Skip to main content

Gents:

Some here may recall an embarrassing incident on the Rattler in NC this fall, Tour d'Smo.  Its not important to the message if you do not recall, but I think I figured out what was happening.  At the time, I suspected loose front wheel bearings introducing a little slop in the steering, and indeed that was the case, although I'm going to say that was minor.  Also I was able to fix that with a small adjustment to the clamping nuts on the front spindles.  But the larger problem was not fixed and indeed got a lot worse.  Driving the car after towing it 500 miles home, something terrible was certainly at play with the suspension.  I did hit a terrific pot hole on the last run on the Tour d'Smo, so figured maybe something important  broke??  @Stan Galat had an issue maybe from a similar encounter: he in fact broke something.  So I started to look around by shaking some wheels.  The rear left wheel seemed a little loose actually, and given what was going on with the front bearings, bearing wear was suspected.  I also noticed that the tire, which normally has a really close (but not touching) fit to the body was actually rubbing the FG. So I began a disassembly.  What I found is in the picture: The brake rotor spline has been worn pretty close to gone.  Talk about wobbly.  So i will need a new one of those, and I'm on to getting one.  All of that is preamble to my question to all you technical wizards, which is this:  Why??   I did notice that before I loosened the castle nut, there was some axial slop in the lash-up.  There is a spacer between the nut and the hub because the axle spline extends beyond the length of the hub spline, and I could turn that spacer with my fingers before loosening the nut; it should be clamped down tight to the bearings by the castle nut.  Hmmm  ... .  So I thought the bearings were indeed loose/worn and must be replaced. I Have not removed said bearings yet, but plan to.  Again the question: how does that outer spline in the hub wear itself down to about gone?  What lets that happen?  Cheap materials?  A natural consequence of worn bearings? Too many burn-outs??   I really did  drive the wheels off the thing on all those twisty mountain roads?  This curious mind  needs to know.

Rotor

2007 JPS MotorSports Speedster

Attachments

Images (1)
  • Rotor
Original Post

Replies sorted oldest to newest

PS: I had some trouble locating all the bearings and seals etc for the rebuild. No luck at my local stores that usually carry what I need; other on-line outlets list the parts, but turn out to be out of stock ... The folks at CB Performance did come through however, with two full sets of pieces and parts, shipped out chop-chop quick.  I mean if one bearing set is bad, the other must be too, right?

Purchased a used VS year ago on CT. While flat towing the speedster it started to sway quite a bit L & R and found the read drum was loose to the point I could turn the Castle nut with a vise grip. I cut some aluminum strips for shims from a aluminum soda can, set them in the drum spline opening and bending over the ends to make tabs. Sat on my butt kicking and forcing the drum on the axle shaft until it seated... Castle nut as tight as I could by pounding the vise crip with a rock and made the 250 trip back home.

I'm sorry for the trouble, @El Frazoo, but glad you found the problem.

I doubt your hub rounded out on the pothole where my suspension broke, but the roads and the way we drive on them surely contributed to both equipment failures. We're pushing well past the design parameters (assuming there were any) for aftermarket VW parts, which in most cases don't even approach the quality of OEM parts in a German economy car last built a half-century ago.

Tearing through the mountains at twice the posted speed limit was never what the Sainted German Engineers imagined, and the suspension and brake parts were never designed for how we use them. The absolute weakest of weak links is how the rear hubs are fastened to the car. It's a perfectly fine setup for profiling, and comically inadequate for how we use them.

Tom Boney had a similar failure in his IM on the same roads a few years back. I've actually torn the friction-welded center out of a CB rear hub several times channeling my inner hick-town, white-trash hooligan. The next setup on the car will be a CSP (I'd do an AirKewld, but the Euro is in the toilet, and Pete's pricing has gone nuts).

Suffice it to say, I feel your pain. The ultimate fix is to spend the long money to try to ensure it doesn't happen again, and to check the tightness of the castle-nuts frequently. When you think about what is actually happening, you can understand why this nut being really, really tight is the only thing standing between you and a pretty catastrophic failure.

The advice Gordon and Al give regarding the torque of the castle nut is good, but I've learned (out here by myself) that 217 lb ft or 250 lb ft (or whatever) is the barest minimum. You do you, but I'm torquing that particular nut as tight as I can get it. The chances of stripping it are near zero, and the downside of it being too loose is exceedingly bad. I go to at least 400 lb ft, but I'm a spot-torque man on this particular piece (torque it until you see spots). In this instance, I'd tighten with a breaker bar and cheater, and just check it with a torque wrench set to 300 lb ft. I just want the wrench not to click.

