Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
OLi

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #1 

I have found like 3 cases where yu need to do balancing, not by moving the CG but moving the CR.
Current cases are gasturbines, first was a aero engine like JT8D and in final testrun it was sometimes not accepted in the cold end and they could only access the outer rotor system so 
when needed they but some weight in that to make that worse but the bearing assy was so flimsy so that moved the CR or maybe the shaft bending? Anyway you put a weight on a shaft with one speed and influenced the balancing of the shaft with the other speed.
They did that by a 3 run method or similar as they only had a tacho from the outer shaft.


Current case it is a industrial machine with similar feature but in the hot end.
I have tacho from both shaft and I can again access the outer shaft but to access the inner I need to split the machine eg separate the gas generator from the power turb and that is a pain.
So right now it seems we only need to improve the outer system so that is fine, experience say that it normally the inner system will get worse.

So in the future if we have bad luck, can we use a testrun w. testweight etc. on the outer system then calculate the response as normal on the inner system and put the calculated weight on the outer system? It does not seem to be really technically good as they are running different speeds but maybe it works anyway and I can't see why?

Worst case I will know one day.

 


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

Sr. Member
Registered:
Posts: 647
Reply with quote  #2 
Just curious, what does TP stand for? 
Quote:

I have found like 3 cases where yu need to do balancing, not by moving the TP but moving the CG.

OLi

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #3 

Sorry I was thinking in Swedish, TP is "Tyngdpunkt" eg center of gravity so I revised as above, maybe minimally more understandable.... Thank you for heads up.
Both cases above are 2 shaft machines with one or more common relative bearings where the inner shaft have a bearing common with 
the outer shaft and then to the frame. So in the current case we only have 1 sensor on the common relative bearing that we filter and measure
relative 2 speeds, one from each shaft.
If I add weight in one system inner or outer both vectors, unbalances are influenced that is confirmed in both cases above.
So normally we like to improve balancing in one system and live with whatever the result will be for the other system/shaft also making it slight worse.
It would be nice to have a estimate of where the system goes that we don't target for the balancing effort so we can optimize somewhat.

Current case is a easy version maybe, we need to balance the shaft/system that we can put weights in so that should work.
Now the bad shaft is 4-5 mm/s and the good 1-2 mm/s so we have some margin at full load.
We will likely see what happens this weekend. If everything works out we will have a chance to enter a test
weight and likely also some balancing runs if needed. Holding thumbs. We will see where the other shaft goes.

So we put weights in one plane, measure in one place but measure at 2 speeds and get 2 vectors for 2 speeds and both are influenced by adding weight.
Same apply if we split the machine and but weights in the other system/shaft and then it is more important as splitting machine is a pain.


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

Sr. Member
Registered:
Posts: 1,915
Reply with quote  #4 
Well lesson learned so far is that if you change shaft properties by working on shaft to fit new bearings and lighter blades due to a small bang or rather 3 due to feeding it to much hydrogen that blew up combustion chambers and more. It will make the critical speed increase and if that is close to full load speed that is normally running slightly above critical speed and now closer. Then you increase the unbalance sensitivity in this case like 20 or maybe 30%. I thought I knew that if someone had told me the rotor modifications. That was all we had time to learn then it was back to operating.
__________________
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.