running a lube system under slight vacuum

[Tmoose]

We have a slow speed machine with a pair of 5 foot diameter journal bearings. the top half of the journal is open, and oil is sprayed on the journal from above. The bottom of each bearing housing forms a sump well below the level of the seal, and has a 2 inch drain pipe pitched down hill to the lube skid. On either side of each bearing is grease filled labyrinth seal. The labyrinth ring is stationary and has a large O-ring on its OD compressed about 20%, intended to seal against a bore in the main housing. One seal leaks one drop every 10-30 seconds. There is suspicion that the housing bore has a scratch or other damage that the O-ring can not effectively seal.

I've been daydreaming about just pulling a few inches of vacuum on the skid breather for the purpose of reducing or eliminating the leak.
Immediate potential downsides are:
- pumping some amount of oily vapor into the shop ( solution - coalescing filter)
- possibly pulling dirt past the seals, or in thru poorly sealed components
- finding a fan and motor suitable for pulling oily vapor safely. ( solution - Using a tiny amount of seal air for venturi extraction )

Any thoughts?

thanks

Dan T

 

[flexibox]

Question: Has the machine been running and developed a leak or have you just installed the laby's and started up with a leak? If it has been running for a while and just developed the leak, then one can assume its not a scratched / damaged bore? What is the material of the Oring and is it compatible with the Oil. ir EPDM is not a suitable material in Oil. I understand that this is a large machine and costly to stop and check but I don't think what you propose is a long term solution. Is the other end is leaking too? At these diameters I would think that the Oring has been joined on site, perhaps the leak is coming from the joint? I would collect the leakage and observe to see if leakage increases over time. if it does change seal and check bore during next shut. Check to see if the lube oil injection is higher at the leaking end it may stop the leak if you can reduce the flow at that end.

trust this adds some value.

 

[Tmoose]

There are 4 of these seals on this machine. All are described as "the same" design, although the drawings reveal "significant" differences in axial offsets within the large main housings (with the bore the O-ring bears against) of slingers and other oil-management features. Probably true even in in the original design, but now at least partly because the machine was re-designed with a little shorter OAL by the OEM for this installation, and all the length was stolen from within the bearings.
This one ( of 4) seals reportedly leaked immediately when commissioned new a few years ago. During commissioning the oil spray bars had the end nozzles blocked with OEM's permission to bias flow away from all the seals.

Last spring new US made nitrile O-ring stock was used to make replacement "o-rings." Cleaning some of the nether regions was "best effort". The (only) joint in the o-ring by design is positioned around 3 o-clock and butts against a small steel anti-rotation block (previously missing ). Thoughtful application of hi-temp silicon was used to enhance some obvious gaps in the sealing around the butted joint. New split cast iron labyrinth rings ( with prevously omitted curious end butted seals) were installed, and (missing) scrapers were added.
The seal replacement requires raising the 14 foot diameter, 16 foot long rotor slightly (there are other system seals within about 1/2 inch of the journals) to allow rolling the iron seal ring lower half out.

The leak is reportedly quite a bit slower after the seal replacement. Maybe something like a drop per minute. More of a housekeeping nuisance Manageable with oil socks or a small drain gutter

The Origin (s?) is hidden by the oil saturated grease oozing from the labyrinth. We won't get a chance to get near this machine for months. Disassembly for another seal replacement (rotor lift) might be a year away. Even then the suspected bore scratch region is virtually inaccessible, except >>possibly<< to confirm with a brass pick and mirrors or bore scope.

If a slight vacuum can be safely added, and it serves to reduce or stop the leak it, would be not so much a band-aid
but more like this -

http://i263.photobucket.com/albums/ii148/GTR1964/A7.jpg

or this -

http://amazingdata.com/mediadata32/...-before-after-makeup-5_200907261737052905.jpg