Half shafts with U-Joints vs. CV Joints

[Ron Earp]

Educate me on some drive line components

Original GT40s use half shafts with U-joints and rubber donuts to couple to the transaxle, correct? And, is there a sliding component to this assembly, like a yoke that can slide into a splined section say half way down the half shaft?

What are the disadvantages of this system compared to a half shaft system using CV joints?

I assume the CV joint might be smoother in operation and use less space? Not sure on the space thing. However, the CV setup has to have a sliding component as well and I assume that is the splined section that can slide into and out of the CV joint.

What is the MTBF of a U joint versus a half shaft assuming both are within correct operating angles? Seems U joints pretty much can last forever properly greased (forget my Jensen race experience for a moment...) and tended to, while the CV joint will have to be rebuilt at some point?

 

[Randy V]

CV Joints are "Constant Velocity" whereas Universal joints are not and they (U-Joints) will introduce a dynamic vibration into the driveline if not phased properly or at an acute angle..

There's an excellent article on Wikipedia about the CV joint;
Constant-velocity joint - Wikipedia, the free encyclopedia

VS. Universal joint

Overall - even when not running at a large angle - the CV joint is smoother and capable of carrying far more torque for a longer period when run at an angle..

 

[John]

Randy,

My KVA has CV joints on the transaxle side of the drive shafts and U-joints on the suspension side. It uses Corvette rear suspension . I've been trying for a number of months to eliminate a drive train vibration that occurs when I'm accelerating in 2nd gear from about 2K to 4K RPM. Someone else suggested that this combination might be the cause but I was not immediately convinced because the KVA's were designed by legendary GT40 engineer Kevin Atwell - I assumed he would have been aware of such a harmonic vibration.

In you judgment, would a design such as my KVA's introduce vibration or would the combination CV/U-joint combination cancel out any harmonic vibration?

John
epper:

 

[Jac Mac]

If you use only one U-Joint with a CV at the other end you introduce an acceleration/deceleration in rotational speed into the driveline at any time the U-Joint is not running dead straight. If two U-Joints are used & correctly phased the acceleration/deceleration is mainly confined to the shaft area between the U-joints and it depends on the amount of material & diameter involved in that portion of the shaft how much it will be felt by you ,the driver.( I say mainly since if the angle in each U-Joint is not the same then the rotational speed change for each joint will be different)

Now as an example if you have a U-joint @ the Transaxle & a CV @ the hub carrier the half shaft/axle/wheel will try to accelerate/decelerate when the U-joint assumes any angle.

The rubber Do-nuts performed two functions-Shock load damping & compensated for some of the accel/decel we have discussed.

Jac Mac

 

[Jac Mac]

Randy,

My KVA has CV joints on the transaxle side of the drive shafts and U-joints on the suspension side. It uses Corvette rear suspension . I've been trying for a number of months to eliminate a drive train vibration that occurs when I'm accelerating in 2nd gear from about 2K to 4K RPM. Someone else suggested that this combination might be the cause but I was not immediately convinced because the KVA's were designed by legendary GT40 engineer Kevin Atwell - I assumed he would have been aware of such a harmonic vibration.

In you judgment, would a design such as my KVA's introduce vibration or would the combination CV/U-joint combination cancel out any harmonic vibration?

John
epper:

John- to eliminate the halfshafts in this example- note the road speeds from 2k>4k in 2nd gear. Now in 3rd gear thru the same road speeds is the vibration present to a lesser degree, if so then your halfshafts are probably the cause.

Jac Mac

 

[Dave Bilyk]

Ron,
looking at Big Foots animations, I can see three factors that might limit the life of a CV joint.
First, the balls operate at a larger radius than the UJ bearings, which means more movement for a given articulation.
Second, the balls try to ride out of the grooves, and require high stiffness in the outer part to control this tendency.
Third, the balls have to transmit force while some sliding takes place on the loaded faces, which means more friction and sliding with the larger movements mentioned above.

As Jac Mac has said there is a significant difference between a UJ / CV combination, and a UJ / Rubber Donut combination. The flexibility in the Donut allows harmonic movement of the shaft without generating significant harmonic torques, while the CV is stiff, generating harmonic torques which are fed into the wheel and back into the transmission.

regards
Dave

 

[Chris Duncan]

CV is the superior piece.

As part of the design they have some telescoping travel and thus the assembly is lighter because it does not have a separate telescope function.

If properly maintained they last just as long as a U-joint.

The only problem they have on hi-perf vehicles is the centrifugal force at high speed will expand the rubber boot way out till it rubs on nearby parts. You need to put zip-ties around the smaller diameters of the boots.

 

[J. Salmon]

I replaced a torn boot on a CV joint on my 355, and though it is fundamentally like the image shown, in reality it was much more robust. I was impressed. Messy job packing that thing with grease, and the friction is so little that it just flops all over the place. Compared to a wheel bearing, it just looks indestructible.

In my experience, it's the boot that fails. As long as it stays sealed and does not get contaminated, they area really solid. And just like the picture, the shaft can be at any angle and it turns without oscillation.

