[fluider]

I know that many of you here have some sort of engineering education and/or experience in mechanical enginnering job, or are enthusiasts just like me who love exploring all aspects of their hobby. I admit that I haven't done enough research myself neither in Slovak literature nor English one, but I haven't been able to find what I want to know in English books which I downloaded.

I'm desperately interested in what makes mechanical engineer to make final decision when choosing bearings or bushings? What's the last grain of sand that leads to resulting decision? Is there any methodology to help in this?

Obviously, you can pick bushings that can take damn high loads, work under water or in the dust and need no lubrication. I know that there's no ideal bushing material and their qualities highly depend on axle material and finish used. We know that Keith Scott quit bushings in Legend Mk2 as they were "eating" 7075 pivot axles on Mk1. Now, the new Banshee Prime runs on sealed bearings.

What would you base your decision on if you'd find yourself stuck, deadlock, in design process?
Thanks for any advice.

 

[HardtailHack]

I have just found out for my Nicolai frame there is now a self lubricating bush kit instead of the usual needle rollers, so when my current bearings wear out I'll give them a go.

I think frames that are made to poor tolerances should stick to needle rollers as they can be a bit more forgiving if pivots don't line up 100%, my previous frame used sealed bearings and because of poor alignment they were ruined in a very short period.

I still can't understand why companies like Mongoose will put a needle roller in a link that doesn't move enough to allow the bearing to rotate enough to prevent false brinelling(correct term?). The bearings end up seizing as the grease is forced out from where it is needed and the bearings usually seize or get extremely tight.

I would prefer a frame built with fairly large sealed bearings instead of needle rollers simply because they are too difficult to seal adequately and you have to replace the axle(inner race?) when you change the needle rollers.

 

[fluider]

I can't understand why anyone is using needle bearings, as they can't take axial load since each suspension pivot is experiencing combined loads.

I could understand the switch to deep groove ball bearings in case of Legend Mk2, it was technical aspect, but why with Prime? Is the public attitude (or dis-knowledge?) to bushings such a strong that Banshee decided to deliver ball bearings rather than losing potential customers?

 

[4130biker]

Is the public attitude (or dis-knowledge?) to bushings such a strong that Banshee decided to deliver ball bearings rather than losing potential customers?

I think that's part of it- even when they're running perfectly, they do have a bit more stiction which people don't like. On the right application a little stiction might actually be ok or good, but for dh I think it's generally looked down upon.

I do think that bushings are a neat idea because of some of the theoretical benefits, it just seems to me that it's never quite worked out perfectly for a dh bike.

 

[fluider]

As for the DH application... Can it be that high angular acceleration (big hits in high speeds) and higher frequency (multiple times more hits per second) was the cause of trouble in Legend Mk1 pivots?

Edit: just an idea. Can above mentioned be the reason why bushings are suitable for small angular movement mainly?

 

[Slater]

Unfortunately it has nothing to do with the characteristics of the forces the suspension will see, service life, or any of the other marketing reasons you have heard.

Bottom line is that machining tolerances for bushings are much much smaller than for bearings. Housing OD that the bushing presses into, axle OD, axle surface hardness and finish are the main ones. When any of those are off by even a little bit (or the bike is not dead-square after welding and heat treat), the bike will either never fit together, have considerable play in the pivots, or eat the bushing(s) and axle(s) in less than a month.

Bearings on the other hand give you much more room for error in axle and housing sizing, and rear end/linkage misalignment.

In an ideal world, for suspension pivots they would all be polymer bushings. Plain and simple. But realistically it is not cost effective for the manufacturers to cut their tolerances in half or thirds in order to use them. Spending a bit more on bearings over bushings is a better fiscal decision. I've built bikes with both and have unfortunately found this out the hard way.

 

[Pslide]

I was one of the Mk1 testers and did a lot of feedback to Keith about the bushings in the Mk1. You are right in that we were seeing more wear than expected...not in the bushings, but in the aluminum, both on axles and flange seats. Keith had found some super hard anodizing that would have probably solved the axle situation, and some simple stainless steel shim washers probably would have solved the flange wear.

But the bottom line - and this is my opinion now - is that bearings are just an overall easier solution in terms of maintenance, frame tolerances, durability, and customer perception.