Other, better, more careful professional mechanics will be aghast at my ham-handed approach, but it's worked for me, and I'm a lowland gorilla who drives like his scalp is on fire.

I don't want to be accused of being a disciple here, but Stan is correct.

I am also of a mind to tighten the bejeesus out of this particular nut.

I use a 3 foot aluminum pipe over my 3/4" drive breaker bar. I lean on it pretty hard and give a couple bounces, so I'm at least 450 ft. lbs.

Same deal on the gland nut/flywheel.

You all know how I drive, and I've never had a failure.

"We've come this far so lets not ruin it by thinking !"  Well...I can't help but think about this one. I also know that many of you guys know way more than me but I'm not convinced that over torquing is a good idea. Especially if it's nearly double what is called for. On one torque chart a 1 1/8'-7tpi , grade 5, bolt with waxed threads requires 397 ft lbs which attains a clamping pressure around 43,000 lbs.  if the threads are clean and dry with no lubricant, it will take about 794 ft lbs to achieve the same clamping pressure.   My point being that any lubricant of any kind on the threads significantly decreases "thread drag resistance"and can significantly increase the amount of clamping pressure to the point that the bolt is over-stretched. Also the threads of the bolt and nut may fail by shearing.

In this particular case, I'm not convinced that even 44,000 lbs of clamping pressure will stand up to a launch.  Also cast iron, such as is the drums or disc's, has a very low shear force resistance. The axle is hardened steel and is capable of eating cast iron on any bad day.

I can't remember what VW says about oiling the threads but many car manufacturers say to use 10wt oil or they specify dry threads.  What ever you do, NEVER use Never-Seize on anything except non critical fasteners in high rust possibility areas........Bruce

The consensus for rear castle nut torque among some friends with fiberglass dune buggies that see challenging off-road several times a year is 280- 300 ft. lbs.  As a group we do 3 weekends, anywhere from 2-4 days a weekend, each summer, some do day trips out of town as well and I don't remember the last time anyone was plagued with stripped splines.  I know some of the local VW street crowd do about the same, and this includes a couple of guys with 11 and 12 second cars.  And don't forget the cotter pin!!

@Stan Galat posted:

...We're pushing well past the design parameters (assuming there were any) for aftermarket VW parts, which in most cases don't even approach the quality of OEM parts in a German economy car last built a half-century ago.

Tearing through the mountains at twice the posted speed limit was never what the Sainted German Engineers imagined, and the suspension and brake parts were never designed for how we use them. The absolute weakest of weak links is how the rear hubs are fastened to the car. It's a perfectly fine setup for profiling, and comically inadequate for how we use them...

The advice Gordon and Al give regarding the torque of the castle nut is good, but I've learned (out here by myself) that 217 lb ft or 250 lb ft (or whatever) is the barest minimum. You do you, but I'm torquing that particular nut as tight as I can get it. The chances of stripping it are near zero, and the downside of it being too loose is exceedingly bad. I go to at least 400 lb ft, but I'm a spot-torque man on this particular piece (torque it until you see spots). In this instance, I'd tighten with a breaker bar and cheater, and just check it with a torque wrench set to 300 lb ft. I just want the wrench not to click.

Other, better, more careful professional mechanics will be aghast at my ham-handed approach, but it's worked for me, and I'm a lowland gorilla who drives like his scalp is on fire.

Stan brings up a couple of great points- Dr. Porsche and the engineers at VW never imagined we'd be doing the things we do with their budget commuter car for the masses, and with a lot of aftermarket parts these days not even approaching OEM quality we have to come up with our own solutions to problems the factories guys simply couldn't begin to anticipate.

And Stan- is that 400 ft. lbs. lubed or dry?

@El Frazoo- Glad you caught it before it stranded you, Kelly!

Last edited by ALB

FWIW, back in the day I had to renew the drums on my old 356, so I know about the castle nut, and the big cheater bar. Still have the 36 mm socket I used back then. fits perfectly. And I do know about feet and pounds, mine being close to 200 at this point.  so me at 2 feet ought to get the job done.  I'll be checking on the torque on the other side too.  No wobble there that I can detect, but if one side went bad ...

