Live spindle with aluminum hubs

[Pavan Dendi]

For those running a live spindle in the front with aluminum hubs, do your bearings ride directly on the aluminum hub? Is there some sort of coating or hardening (anodizing?) that you use so the aluminum doesn't get damaged? Or do you have some sort of steel sleeve pressed on to the hub so the bearings have a steel surface to ride on?

Also, what advantages/disadvantages have people found in running a live spindle in the front?

_____________________
UT Austin FSAE
http://www.me.utexas.edu/~fsae

[Denny Trimble]

I'd be slightly worried using aluminum in a rotating bending load case. Steel or titanium might be more appropriate here, in my opinion. You would need a much larger diameter (thus larger, more expensive bearings) to have the same fatigue life with aluminum compared to, say, hollow 4340 at RC40.

I've only designed fixed front spindles, but we're considering live spindles for next year.

University of Washington Formula SAE ('98, '99, '03, '04)

[Marc Jaxa-Rozen]

Sorry for the newbishness, but's the difference between fixed and live front spindles?

Thanks-

Marc
ENA

[Charlie]

I wouldn't assume bigger bearings are more expensive, at least thats not always our experience. http://fsae.com/groupee_common/emoticons/icon_wink.gif

We've run 7075 aluminum live spindle hubs since 1997. Actually the same units were used on the 1997, 1998, 1999, 2000, and 2001 cars! And the rotating bending did not cause failure of the hub ever ('01 car is still around). Its definitely something you need to consider in design. You will have to have a bigger diameter bearing, and there is a weight penalty there, its all a bunch of compromises like anything else.

This year we are losing the aluminum live spindle and going to a steel fixed spindle in an aluminum upright for various reasons...

-Charlie Ping

I just need enough to tide me over until I need more.

[Pavan Dendi]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Charlie:
...

This year we are losing the aluminum live spindle and going to a steel fixed spindle in an aluminum upright for various reasons...

-Charlie Ping
<HR></BLOCKQUOTE>

We are using steel spindles in an aluminum upright on the 2003 car. Ours are attached by pressing them in and then retaining them in the upright with a giant bolt in the back. We've noticed a few problems with this...the steel has elongated the hole in the upright and there is some play. I think this might be due to the combination of bending loads and the differences in toughness and thermal expansion properties between the two metals.

How are ya'll planning on attaching the steel spindle to the upright? Epoxy?

_____________________
UT Austin FSAE
http://www.me.utexas.edu/~fsae

[Schumi_Jr]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>This year we are losing the aluminum live spindle and going to a steel fixed spindle in an aluminum upright for various reasons...
<HR></BLOCKQUOTE>

We are also using a steel spindle pressed into an aluminum upright. Care to share why you are moving away from the rotating spindle?

Aaron Johnston
University of Waterloo FSAE

www.eng.uwaterloo.ca/~fsae (http://www.eng.uwaterloo.ca/~fsae)

[Charlie]

It's not my area, and right now I'm knee deep in Engine parts, but I know that by placing the bearing outward it reduces the required bearing size, and like I said earlier with a steel inner plug your bearing ID can become much smaller.

-Charlie Ping

I just need enough to tide me over until I need more.

[Steve-Imperial]

Im currently designing a fixed splindle upright for our new car. Its gonna have a steel axle pressed into a steel fabricated upright to avoid destortion of the bearing surfaces during welding.
On our previous car we used a "live" ali spindle with ali upright and 35mm ID bearings and so far they have been OK. The main problem I see is that the bearings are so far from the wheel centreline that their loads are pretty much doubled I think. With the fixed axle the bearings can be either side of the centreline greatly reducing loads. This means lighter beaings can be specced so the package is gonna be lighter overall (hopefully!!)

[ben]

I would weld the stub axle into the upright. We've done that for the last 5 years and had no problems with the bearing surfaces distorting.

Ben

University of Birmingham
www.ubracing.co.uk (http://www.ubracing.co.uk)

[Big Daddy]

When you weld it just tack it all the way around in 4-8 places this will minimize pull when you lay the beads down. I would weld a bead between two spots then go 180 degrees and weld a bead there then let it cool. Repeat and viola http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

"A woman is a lot like a beer, they look good, smell good, and you would run over your own mother to get one." Homer Simpson

[Steve-Imperial]

So do you weld the upright parts directly onto the spindle or weld the parts onto a "ring" and then weld the stub axle into that? If the second option then thats pretty much what im planing on, except without the welding....

