Does anyone have any experience with taking phase readings on the second order of turning speed on a 3600 rpm motor with an electrical frequency at 120 hz. When an attempt to take phase readings of the second order, the phase was unstable possilbly due to the electrical frequency at 120 Hz. This is on a 60 cycle system. I would think this same situtation would hold true for 3000 rpm motors that operate on 50 Hz. with an electrical frequency at 100 Hz.

Are you utilizing a photo-tack, laser, to identify 1x TS? Alternatively, are you using comparative phase. I have used comparative to look at two times, but while using a 1x trigger such as a laser tack or photo tack you are tied to 1x for phase measurements. Unsteady phase would indicate a resonance of some type I would think, but still at 1x TS because of the trigger. The 120 Hz would indicate an electrical problem is amplitude is excessive. Is your resolution sufficient to separate 2x TS from 2x line frequency of 120 Hz? Sorry for so many questions just trying to get a handle on what you are trying to accomplish.

Ronnie,
I might be able to help.

quote:

---

Originally posted by Ronnie Lynn:
...turning speed on a 3600 rpm motor with an electrical frequency at 120 hz. ...

---

Is your motor unloaded?

If it is loaded there should be some slip frequency i.e. 3600rpm nameplate machine running at 3565rpm loaded, slip frequency = 0.58hz and actual running speed = 59.42hz.

If you can look at a spectrum and with enough resolution you should be able to distinguish the difference between 2x running speed of 118.84hz and the electrical component of 120hz.

My recommendation is to determine actual running speed (either with your spectrum analyzer or with a quality strobe) as accuratly as possible and we will move on from there.

Hugh Hackett

Thanks for the replies. I can't answer your questions because I wasn't actually doing the testing. The person I work with asked me to post the question. I do know the motor was under a load and I feel pretty confident he would have separated two times running speed from line frequency. I do know that he suspected resonance and that he did bump tests to confirm it. I will have to ask him if the one times turning speed had a stable phase. I also know he was using a two channel analyzer and was doing comparitive phase readings. JB, you have done comparitive phase readings on the second order? Did it prove or confirm what you were testing for? I am curious, what were the results?
Thanks,
Ronnie

I"ve used an optical pickup and done absolute phase comparisons of the 2nd order. I've seen phase fluctuate all over the place because the measured 2x amplitude was low and a solid phase reading couldn't be obtained.

I've seen the same thing with Differential phase, where you don't use an absolute reference on the shaft, but use another probe as a reference to see the difference in phase from one point to another. In this case the 2x amplitude at the reference point wasn't strong enough to provide a solid reference to determine the phase of the second point.

That's the best I can do with the info at hand.

We suspected a flexure mode in the bearing mount driven by two times TS in the horizontal direction. 180-degree shift on the unit confirmed. We isolated the anti-node points and stiffened to raise above two times. Our unit allows a print out of all frequencies above a certain value with degrees of phase; this also depicted a 180-degree shift. Hope this is of some value.

Mr. Hackett:

My unit allows with a photo-tach to reference a 1x trigger. Would you enlighten me on how you are able to trigger off a two times turning speed signature? Does your equipment allow this? I have used sync-time averaging when balancing for a good lock on 1x, but still held to a 1x trigger. Your information would be quite welcomed.

IN the first case I was speaking of triggering off of the synchronus speed and having trouble getting phase on the 2nd order because of low amplitude. Bently Nevada ADRE or CSI.

In the second case I'm talking about the Cross-Channel Phase feature of the CSI 2120 2 channel hand held analyzer. This feature lets you enter the freq. you want to trigger on and it will give you a difference in phase between that freq. peak on the channel 1 probe to the freq. peak on the channel 2 probe. In this case if either peak is weak or fluctuating you can't get a steady phase.

Thanks for the input guys. I will pass this on to David, the person who asked me to post the question. I do appreciate you taking time to share your experience and knowledge.
Thanks,
Ronnie

My name is David and I am the one who initially posted the question. I am using a CSI 2120 dual channel analyzer there is a resonance problem on the motor base and what I was trying to do was to take multiple phase readings on the base to determine the movement of the base so it could be stiffened in the right places to move this resonance frequency. When I attempted to take cross chanel phase readings whith two accelerometers my coherence would drop almost to zero which was an indication to me that my electrical frequency of 120 hz was effecting my phase readings. When I attempted to take phase readings whith a photo cell the phase readings would continue shift 360 derees. I did take a high resolution spectrum and had a two times component along with an electrical frequency of 120 hz. The motor only has a 15 rpm slip so I dont know if there is a test taht can isolate my secound order from the electrical frequency and give me accurate phase readings.

David:

I have had sucess with live readings and stiffening at the highest amplitude of the offending frequency. I guess this would be 2x on your motor base. I plot on graph paper for each axis and form a poor man's model graph. This has been very useful, hope this is of use. Mabey you have already utilized this and it has not helped, just a thought.