I have now ordered parts sufficient to rebuild both sides top to bottom, bearings included.  [Do you think the bearings might also be affected by that sloppy spline?] Had a lot of trouble locating the rotors.  Many local and on-line shops list these but do not actually have them. Jbugs has them.  They also list the four hole rotors as both long and short spline, which I find interesting. The one I took off uses the short spline and a spacer, what they will send is a long spline.  Seems to me longer is better. When I ask, they admit these rotors are Empi parts. About the only game in town so far as I can tell.

I get aircooled (Bruce)'s technical essay on bolts, nuts and threads, torque etc. .  Its all been designed and spec'ed out by folks who build and understand such things, strength of materials and all of that. I'll be going for 220 + what it takes to get the cotter pin in place.

I note also the admonition to recheck these torques as a routine maintenance item. Hmmm.  I seem to recall such a note in the paperwork from JPS.  I did not do that, so maybe this is all on me.  And all those burn outs . . .

And go with 250 ft/lbs. plus the next hole, not 220 ft/lbs.  Experience has shown this to be prudent.  Just stand on the bar at the calculated distance from the center point with your 200 pounds (looks like 1'-3" out).  I've not found it necessary to bounce on the bar, but if you're feeling younger, so be it - one bounce, that's it.  Any more and "No Beer for YOU!"  We don't want to hear that you slipped off the bar and broke something that, at our age, won't readily grow back.

Also, add "rear hub torque check" to your growing list of Spring Maintenance items.

I also subscribe to Bruce's fear of over-torquing and potentially stripping threads or worse, but I'm either too ignorant or too lazy to look up different torques for waxed or lubed or dry fasteners (but the people at "Fastenal" know that stuff inside out).  So I always just assemble them dry, unless told otherwise, to the specified torque, more-or-less (but more for rear VW hubs).  If that was good enough for my Navy Sea-Bee uncle it was OK by me.

So the consensus is 217 Ft. Lbs. is NOT enough. I will submit that 250 isn't enough either. I like Al's 280-300 setting. A nice round number and a lot more than factory(which was only to hold 60 hp!).

As I said before, I've never stripped a hub. I do know that it is an often-checked item on the Formula Vee cars. Most guys use a hitch pin in the cotter pin hole. If it goes in easy, but comes out hard, you KNOW it's starting to back off. They recommend 250-300, the factory spec of 217 is NOT ENOUGH! And that's for 60 hp on a good day, but with slicks on a track pulling high G-loads.

Last edited by DannyP

@aircooled - Bruce, if you can stretch the threads on a stub shaft, remind me not to shake your hand (for fear of being crushed).

400 lb ft isn't gonna do it. I know this from experience.

Every time I climb up on the scale, it's like roulette - round and round she goes - where she stops, no one knows. Let's call it an even 250 lbs of rippling romance. I don't lube the threads, but they aren't "rusty dry" either. Me standing on the end of an 24" breaker bar is 500 lbs, but that would require balance I don't have, so let's just call it 400 lb ft. I bounce a little, leaning hard on the bar, and then go to the next hole. I may stand on it, but I'm balancing on something, else I crack my ginormous pineapple like a grape.

This gets me to the torque settings you guys are reporting with great specificity. I know you can math and all that but when you say, "I weigh 150 lbs and I have a 2 ft cheater, ergo: 300 lbs, I wonder - how do you pinpoint exactly the point your weight is centered on the bar. Your foot is 4" wide. You're guessing even with a torque wrench and a toque multiplier (there is loss in the gears of the tool)

I've used a hitch-pin forever. It looks like you care when you use one, and like you never check your hubs when you have a cotter pin.

I continue to contend that a torque wrench is for assembling an engine. I'll not be using them a bunch of places elsewhere. You get a feel for fasteners. My engineer son uses a torque wrench to tighten his lug-nuts on a Ram 1500.

I can't imagine.

Last edited by Stan Galat
@Stan Galat posted:

@aircooled - Bruce, if you can stretch the threads on a stub shaft, remind me not to shake your hand (for fear of being crushed).

400 lb ft isn't gonna do it. I know this from experience.

Every time I climb up on the scale, it's like roulette - round and round she goes - where she stops, no one knows. Let's call it an even 250 lbs of rippling romance. I don't lube the threads, but they aren't "rusty dry" either. Me standing on the end of an 24" breaker bar is 500 lbs, but that would require balance I don't have, so let's just call it 400 lb ft. I bounce a little, and then go to the next hole.

I've used a hitch-pin forever. It looks like you care when you use one.

