Steering Tie Rod Fracture

[christopher storey]

Today, after 60 miles on my newly rebuilt S1 4.2 FHC, fortunately whilst reversing on full lock in a friend's driveway, the left hand tie rod suffered what looks to be a fatigue failure ( crystalline appearance under medium magnification) at the point where the rod joins the ball in the rack housing . Has anyone else had this? and more to the point, can anyone suggest a cause - I am concerned that there is some other suspension defect which has put excess strain on the rod

[daverawle]

Wow! that's mega scary. While not having any firm idea about your tie rod failure I have had problems with track rod ends which "neck out" after a small deflection of the suspension as the wheels drop. If your car is riding a bit high that may be worth checking. Let us know what you find.

Dave

[p1vickers]

Christopher

This was probaly caused by the track rod ball joint being adjusted too tight. This will cause the weld to the ball end to be under great stress when the suspension flexes.
It also may just have been a poor weld, but not very likely.

Please let us know the outcome.

Pete
1967 2+2

[Heuer]

Christopher

That is worrying. If it is an original part I can only think previous accident damage left a hair line crack which eventually gave way when the rod was under most stress. If it it a repro part maybe worth talking to the seller! A good friend of mine raced a Lister (now runs a Lola T70) and he insists all parts are crack tested every couple of seasons regardless of no visual damage. You will not believe the number of parts he has replaced as a result, some of which could have proved fatal if they expired at track speeds. Not practical on a road car of course but illustrative of it not being a rare problem. Just as well you discovered it when you did.

Was there any obvious damage, corrosion or scratches and any chance of a picture? Replacing the right hand tie rod would be sensible.

When I was researching the Jochen Rindt E-Type for this thread: http://etypeuk.com/forum/viewtopic.php?t=1311 I was reading the cause of his untimely death at Monza in the Lotus. "On investigation, an Italian court later ruled the accident must have been caused by the right front brakeshaft breaking under the torsional load, thus explaining the sudden jinx to the left under heavy braking, with cold tyres exacerbating the problem. He would probably have lived if just one of the following had happened: the brakeshaft manufacturer not scratching the surface of the shaft during service and weakening it or the Lotus team spotting the defect during quality control ..........."

[christopher storey]

Pete and David. It was an original tie rod, put back in by me when I rebuilt the rack . I don't think I overtightened the ball housing, the rod was fairly easily moveable by hand, but not floppy . Interestingly, though, there was bluing on the outer part of the ball housing, suggestive of heat, but my engineer friends think this was part of the hardening process, and the ball was still liberally coated with grease and came out easily. For the metallurgists amongst us I show the fracture surfaces. What worries me more is the resting position of the unloaded suspension on the unbroken side ; it does not appear to be necking, but it is perilously close to it

[Heuer]

Christopher

Did some searching via Google images and turned up this interesting picture and description:

As indicated previously, a fatigue crack will grow as the component is used. Conversely, without cyclic/alternating stress (or a sufficient level of cyclic stress) a fatigue crack in a component or structure will not grow. Thus, if a record is available and/or can be predicted, i.e., number of flights, hard landings, severe winds, etc. and when and where severe stress occurred in the structure and then periodic inspections can occur and those inspections will be timed to catch fatigue cracks. That is to say, things, components, structures (aircraft) can and will have cracks in them. Isn?t it comforting to know the airplanes we fly have fatigue cracks in them, yet they continue to function perfectly? The secret to safe use and a most important concept in fatigue analysis is critical crack size. That is, a fracture (complete separation) will not occur until the crack grows to a size that is critical. Another way of saying this is, a crack can exist, and even grow, but the metal will not separate or fracture just because a crack is present. The metal will separate when the crack reaches critical size. This is the ?whole? concept behind the inspection of aircraft. You find/catch the cracks, remove/repair/blunt the cracks before, in fact, long, long before, they reach critical size. That is, the (NDI) inspector might miss the crack during the first inspection (say February), but the next inspector will find it on the second inspection (say June). Even if the second inspector also misses the crack, a number of additional inspections will occur before the crack reaches criticality.

Let?s try to visualize the concepts behind metal fatigue. Photograph A is a pin, used to swing or pivot a large electronic gate at a very large, high security facility in the DFW area.


