Bearing Selection and Loads

[pereira1]

Hello Everyone,
We would like to verify that the forces we calculated for our wheel assemblies are reasonable. We will be using these forces to select our hub bearings.
With our current suspension and chassis setup , we have calculated that the highest bearing loads are sustained in the right rear wheel during 1.5 g cornering.
Assuming that only 1 bearing takes the all the lateral load, the maximum radial loading is around 36 kN (in one of the bearings) and the maximum lateral loading is 7kN(in one of the bearings). We also assumed that the car is experiencing a 3g bump in all four of its wheels.
Through equations provided by SKF catalogs, the equivalent bearing load, that has to be used during the bearing selection is about 42.5kN.
Do these numbers seem reasonable to you ?

With those loads in mind, I believe a skf 71915 series bearing will work for us (provided that we increase the distance between the two bearing to around 2.5 inches --- doing that would decrease the loading we calculated earlier). An SKF 7015 series would also do the job ( with the current bearing distance of 2.0 inches)

Despite these calculations, I'm still not very confident in the numbers we found. A quick search of the forums revealed that even the SKF 71912 series bearing (a bearing with a smaller bore diameter -- rated to upto about 34.5kN) is overkill (if you want to check out the discussion I'm referring to, the website is - http://www.fsae.com/forums/showthread.php?9508-Upright-Bearings&highlight=71912 -).

What do you guys think ?

Let me know if you need additional details to provide a better answer,
Thanks,

Amon Pereira
Purdue Electric Racing
Purdue University,
West Lafayette,
Indiana,
USA

[Z]

We will be using these forces to select our hub bearings.
...
What do you guys think ?


Amon,

IMO, all your above choices of bearings are BAD (for similar reasons to those on your linked thread...). This is excusable because you probably don't have any good teachers who can point you in the right direction. Likewise, the Tech Reps at the bearing companies can only tell you what their bearings can and can't do, but, quite reasonably, they have no idea what is good or bad for an FSAE car, all things considered.

So;

1. Why such a large diameter bearing? I am guessing that you want to fit the CV inside it. Why? Because everyone else does it that way?

2. Why "7xxx" series bearings (ie. "Angular Contact Ball Bearings")? What do you know about preloading these ACBBs? How will you seal these? How effective (= strong, stiff, low-weight, low-cost, easy-to-manufacture, and RELIABLE+++), will the whole package be, including inner-hub and outer-upright?

3. Will you use the same large diameter arrangement at the front? Why (or why not)?

4. You mention the working loads, but have you considered the working compliances? Have you followed the many discussions on this Forum regarding the excessive compliances (mainly of toe and camber) of so many FSAE cars, and how this totally ruins their "theoretical" performance?

5. Perhaps most importantly, have you "benchmarked" the "prior art"? That is, (translating those Sys-Eng-buzz-phrases), have you had a look at the wheel-bearings of ride-on-lawnmowers, ATVs, small-cars, big-cars, old-cars, box-trailers, and all the other roughly FSAE-sized vehicles out there, so that you have an idea of what actually works, outside of the very small FSAE world? (And note that the FSAE-world is dominated by cars that CAN NOT drive 30 kms without breaking down...)

Z

[pereira1]

Z,
1. We are choosing a large diameter bearing in order to accommodate a tripod inside of our hubs. The smallest size we can go down to would be a size 14. The advantage with this design is that we can minimize the amount of parts in our drive train and that way also minimize the weight of the wheel assembly.

2. If you check out the skf bearing selection matrix (available at http://tinyurl.com/no359gs), you'll notice that angular contact ball bearings maintains a good combined loading capability while still having a good low friction rating (unlike the deep groove ball bearing and the tapered roller ball bearings).

According to skf, a class B light preload would work for our application. We are planning to have the bearing dimensioned to that specific preload. That way, all we'll have to do is make sure that the inner and outer race spacers have the exact same axial lengt. We are also planning on having the bearings sealed at the factory, so that we won't have to worry of any contamination.
In terms of axial stiffness, unless we increase the preload, we will be probably be looking at about 6.5 micro meters of axial deflection during a 1.5 g corner (assuming that one bearing takes all the axial loading).