 

[Phil]

Ron
Those inner couplings on the original 40 were called Rotoflex Couplings, and I have also seen them on some vintage racers of the same era. Essentially the coupling is a rubber donut with a steel band molded inside, which has thru-tubes for boltup. They absorb a lot of shock, but are prone to breakage if overstressed or get oil on them frequently. The coupling also absorbs linear motion to a degree to compensate for the suspension going thru its arc. They worked back then, but as was expressed in the other posts CV joints are far superior. Most front drivers on the road today using CVs never need a replacement unless the boot is compromised, or they have mechanical damage from an accident. I have had two cars with CVs go 250,000 miles without a problem. Just choose the right joint for the power you plan to run, and keep them sealed with the proper grease inside and they are relatively hands off.
Hope this helps
Phil

 

[Randy V]

Porsche 930 CV

 

[BruceA]

Its all been said above. You dont mix Constant Velocity and Universal type joints. The UJ type speeds up and slows down during one 360 degree turn of the axle. This is why you put a UJ on each end, and out of 'phase, so while one is speeding up, the other is slowing down, to give the impression of 'constant' rotational speed.
On another note, I didnt think Ken Attwell had anything to do with the design of the GT40? Correct me if Im wrong. He was an engineer (which could be anything from a millwright to stress anyliser) at the UK based Swansea Ford car plant.

 

[John]

John- to eliminate the halfshafts in this example- note the road speeds from 2k>4k in 2nd gear. Now in 3rd gear thru the same road speeds is the vibration present to a lesser degree, if so then your halfshafts are probably the cause.

Jac Mac

Jac,

I'll try your suggestion - I generally don't notice the vibration when accelerating thru that rpm range in 1st gear!

John
epper:

 

[Dean Lampe]

I have had my u-joints north of 150mph without any noticable vibration. For what we do they will both serve the purpose well. If you can tell the diffence, I would guess you had a worn part or a problem else where. I like the vintage look of the u-joint. It depends on what look you are going for with your car more than anything else. I just don't see a lot of GT40's with 50,000 miles on them.

 

[Eric A.]

According to Carroll Smith in one of his books he stated that combining u-joints and CV joints was definitely not a good idea. He much prefered the CV joints and gave the tip of using the zip ties around the boots to keep them from expanding. On the halfshafts that I have worked on, the inner joint usually has slotted tracks for the balls to allow for the in-and-out movement of the shaft during suspension travel. On halfshafts with splined joints the friction on the splines as the parts slide puts a load on the joints on either end. This occurs mostly during acceleration. On the Datsun 810 halfshafts that I worked on they used splines with a radius and had ball bearing in the splines between the two parts to reduce this friction. I think the same thing was done on the Z cars with that suspension. I have seen u-joints break, but usually something else will go before a CV joint comes apart.

 

[Ron Earp]

This is a very helpful discussion I think, good replies. That CV joint animation is really useful

Dean, you have U joints on a ZF, correct? Did you source those or are they RCR parts? I assume you have some sort of sliding yoke to give it some lateral extension when the suspension moves?

The CV joints do seem to make good sense. On the CV joints then all you need are two CV joints that will fit whatever flanges you are working with, then, a half shaft of the correct spline type and length? I understand that at least on one end a clip is used to make sure it doesn't slip out of the joint. Do both ends have clips?

 

[Mike Dozier]

If cost/authenticity are not a factor...then CV's are the way to go.
Corvettes ran u-joints for 44 years...but switched over to CV's in 1997 (?).

IMHO u-joint/half shafts are more period correct...my set cost $ 650 and will handle 500 hp.
At that time custom CVs/shafts would have run me twice that amount or more.
So it was as much an economic decision as visual.

I have no idea what the weight differential or driveline loss factors are for UJ vs CV.
I assume CVs are superior on those issues since all modern race cars run them.

MikeD

 

[Jac Mac]

The TVR that I work on has U-Joints with sliding splines, zero problems with this & it tops out around 150/160 mph on current gearing & has raced for 7 years . Shafts were shortened slightly to accomodate a rear end change 3 years ago at which time U-Joints were removed-checked-refitted. The main thing is to optimise the position-length of the half shaft along with transaxle position to minimise the plunge distance on the splines/CV's.
I see too many cars on this site where this has not been done due to cost of remounting transaxle or altering the chassis to do so.

 

[Dave Wharran]

Ron,
If it is a ZF that was sourced from or for a Pantera, it should have the U joint axles/flanges. The half shafts are made up of all standard Spicer brand parts including the sliding spline, with the exception of the tubing that sets the length.

Dave

 

[Randy V]

Dave - That looks gorgeous!

 

[Cliff Beer]

One recommendation with regard to CV joints: stay away from "rebuilt" CV joints. There's nothing "rebuilt" about a rebuilt CV joint. Without exception the "rebuilding" consists of simply disassembling the the race and the balls and replacing obviously worn balls with slightly less worn used balls. And, perhaps, where the race is obviously and significantly scored/worn then a less scored/worn used race will utilized. New CVs aren't that much more expensive than "rebuilt" so go with new!