[ben]

We weld the upright bits straight to the stub axle. I would have thought you'd have to make the upright pretty hefty to not deform with the sort of press fit you'd need to locate the axle as well as a weld could.

Our uprights are 1.2mm wall thickness BTW.

Ben

University of Birmingham
www.ubracing.co.uk (http://www.ubracing.co.uk)

[Brent Howard]

We have always used the set-up that nearly everyone else is describing, but I thought I would add one note.

Last year we had both uprights fail through fatigue/shear on the weld surface. They had been run about 40 hours - 45 hours each when they failed (one in endurance, one in testing afterwards). The weld was porous...very porous with little penetration. However keep this failure mode in mind when considering your design. Also, we did run 2-3 endurance laps with that broken and only the caliper holding the wheel on. The only thing that saved us was that the 10" rims ware close enough in size to the rotors that it wasn't wobbling enough for us to be pulled off. Our rim sure didn't appreciate it and the driver wouldn't touch the brakes, but we didn't get passed.

http://www.enme.ucalgary.ca/fsae/Pictures/03/PM1.gif

Brent

www.ucalgary.ca/fsae (http://www.ucalgary.ca/fsae)

[Steve-Imperial]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by ben:
We weld the upright bits straight to the stub axle. I would have thought you'd have to make the upright pretty hefty to not deform with the sort of press fit you'd need to locate the axle as well as a weld could.

Our uprights are 1.2mm wall thickness BTW.

Ben
<HR></BLOCKQUOTE>

Well I was planning to leave a shoulder on the inward end of the spindle and the end nut to hold the assembly together so only a slight interference fit would be needed.

1.2mm wall thickness seems really thin, im using 3mm in my current design!! Any chance of seeing a photo of your design? Photos of mine:
http://img27.photobucket.com/albums/v81/steve-o7628/upright26204no2jpg.jpg
http://img27.photobucket.com/albums/v81/steve-o7628/upright_26.jpg

The total weight is about 1.4 kilos

[ben]

You can kind of see it in this shot from our FS2003 design event judging. It's just a 1"*2" box section with a 30mm od steel stub axle welded in as discussed.

http://ubracing.markdooley.com/gallery/car_17/DVC00266

Ben

University of Birmingham
www.ubracing.co.uk (http://www.ubracing.co.uk)

[Steve-Imperial]

looks a lot simpler than my design and probably lighter too....

[Pavan Dendi]

So no one else has had any problems with steel spindles in aluminum uprights developing play over time?

Also, for a live spindle in the rear, how are people attaching the outer CV to the hub? Currently, our design calls for a two part hub. The inner part has male sinusoidal splines and a flange that bolts to the tripod housing. The outer part has the female splines and the two parts bolt together capturing the hub in the upright.

Does anybody else do it this way, or is there a simpler way to do this? Regular aluminum splines don't really sound like they will last long and are a pain to machine. Bolting the two parts together with several bolts would make assembly/disassembly more complicated but the machining is easy. It's also pretty tough to fit decent sized bolts in there.

Any thoughts?

_____________________
UT Austin FSAE
http://www.me.utexas.edu/~fsae

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Steve-Imperial:
looks a lot simpler than my design and probably lighter too....<HR></BLOCKQUOTE>

The key is to make the caliper mount stiff enough. I wasn't happy with that aspect of our 2003 design (I ended up doing someone elses job in a hurry). We will be rectifying that this year.

Ben

University of Birmingham
www.ubracing.co.uk (http://www.ubracing.co.uk)

[Omer]

schumi_jr and all

"We are also using a steel spindle pressed into an aluminum upright."

did you use a steel sleeve inserted into the aluminum upright to prevent the creation of freedom between the steel spindle and aluminum upright.
we are looking on building our own magnesium upright does anyone has experience with that?

[Cement Legs]

We did the same thing to our current uprights and as long as the interference fit is close I dont think you will have any problems. I would also consider heating your upright to around 150 F and putting yourspindle in the fridge or freezer before hand to help reduce the materail displaced from your interference fit during pressing.

