can also be calculated, but is that typically used, or do you just state the "standard" grades?
__________________
"The trend is your friend"
fburgos

Sr. Member
Registered:
Posts: 670
Reply with quote  #7 
from the reports I've seen, include initial and final vibration/unbalance readings and tolerance for each plane, for example.



balancing report.png 
PS. this was accepted because the service speed is 900rpm.

OLi

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #8 
I don't state grades achieved only that G 6.3 is withheld but nowadays having a complete certificate as above the end user can evaluate as he like. He can even set it up in a different machine like stiff versus soft, different drive system etc. and be surprised, the numbers are not the same. 
__________________
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  #9 
I am a bit late to this topic as I was away...but maybe this might help.....

I had the same definition in mind as fburgos........ but for an hour I couldn't remember where I saw that information.  In reviewing VI's Level IV handouts, I came across a paper by Ray Kelm that talked about residual unbalance.   You can find it here:

   https://www.kelmengineering.com/documents/AdvancedFieldBalancing.pdf

In addition to his take on the definition, Ray also talks about shop balance specs (I would agree with you in that you can equate to Balance Grade levels) but as Ray indicates, "...these specifications generally do not specifically define level of vibration or residual unbalance in an assembled machine at operation speed".   

One would assume the end user will take care of that upon final assembly?  [biggrin]

I don't know enough about OLI's comment....When I first started in vibration field, our facility used an IRD B-50 shop balance machine (i.e. soft-bearing balancer) which worked just fine. Eventually, we obtained a hard-balance machine. I can't comment on how different the results would be.  

Ron Eshelman wrote a paper on shop and field balancing in which he describes a procedure ("Charlie Jackson initiated?) to perform a Residual Unbalance Check - the purpose was to check the sensitivity of a balancing machine.   Essentially, a rotor is balanced to the "best it can be"  and then a known weight applied at various angle locations  (same radius) and a polar plot is generated to estimate the balance tolerance.

I recall a Westinghouse specification for large motors that implements this same procedure.  If not for all cases, perhaps for initial certification of a new service company.

Hope this helps
Regards
Jim P 




fburgos

Sr. Member
Registered:
Posts: 670
Reply with quote  #10 
This known weight at known radius using a polar plot was used in my previous job for steam turbine and turbochargers on soft bearing machines and hard bearings.
OLi

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #11 
Unbalance sensitivity can be tested according to either API or ISO or a mix that you agree upon, they are not exactly the same but similar and they include defining a resonance passage, adding testweight(s) and moving 180 DEG. Testing in several angle positions looks like the balancing machine testing procedure that is also standardized.
I am proud to say that the standard for unbalance sensitivity was among others initiated by my father long time ago and it eventually ended up to be a ISO standard. It is a way to sort of backwards make sure you don't run on or near a resonance as that normally will give unreasonable hi sensitivity for unbalance. It did give me a nice setup of work during 4 years about 35-40 years later to check a couple large of motors during various FAT stages.
I think that moving a rotor and as it was in the case I thought of adding shaft extensions to be able to run the rotor in a balancing machine is enough to move the center of rotation. How much and how good it is to have stiff or soft bearings is a tricky question it seems and also depend on rotor, speed etc. I can only say that personally being "backwards" soft machines give the operator a direct indication when balancing is not finished, it vibrate. In hard machines you don't get the same feedback unless it is real bad. I also get better resolution and lower noise level in a soft machine and can reach better results. I then overlook the ISO 9000 easiness to just enter numbers not knowing what the machine do and operational benefits and increased throughput found by using a hard machine because I don't care about that as I don't do such jobs.
Adding one of my sons better jobs, just to impress....

 
Attached Files
pdf balcertmetr.pdf (466.08 KB, 9 views)


__________________
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.