Play in steering rack

[Luckyone]

I found the excellent guide in the useful sticky section for the rack rebuild, but one of the replies stood out for me

GT6k said:

Brilliant job thanks for the info. Just as a warning to others I sent a leaking but not worn rack in and got a newly sealed but worn rack back.

Many many years ago now, in the days I was starting to try to do a partial body lift I changed the input seal



I'd kind of forgotten I'd done that by the time I had finished the full rebuild, but it didn't leak so I was happy for a short while. Only there was, right from the first drive, a lot of play in the rack. I'm sure there wasn't any before I changed the input seal.

I seem to remember the shims falling off as I pulled the input shaft out but I'm sure I put them back if the did fall off so I wander what I messed up? I can't think it was muck getting in & causing wear as the free play was evident right from the first try with the new seal.

I tried tweaking the adjuster as soon as I could but it didn't take out all the free play. If it wasn't knacked before, I guess it likely is now as it had a bit of use with the play there.

Years back I seem to remember other people saying they had problems with refurbished racks, what are the results like these days?

I have a major lack of time these days so would be more likely to send it off for a referb, though there is a non running Scooby sat next to it in the garage......

[pmessling]

I brought a scooby rack but never trial fitted it. Went down the rot of getting another used rack and making good one. Mine was a very worn pinion. Water had gotten in as you saw in the sticky. TVR parts does have the pinion but it's not complete and I couldn't figure out how the valve part separated from the pinion and didn't want to damage the inner torsion bar.

[Luckyone]

pmessling said:

I brought a scooby rack but never trial fitted it. Went down the rot of getting another used rack and making good one. Mine was a very worn pinion. Water had gotten in as you saw in the sticky. TVR parts does have the pinion but it's not complete and I couldn't figure out how the valve part separated from the pinion and didn't want to damage the inner torsion bar.

Ah I see, has anyone fitted a Scooby rack to a cerb then? From memory I think it's only a chim I've actually heard of, but I've been out of touch for quite a while.

Interesting new pinions are available, I fairly sure mine wasn't worn, the pic isn't the best but it looks a lot better than your very worn one.

The amount of free play when I first tried it was silly, but all I did was take the pinion assembly (in my old pic) above then put it back with the new seal right way. I didn't touch any other part, the first time I moved the adjuster was after I found the highly excessive free play on the first test drive.

Maybe I'll try taking the pinion out again & look more closely at its condition.

[pmessling]

Just compare the middle section where all the wear happens to the upper and lower parts. My worn one was like a knife edge.

[Luckyone]

Can anyone see the pic in my first post? They are hosted in the same place as other old ones that still seem to work but I'm having trouble getting any new ones to post properly.

Anyhow when I went to start try pulling the pinion shaft out again (as in the old pic if anyone can see it) I remembered the steering rack was the only thing left in a fully stripped chassis the last time I tried. There is a lot more in the way like engines & water pipes now, so I've put that on hold for the time being.

Instead I double checked all the UJs & got quite excited when I found a bit of play in the splines, so I pulled the UJs both off cleaned all the splines then reassembled it all only to find there was.... no difference at all still too much play

I can very clearly see the pinion shaft moving exactly with the steering wheel but the rack shaft not so much.

I found the video on the excellent rack rebuild thread in the stickys, mine hasn't got that much play, its 10mm on the rim of the steering wheel.

I took the adjuster assembly out & cleaned every thing I could, it didn't look to bad but still more rust than you want to find! However with that back in there is no difference in the 10mm of play with the adjuster wound in to the point the rack wont move to the adjuster backed right out so it's just the spring. Seems a bit odd but I guess it just points to the pinion being worn.

I spoke the new official TVR parts suppers, they said they had the new pinions remanufactured but they aren't cheep here is the new pinion. They offer exchange racks but he said they don't really like doing that & would much rather you sent in your own rack for a rebuild, the basic rebuild being £477 then if you need a new pinion they would add that to the bill so 477 + 270 = £747

If the rack bar is also worn that would need replacing but that is currently showing unavailable https://tvr-parts.com/tvr-parts/part-details/tvr-h...

HOWEVER copied directly from that link "This rack bar is for the 7 tooth pinion and the rack bar length is 602mm. This rack bar will not fit TVR Tuscans, Tamoras, T350s or Sagaris which all have longer rack bars."

So it sounds like later racks are not a good idea, that is my worry with a Scooby rack too. I happen to have a WRX sitting here so measured the rack movement for one revolution of input & surprisingly its almost exactly the same as the Cerbs at about 50mm give or take a couple of mm. I also have a 200SX S14a & that rack only gives about 30mm.

But with the WRX it's the rack bar length that bothers me. I didn't try pulling the boots of a perfectly good rack on the WRX just to check the rack bar length but the rubber boots on the ends of the rack are about 420mm on the WRX & 445mm on the Cerb.

