Shaft Centerline 2

[Monrooy]

I want to know how I can interprete a Shaft Centerline Graphic?

I find one in a report for a Generator with 40MW at Geotermal Power Station

 

[drawoh]

Monrooy,

Can you provide us with a picture of what you are trying to interpret?

JHG

 

[Monrooy]

OK

 

[GregLocock]

Over 11 minutes the shaft lifted by approximately 100 microns.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[unclesyd]

Not quite like some I've seen but the squiggly line is the actual track of the centerline of the rotor shaft moving.

GregLocock correct me if I'm wrong but we used to call this a search for rotor's magnetic center.

 

[GregLocock]

Has all sorts of uses, until the OP expands on his problem, we'll never know.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[electricpete]

This resembles a startup or shutdown plot where shaft speed is changing.

The phrase "search for rotor's magnetic center. " applies in axial direction, but has no business being applied to the radial direction imo: magnetic force in radial direction tends to pull the rotor further off-center. (acts like a negative spring constant)

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[byrdj]

not sure
but it looks like a calculation of the "center of shaft" taken from X/Y proximity probes. that is the center of the vibration orbit is calculated to represent the position of the rotatating shaft in its bearing. I would guess the circle shown represents the bearing clearances of 0.014"V X 0.018"H of a 9" diameter shaft.

as the unit was brought to speed and an oil wedge was developed the shaft was lifted 0.005" and pushed sligthly to the right. the expected bearing loading could be derived my knowing realitive bearing rise on the others.

I'm not a real fan of promity probe predictive anologies, nor any experance, but that rise could indicate a slightly under loaded alignment of that bearing, or an oversized bearing. my gut is a 0.003" rise would be a good loading

 

[Monrooy]

Thanks for your answers .

byrdj I agree with you.

And now I have more questions , Is it this motion common in a shaft with a journal bearing during start up?. Is possible that the shaft may have a mis alignment?

 

[byrdj]

again, I'm not experaince in the use of prox probes for rotor dynmic diagnostics.

So I would want to refer back to more basic information to determine if this is a "Problem"

I'll assume this is the inboard bearing, since bearing loading there has beeen a problem with many units.
Is there an increase vibration
what are the bearing temps (metal and oil drain) for this bearing compared to the others, especially the inboard turbine (assuming 4 bearing machine)
AND was there recent maintance where a bearing elevation change made.

I interput the curve as
the little sqibble at lower 1/3 was at critical, and the knee at the upper 1/3 is when oil wedge hight increased to a point there was reduce horizonal force and the rotor came back to horizonal center line. but I do not see a sqibble at top end, so I would assume vibration was acceptable.

I would also assume since there is not a very gross sqibble at the top, the bearing loading is notlessen to a value that would cause oil whip.

Now if this was just data from a run up to speed, you might expect the loading to increase as the support pedistal grows

 

[byrdj]

electricpete brings up a point i have NEVER considered in all these years....correct me if I am taking your comment wrong in that the magnitic forces would "support" the weight rotor?
very interesting, thanks

 

[GregLocock]

I don't know about the magnitude of the rise, but the trend seems consistent with the development of hydrodynamic lubrication, and any misalignment issue seems quite small at this location.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[JJPellin]

I use this exact plot on a regular basis. This is a shaft center-line plot from a set of X-Y radial proximity probes in a Bently-Nevada continuous monitoring system. The plot looks identical to the ones that I generate in the Data Manager 2000 (DM2000) software. The plot represents the DC component of the dynamic signal from the two probes. This represents the average shaft center-line as the machine is started up from a dead stop. The circle represents the total bearing clearance (on diameter). The plot looks perfectly normal to me. The shaft came up about 1/3 of the total bearing clearance and moved off slightly to the right based on the expected oil wedge location for this rotation. If the bearing is a plain journal bearing, I might have expected more movement to the side. This is probably a tilt pad bearing of a more sophisticated design. A bearing of this sort can be designed to have the shaft come up almost perfectly straight from the bottom of the clearance. Since the plot is labeled for points 1X and 1Y, I believe it is from the outboard bearing. It would be typical to number the bearings from the driver to the driven for most systems. But, it is common with Bently-Nevada to number them from the driven to the driver. That is the standard that we use with all of our systems. I cannot comment on the condition of the shaft alignment based on this plot. An orbit and waveform plot would be more helpful there. I would base alignment evaluation on a spectrum, as well. But Bently-Nevada has generally emphasized orbit/waveform analysis over spectrum analysis. Without information about the bearing design, verification of the location (inboard or outboard), coupling design, speed, etc. any sort of analysis would be pure speculation. You might get better feedback in the Mechanical Acoustic/Vibration Engineering forum.

Johnny Pellin

 

[JJPellin]

Since this is just the DC component of the signal, nothing can be said about the vibration of the shaft during this start-up. The squiggle along the line is almost certainly the slow-roll phase of the start-up. A train like this would normally be started up to some minimum slow roll speed (ours are usually 700 to 1000 rpm). They would hold there for a period of time to allow the machine to warm up and grown into alignment. Then the machine would be run up to full speed (probably 3600 rpm if this is in the US). This could be a stiff rotor machine and not pass through rotor critical. If it is a flexible rotor machine, then it most likely passed through critical speed in the ramp up between slow roll and full speed. That would explain the bend in the line along that section. If I had this plot during the start-up of one of my large steam turbines (we have no generators) I would be quite happy.

Johnny Pellin

 

[Monrooy]

JJPellin, thanks for this good explanation.