Axles - floating or not?

[Andrew Robertson]

Hi All,

I'm in an unfortunate situation with the firm who made me some custom axles. I am using Porsche 930 box with 930 CVs on both the inner and outer ends, so nothing unusual.

I have some porsche race car axles of a different length, and the CV inner section is a nice engineering fit on the splines of these original axles, and will slide smoothly on the splined section at just a light hand touch. I supplied both as a sample.

The new axles arrived and the CV inner is a light to medium interference fit on the spline - i.e. needs to be hammered on.

I want the CVs to be centred and happy through the travel of the suspension, and have always assumed that the axles float axially to allow this to occur (hence the circlips on each end). I have consulted a business that race prepares and manages a variety of competition porsches - he always uses floating axles on purpose built track cars.

The owner at the firm says he always does axles for CV cars non-floating (i.e. interference fit). At this stage I'm wanting him to remedy the situation and do them again in floating. He seems a reasonable guy and is prepared to entertain that.

Am I on the right track?

Cheers, Andrew Robertson, New Zealand

 

[Trevor Booth]

Yes Andrew,
as the suspension travels thru its up and down motion the axles will need to float.

Some CV locate on OD of spline, some locate on ID of spline.

you can check this from your sample axle and make your new axles the same way.

If they locate on OD of spline you could have the axles that are already made ground down slightly to give the required fit.

 

[Randy V]

My Axles do not float inside the CV Inner Race. Rather the inner race and axle are assembled tightly and the position of the inner race inside the CV Outer race at ride height is neutral (centered) and as the suspension moves through it's range of motion, the inner race floats within a tolerance of either side of the outer race..

 

[Terry Oxandale]

I set up my 240Z with 930 axles and they did NOT float. Two clips on each end (of an axle designed to float using 3" of splines on each end) once I determined the static ride height and where the axle, grooves, and clips would be located. I never had an issue in 10 years with a "fixed" set-up. The plunge on a 930 CV is about 10mm each way (inner race movement relative to the outer race), which is 20mm when you count both sides. Unless you have some large amount of suspension travel using short control arms, or have some high angularity in the shafts (which reduces the plunge due to the inner race being cocked over to the side), you'll never need the axle to float (slide) in the CV's inner race. The OEM 930 donor axles/CVs I used when I fabricated all of this appeared to me to also be fixed.

 

[Scott]

The "Float" in the axle is built into the CV joint itself. The inner race (axle) is allowed to move in and out relative to the outer ball race (transmission or wheel).

 

[Cliff Beer]

I can't imagine a race car using CVs with sliding axle shafts (axle shaft sliding within the inner CV race) as the thing is going to bind with power-on/power-off through the range of motion of the travel. Also, eventually the splines are going to hammer out and you'd end up sitting by the side of the track with stripped axles, watching your race mates go by.

Just make sure you're axles move in relative harmony with the rest of the suspension. If you're using double wishbones for example then the pivot points for the wishbones should be roughly in the vertical plane of the center point of the CVs (adjusted for intentional camber changes throughout the range of motion of course).

As others have pointed out, the CVs themselves "float" sufficiently.

 

[Trevor Booth]

depends on CV type, I should have qualified what I posted, however,
some CV are made to "plunge" within themself some are not. They are more correctly described as TRIPOD joints having 3 balls only wherein the outer race has three semicircular straight tracks.

A true CV joint (RZEPPA) is not designed to plunge.

I have done some work on a 900HP LOLA indy car and it has sliding splines. The CV is not designed to "plunge" , the splines do not "hammer out " or strip.

have a look at road cars as well - you will find they have a mixture of sliding splines and or tripod joints designed to "plunge"

 

[Randy V]

I should clarify that the CVs shown in my picture above are Porsche 930.

 

[Terry Oxandale]

Many dune racers use the full 3" of spline as a sliding expansion of the shaft. Yes, in a good race the splines turn blue from all the heat built up, but it does work. But I can't image any decent road car suspension that needs more movement than the typical 930 CV plunge will provide. As Randy has shown, the 930 CVs in the OEM application are fixed on the shaft.

 

[Andrew Robertson]

Thanks guys - I have an all-930 set-up so I'll go with the tight fit, set them up so CVs are centered at ride height, and let CVs take care of the effective length change that occurs with suspension movement. Thanks for your help.

Cheers, Andrew

 

[Cliff Beer]

depends on CV type, I should have qualified what I posted, however,
some CV are made to "plunge" within themself some are not. They are more correctly described as TRIPOD joints having 3 balls only wherein the outer race has three semicircular straight tracks.

A true CV joint (RZEPPA) is not designed to plunge.

I have done some work on a 900HP LOLA indy car and it has sliding splines. The CV is not designed to "plunge" , the splines do not "hammer out " or strip.

have a look at road cars as well - you will find they have a mixture of sliding splines and or tripod joints designed to "plunge"

Trevor, good to hear you've had such good experience running your 900hp Lola.

My own first hand experience with track time, sliding splines, and a mere 300hp has been quite different. The splines do indeed hammer out over time, exactly as I mentioned, and leave one (that'd be me) sitting on the verge with stripped splines. If you're checking the fit after every race you'll notice that they become progressively looser, from very slight interference to no interference then to lots of slop. Those axles were made by one of the best CV/axle shops in the country too, so it wasn't a matter of poor engineering or quality.