In terms of strength, the bearings will definitely work for our dynamic load cases. In terms of weight, the bearing are very light (around 0.3 kg/bearing) compared to other options.
In terms of reliability, the dynamic loadings are well below the C_0 rating of the bearing.

3. In order to use the same bearings and to manufacture similar dimensioned front and rear, We'll be using the same larger diameter arrangement in the front. By doing this, we are looking to keep complexity at a minimum.

4. No I've not considered working compliances. I apologize in advance if this question has been asked before but how much of compliance is enough to " ruin performance"? The axial deformation of the bearings during cornering I calculated (see answer to question 2 for value) seems very minute.

5. We didn't specifically benchmark our bearing arrangements to designs out of the FSAE world. However we did some benchmarking to designs done by other FSAE teams. We studied the hub designs of high ranking teams and tried to see what we could do better. We also made sure packaging issues were taken into account when making decisions.

Why do you think the 7xxxx series is a bad choice ? And What do you think about the loading we have found ?

Regards
Amon Pereira
PS: Before I end this post, I would like to make a comment about us not having good direction. We do in fact have good direction given to us by professor, however in our case our professors do not tell us exactly what to do.

[luxsosis]

Z,
We are planning to have the bearing dimensioned to that specific preload. That way, all we'll have to do is make sure that the inner and outer race spacers have the exact same axial lengt. We are also planning on having the bearings sealed at the factory, so that we won't have to worry of any contamination.


What will happen to this preload if the parts change temperature?

[kcapitan]

Hi Pereira

I worked on the same project last year and picked very similar bearings based on similar criteria. For our rear hubs we also put the tripod joint inside the hub and our front hubs were just a simplified version of the rears with the same bearings.

Your loads are similar to what I used, although I didn't consider a 3G bump on the wheel, just the increase in normal load from weight transfer. If you have wings on your car you might be able to hit higher than 1.5G in the corners, depending on what speed you're assuming.

I picked unsealed bearings and trying to design seals was a giant pain. I was never really satisfied with the design and I don't think the bearings will have the life of a design with better seals. Definitely make sure you can get those bearings sealed and the seals will be both water and dust tight.

In regards to the stiffness, you've done your homework better than I did. I spoke with engineers at INA and they recommended we use two angular contact ball bearings in an O-configuration (line of contact through the bearing points out of the hub, as opposed to an X-configuration which is the opposite) as the O-configuration would be much stiffer than two deep groove ball bearings. From what I saw, most passenger cars use a single, double row ball bearing, but these tended to be heavier than two single row bearings and I couldn't find anything in the size I wanted.

All-in-all it sounds like you've done your homework. If you're unsure about your design I recommend you talk to your professor or see if your uni has contacts at SKF. The guys I worked with at INA were Formula alumni and very friendly.

Regards
Kyle

[Z]

Amon,

Thank you for thinking about and answering each of my questions. Here are my comments on each of these issues, with these comments intended to improve your understanding of Engineering.

I list the following in reverse order, namely from the more general to the specific.
~o0o~


We do in fact have good direction given to us by professor ...

I can not see how your professor letting you get this far with your particular choice of bearings, without you also covering the other options, is a good thing for your education.
~o0o~


5. We didn't specifically benchmark our bearing arrangements to designs out of the FSAE world.
... we did some benchmarking to designs done by other FSAE teams.

As I noted before, the distinguishing feature of FSAE cars is that they fall apart a lot. I am sure that if you did a more detailed "benchmarking" of most FSAE wheel-bearing designs, then you would find that, to a large degree, they are very UNreliable. (Anyone care to share their stories? :))

I am familiar with one FSAE team that uses a similar design to yours. In fact, their's is somewhat more "optimised" in that they use 6816 DGBBs (IIRC), which are less than half the weight of your proposed 7915s or 71915s (ie. about 160g vs your 350g). In one recent year when the team got the car finished early enough for some genuine testing, the bearings failed TWICE before they even got to comp (again, IIRC).