I continue to contend that a torque wrench is for assembling an engine. I'll not be using them a bunch of places elsewhere. You get a feel for fasteners. My engineer son uses a torque wrench to tighten his lug-nuts on a Ram 1500.

I can't imagine.

I use a certain number of "ugga-duggas" on the air impact wrench LOL!

Here's a test to play with.  Take a 3/8 - 16 X 1" bolt, put 3 washers on it and then a nut.

Put some never seize on it and torque it to spec. (about 32 ft lbs) or what ever the chart says.  The bolt will fail before you get to the torque spec. Wheel bearing grease works well too !  Not trying to be a smart ass here. We just need to be careful/mindful of what we are trying to do.  Sometimes more is not better.......Bruce

@aircooled posted:

Here's a test to play with.  Take a 3/8 - 16 X 1" bolt, put 3 washers on it and then a nut.

Put some never seize on it and torque it to spec. (about 32 ft lbs) or what ever the chart says.  The bolt will fail before you get to the torque spec. Wheel bearing grease works well too !  Not trying to be a smart ass here. We just need to be careful/mindful of what we are trying to do.  Sometimes more is not better.......Bruce

Most torque readings are done dry (unless specifical lubrication is called for), are they not?  I was under the impression a lubricated nut/bolt normally used 10-15% less torque than the dry spec for that reason?

Here is a link to an old Mechanic’s Handbook which was standard issue for all hourly mechanics and technicians that worked at the Santa Susana Filed Laboratories.  This does not appear to be the most recent version, but it might be close.  Clearly some of the information is a bit outdated, or unique to the world of testing rocket engines. But there is a lot of good stuff on piping , tubing and fasteners and the art of measuring torque as a method for establishing a desired preload on the joint being torqued.  Torque values listed do assume appropriate lubricant is applied.

Jon

@ALB posted:

Most torque readings are done dry (unless specifical lubrication is called for), are they not?  I was under the impression a lubricated nut/bolt normally used 10-15% less torque than the dry spec for that reason?

The only fasteners I've used lube on for torque are internal engine parts.

ARP rod bolts come with specified grease to be used to torque the caps on. This gives just the right amount of bearing crush to give you 0.001" of gap all around for an oil film.

The other time was to torque a head gasket. I don't remember what brand of engine, but fresh engine oil was used on head bolts/nuts to achieve the correct clamping pressure. This may have been ARP studs and nuts on my Corrado G60, but it was a long time ago and I'm not sure.

@aircooled posted:

Here's a test to play with.  Take a 3/8 - 16 X 1" bolt, put 3 washers on it and then a nut.

Put some never seize on it and torque it to spec. (about 32 ft lbs) or what ever the chart says.  The bolt will fail before you get to the torque spec. Wheel bearing grease works well too !  Not trying to be a smart ass here. We just need to be careful/mindful of what we are trying to do.  Sometimes more is not better.......Bruce

If it's a cheap hardware store bolt, yes. But if it's grade 8 or AN(military or race car spec), it will not fail.

I'd love to see a stripped axle nut. Just try to do it. I'd bet on that not happening.

Last edited by DannyP

Well here's my take, sorry for being off the air for a little bit.  WRT Torqueing bolts and stuff: I think one should not be trying to second guess the amount of friction you need to overcome to get to a torque reading, i.e., "dry threads". Entirely too variable. Some lubricant seems correct to me.  As it is, there is Liquid wrench on there now, and so there is that.  I am a BIG fan of Neversieze, but sounds like that might be going too far.  Lets say I'll clean those threads up, they will be clean, not rusty, and I'll see how that all works out.  What I think I want is to develop the correct pressure on those threads (which, as a simple machine, is really a wedge) so as to put the bolt in the correct amount of tension.  Further, I think of this bolt in tension as a very very stiff spring, which means that it will stretch in proportion to the tensile load applied.   And you want the working load on that bolt to NEVER come close to counteracting the tensile preload in the bolt and bring it to zero, which implies that there will develop a gap in the lash up.  I do not have a problem with using 250-300 ft-lbs, and will report back about how that goes.  Should be fine. I have no idea how much torque on that large fine thread amounts to how much tensile stress in that bolt, but I do know that the yield stress for that bolt is probably about 30,000 psi, and the section of that big bolt might be about a square inch, so that's ~30,000 lbs tension to get to the elastic limit.  Even if only 1/2 in^2 root area, that's still 15,000 lb. The breaking stress would be higher.  That shaft is probably pretty decent steel. Likewise the nut.  The more likely failure mechanism here for overtightening would be due to shear stress in the thread.  And thats a whole other technology -- read as "black art". Further note: I have learned that the replacement disk hubs come with a long spline or a short spline.  Why this might be, I do not know, but I do know that my axle stub has a long spline and the damaged hub in the picture is a short spline.  Which requires a spacer be applied so the castle nut has something to push against.  And which also implies that the shaft has more bearing area in the spline than the hub can provide.  Does not sound like a good thing.  I will also inform that the replacement, due in next week, will have a matching long spline.  Perhaps the short spline gave too little bearing surface for the spline, and the longer spline will fix that -??-