As the gate swings open and closed, cyclic stresses are applied to the gate hinge pins. Because of a machining/design/manufacturing error, a fatigue crack starts in one of the hinge pins at the 10 o?clock position. In operation, this pin is in a two-way bending. To help visualize two-way bending, take the ends of a pencil, one end in each hand, and bend/bow it down (without breaking) and then up. As the pencil is bowed up, the top of the pencil will be arched. Think of fibers oriented longitudinally along the pencil. These fibers are stretched and thus they are under a tension/pulling stress.

Actually, another fatigue crack did start at the 4 o?clock position (see arrow at 4 o?clock position in Photograph A). Every time the gate cycled between open and closed the subject gate pin, experienced tension first, (lets say, as the gate is opening) on one side of the pin and then as the gate swings closed, tension is experienced on the other side of the pin. The fatigue crack grew first because of the cyclic tension stress, from the 10 o?clock position and then from the 4 o?clock position. The fatigue crack grows at each position only when a tensile stress is present, i.e., the top or bottom fibers in the bent pencil are stretched. As the gate is used, the cracking grows into the pin, from the surface toward the center. As the crack gets larger, there is less metal to take up the applied stress each time the gate cycles and the crack grows faster and faster. Finally the crack, or in this case the two opposing cracks, reach critical size and there is insufficient metal left in the pin to take up or transfer the stress from opening and closing the gate and fast fracture occurs (the remaining metal rapidly fails in an instant). The pin separates unexpectedly and possibly catastrophically. The final fast fracture is the darker band in the center of the pin fracture, which is oriented from about 7 o?clock up to 1 o?clock.

Site was here: http://metallurgist.com/html/MetalFatiguePt2.htm

Picture looks the same as yours suggesting the fracture started on each side of the rod, flexing occurred which eventually caused your failure. Suggests the rod was damaged either on two opposing sides or there was a circle of damage. Just a matter of relating that back to its relative movement on the car. I reckon the answer lies in your last picture above.

[christopher storey]

David : thank you, that is a most interesting post . The orientation of the diametrical band in my tie rod is precisely at right angles to the up and down bending moments which form the bulk of the cyclic stresses on the tie rod i.e. exactly the same as illustrated. Also, there are quite heavy corrosion pits on the neck of the rod for about 3mm away from the fracture line, which might have been the original propagation source for a crack. New rods and housings I think!

[christopher storey]

Many thanks to those who have contributed to this thread. I measured my centre to centre extension of the dampers at full droop and found it to be an average of 420mm. Clem ( from the Jaglovers board ) in South Africa kindly measured his and found it to be at least 10mm and probably nearer 20mm less than this , which rather suggests that the neck of the rod has been able to impinge on the ball housing. Accordingly I am replacing the dampers, the tie rods and the housings and hoping that this will restore safety. It also probably explains apparent rack vertical movement, since after full (hyper) extension of the suspension , an upward thrust is probably being imparted to the rack end which would not be present ( or at any rate would be of lesser degree) if the proper geometry were preserved

[christopher storey]

Just to round this one off: when I got the dampers off the car, and compared them with the new ones, I found that the old dampers had an extended length of 435mm compared with the new ones 400mm. It is no wonder that the wishbone was dropping so far at full droop that the tie rods were being subjected to excess strain - in fact I think it was the tie rods which were restraining the suspension from any further droop, rather than the dampers!

All ( tie rods, ball housings, ball cup and dampers) have now been replaced on both sides, and I hope this will mean that I need not fear a repetition

[Echezeaux]

While checking the front wheel bearings free play, I noticed that with the suspension totally relaxed the angle of the steering track rod ends was not quite right.

I heard about track rod ends sold for E-type not being the proper ones. Indeed, after removing them from the tie rod levers, it was obvious that the angle needed to match the lever by the tie rod was not wide enough.

Having replaced the old Girling shocks by new adjustable GAZ not too long ago, I went on to measure their extended lengths.
Girling : 390 cm
GAZ : 430 cm
Similar to what Christopher found on his car.

BINGO ! Maybe wrong shocks ! But no, even with the stub axle carrier lifted to match the length of the Girling the tie rod angle is still not right !
So now, still wrong shocks plus wrong tie rod ends.

