Hi

You will have to excuse the lack of information, but there is a very good reason behind this.

Basically i have a 350 MW generator rotor that when you hold speed unexcited the vibration levels start to increase (this is with no real change in temperature). the interesting part is that the vertical and horizontal proximity probs shows the vibrations levels increasing out of phase E.G:-
OBH increases from 45 Um p-p at 180 degrees to 200 um p-p at 180 degrees
OBV increases from 20 um p-p at 0 degrees to 150 um at 0 degrees.

there is a good chance that this is operating on / near third critical.

so my question is why / how is OBH going out at 180 degrees from OBV when the probes are 90 deg apart?

again sorry for the lack of info, but my hands are tied here

In circumstances like this it is always a good idea to confirm the wiring. I have seen people looking at orbit/timebase plots based on miswired/mislabeled systems. An orbit based on an inboard vertical transducer and an outboard vertical transducer can leave you wondering what the issue is when it is simply a wiring error.

Having said that, the only time when the phase lag angles will be exactly 90° apart is when the major axis of the filtered orbit is directly in line with a defined transducer angle. If the major axis of the orbit is "tilted" and does not point to a transducer, then the difference between the phase lag angles will not be 90°, even though the probes are physically at 90°.

John from PA

Chris,

I would not be as concerned with the phase difference between the two probes as how the amplitudes and phases change over time. If you plot the vibration vector (amplitude and phase) over time and it shows a Spiral pattern, then you have a distinctive problem. I worked on a generator of similar size (60 Hz machine) that had Spiral vibration at OB bearing. It was operating close to a shaft critical and the excitation was from a rub induced by the exciter brushes. My advise is to look at the big picture of your data first before looking at the small stuff. By all means do as John suggested about verifying your probe connections and labels.

Walt

UK Chris, attached shows some examples of varying phase lag angles and the affect on the Orbit.

John from PA

This message has been edited. Last edited by: John from PA,

OTB_and_90_deg_relationship.pdf (61 KB, 51 downloads)

John,

I am unable to doownload your file. It is displaying goofy characters for the "degrees" in the file name. Could you change the file name and post again?
Thanks,

Walt

Thanks for the reply s, I also can not access that file.

I did miss out one vital part the problem is at the OB of the DE bearing of the generator, also the vector change is basically in a straight line. There is also no possibility of producing an orbit.

The only other information I can give is that this effects the run down through first but not second.

Again sorry for the lack of information, I feel very frustrated by how my hands are tied.

This message has been edited. Last edited by: UK Chris,

to be able to down load the file you need to add  after the degrees sign.

this may work

http://maintenanceforums.com/g...5B0_relationship.pdf

Reattached without degree symbol. By the way I just randomly threw in the numbers and they generated a pecularity in the data. TEST QUESTION - Anyone notice?

John from PA

This message has been edited. Last edited by: John from PA,

OTB_and_90_deg_relationship.pdf (61 KB, 26 downloads)

Thank's John,

Is that a new version of Bently software or another product?

Walt

Walt, it is a Bently utility that is built into the Machine Library CD. Unfortunately, although it can run as a "standalone" we can't give it out. Sorry.

John from PA

Test question - the orbit is in reverse precession. Why is that????

vibramac,

Yes, the orbit shows reverse precession. That is because the phase lag angle of the "X" transducer is defined as 210 deg and the "Y" transducer 120 deg. If you understand the basic definition of the phase lag angle and relative phase between two signals, it can be seen that this is reverse precession. Take the case where I put in the signals as "X" - 6 mils at 120 deg and "Y" = 2 mils at 210 deg. Obvisously here the peak for X at 120 deg is reached prior to the peak of Y by 90 degrees. This would be a forward precession. See the new attachement for this specific case.

This raises another question? As a machine ramps up can you sometimes observe both forward and reverse precession? The answer is Yes as a rotor commonly goes through both forward and reverse whirling. In particular, the orbit will generally be reverse if you are in-between split criticals. Next time you run up a machine that passes through several critical speeds watch the filtered 1X orbit go through its gyrations!

John from PA

forward_precession.pdf (14 KB, 8 downloads)

Chris,

quote:

I did miss out one vital part the problem is at the OB of the DE bearing of the generator, also the vector change is basically in a straight line.

Just to clarify, are you saying that the amplitude increases, but the phase angle at each probe remains the same?

Walt