Why was the Legend Mk1 worse than other bikes using bushings? I think it was both the higher loading, and probably moreso the frequency and speed of the bushing surface against metal. Some of the Legend bushings were larger in diameter, so your linear speed between your wearing surfaces will be greater given the same amount of rotation. And DH bikes travel at much higher average speeds, meaning the cycling frequency (and thus surface speed) will again be higher.

If I were building a one off race machine where every gram of weight was critical, and it would be meticulously built with exact tolerances, and it would be maintained by a race mechanic with fresh lube regularly, then I would use bushings. I don't think stiction is a big issue.

But if were design a DH bike for everyday use and mass production, I'd make the same decision Keith did with the Mk2.

 

[fluider]

So my thinking was heading the right direction. Yes, you want larger diameter bushings to decrease the pressure at axle and bushing housing, but then you get high surface speed, not mentioning the higher angular speeds.
Pslide, thanks for comprehensive answer.

 

[Pslide]

In thinking about it, you'd be better off running hollow steel axles and flange surfaces if you could keep it light enough. Maybe put a nice high polish on them as well. (EDIT: You could keep the diameter smaller as well.)

I wonder why no one has come up with a hydraulic bushing/bearing that is sealed but runs on a thin layer of fluid? I'm guessing the sealing is the difficulty. Engines obviously run on this principle, but using pumped oil. If you could self contain the principles in a sealed bearing it would be a pretty neat invention.

 

[fluider]

Ok, I come up with my dilema. There's about 7000N at each pivot, or pivot pair, part is to be underneath the BB and between BB and rear wheel, so dust, mud, sand and water are trully expected. Part is quite wide so lubrication through a grease port in centerline is not the way to go. Maintanance can be done regurarly but 1x/year sounds much better. And there is rotation by circa 45° at each end of part. Just a prototype design, but manufacturing cost can be big issue . Axle OD is 12mm if I remember correctly.
What do you think ?

or

 

[Pslide]

It looks like you'd run into some of the same issues that the Legend Mk1 had unless you are using steel for the axle and flange surfaces. And I'd personally want a lubrication system along with some clever sealing.

Assuming you addressed the above, I'd still be hesitant. The only way I'd go bushings in that location is if this is on a dual link bike, and both the main frame pivot locations are part of the same CNC'd piece. In other words, no welding in between the pivot locations that could throw off your alignment. Same rule for the swingarm.

In all other cases I'd go bearings.

Just my opinion though, and I am not exactly drawing on a wealth of experience!

 

[roel_koel]

as previous posters have commented, bushings are ideal when other factors are "ideal" mainly manufacturing tolerances and regular maintainence

if the tolerances are out (as they often are on many bike frames) then bushings are not the best choice?

on paper or in others terms 'ideal engineering practise' they make much more sense than ball bearings, especially considering the limited rotations angle of most suspension pivots

I've just had a customer ruin a frame running bushings because he rode it in mud and wet for several months, and it also looks like manufacturing tolerances were out as there is exaggerrated wear on one side of a bushing and no wear on other side


but rubber sealed ball bearings (ideally common industrial stock) in a radial package are ideal, easy to design pivot hardwear around, easy to work on and replace cheaply, and deal with slightly 'off' manufacturing tolerances

 

[fluider]

And what about the bearing sealing? What's the functional/reliability difference between lip seals and gap seals? Which one of them can be easily pulled out in order to press new grease into bearing and then pressed back into bearing? For example SantaCruz or Turner on new DHR they use one-side sealed bearings with custom-made removable washer that covers bearing interior. Is it worth of designing custom sealing and regreasing system when you can cheaply get a double-side sealed, life-time lubricated bearings ?

 

[fluider]

This is the suspension part of UncaJohn's prototype frame that was only using IGUS bushings. The small pivot was shock eye mount. The big pivot at the bottom housed BB shell. He was riding the bike for 7(?) years until this link snapped. As you can see there is no wear, no scratches on the bushing contact area, also he told me he didn't notice any damage to pivot axles. All Alu parts were Fortal 7075, though I don't know bushing type.

 

[Pslide]

Are there more pictures of "UncaJohn's" frame out there somewhere? Looks interesting...

 

[fluider]

His website is after all those years still up & running: http://prototype.gr/prototype.htm