Gearbox labyrnith seal melted

[Abualnassr]

Hi.

When we were overhauling a centrifugal compressor which is connected to a motor through a speed increaser (gearbox), we noticed when we opened the gearbox cover that there as some melted metal on the shaft just underneath gearbox labyrnith seal of the high speed shaft (5800 rpm). Why would the seal melt ?!

Regards,

 

[PeteDB]

The answer is obvious. Because it got hot.
Not so obvious is why it got hot. A labyrinth seal should be non-contacting. You have to figure out if you have alignment issues, not enough clearance between the shaft and seal, or some other reason why the shaft would rub against the seal. I am assuming you saw rub marks on the shaft. If no rubbing was seen, did someone do welding on the gearbox without proper grounding?

 

[Abualnassr]

Well, actually there was some misalignment between the compressor and the gearbox. And there was no welding don on the shaft, but there was slight rubbing on the shaft. BTW, we had really a rough time trying to grind the melted metal off the shaft.

Just a question, do you think that misalignment between the compressor and the gearbox might be the root cause ?!

 

[DesignerGuy16]

Was this mis-alignment through a fixed coupling, or one that allows for some mis-alignment? If it was a fixed coupling that doesn't allow for mis-alignment you could end up with a lot of side-load on that seal, which is probably not designed for any side-load at all.

It probably melted shortly after it seized from galling. Like PeteDB mentioned, get the alignment fixed and make sure you're not side-loading that seal. It may be as simple as adding a spider coupling in line if the motor doesn't face mount directly to the gearbox. Again I don't know what's already there.

James Spisich
Design Engineer, CSWP

 

[Abualnassr]

Thanks guys for your comments. The coupling is a flexible disc type.

To see the system configuration, please have a look at the attached image.

 

[DesignerGuy16]

Do you have more detail on that flexible disc coupling? Ie, max angular or diametrical misalignment?

Have you measured this with your current setup?

James Spisich
Design Engineer, CSWP

 

[Abualnassr]

For the coupling :

Max axial misalignment (in) = 0.08
Max parallel misalignment (in) = 0.024
Max angular misalignment (degree) = 0.50

When we did the laser alignment we didn't record the amount of misalignment, we only corrected it.

 

[DesignerGuy16]

Was this correction after melting the seal? Any luck with it back up and running?

James Spisich
Design Engineer, CSWP

 

[Abualnassr]

Yes, after the melt of the seal we've done the alignment. And, now it's running smoothly. But you know what, I think the parallel misalignment was more than 24 mils when we checked it the first time with the laser alignment tool.

 

[diskullman]

Disc couplings handle angular misalignment well, but are very unforgiving on parallel misalignment.

Russell Giuliano

 

[Tmoose]

I'd expect the disc coupling is really 2 disc couplings with a spacer or spool piece, because as diskullman said, a single disk element is radially very stiff.

http://www.rexnord.eu/fileadmin/rexnord-bsd.alt/gb/c6idx.html

Now you are faced with Quite a few "if"s, and they are all un-answerable now that repairs are made. A careful autopsy is important if a reasonable root cause is to be ascertained. A list of "coulda-beens" is not all that useful in preventing it happening again.

I'd use a feeler gage to probe around the labyrinth seal gap and test for fairly equal gap. If the g'box housing machining was done badly ( new or rebuilt) the labyrinth bore may not be so concentric with the bearing bores, and will rub even with nicely aligned shafts.