bushings vs needle bearings for oscillatory motion

[qnen]

We are currently designing some linkages and the question of plain bushing vs needle bearings at the joints has started a small debate. The mechanism in question could be considered a 6 bar linkage, the links are in the range of 4" in length, weights are in the range of a couple of pounds. Most joints will oscillate only about 100degs, back and forth, 100 times a minute. Looking to get about 5 million cycles of life. Lubrication is tricky since this mechanism is in close proximity to a food contact surface.

Can anyone suggest some good resources for designing/ selecting bearings for oscillating joints?

 

[eromlignod]

I think engineers overuse bushings these days. If this is for capital machinery (not a production item) then I go with needle bearings anymore. The difference in cost is small and not worth fooling with on a machine that might cost several hundred thousand dollars anyway. Bronze bushings are for intermittent-duty stuff, like hand levers or adjustment leadscrews, etc.

Remember that the journal of your shaft has to be case-hardened and ground, though.

Don
Kansas City

 

[hydtools]

Garlock DU bushings are prelubed. No worry about lube washout when equipment is cleaned.

http://www.ggbearings.com/

Running fits are as tight as to looser than needle bearings.

Is space an issue? Needle bearings require more radial space than bushings.

Ted

 

[btrueblood]

Greg Locock gave a reply to a similar question (from me), in which he pointed out that less than full rotation is not really an issue for rolling element bearings, provided that the rolling elements turn thru more than one rotation. So, Don's advice is a good one.

 

[John2004]

"Full compliment" needle roller bearings are better than caged needle bearings for oscillating motion, according to Timken.

http://www.igus.com

has some plastic bushings that outperform bronze bushings and work well for certain oscillating applications.

Personally, I like needle roller bearings, they can handle very high loads for a given bearing size, are compact, and are tolerant to some misalignment. Drawn cup needle roller bearings can be very cost effective if you can use those. My experience has been with Timken needle rollers, they have a good catalog on CD they will send you.

John

 

[unclesyd]

Could you use a spherical bearings or a rod end bearing. There are plenty of them not lubricated and if you need a lubricated one there are many food grade lubricants.

http://www.igus.com/igcat/ig_cat1.asp?p=ig_23_4.gif

http://www.aurorabearing.com/products/commercial-industrial/rod-end-bearings/inch/default.html

 

[John2004]

You would have to check the food grade requirement with the IGUS catalog or website , but if the loads are only a few pounds perhpaps the plastic IGUS bushings would do very well.

They have an online bushing life estimator "called expert system". You can enter the loads, oscillation angle, speeds, etc.. and it will tell you how much wear there will be on the bushing ID after a certian time or a certain number of oscillations.

Some of the materials they have are pretty impressive as far as load capacity and life expectancy. They claim the results of the life estimator are based on both real world tests and theory.

Rod ends are also good as unclesyd mentioned, also available from IGUS and they tolerate more misalignment.

The IGUS bushings would never need any type of lube as they are self lubricating. They are also very cost effective depending on which bushing material you use.

John

 

[MadMango]

Don's question about this equipment being a one-off machine or a production widget would help greatly in providing a proper answer. I'm all for plain bearings (plastic or DU) when they are applicable, which this could be.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the best use of these Forums?

 

[qnen]

Thanks for all the replies so far. To fill in some of the blanks others have pointed out:
If all goes well, there might be about 15 of these machines made
At several joints, radial space (dia) is limited but width (len of brg) is not.

The community at large seems to be reflecting some of our internal thoughts and ideas, so we feel we are on the right track.

Many thanks to those who pointed out that IGUS has the online calculator that is set up to handle pure oscillator motion. That is a hugh help compared to all the constant rotational information out there.

 

[tbuelna]

gnen,

With regards to oscillatory motion, a well designed and maintained plain bearing will usually outlive, cost less, have less friction, be more durable and weigh less than a rolling element bearing.

The fatigue life of a bearing is usually limited by contact stress. The contact stress of a plain bearing operating in boundary or EHL conditions will yield a longer fatigue life than a similar sized rolling element bearing.

The plain bearing costs less because there are simply fewer components.

The plain bearing will have less peak friction losses in an oscillatory joint because both bearings are operating in boundary conditions at the instant the link/joint stop and reverse direction. The plain bearing has less friction because its journal radius (ie: the radius at which the friction loss is incurred) is much smaller than a rolling element bearing of equivalent load capacity.

The plain bearing is more durable because it is more tolerant of debris passing through the contact zone.

And finally, the plain bearing installation will weigh less because there is much less steel involved.

While all of the above is true for oscillatory motion, the rolling element bearing will be more efficient in continuous rotary motion.

Good luck.