[Akos]

Over the FSAE cars I worked on we ran practically every permuation mentioned above.

Steel spindle pressed into an AL upright works only if its very tight fit and with the upright wide enough so the stress near the spindle base is not too high. The best method is what Cement Legs mentioned. Trying to use a single bolt to hold the spindle in alway resulted in sloppy spindle over time.

You can run the bearings directly on a AL hub, this is the setup on the Formula Renault. The hub should be anodized and a pretty good press fit into the bearing. Also make sure that there is no chance for any play between the two surfaces, as the bearing will chew up the hub in no time.

Steve-Imperial: Bearing loads have not much to do with the bearing distance from wheel centerline. Your biggest load on wheel bearings comes from cornering. Here what determines the load is the distance between the two bearings. Live spindle or not, you can use the same size bearings as long as they are the same distance apart.

[welder]

Just an aside regarding the alloy hubs - I am in the process of building a similar type of motorcycle powered car.

I had originally intended to hard anodise the 7075
hubs, but was advised on another forum that the result can be a 50% reduction in fatigue life.

So unless you subscribe to the Chapmanesque school of design, might be something to reconsider. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Jersey Tom]

For cyclic loaded parts I would go 7475 over 7075.

[Micko..]

hi welder

so what is the go with the 50% reduction in fatigue life for the anodised parts? i had always thought that the anodising improves fatigue life? maybe i have to do some more reading. we are running a live spindle with ally hubs and are getting the hubs hard anodized (mostly for the bearing surfaces) this is not my area but i am interested to hear more about the impact of anodizing on fatigue life.

cheers

[Jersey Tom]

Anodize is "good for mitigation of stress-corrosion cracking failures", from what I'm told by a design engineer at Lockheed.

I have heard that it can have an impact on fatigue life. I'm actually going to contact Denver Metal Finishing later today for a couple things. They're pretty solid guys. I'll see what they have to say.

[Denny Trimble]

Jersey Tom,
Where do you get your 7475?

A quick matweb search showed you're right on fatigue strength:

All units KSI

Alloy__HT____Sy____Fatigue Strength (10e6 cycles)
6061___T6____40____14
7075___T73___63.1__21.8
7475___T735__61____31.9

[Jersey Tom]

Yea. 7475 and 2124 I'd take over 7075 and 2024 respectively.

I haven't gotten any 7475 yet, Denny. This is my first year on the design team. Doing a redesign of our uprights and most likely it will be one of those two alloys. Neither of the are too exotic/scarce.

There are a couple places I'm looking into though. There's a place in Denver, ALRECO. Supposedly have a ton of recycled and stock aluminum alloys. But I've never dealt with them and theres no chance in hell I'd get any sort of certs with material.

Other options are Reliance Steel and Aluminum (rsac.com) and Transtar Metal (transtarmetals.com)

[welder]

Jersey Tom, over here in the third world http://fsae.com/groupee_common/emoticons/icon_razz.gif I'm suffering from offcut envy, one of the local suppliers over here considers 6061 exotic so 7475 would be other worldly. http://fsae.com/groupee_common/emoticons/icon_smile.gif The remaining outlet, since my "tame" storeman retired no longer permits the likes of me to scrounge due to OH&S and liability concerns, so everything is sold as scrap - wouldn't you just love to punch a lawyer in the throat!!

Miko, the fix suggested to me, was to machine the anodised finish from stressed areas. As Jersey Tom says you then have the corrosion considerations. Perhaps the alternative is a regime of crack testing and lifing - the engineering department would have NDT facilities I assume?

Anyway, one of Ron T's old crew has told me of another coating to try, I have my doubts about it's ability to withstand bearing loads but time will tell.

[Draksia]

I am thinking about some 7475 aluminum drive line parts and I was wondering is any one found more about what fatigue life is after hard anodization. I have little info on the process. So any more info would be helpful.

[Joest]

Just out of curiosity, why don't you guys use bearing that come complete with steel inner, outer races, and seals? I wouldn't even dream of allowing the roller element to contact the aluminum surfaces inside the upright. It seems that the perfect bearing for FSAE type car would be a fully sealed double row angular contact ball bearing. They are available from supply houses like Pegasus. Many formula cars use them and they are relatively inexpensive. Try this link:

http://www.pegasusautoracing.com/ProductDetails.asp?RecId=294

This particular one, which is for a Swift chassis, might be overkill for FSAE, but companies such as SKF, Koyo, FAG, and Timken make other sizes that will work. Of course a sufficiently large ID is required on the driven wheels so that the axles will slide through the bearing.

