Bearing fully submerged in hydraulic oil

[MSoap]

I was wondering if anybody might have any comments on the possibility of using a cylindrical roller bearing which is completely submerged in a hydraulic oil bath.

The bearing will be used on a test stand, where the inside of the bearing housing will have a circulating oil bath, which will be cooled so that it does not exceed 70?C (160?F). The oil would have a pressure of 25 psi maximum.

The bearing is in a stationary housing with a horizontal rotating shaft. A typical cycle will be accelerating the shaft up to a maximum of 3000 rpm (often only up to 1500 rpm max) over a period of about 30 minutes. The speed will be maintained for about 1 minute. Then the shaft will slow down to a complete stop over a period of about 1 hour to 1.5 hours. At most, one of these cycles will be conducted a day.

My primary question is how feasible would this setup be?

I know I would be getting a power loss from this setup due to the extra friction on the bearing, but are there any other concerns I should be worried about?

From what I have gathered, the oil that would be used should be very close to the recommended viscosity for the bearings, but are there any guidelines as far as the filtering that would be needed for the oil?

Thanks for any advise or thoughts you might have.

 

[electricpete]

Normally in a horizontal machine the oil level is maintained at 1/2 the level of the rolling element or lower. Above that you are not following standard application guidelines and need to evalute the thermal performance. What is the size of this CRB?

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

 

[MSoap]

The bearing has a ID of 2.25" and an OD of 3", with a width of 0.375" wide.

 

[romke]

you might see a local temperature rise due to excessive churning of the oil in the bearing. as electricpete pointed out the oil level is quite a bit higher then normally would be advised.

this will not only lead to powerloss but possibly also to slowing down the acceleration envisaged.

it may also lead to a local temperature rise of the oil in the bearing to such an extend that deposits may be formed on rollers or raceways. if that happens bearing life will be substantially shortened.

if you can't change the oil level, would it be an idea to use a completely sealed grease lubricated bearing?

 

[MikeHalloran]

Or build a sort of leaky bucket around the bearing, maybe with slingers adjacent the bearing, so it can throw the oil it doesn't need out of the bucket.

... or, not. That could make foaming more of a problem.



Mike Halloran
Pembroke Pines, FL, USA

 

[diskullman]

Use a synthetic oil, it will run cooler.

Id suggest using a standard off-the shelf pillow block set-up for oil, such as the SN type or SAF type housing.

Russell Giuliano

 

[MikeHalloran]

... Wait a sec.. Hey, the bearings inside hydraulic pumps run completely submerged all the time and don't give trouble.

Maybe it's not a real issue.



Mike Halloran
Pembroke Pines, FL, USA

 

[electricpete]

I'm not sure I get your point Mike.... maybe we should ignore the bearing manufacturer's application guidelines? What size and speed bearings was the application you mentioned?

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

 

[electricpete]

Sorry, my post didn't come out right Mike. My apologies.

Here is a post from FAG Bearing Lubrication document WL 81 115/4 EA/94/2/00. It is not particularly specirfic... leaves a lot of wiggle room for user judgement:

4.2.2 Oil Sump Lubrication
In an oil sump or, as it is also called, an
oil bath, the bearing is partly immersed in
oil. When the shaft is in the horizontal
position, the bottom rolling element
should be half or completely covered
when the bearing is stationary, fig. 47.
When the bearing rotates, oil is
conveyed by the rolling elements and the
cage and distributed over the circumference.
For bearings with an asymmetrical
cross section which, due to their geometry,
have a pumping effect, oil return
holes or ducts should be provided to
ensure circulation of the oil. If the oil
level rises above the bottom roller and,
especially, if circumferential velocities are
high, the friction due to churning raises
the bearing temperature and can cause
foaming. At speed indices of
n • dm < 150 000 min–1 • mm, the oil level
may be higher. If complete immersion of
a bearing in the oil sump cannot be
avoided, as is the case with the shaft in
the vertical position, the friction moment
doubles or triples depending on the oil
viscosity. As a rule, oil sump lubrication
can be used up to a speed index of
n • dm = 300 000 min–1 • mm; if the oil is
renewed frequently, a speed index of up
to 500 000 min–1 • mm is possible. At a
speed index of n• dm =300 000min–1• mm
and above, the bearing temperatures often
exceed 70 °C. The oil sump level
should be checked regularly.
The oil renewal schedule depends on
contamination and ageing of the oil.

I know other publications steer toward lower levels for higher D*N (like 1/4 to 1/2 of lowest rolling element while static for 3600rpm 6313).



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

 

[electricpete]

Whoops - looks like parpticular this pub wouldn't even allow oil sump lub for 3600rpm 6316 based on D*N. At any rate it becomes more imoprtant to avoid too much oil as D*N increases.

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

 

[MSoap]

Thanks for all of the replies...

After looking over the thread again, I realized that I gave the wrong size for the bearing. It is actually a 35mm bore, 80mm OD, and 21mm width. Basically the equivalent size of a 6307 bearing, except that it is a cylindrical roller bearing right now. However, it could easily be replaced with a standard ball bearing.

So, after adjusting for my error, that would mean that the highest dN that would be reached would be 105,000 (a max speed of no more than 3000 rpm). So as I understand the guides given by FAG, this might be within the acceptable range of using a higher fill level. The acceleration/power loss resulting from complete immersion are not extremely critical. I guess my main concern would be damage to the bearing due to excess heating. However, I know that I have run across a few publications that specifically mention that vertical shaft applications can run in an immersed state. So if there is fluid circulation and cooling, as well as temperature monitoring of the fluid, maybe it is worth a shot?

I like Romke's idea of a sealed bearing, but since the hydraulic fluid is slightly pressurized (~25 psi) would I need to worry about the oil seeping in or the grease seeping out around the bearing seal over time?

Any other thoughts?

Thanks

 

[Eratosthenes]

MikeHalloran is correct. The roller bearings in hydraulic gear pumps do indeed run fully submerged in hydraulic oil and work well despite the low viscosity of the hydraulic oil, which implies a low film thickness. Typical speed is 1800 rpm at loads of several thousand lbf. The load cycles on and off depending on demand. There is no known high temperature issue.
Most people use a 10 micron filter but a 5 micron filter would be preferred.