The rack bar needs to have the correct correlation of distance between the inner wishbone mounts (& the rest of the suspension points) to work correctly or you get bump steer. Our Cerb had not got any bump steer & I'd rather keep it that way! I know there was a thread some time ago about bump steer & I think a group buy for some bits to correct it, but there is something about our 98 "GT" that means it hasn't got any bump steer. I forget now if its different up rights, or maybe some of the later Cerbs had the wrong rack bars from new (not too surprising knowing TVR!) but I carefully checked mine when I rebuilt the car & there was virtually no movement of the uprights with in the normal range of suspension movement.

That rack bar difference is a new one on me, I'm not sure where they are measuring it or I'd check mine.

So I'm a bit stuck for ideas now, I think I'll try to get my fuel leaks fixed & take it for a drive to see of the UJs have made any difference to the feel. I did notice the 200sx has about the same about of play at the wheel (probably in the rubbers) but that doesn't notice as it cushioned.

I know pre rebuild I had too much tow in on the Cerb so that would to a degree mask the play in the rack I guess as the car is making its self drive straight more, it's running very little now so the steering has more immediate effect, thus showing up the free that was always there I now assuming. It's done 40k miles & I see quite a few others had problems before that.

[GT6k]

Two things i would add

You absolutely must have the joints bolted to the column with no play, part of the issue with mine was that the installer had used all the length up in the top joint and had the very bottom joint was not seated fully in. Then you have to do the bolts up super tight to eliminate play. If you allow play to occur for any length of time it seems to sightly oval the hole in the UJ and it never goes properly tight. I eventually put a further new set of UJs in and got the total play to an acceptable level although there was still some in the rack.

The second idea is to go and talk to the people doing electric rack conversion, I know of one converison http://www.lloydspecialistdevelopments.co.uk/ which seemed to work well.

[pmessling]

The sticky is of my rack and I used a tamora rack

I think the difference in length doesn't make a difference in later cars because of different wish bones. Same as used on the T cars. If you need any parts measured I have a second rack in pieces.

I'm not sure if you could swap to the later wishbones to be able to use the later rack.

I opted for a used tamora rack which on stripping it down was in very good order. All new seals etc were fitted. I also packed plenty of lithium grease in the adjuster.

[Luckyone]

GT6k said:

Two things i would add

You absolutely must have the joints bolted to the column with no play, part of the issue with mine was that the installer had used all the length up in the top joint and had the very bottom joint was not seated fully in. Then you have to do the bolts up super tight to eliminate play. If you allow play to occur for any length of time it seems to sightly oval the hole in the UJ and it never goes properly tight. I eventually put a further new set of UJs in and got the total play to an acceptable level although there was still some in the rack.

The second idea is to go and talk to the people doing electric rack conversion, I know of one converison http://www.lloydspecialistdevelopments.co.uk/ which seemed to work well.

Yes I spent quite some time with my UJs, just cleaning up the splines wasn't enough, in the end I put them in my hydraulic press to push them properly down on to the shaft, they were a nice fit without the bolts then. I'm happy there is no play there at all now.

An electric rack does seem like quite a good idea, but we had a BMW F13 640d MSport for a while & the steering in that was horrible absolutely no feed back at all, I complained to BMW & they said all the new BMs have electric power steering now apart from the M5 & M6. We didn't get another BMW! (A real M would be nice but my other half does too many miles for one).

Out of interest what do you think is an acceptable level of play to be left with? As I said I now have 10mm on the steering wheel rim.

[Luckyone]

pmessling said:

The sticky is of my rack and I used a tamora rack

I think the difference in length doesn't make a difference in later cars because of different wish bones. Same as used on the T cars. If you need any parts measured I have a second rack in pieces.

I'm not sure if you could swap to the later wishbones to be able to use the later rack.

I opted for a used tamora rack which on stripping it down was in very good order. All new seals etc were fitted. I also packed plenty of lithium grease in the adjuster.

Thanks yes it would be intresting to know what your rack bar length is, I assume the one you have in bits is the original from your Cerb? I tried to measure mine but is very hard to do it accurately in the car, I think it something like 675mm between each of the outer ends of the ball joint cases of the track rod. That is odd though because those ball joint cases are about 30mm which would make my rack bar 615mm, I guess I could be off by that much in my poking though the gators guessing measurements.

Similar poking & guessing shows the WRX rack is about 660 in the same places, so possible quite similar to the Cerb.

Lifted from the other rack bar they sell

https://tvr-parts.com/tvr-parts/part-details/tvr-h...

"This rack bar is 665 mm long, please check your rack bar! The rack bars are only likely to be 689mm, 665mm or 602mm. This rack bar is for the 7 tooth pinion and is the rack bar as fitted to the vast majority of later cars."

As they say its only likely its one of those three lengths it's possible the "GT"s had a different length bar again....

[Luckyone]

pmessling said:

TVR parts does have the pinion but it's not complete and I couldn't figure out how the valve part separated from the pinion and didn't want to damage the inner torsion bar.

Rebuilding the rack as you did is an option but if a new pinion is required that was the other problem he mentioned, getting the valve part off is apparently very complicated & fiddly with a good margin for messing it up. Correct Setting up of the inner torsion bar etc was also critical he said.