When you machine your upright, leave a shoulder on one side and retain the other side with a c-clip or backing plate (low-load side). You will also need a slight interference fit between the bearing outer race and the upright (maybe a 0.001"�) so when the upright is heated to say 200-250F the bearing will slide right in. Be careful how you heat 7075-T6 as it has been show to loose up to 50% strength if held at over 300F. You may want to keep this in mind when designing the uprights since they are exposed to heat from the brakes. Luckily uprights are typically designed for stiffness as opposed to stress limits therefore it should not be an issue in most cases.

-Joest

[Danieler]

Nonsence!!!
WE all know that you are biulding your upright from a regular steel that you stoll from the lab of prof` Zaretsky... http://fsae.com/groupee_common/emoticons/icon_biggrin.gif
http://fsae.com/groupee_common/emoticons/icon_biggrin.gif
Originally posted by Omer:
schumi_jr and all

"We are also using a steel spindle pressed into an aluminum upright."

did you use a steel sleeve inserted into the aluminum upright to prevent the creation of freedom between the steel spindle and aluminum upright.
we are looking on building our own magnesium upright does anyone has experience with that?

[Jersey Tom]

I was unable to locate a source from which I could procure any 7475. Instead I will be machining our uprights this year from a different alloy with 10-20% higher yield and ultimate tensile stress than 7075. Or at least that's the plan if there's enough in stock. If not, it will most likely be 7075 in the T7 or T6 condition.

I couldn't find anything concrete on anodization and fatigue life effects... though I'd think hard anodize would be less liable to create microcracks than convention anodize, since its a cold-water bath process, as opposed to dunking it in acid.

In any case I'll be using electroless nickel instead of anodization. Looks beautiful, and a much more even coverage in complex geometry than anodize.

With regard to bearings, we had been using double-row angular contact bearings in our wheel assembly for the past couple years. Got them straight from Taylor Race, as they fit their drive components perfectly. Big and heavy though. Going with a different type of bearing this year, should be much lighter and much more compact, while keeping the same ID.

Edit - Actually what we had used previous years looks the exact same as what Joest pointed out.

[Chris Allbee]

I once had a reliability engineer from Honda R&D tell me that anodizing reduces the tensile fatigue strength (thats the one most often quoted by the way, shear is usually a lot less) by about 15-20% depending on the alloy. However, polishing or shot peening critical surfaces can raise the fatigue life by about the same amount. I'd have to dig up my notes from the conversation for the exact figures and the order he recommended that they be taken.

[Jeremiah]

This is from "Mechanical Engineering Design 7th edition" page 335 under electolytic plating...

"Metallic coatings, such as chromium plating, nickel plating, or cadmium plating, reduce the endurance limit by as much as 50 percent. In some cases the reduction by coatings has been so severe that it has been necessary to eliminate the plating process. Zinc plating does no affect the fatigue strength. Anodic oxidation of light alloys reduces bending endurance limits by as much as 39 percent but has no effect on the torsional endurance limit."

[Jersey Tom]

Yea, the question though becomes -

What is the effect of each grade and class of coating on fatigue life?

And what of electroless processes? I haven't been able to find this kinda info anywhere.

Looks like it might not be a bad idea to glass bead shot peen these things, then electropolish, before going for the plating. I'd really rather not run bare aluminum, especially 7000 series.

[twpvw5]

Back to the original question, here at MU we use fixed front spindles made from 4130 pressed into Al6061 uprights, we have never had one fail or get loose. As for the rears we press the bearings, used to be double angular contact now two seprate sealed w/ a spacer in between, into the Al6061 uprights then Al hubs with a keyway then more 4130 for the inboard side with the brake rotors and CV housings.

[Omer]

Originally posted by Tyler Pickett:
Back to the original question, here at MU we use fixed front spindles made from 4130 pressed into Al6061 uprights, .

did you use a big nut at the end?
or a different method.