Checking with the usual parts retailers, all their tie rods appear to be the right one, but how to be sure ??

As for the shocks I need to have their travel reduced.

Anyone with experience in this matter ?
Any help appreciated.

Pics to follow.

Thierry

[daverawle]

Checking with the usual parts retailers, all their tie rods appear to be the right one, but how to be sure ??

Thierry,

You could take your incorrect one to the supplier and compare it, the difference in angular movement is quite noticeable when placed side by side. These were on my car, the correct one is on the left



I replaced them with Quinton Hazel QR1178S which I found on ebay.

Dave

[Echezeaux]

Thank you Dave for the part number and the pic, something to show as a picture talk much better than word.

Thanks again for your time.
Thierry

[Echezeaux]

two pictures of my problem:

[christopher storey]

Thierry : this is almost identical to the problem which led up to my tie rod fracture, although I do not know whether my track rod ends were incorrect. The fundamental problem is the excessive downwards travel of the shockers. I suggest that you undo the clips on the gaiter and pull it back and see whether the inner ball end of the tie rod is "necking" on the ball housing on the rack i.e. the lower periphery of the rod touching the ball housing. If it is , then fracture is a definite possibility

[daverawle]

With the front jacked up and the suspension at full droop it should be possible to turn the steering from lock-to-lock without it jamming, mine wouldn't do that with the incorrect track rod end. I also have Gaz shocks and as best I could measure with them on the car they were around 410mm between the bolt hole centres.

Dave

[Echezeaux]

Thanks all for the replies, I think that we have a similar problem.

Christopher, as you mention, the problem seems to lie with the shock, and like Dave, I went on doing some measurements :

The suspension fully dropped with the Gaz shock : 410mm
The same without the shock : 410mm ,412mm

The gaiter of the steering rack removed shows what I am calling the ball pin
touching the housing. The wheels being straight.

I am using the names from the Jaguar manual as I get confused.

Now, if I lift the suspension to 400mm, shock's length, the ball pin is free even turning the wheels side to side.




my none original tie rod ends are not binding at 390mm

Checking the length of my old GIRLING Vs GAZ :
GIRLING collapsed = 260mm Extended = 392mm
GAZ collapsed = 300mm Extended = 412mm

Keeping in mind that the mid laden position is 343mm

So... the GIRLING are more at the proper lengths and the GAZ are too long.

I contacted GAZ about this, their answer is that they can reduce the down stroke of their shocks, at a cost of 15 pounds each, well ! living on the other side of the planet, I'll prefer to find someone a bit closer.
I will shop around tomorrow to find the proper tool to open the GAZ and fit a spacer of around 30mm inside.

Let me know what you think about it.
Thanks again for your inputs.

Thierry

[daverawle]

You have me worried Thierry, so I have just removed the Gaz shocks to make an accurate measurement = 418mm. They are now on their was to Gaz to be limited to 400mm.

Hope you get yours sorted without having to return them.

Dave

[christopher storey]

Thierry : that appearance is very similar to mine. I think you are wise to limit the GAZ travel by some means. It is , however, very surprising that these large and apparently reputable damper manufacturers do not seem to realise the hazard that the excessive travel causes , and I think it needs to be pointed out to them

[1954Etype]

Right OK, I'll be the one to ask to dumb question. What difference does it make? Providing your track rod ends don't neck on the road (and I am assuming that only leaving the ground would cause this if you have the wrong ones fitted) and the same applies to the shock absorber I would think. Although, I cant see what having too much travel in the shock would make.

Clearly I am missing something here!

[daverawle]

Angus,

The problem only occurs when jacking the car; overly long shocks allow the suspension to droop to the point where it is stopped by the steering components. They were never intended to be stressed in this lateral way and in Christoper's case stress fracture of a track rod ensued. What we potentially have are 2 after-market components that are incorrect; track rod ends that don't have sufficient angular movement and overly long shocks. I have already replaced the track rod ends and my shocks are on their way back to Gaz, who have agreed to limit their extension to 400mm. I will point out the problem to them but I suspect they will need a picture of an original shock and a tape measure to be convinced - anybody got one?

Every MOT time when the tester raises the car I'm asked if it's correct that the steering is so tight and I always told them that they are all like that, but they shouldn't be so mine's getting fixed.

Dave