And now the punch-line. When I suggested that perhaps they should redesign the bearing-package to something more robust, the response was "Oh no, ... they are supposed to do that!" This was something to do with Colin Chapman's credo of the perfectly designed racecar being one that falls apart just as it crosses the finish-line. Except that this particular car was regularly falling apart long before it even got to the finish-line, and at the expense of significant damage to other parts of the car, and even less testing than the meagre amount that they might have got...

Somehow, I think that your professor, and probably far too many other FSAEers, believe that this is how "real racing" is supposed to work. Please keep reading this post to see how stupid this is...
~o0o~


4. No I've not considered working compliances.
... how much of compliance is enough to " ruin performance"?
The axial deformation of the bearings ... seems very minute.

The types of compliance (given in my first post, and NOT "axial"), and how much of them is bad, has been covered many times before on this Forum. It is basic VD.
~o0o~


3. ... We'll be using the same larger diameter arrangement in the front ... to keep complexity at a minimum.

You should realise that the real reason that you are using large-diameter bearings front and rear is because you have seen many other teams doing it, and "it looks cool". There are many simpler (ie. LESS complex) arrangements that you can use F&R.
~o0o~


2. If you check out the skf bearing selection matrix (available at http://tinyurl.com/no359gs), you'll notice that angular contact ball bearings maintains a good combined loading capability while still having a good low friction rating (unlike the deep groove ball bearing and the tapered roller ball bearings)...

How "good", and how "low", is a "good low friction rating"???

I also noticed on the SKF page,

"The matrix permits only a relatively superficial classification of bearing types.
... some properties do not depend solely on bearing design. For example, the stiffness of an arrangement incorporating angular contact ball bearings or tapered roller bearings also depends on the applied preload and the operating speed which is influenced by the precision of the bearing and its associated components as well as by the cage design."

An example of where you will go wrong is here,


..., all we'll have to do is make sure that the inner and outer race spacers have the exact same axial length.

"Exact"??? And your professor is teaching you Engineering???
~o0o~


1. ... The advantage with this design is that we can minimize the amount of parts in our drive train and that way also minimize the weight of the wheel assembly.

As noted before, there are many wheel bearing arrangements that are better than the ones you are proposing. But YOU will have to go looking for them outside of FSAE.

As just one hint, for your front-wheels you could use an easily designed and manufactured "small box-trailer" type package that will be stronger, stiffer, lighter, and far more durable (it could outlast the rest of the car) than your 71915s. And I can buy the full bearing-and-seal package for $9.99 at any Cheap-Auto-Parts-R-Us store in Australia (don't know about your country).

By contrast, your 71915s are described by SKF as "super-precision" bearings, and have a speed rating more than 10x faster than you will ever need (they are intended for high-speed machine tools, with the corresponding precision required for pre-load, etc...). So how much do you think they will $$$cost? (Hint: your first post linked to a similar discusion). Or does your professor believe that $$$s do not matter in Engineering?

Bottom line, you can have a much better bearing package in terms of FSAE performance (ie. lighter, stronger+++) than the ones you are proposing, with this "better" package ALSO being much cheaper and more reliable. But you are unlikely to find such at the "$exy" (ie. "optimised") end of FSAE.

Z

[Markus]

As I noted before, the distinguishing feature of FSAE cars is that they fall apart a lot. I am sure that if you did a more detailed "benchmarking" of most FSAE wheel-bearing designs, then you would find that, to a large degree, they are very UNreliable. (Anyone care to share their stories? :))

I am familiar with one FSAE team that uses a similar design to yours. In fact, their's is somewhat more "optimised" in that they use 6816 DGBBs (IIRC), which are less than half the weight of your proposed 7915s or 71915s (ie. about 160g vs your 350g). In one recent year when the team got the car finished early enough for some genuine testing, the bearings failed TWICE before they even got to comp (again, IIRC).