I also plan to renew the bearings, just because why not, It's all apart anyway, so a few more bolts and selected bad words to R&R those bearings. I think me nemesis might be that snap ring on the inner bearing.  We will see.

As to precise torque readings, I'm kinda in the same camp as Stan and Pip: precision here is exactly what you need, for all the reasons provided above, when assembling an engine where the stretch of the bolt (remember, the bolt is a very stiff spring)  must be just so to preserve close tolerances in parts that need close tolerances.  This rear axle castle nut does not require these characteristics, so far as I can see.

Lastly, I do have a pneumatic impact wrench, which was handy to get that nut off, but I don't think I can use that device to "torque" the nut on.

@El Frazoo - May I call you "el"??

No, I wouldn't use an impact wrench to "torque" the castle nut on.  Off, yes....  On, no.

You really won't know what torque it is set to.  But then, I am not a mechanical Engineer, even though I've stayed in any number of Holiday Inn Express places over the years, but I think I can get you a bit closer to enlightenment:

Here is a handy-dandy torque guide for various size English and Metric bolts, from our friends at Fastenal, for common fasteners:

https://www.fastenal.com/conte...eference%20Guide.pdf

You'll need to know the diameter of the drive/stub axle which, for a 36mm nut, should be 24mm.  That's shown on here down towards the bottom:

https://www.insight-security.c...ow-metric-bolt-sizes

Now, go back to the Fastenal chart and scroll down the left side to Metric fasteners, then find 24mm and scroll over to the right to the Grade 8 section and that gives you 523 Ft/Lbs. Max. Dry.

(BTW: It shows 34,450 for the clamp load on a 24mm shaft so your guess was very close - I guess all those ME/Physics classes you took somehow paid off.)

The K=Factors are in the notes in the lower right corner and you should be using k = 0.15 (lowest torque) for a lubed connection and K = 0.20 for a Zinc/dry fit.

Stub axles are hardened and may be harder than Grade 8.8.  If so, go to the 10.9 section and the torques go up significantly.

Remember - These are MAX tightening figures, not to be exceeded.  Working torque ranges will be lower and determined by metal hardness, thread mating area, thread pitch, lubes used, if any, so the range of 250 - 300 ft/lbs is in about the middle of the range and should not damage stub axle or nut threads - I can't speak to the hub splines as I don't know the material.

I also think it's a great idea to go with the longer reach hubs - It makes the stack-up more logical.

Hope this helps.  gn

Last edited by Gordon Nichols
@Stan Galat posted:

400 lb ft (+/-) doesn't stretch, strip or otherwise harm the fasteners in any way. I'm struggling with why you all want to keep them looser than that.

Spot-torque.

You'll never damage the hardware. Check with a torque wrench to make sure you didn't get weak last night. Set for 300 lb ft the wrench shouldn't click.

If the wrench doesn’t click at 300 lb ft, then you don’t have 300 lb ft of torque on the fastener.  The click is the breakaway torque setting.  Maybe I’m not understanding what you mean?



BTW, how many people actually own a torque wrench that goes to 300 lb ft or more. I don’t know if they even make a 1/2” drive that goes that high, probably only comes in a 3/4” drive.

Last edited by LI-Rick

@LI-Rick, if your torque wrench doesn't go to 300 that is a good reason to use the Torquemeister where it would only need to go to 33.

I’ve never torqued an axle nut.  I put it on with an impact wrench and then use my axle nut wacker. I give it a couple of good hits with a hand sledge, then put in the cotter pin.  Has worked for me, even in my 10.90 drag car.  

55CB6097-F92B-4D33-8A2A-E2CE01E0DEBC

Attachments

Images (1)
  • 55CB6097-F92B-4D33-8A2A-E2CE01E0DEBC
Post Content
×
×
×
×
Link copied to your clipboard.
×
×