Updates on replacement brass oil rings in overhung pumps with alternate material 5

[electricpete]

Pump users are familiar with problems of brass oil rings on overhung pumps.

http://turbolab.tamu.edu/proc/pumpproc/P22/11.pdf

We are fighting some that give us discolored oils with brass particles on some older Flowserve overhung pumps - 3600rpm, 65-70mm shaft size.

We recently found out the Trico flexible flinger line has just been discontinued this month. Apparently there was not enough demand.

There was some interesting discussion of alternative fiber reinforced composite oil rings here:

http://www.eng-tips.com/viewthread.cfm?qid=163919

As I understand the OEM now uses stainless steel oil rings in new pumps.
Not sure if installing stainless steel would create potentially worse problems (now we have brass particles in oil... but if steel o-ring should degrade it could put much more damaging particles into oil)

I'm interested in any further thoughts and experience using alternate oil rings for this type of pumps.



=====================================
(2B)+(2B)' ?

 

[JJPellin]

Our standard for API pumps is still to convert all brass oil rings to non-metallic, fiber reinforced composite (Luytec). The vast majority of our centrifugal pumps have been converted to the non-metallic rings with very good results. We recently installed a new Flowserve pump that came with a single brass oil ring in the center of the bearing housing. We immediately had problems with dark oil. We have been changing the oil twice per week for the past 6 months because we can’t take this pump out of service to replace the ring. When we dug into the problem, we had some surprising findings. First, for their RVX bearing housing upgrades, Flowserve recommends fixed disk flingers rather than oil rings. The only reason our new pump came with an oil ring is that our specifications do not allow for fixed flinger disks. Second, Flowserve did not standardize on non-metallic rings like I had thought. They still use brass rings when an oil ring is specified. Lastly, brass rings have another problem that I did not recognize. The brass particles cause rapid oxidation of the oil. The oil is turning black because of the brass particles. Stainless steel would not produce this effect.

So, my preference for oil rings would be as follows (in order from best to worst):

1. Pure oil mist with no oil rings.
2. Non-metallic oil rings
3. Fixed disk flingers for bearing housings that were designed for flingers.
4. Stainless steel oil rings.
5. Brass oil rings.


Johnny Pellin

 

[electricpete]

Thanks Johnny, great input as usual.

Fixed flinger is what Floserve had recommended to us. However the flinger has to be so small (to be installable in that housing) that the oil level needs to be raised (and new seals to accommodate higher oil level). We estimate the new proposed oil level is close to middle of the lowest ball of the back-back angle contact bearings (lower on the radial bearing). Do you think that would cause problems for this application (3600rpm, 65-70mm shaft size)

=====================================
(2B)+(2B)' ?

 

[JJPellin]

The pump I referred to with the dark oil problem only runs at about 1500 rpm. I would be worried about overheating if this was a 3600 rpm application. For all of the fixed flinger disk applications, I thought that they usually set the oil level just about at the ID of the outer race of the thrust bearing so that there would just barely be any oil up into the balls of the bearing. But, this makes the oil level very touchy. If it drops by 1/8", you could lose oil to the flinger and fail the bearings. If it rises by 1/8”, you could flood the bearing and overheat.

I don’t have any bearing housings of this arrangement running at 3600 rpm with only the fixed disks. But, I would feel comfortable installing this configuration if it was the manufacturer’s standard design. We have run into more trouble when we try to force to OEM to deviate from their standard design to satisfy our standards.


Johnny Pellin

 

[GPRnD]

electricpete, if you do what FLS recommend, you basically end up with an oil bath arrangement with the slinger augmenting the oil circulation in the sump. My personal experience with shafts of 65-70mm is that churning losses will be significant at 3560 RPM. My OH2 bearing designs would not employ an oil bath arrangement as a result.

That said it can work as long as you have enough cooling on hand. I'd say a DE fan is a must. What are your maximum ambient and pumpage temperatures ? Also is the pump outside or not ?


A couple of other papers that might be worthy of review:

1. The original paper on non metallic oil rings that prompted Sulzer to switch [URL unfurl="true"]http://turbolab.tamu.edu/proc/pumpproc/P17/P179-16.pdf[/url]

2. An update on testing metallic oil rings vs non metallic oil rings vs Slinger discs, that was presented at the 2014 Pump Symposium. This paper is not available online but I have a copy so if there is interest I will upload it. One takeaway from it is that slinger discs are extremely sensitive to submergence depth (to approximately the 6th power), as speed increases. If you don't achieve sufficient submergence they end up delivering very little oil.

 

[ccfowler]

Johnny, I like and agree with your preference sequence!

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[Artisi]

Not having much experience with oil bath / slingers coming from a 50Hz area where grease usually sufficed for lubrication - just a thought, could you use excentric fixed stainless steel slinger - I probably needs some design and thought but would likely overcome the concern of a low oil level in the sump.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)