Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Sulzer Blower - High Vibration Issues

Status
Not open for further replies.
Joined
May 26, 2010
Messages
2
Location
CA
The attach file shows pre and vib data for the inboard bearing (page 1 - 3) and outboard bearing (page 4 - 6). The blower runs at appro 6934 RPM. The rotor was changed out at shutdown, and high vibration was experienced following start up. The 1X filtered orbit plot on the on the outboard bearing barely straddle the Y-axis (see attached documet) on May 21st, and changed to a straight line on the My 25th.
Base on the data/information provided in the attached doc, I will highly appreciate contributions highlighting what the issues might be as well as the risk of running the equipment for the next few months in this condition.
 
You have 3000 g's acceleration!!! If this is a real number, then it should be very audible. Looks like serious impacting, if accelerometer is firmly mounted. You need to act quickly with further test/analysis and plan on a shutdown. Is your accelerometer actually rated to 3000 g's; perhaps a scaling/sensitivity error like 10 mv/g instead of 100 mv/g.

Walt
 
Adding to Walt's comments:
- generally Accels are limited to +/- 5 Volts.
3000 g's @ 100 mV/g would be 300 V.
(your route based analyzer + Accel would clip the data.)

was your sensitivity setting: 1 mV/g.?

- the time domain shows alot of high frequency content with impacting. whereas, your spectrum is limited to <3 orders. can you capture High Frequency spectrum.?

- Spectrum shows: ~6 mils p-p @ 1x (115 Hz)

converting to acceleration:
A = D * (2*pi*f)^2
A = 6 mils p-p * (2*pi*115)^2 /(386.1 ips/g) / (2 p-p/pk) / (1000 mils/in)

A = 4.06 g's peak @ 115 Hz.

The TWF shows alot more acceleration than 4 g's pk.


Mike.
 
Hi all,

The accelerometer data is obviously wrong by a factor of 100 as 30 or 40 g's peak is the realistic highest the bearing housing is going to be with a journal bearing.

Working out the shaft g's and comparing them to the housing g's is close to meaningless as there is an oil film between them etc.

You will also find that there is no significant relative displacement at the frequencies that the bearing housing higher harmoncis occur at, which shed light on the looseness.

The fact that 1x and 2x vary so much and there is obvious impacting signals looseness or a bearing design fault if there has been a design change.

Orbits go from round to flat due to many reasons, one being resonance differences in the horizontal and vertical planes and another being gyroscopic effects which occur if you have 2 rotors between the bearings or an overhung rotor.

The big picture here is finding the source of the impacts and that does not come from looking at the eliptical nature of a 1x orbit after slow roll subtraction.

Cheers

John
 
Thank you very much guys for throwing some light on this issue. However, the equipment is still running with overall vibration of approx 3.6 Mils pk-pk on the bearing housings.

I intend to do dry ice-blasting on the rotor to get rid of any imbalance; and also check the bearing clearance; and mic journal when we take a downtime in the near future

Once again, I appreciate all your contribution. Suggestion are welcomed - keep the discussion going. I can provide additional data if needed

Thanks,

Vic
 
I am not familiar with the particular machine you are working on. But I have seen some problems arise after overhauls that had similar symptoms. In all of these cases, we had higher shaft deflection in the vertical and lower in the horizontal. We may have shown some signs of looseness in the vibration on these events, as well. These could be relevant to your problem.

We replaced the rotor and bearings in an Elliott barrel compressor. Following the overhaul, there was higher vibration in the vertical and lower in the horizontal as measured with proximity probes. When the machine was shutdown, they found that the tapered dowel pins for the inboard and outboard bearing housings had been reversed. The pins were similar but not identical. When the IB pins were placed in the OB housing cover, they did not extend all the way through the top cover and engage the bottom of the housing. As a result, the outboard bearing housing cover was not properly positioned. It was setting off to the side creating a bearing clearance that was squished side to side, but full clearance up and down. We added a dowel pin check to our repair procedures. We place the pin in the top cover and mark where it protrudes. Then we place the pin the bottom half and mark where it protrudes. The two marks should be very close side by side.

We had another machine with unusual orbits and higher vibration than expected. We found a similar problem with the tapered dowel pins. We did a blue check of the pins to the holes and found that they were getting tight in the holes in the bottom housing when they were still loose in the top. We believe that at some point in the past, someone reamed the pin holes in the upper cover while cleaning it on the bench. Since they only reamed the upper part of the hole, the pin was no longer properly locating the cover.

We had a new gearbox for a large fan that had higher vibration in the vertical and lower vibration in the horizontal. We found that the original construction of the gearbox had missed a step. In this particular gearbox, new bearing liners are made with excessive crush in the vertical. They are supposed to be dressed down at the split line to achieve the target housing crush. The factory had failed to do this. As a result, there was excessive vertical clearance in the bearing and very high crush in the housing. The bearing was actually holding the gearbox cover up. We found this with a lift check in the field. The vertical lift was over 0.012” and should have been about 0.007”. When we opened it up, we expected to find the bearing wiped. But it was not. The manufacturer makes this gearbox with bearing liners that have to be custom fit each time. Or, as an alternative, they will make it with precision bearings that do not require hand fitting. We converted to the other bearing arrangement.

I would add the following checks to the next repair:

Check the dowel pins for proper engagement in the upper and lower housings.
Bolt up the housings with no bearings and no rotor and check the bores for out-of-round, taper or offset at the split line.
Check the crush between the liner and the housing.
Install the bearing liner with no rotor and check the installed bearing bore for out-of-round, taper or offset at the split line.
Blue check the shaft to the lower half bearing liner to verify bearing housing alignment


Johnny Pellin
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top