Steering problem after rebuild

[quartetchiang]

Esteemed Friends:

Now this puzzles me. After I had my steering shaft/column rebuilt on my '89 Turbo, it steers extremely awkward. As you turn either clockwise or counter-clockwise, instead of a progressive motion, steering travels in alterations of being too tight and too loose for several segments! What could I have done wrong? I changed both Universal joints, as well as the lower steering shaft with rubber...

Any inputs welcomed, this is so odd!

Yous,
Paul

[brp914]

I experienced same. As part of a suspension rebuild I replaced both steering column u-joints. I bought the joints from Porsche. I had the same awful symptoms you describe. I put my old joints back in and all was well. I theorized that I hadn't put the joints 90 degrees out of phase, but it turns out there is only one way to put it back together.

[Bill Douglas]

Umm, I don't think this is the solution, but worth a try because it's easy... Try putting some 3 in 1 type oil on the universals in the steering column just in case they need more. I seem to need to do it on my car every six months.

Also you could jack up the front of the car and have someone turn the wheel so you could view what's going on when it gets hard then easy.

[quartetchiang]

BRP914:

Only one way of putting it together? Can you be more specific?

[Walt Fricke]

Universal joints do not rotate uniformly (that is, the output side does not turn at the same rate as the input side) unless everything is in a straight line.

Porsche dealt with this by making sure that the two joints were at just the right angle to each other.

I learned about this when I cut the ends off of the middle shaft and welded them onto a hollow tube (got to save weight, don't you know), but didn't get it quite right (because I didn't appreciate the need to do that). I could feel the uneven motion, although it wasn't so much like a bind as I recall.

Jacking up the front, as suggested, should help you figure out if you have this kind of motion.

You'd think you couldn't go wrong by using the Porsche parts only here, because the splined rod ends have a flat ground onto them, and the cross pinch bolt can only fit across if the splines are lined up just right. As I recall, these flats are at something like a 30 degree angle to each other on the rod part.

Mine were close enough to being right that I could grind a bit and change a spline or two. But you ought not to need to do that, because you didn't mess with anything.

You can, however, see if you can get any of the four cross bolts to go on in a different position than what it is now in. Maybe you were able to force one in?

[brp914]

As Walt said, a u-joint is not a CV joint. I've seen the horrible trig function that describes input versus output speed as a function of angle. But no matter. Porsche machined grooves into the shaft to assure proper phasing of the u-joints so that the system can only be assembled correctly. And as I recall, the joints are symmetrical so swapping them end-for-end should not matter. I never could figure out why my dirty, grimey 40 year old u joints work great and the shiney new Cad plated joints from Porsche would not.

[DRACO A5OG]

For me, I had slop in my original U joints and swapped with new Porsche Shiny Cad ones. The trick was to make certain they were on the "flat" spot of the splines correctly and seated deep enough as not to bind. I had to grind down the flat spot to be a bit longer as I recall. I might have been lucky but it worked for me.

Compare the old and the new one's alignment of the holes.

Again, I got lucky.

[quartetchiang]

Dear Walt:

That is fascinating experience! I think the answer is there somewhere!

[quartetchiang]

Dear BRP914 and Draco:

Cad-plated new ones! Like the one in attached photo?! That's the one I have!
Has Porsche ever issued any bulletin explaining why the new ones are made to annoy?

Paul

[DRACO A5OG]

Paul, post a pic of yours maybe we can tell if it is binding some how.

What I would try to do is to remove and compare the old and new just to be certain or at least do as BRP914 did and test to see if old ones will bring things to normal then observe how the new ones fit up.

This is something you want to correct because a binding U joint will soon fail. Not a good thing :-(

[quartetchiang]

Thanks guys!! I am hoping to get the problem sorted next weekend. If anything proves to be positively the cause, I will let you know!!

[JohnJL]

How did this go?

[chris_seven]

They don't really bind but as has been said they are not CV joints and the output side will try to turn more slowly than the input side unless they are in a straight line ( Newton's Third Law - Conservation of Energy etc)

The result is that when the shaft rotates a torque is generated.

By selecting the relative position of the UJs to torques generated can cancel out but if they are not aligned correctly they will feel unpleasant.

This is common problem of LHD to RHD conversions when the steering shafts aren't modified.

Early steering shafts had indexing pins so it was easy to replace joints but later shafts rely on the bolt position.

The easiest way is to mark up positions before swapping the joints.

[Walt Fricke]

Chris - I think that if the input rotates at X revolutions per unit of time (T being long enough so that X is a round number of revolutions >=1), the output will also rotate X times.

But the rates of rotation for one revolution on the output will vary - sometimes being faster, sometimes being slower, during one rotation. The result is, for want of a better word, a lumpy output because of the speeding up and slowing down. The greater the angle, the greater the effect. I could notice this when I used U joints instead of CVs on an early 911 race car I purchased which had been modified in this way. You didn't notice it on the track, just putting around in the paddock, but I changed to CVs.

Had I paid more attention to this effect I might have got my steering system right to start with, rather than feeling this odd motion driving the car. Again, it was in the paddock maneuvering that I noticed it.

U-joints are unconstant velocity joints (absent the special case where their axes are at 0 degrees to each other)?

Walt

[quartetchiang]

Ok guys, I'm back. Was occupied with job and no time for this "hobby". Take a look at the photo I took of my current lower joint! This can't be right?

[quartetchiang]

And upper joint output end has scratch mark as well..

[quartetchiang]

Lower joint input end is rubbed-off also..

[DRACO A5OG]

Oh no, That is not good at all. There is your inconsistent steering feel.

Most likely the shafts are not seated in all the way into the U-joints.

[quartetchiang]

Draco:

Enlighten me: you stressed the importance of the shaft being all the way in, but it looks to me (certainly I could be wrong) that at least the the output end was compromised because the shaft was too much inside, causing interference, even damaging the actual joint..

[pete3799]

So the end of the yoke is hitting the splined shaft? Can you center the joints on the shafts so that you have an equal amount of shaft protruding? Not sure if that's possible given the pinch bolt location.