Of course they can do it for you, that's part of the £700 odd I mentioned above...

[pmessling]

I was trying to figure out how the torsion bar separated from the pinion. Looks like some drilling pins out.

My pinion is unusable now

I used my original rack with the tamora pinion. I did measure and they are the same. I'll get a measurement tomorrow. When I'm in my garage.

Rebuilding it wasn't that bad after my first mistakes. But worth pressure testing before putting it back in.

[Luckyone]

pmessling said:

I was trying to figure out how the torsion bar separated from the pinion. Looks like some drilling pins out.

My pinion is unusable now

I used my original rack with the tamora pinion. I did measure and they are the same. I'll get a measurement tomorrow. When I'm in my garage.

Rebuilding it wasn't that bad after my first mistakes. But worth pressure testing before putting it back in.

Ah that's interesting, do you mean you reused just the rack body or the rack body with the rack bar? If so I guess the teeth on the rack bar were ok then? I was wondering if that could be an issue.

It will be very interesting to see what your measurement is!

[pmessling]

Original rack bar and body but with the tamora pinion. Very little wear on the rack.

[ukkid35]

Where is the play?

At one point in the rack, or everywhere?

Logic suggests that if it is everywhere then it can't actually be the rack and it is very unlikely to be the pinion.

[Luckyone]

ukkid35 said:

Where is the play?

At one point in the rack, or everywhere?

Logic suggests that if it is everywhere then it can't actually be the rack and it is very unlikely to be the pinion.

Indeed, a good point, I was thinking that, but its definitely more pronounced with the steering wheel in the straight ahead position be that in the centre of the rack or one full turn in either direction.

There is play between those points, with the wheel upside down but it's much more cushioned. When the wheel is in the straight ahead position it just moves with no resistance in the region of play, stopping hard at the ends (within that 10mm of movement at the rim)

I've no idea how the torsion bar in the pinion works or even how the power assistance is activated, I assume there must be an amount of movement against the torsion bar for activation & a total guess had me wondering if any play could develop there.

The change in the feel of the play had made me discount the idea but you have me wondering again now!

[ukkid35]

There is only one 'correct' position for the pinion relative to the bar (because of the flat on the spline where the UJ attaches).

But you can fudge it if you don't mind the UJs being either off centre or misaligned (or you create a new flat).

The UJs should be perfectly aligned with the flats all matching, and with the pinion flat horizontal when the steering is centred.

This made no sense to me until someone explained that the UJs are not Constant Velocity joints, so the steering feel would be uneven/unbalanced side to side if they are not aligned correctly.

However, you can fudge that to test whether there really is play between the pinion and rack, especially in the centre position.

[Luckyone]

ukkid35 said:

There is only one 'correct' position for the pinion relative to the bar (because of the flat on the spline where the UJ attaches).

But you can fudge it if you don't mind the UJs being either off centre or misaligned (or you create a new flat).

The UJs should be perfectly aligned with the flats all matching, and with the pinion flat horizontal when the steering is centred.

This made no sense to me until someone explained that the UJs are not Constant Velocity joints, so the steering feel would be uneven/unbalanced side to side if they are not aligned correctly.

However, you can fudge that to test whether there really is play between the pinion and rack, especially in the centre position.

Now that is very interesting.

I assume when you say bar in "pinion relative to the bar" you are talking about the the link bar between the two UJs & not the rack bar (that slides side to side inside the rack).

I knew about UJs needing to be in alignment as in prop shafts, then as you say found the flats make sure that they are when you assble the UJs & shaft.

I also knew about the rack bar needing to be exactly centred with the steering wheel in the straight ahead position to avoid bump steer issues.

BUT I hadn't heard about "with the pinion flat horizontal when the steering is centred." Mine is a good 90 degrees off that as I pulled the pinion out to change the seal & had a hell of a job getting it back in again. (I now think it may well have been a lot easier if I'd removed the adjuster!) any how I'd lost track of the original orientation of the pinion in my many goes to get it in.

So you say the UJ on the rack should be something like the pic with the steering wheel in the straight ahead position?



The play is certainly in the less pronounced area with the UJs like that.

I don't really want to make a new flat on the pionion & it would mean my rack bar was sticking out more on one side than the other so I'd have different length track rods.

I have a better look at how hard it is to get the the pinion assembly out with everything still in thr car...

[ukkid35]

I may have got the orientation of the flat on the pinion wrong.

With the bar centred, there is only one position of the pinion where the flat is either horizontal or vertical (ie where the UJs will be perpendicular).

[ukkid35]

Also, I ended up using 12.9 cap head bolts and clamping the UJs beyond what seemed possible.

I was using standard nuts, found the torque at which they stripped, and then tightened slightly less.

[ukkid35]

Luckyone said:

So you say the UJ on the rack should be something like the pic with the steering wheel in the straight ahead position?



The play is certainly in the less pronounced area with the UJs like that.

Yes, that is what I think is the factory position, but that is also where I would expect most play.

I would expect less play if you reseat the pinion differently so there is a different straight ahead position on the pinion.

But then you have the flat issue to deal with.