Between thumb and pointer, eight 40 mm conventional angular-contact bearings, each subjected to 5000 N of load at
1500 Rpm (180 km/h), would have a combined power-loss of some 250 W (1/3 of a Hp). But everything counts I guess?
"I spent most of my money on wine and women...I wasted the rest"
A ceramic hybrid rolling element bearing would have somewhat lower inertias, but it would not likely have lower friction losses when used in an application like a wheel bearing. A properly designed and lubricated rolling element bearing is operating in a hydrodynamic contact regime, so the static friction characteristics of the bearing materials does not have an effect.
As I noted, a ceramic roller/steel race hybrid bearing has less load capacity than an all steel bearing. This is due to the much greater MoE of the ceramic rollers producing a higher proportional contact stress against the steel race surface. As a result, the steel race will have a reduced (L10) fatigue life with ceramic rollers. In order to compensate for the lower load rating, a ceramic hybrid bearing naturally has to be slightly larger than an all steel bearing, offsetting some of the weight savings.
Regards,
riff_raff
"Q: How do you make a small fortune in racing?
A: Start with a large one!"
Friction, or powerloss, of a rolling bearing can be described in two; Load-independent viscous losses from the lubricant
and Load-dependent losses from hysteresis in the material and partial sliding contacts (compare with the tyre-thread).
The lower friction of the hybrid bearing comes from higher stiffness of the roller material, thus less hysteresis.
"I spent most of my money on wine and women...I wasted the rest"
Sounds like you have a good knowledge of rolling element bearing principles.
I agree with your statements regarding the main sources of mechanical losses in roller bearings. Which type of loss predominates is dependent upon operating factors such as DN, lubrication, shaft and housing deflections, etc.
With a high DN, churning losses can be significant. When there is substantial shaft or housing deflection, skidding/skewing/sliding can produce big losses.
But as you noted, when the bearing is operating at low DN, properly lubed, and without misalignments, hysteresis (or rolling) losses produced as the rollers are elastically strained passing in/out of the loaded sector of the bearing race likely predominate. But as I noted, you must also consider the impact of the slightly larger roller pitch diameter the hybrid bearing would have versus an all steel unit. The rolling friction losses due to hysteresis effects are a function of the pitch radius the strain loss occurs at. All other things being equal, a radial bearing with a larger pitch radius would incur proportionally larger rolling friction losses for a given load.
The fatigue life of a hybrid roller bearing used in a suspension upright (ie. a fixed outer race with respect to load) would be limited by the surface contact stress capability of the outer race for the given load and number of load cycles (ie. the number of rollers multiplied by the number of shaft rotations). So a hybrid bearing's capability would still be determined by its steel races.
Enjoying the discussion.
Regards,
riff_raff
"Q: How do you make a small fortune in racing?
A: Start with a large one!"
I'm a tribologist for a living, Terry. Popularily speaking, you can say that viscous losses dominates at low load and high rpm, while "structural" losses is more important at high load and low speed.
With a 40/80 mm angular contact bearing, which has less mass and losses than the tapered roller bearing you find in your car, the steel version has a reference speed of 10 000 Rpm, weighs 370 grams and a dynamic load capacity of 36.5 kN.
The latter decides the nominal lifespan, L10h, which I would set to 500 h for good measure, which then allowes for an equivalent dynamic load of 10 kN at 1500 Rpm, in turn resulting in a powerloss of some 60 W. Price about 20 USD.
Anyway I chop this, the remaining reason for hybrid bearings in an F1 car is weight and possibly lubrication needs.
Another case of senseless spending, I bet USF1 won't bother with it, as little as they do with 1000 USD wheel-nuts.
"I spent most of my money on wine and women...I wasted the rest"
=D>
thanks for sharing you expertise here, in the light of these words ,I assume the real trick with the bearings and their lubrication will be just how to seal the bearings without creating drag and still resist the really harsh environment with carbondust ,heat soak etc ,dealing with air volume in the bearing housing etc....
Just look at those speeds and you get an idea, I know that Penske used them in their uprights for CART anyway.
Shamban was an American success story, sealed the Discovery btw, which merged with Busak to Busak+Bhamban in the 90s, was later bought by Swedish Trelleborg.
"I spent most of my money on wine and women...I wasted the rest"
A professional tribologist? Wow, I'm impressed. That's a field of technical expertise that is somewhat rare.
I design aircraft power transmissions for a living, so I can appreciate what a tribological specialist has to say about bearing design.
I agree with you about hybrid bearings and weight in F1. In the aerospace world, the breakpoint for cost-to-weight is around $300/lb for materials, so ceramic bearings at $600/lb are still not cost effective. But for F1, even $600/lb is apparently a bargain.
Regards,
riff_raff
"Q: How do you make a small fortune in racing?
A: Start with a large one!"
In the case of the driven wheels and the free wheels, i guess we are looking at different bearing requirements. The only similarity would be the rotating speeds, other than that the loading should be different right?
The rear has to deal with static loading by the weight of the car, and then there is the shaft loading, axial and radial, on the internal housing.
The fronts now, you have the static loading then axial loading from scrub forces, which only act on the outer housing.
Will the differences warrant the uses of 2 different bearings, or can that be overlooked for the sake of simplicity and interchangeability?
ringo wrote:In the case of the driven wheels and the free wheels, i guess we are looking at different bearing requirements. The only similarity would be the rotating speeds, other than that the loading should be different right?
The rear has to deal with static loading by the weight of the car, and then there is the shaft loading, axial and radial, on the internal housing.
The fronts now, you have the static loading then axial loading from scrub forces, which only act on the outer housing.
Will the differences warrant the uses of 2 different bearings, or can that be overlooked for the sake of simplicity and interchangeability?
Not following you here... the load cases shouldn't be dramatically different.
Grip is a four letter word. All opinions are my own and not those of current or previous employers.
A question for you JT, or anyone else for that matter, I would like to try some realistic numbers on a front upright,
with two "angular contact ball bearings" in an arrangement like this;
The image shows an X-arrangement with the 40 degree contact-lines intersecting, while we would probably prefer the reverse, an O-arrangement. But what would a reasonable inner/outer diameter, and distance between them back-to-back, of the bearings for an F1 car be?
Also, if anyone can help me with a suggested load-case?
"I spent most of my money on wine and women...I wasted the rest"