[twpvw5]

Yeah, we do use a big nut to hold everything together on all four corners, 3/4x16 IIRC.

[consig]

just wondering how you guys pick your interference fits. Do you just machine the parts yourself and leave the spindles a few mills over or do you use a certain ISO grade interference?

I am plaaning on using a grade S6 fit for the shaft. This makes a 24mm diam. shaft oversized by .021-.048mm

[Jersey Tom]

You guys are still using the metric system? C'mon.. get with the times..

.048mm on a 24mm shaft is HUGE.

Rule of thumb I use for light to medium press fits is .0005-.0008" interference per 1" shaft diameter. Been very successful with it.

So for you, 24mm shaft would be oversize by about .013-.020mm.

Just need to be aware of how you're manufacturing the hole and shaft. The pocket for our wheel bearing to sit in I circ. interp'd on a fairly new (2002) VMC. Hurco VM1. Pretty good machine, but even then, cuts circular pockets about .0005" wide in one direction!

[consig]

I agree thats quite heavy. Meant to be a medium to heavy fit. I guess its a good idea to find what kind of stresses that puts on the upright to make sure it is shaped correctly in order to stay in the elastic deformation range.

Very good point about the circular interpolation on CNC jobs. Might try to go find a 24mm end mill, hah

[Jersey Tom]

24mm, I'd just helical bore it with a nice center cutting EM and then come back with a boring bar. Then again its possible you can circ interp it and get it good enough.

[buggaero]

Originally posted by Steve-Imperial:
On our previous car we used a "live" ali spindle with ali upright and 35mm ID bearings and so far they have been OK. The main problem I see is that the bearings are so far from the wheel centreline that their loads are pretty much doubled I think. With the fixed axle the bearings can be either side of the centreline greatly reducing loads.


Originally posted by Akos:
Steve-Imperial: Bearing loads have not much to do with the bearing distance from wheel centerline. Your biggest load on wheel bearings comes from cornering. Here what determines the load is the distance between the two bearings. Live spindle or not, you can use the same size bearings as long as they are the same distance apart.


I am in the process of designing our live spindle in aluminium hubs with Al uprights using a DRACBB. And sure i have my lateral loads much more than the vertical reaction from tyre contact patch. Been through a number of posts in the forum and ended up confused with what to include and what not in the FBD for loads on the bearing. The bearing runs on the hub and is 13mm outward of the wheel centerline.
I may be asking a very stupid question but if only i knew how stupid it is (like Claude says, 10 mins of shame is better than lifetime of ignorance), do i have to assume reaction from ground at the wheel centerline even when i run a positive offset of 2.5" . What's confusing me is, should the bearing overhang be taken from wheel rim or from wheel centerline? Does it work like, the reaction from the ground applies a bending moment at a distance equal to the distance of hub face (attached to rim) from centerline. And a separate moment in opposite direction going through the shaft at the distance of bearing from the hub face. And the reaction due to unsprung mass be separated from it.And these moment couples be resolved to find out the overall bending moment. Just a little bit of light on this and i will be able to proceed further in it. I had supposed earlier that the reaction due to sprung mass be passing through the bearing and the farther the bearing from the hub face, the more the bending moment.

And in case of a fixed spindle setup, the two tapered rollers being on the same side of the centerline is inferior to one with bearings on either side, giving more support..

[sTaRk]

Originally posted by Pavan Dendi:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Charlie:
...

This year we are losing the aluminum live spindle and going to a steel fixed spindle in an aluminum upright for various reasons...

-Charlie Ping


We are using steel spindles in an aluminum upright on the 2003 car. Ours are attached by pressing them in and then retaining them in the upright with a giant bolt in the back. We've noticed a few problems with this...the steel has elongated the hole in the upright and there is some play. I think this might be due to the combination of bending loads and the differences in toughness and thermal expansion properties between the two metals.

How are ya'll planning on attaching the steel spindle to the upright? Epoxy?

_____________________
UT Austin FSAE
http://www.me.utexas.edu/~fsae (http://www.me.utexas.edu/%7Efsae) </div></BLOCKQUOTE>


@PAVAN
I am considering aluminium hubs with steel spindle for the front. Can u post any picture of your bolted configuration of aluminium uprights with steel spindle. This will help us a lot..