Keep those inner races firmly located and let the CV take the lateral movement of the axle. There are one or two notable shops that do some very interesting stuff with the races and balls to make them virtually indestructable.

 

[Trevor Booth]

Cliff,
I guess it depends on the shaft material and any hardening. I recently made two new shafts as the original broke. (standing starts in Oz!!) The original was a manganese steel with induction hardened splines and ground over the full length of the shaft.
There was no discernible wear in the spline
I made the new shafts exactly the same way and same material.

The CV's are 6 ball (25 dia) selective fit with NO movement in the CV

A true CV cant take any axial float unless it is worn out or the cage is broken.

I have a friend with a Penske indy car and it is exactly the same

I am not about to change it

 

[Terry Oxandale]

Thanks guys - I have an all-930 set-up so I'll go with the tight fit, set them up so CVs are centered at ride height, and let CVs take care of the effective length change that occurs with suspension movement. Thanks for your help.

Cheers, Andrew

I'd find out first how much plunge you'll have during full suspension travel. In my specific case, static ride height provides the minimal length, so anything inside of "centered in the CV" is movement left on the table and unused. On good thing about this though is that you can reverse the CV later to account for any wear. Other configurations may have the plunge work in and out from "static" center. It's one of those situations where I measured 3 times before committing to cutting grooves in the splines for the CV positionl.

 

[Terry Oxandale]

Oops, make that maximum length at static right height rather than minimum length
.

 

[Andrew Robertson]

Randy, Trevor, Scott, Cliff, and Terry: Thanks for the sharing all your knowledge and experiences. My axles are 375mm long (14.76in) and from memory I have about 10mm total effective length change as the suspension moves through its full travel. It sounds like the CVs could comfortably chip in 5mm plunge each. I will take great care in understanding the nature of the plunge and will centre accordingly.

It's unclear to me whether those running tight fit use inside and outside circlips as a back-up retention/location measure, or just one on each end of the axle.

Thanks, Andrew

 

[Cliff Beer]

Cliff,
I guess it depends on the shaft material and any hardening. I recently made two new shafts as the original broke. (standing starts in Oz!!) The original was a manganese steel with induction hardened splines and ground over the full length of the shaft.
There was no discernible wear in the spline
I made the new shafts exactly the same way and same material.

The CV's are 6 ball (25 dia) selective fit with NO movement in the CV

A true CV cant take any axial float unless it is worn out or the cage is broken.

I have a friend with a Penske indy car and it is exactly the same

I am not about to change it

Hi Trevor,

Sounds like I could learn a few things from what you've been doing with CVs. That particular car has been sold on since however (CVs were part of that decision) but I've been looking at old Indy cars as a platform for a fun (and outrageous) track car so it's interesting that you mention sliding spline axles on another Indy car. In other words, I might be looking at the very same issue again sometime soon with an old Indy car. Might have to ping you over PM for a little advice if OK with you. I suspect you're spot on with the necessary metallurgy spec to make it all work in practice.

Thanks!

 

[Cliff Beer]

Randy, Trevor, Scott, Cliff, and Terry: Thanks for the sharing all your knowledge and experiences. My axles are 375mm long (14.76in) and from memory I have about 10mm total effective length change as the suspension moves through its full travel. It sounds like the CVs could comfortably chip in 5mm plunge each. I will take great care in understanding the nature of the plunge and will centre accordingly.

It's unclear to me whether those running tight fit use inside and outside circlips as a back-up retention/location measure, or just one on each end of the axle.

Thanks, Andrew

Andrew, a lot of CV axles have a shoulder on the inner side, with the circlip obviously going on the outboard end, thus making the lateral location of the CV somewhat fixed. Not sure if yours are like this or not. If you have the option of making lateral adjustment by way of cutting a new groove for the circlip on inboard and outboard sides that's good (assuming no shoulder (or turn down the shoulder on a lathe) and sufficient length of splines). Just for experimentation you might want to hit the local junk yard and pick up a few CV axles for comparison. In my (limited) experience, many Lobro CVs, for example, fit onto a common spline size.

Good luck!

 

[Trevor Booth]

Andrew,
why dont you get a hold of a 930 manual and follow it
you have all 930 bits

 

[Randy V]

Andrew - I no longer have the axles in my possession or I would go out to my shop and look to be certain. As I recall, there is a shoulder on the axle that the inside of the CV's inner race butts up against. Odd about the circlips though.. As you can see in my picture above, on one end of the axle there was a single circlip and on the opposite end, there were two circlips.

It sounds like your suspension travel is within the limits of the CV.

Remember also, that the distance change is split between both the inner and outer CV.
This way, a 10mm travel really works out to be 5mm per CV.

 

[Terry Oxandale]

I just measured the "plunge" on my 930 CV, and it is very close to 12mm (24mm for two CVs). This is limit to limit of the inner race moved from one side to the other side while keeping the inner race axis aligned to the outer race axis. Obviously one does not want to plan on using this full amount because of obvious reasons, but it is important to remember that this distance decreases with increased angle of the shaft.

Randy's memory of the shoulder on the inside of the shafts is correct. On my current axles I had four grooves (two per CV) cut to fix the CVs in place. I'm guessing that perhaps the used of two clips (in one groove) was to remove any slop. I was told by a Porsche mechanic that at times, the inner race needed to be "popped" with a hammer to get the outer clip to seat correctly, so there must be some pretty tight tolerances involved.