I can share our story. We used a very similar arrangment to what the OP has come up with, but with lighter bearings (140g per bearing).

We did some ~1500kms of testing and racing the first season. After checking the bearings they were still like new. We've done some ~700kms more after that, and have no intention of stopping.

About actual loading: we do have a very lightweight steel spaceframe, 13", 4-cyl car. Because of our non-aero "piece-of-junk" we only hit ~2g's in FSH. This car might be almost everything Z despites but it still proved very un-succesful... :rolleyes:

And because it's a fragile FSAE car just like the rest of the top cars, we have also taken it to do some ice racing. After the ice track was driven with rally cars. Talk about bumby! :)

[Z]

Old saying;
"A good Engineer is someone who can do with $10, what any Fool can do with $100."
~~~o0o~~~


I can share our story. We used a very similar arrangment to what the OP has come up with, but with lighter bearings (140g per bearing).

Markus,

Thank you for sharing your story with us. You must be very proud that your wheel bearings managed to complete OVER 2,000 km (!) without falling apart. That is certainly better than the other parts of your car that you have told us about in the past. :)

I am quite sure that a WELL DESIGNED bearing package such as yours could manage that sort of lifetime. Given your quoted 140 g weight, I am guessing that they are 78xx series bearings, namely "Xtra-Lite ACBBs"? (Note that Amon's suggested 79xx is the heavier "Lite" series.)

(I should stress here to other FSAEers that the key to getting these "thin-ring" ACBBs to work reliably is to have;
1. Just the right amount of axial preload (ie. requires accurate machining), and,
2. Support of the inner and outer races that is stiff enough to remain very close to circular, because when the races get squashed out-of-round only a few balls are forced to carry all the load, so they fail...
I guess Markus's team managed to get this right. Many others haven't.)
~~~o0o~~~

But here is the key issue;
Is Markus's bearing package really good Engineering?

Let's compare Markus's package with the Box-Trailer example I gave earlier (and there are at least two more broad ranges of wheel-bearing types I would also consider).

(FWIW, in Oz the BT-wheel&bearings have been de-facto standardised to either the older model Ford, or GM-Holden, front-wheels, with their old-style tapered-roller bearings on a fixed spindle. Modern Fords and Holdens (and most other makes...) use "non-serviceable" hub-axles-with-integrated-ACBBs, which are generally LESS reliable, and more costly to repair. Ah, progress...)

MASS.
======
1. M's bearings are 140 g each (x 2), plus seals(?), plus large diameter inner and outer mountings.

2. BT bearings are smaller diameter, but on average similar mass to above. Their smaller diameter means that the necessary mountings should weigh less than M's.

But let's say, EVEN on MASS.
~o0o~

LOAD CAPACITY and DURABILITY.
============================
1. M's bearings have carried an FSAE car, with, say, about 100 kg/corner, for 2,000+ kms, so far.

2. BT bearings (as noted, originally fitted to medium sized passenger cars) typically carry 500+ kg/corner, and have been proven, many millions of times over, to last for many hundreds of thousands of kilometres (ie. 100s x M's).

So BIG, PROVEN, DURABILITY ADVANTAGE to BT.
~o0o~

AVAILABILITY and COST.
====================
1. "Xtra-Lite" bearings are rarely stocked at your local, neighbourhood, "Bearings-R-Us" store. Their rarity also makes them more expensive than the more standard sizes. That is what mass-production is all about. They are often MUCH more expensive. Especially so, when bought as "super-precision matched sets".

2. As noted before, BT bearings are available almost everywhere in OZ (even in Fishing supply stores!), at ~$9.99 per shrink-wrapped plastic package (2 x bearings + seal). Well, those cheapies come from China, but similar standard sizes from a reputable bearing company might be, say, ~Euro20, for the complete package.

So, Markus, how much did your bearings cost (true RRP, please)?

And is that extra cost really good Engineering?

Z

[mech5496]

Ζ,

We use DGBB (thin section ones) from SKF, that are around 120gr each, sealed and cost around 20EUR each, never ever caused us any trouble. We had used ACRB's a few years back, and IMO they sucked. Yes, they were light (about 150gr each if I am not mistaken), but they were not sealed, their small diameter meant we had to have a much more complicated hub assembly, in rears we could not get too small because the axle needed to transmit the diving torque, small diameter meant the splines had to be steel etc etc. Long story short, our "bad" DGBB assembly weighs less than half of that old design now...

[Z]

Here is the link to the "Upright Bearings" (http://www.fsae.com/forums/showthread.php?9508-Upright-Bearings&highlight=71912) thread again. There is some relevant discussion there regarding costs, etc. (ie. Francis gets it).

Also, the link from that thread given by EPMPaul for their 61913 bearings (http://www.vxb.com/page/bearings/PROD/Kit8739_1). Note that these 65x90x13 Deep-Groove-Ball-Bearings have a load rating of ~17 kN, a mass of ~200 g each (from my catalogue), and are quoted at $50+ each (wholesale price!).

This is what Paul had to say on that thread.

"Another possible worst load case might also be when you (over)preload your bearings so you might want to be careful about that.
...
...doing the calcs, it came out to a life of around 1700 km based on some lapsim estimates of how much time you spend on each loadcase.
...
Never got around to really checking stiffness(will probably be my downfall in design).
...
Also, keep in mind how you're gonna assemble these things. Representatives sometimes assume that you have access to an induction heater and the whole setup..."

So, ..... yet another DNF in Endurance?

No problems! You just say, "Well, this is FSAE, and the bearings are supposed to do that...". :)

Z

[jpusb]

At my 3rd world team we used large (at the back) and small (front) diameter DGBBs for years. I won't say we were a top team but at least we were consistently in the top 15 in AutoX. I never saw a bearing fail from 2007 to 2010 (and plenty of kms in between). I have even jumped on the car (have pictures, bumpy track, landing sideways sometimes BTW, 3 to 4 inches for about 15 laps over and over again). I have hit countless types of bumps. In MIS 2008 the endurance track forced you to hit the oval bank diagonally, really hard (right front felt like hitting a curb or worse). wing teams were even advised that they would rip the wings off because the angle was significant. On the car it felt like a hard impact curb (but for longer time since the bank doesn't level as a curb does). In MIS 2010 again, but not so bad.

I won't say exact code but they were about 130g each, inside diameter in the range of what people have put here. So with that evidence I can confidently say that you can go light, strong, and stiff. I won't say they were cheap (they were NOT $20 each), but (as with the rest of the car) you have to choose between weight, strength, and reliability. It is not like they costed $20k so in the end I would think cost should not be a problem for bearings, as long as you don't go crazy and buy some super bearing. To save cost I would say go to sheet metal welded uprights. You can get them light as hell, and cheaper than your cereal.

My guess and advice would be that, if you choose DGBBs (regardless the weight and maybe even the load rating, for the sizes we are talking here a large group of options would handle the loads), the real reliability comes from the design concept: how they go in, tolerances, preload, axial spacing between them, locking system, how stiff their housing is, etc. I have seen really bad designs in some cars and in such case the problem is not with numbers but with concepts, and there is nothing a super bearing will do for your compliances.

Hope I helped in some way.

JP

[yeungk]

After reading all these useful inputs, my understanding so far is that, correct me if i'm wrong,

1) DGBB's greatest advantages are its economical cost and availability

2) ACBB may take more axial load, but costs more

3) super precision ACBB costs even more, and at FSAE's racing speeds, is it more than what we needed

I agree that using commercially available hub bearings/hub assemblies may drastically reduce the cost on these particular parts and the small package may have less weight, such as the box trailer bearings mentioned by Z. But it seems like when paired with the upright, the whole wheel assembly would be of similar weight for both kinds of bearings. However, would the smaller bearing introduce more compliance compare to the larger bearing?