Another de-dion tube failure

[Tony C]

FWIW, just put mine together, though it's an SV and may be different. The lower damper attachment is a threaded boss under the DD tube into which goes a bolt, lockwasher and two big radiused washers either side of the damper bush.

 

BRG SV 😬 Boxes empty, car full

[stevefoster]

Arnie the lower mounting is a blind nut lug. The bolt just locates into it, no nyloc.

With the required shimming for the damper to clear the tube there is not that much bolt going into it.

Any loosening would spell disaster for sure. I have always threadlocked all these bolts.

 

I agree the centre mounting shorter damper that goes through the tube using a nice big nyloc should not give any problems.

 

 

My racing pics, 7 DIY, race prep. Updated often here

Photo's of the year here

Hants (North) and Berkshire Area club site here

 

[stevefoster]

I have just replaced my dampers that bolt into the lower lug with shorter ones that go through the centre mounting. The bolt thread length that went into the tacked on lug was 12mm. The bolt diameter 1/2".

This damper was shimmed as instructed to allow the damper to just clear the tube.

 

The new dampers and springs together weigh 1 kg each less.

 

 

My racing pics, 7 DIY, race prep. Updated often here

Photo's of the year here

Hants (North) and Berkshire Area club site here

 

[Bare]

Upon reading all the Talk about the De Dion failures.. It occured that perhaps this is the perpetual problem of the Infamous/ridiculous A Arm locator.

Same basic problem as in my Series 3 back in the early 70's :-(

The Central A Arm mount allows drivetrain torque to end up at the axle ends .. where there is NO support (A Arms are a bit Cheaper to manufacture.. and every Penny counts apparently) twisting whatever is weakest at that point into eventual failure.. NOT news tho.. this has been V well documented/understood for at least 25 years!!

Witness the myriad live Axle failures generated by this ill concieved design... and the various laughable "fixes".. usually comprising a section of Rail Road track (or equally elegant Bodge) welded onto the Axle as "reinforcement".. rather than a reversion to the Series 1 .. 5 link/upper and lower track rod locating method.

Check out the Lotus 11 "Lemans" variant.. a beautifully elegant 5 link De Dion system...Nothing.. like the Cat setup is it ??

V strange to see that the Geniuses at Cat are so reticent (terrified?)to "stray" from the Sainted Colin's Series 2/3 design .. that they refuse to fix even the Blatantly obvious.

In short: if you have a problem: add the lower Track rods and remove the A Arm thing .

End of Rant

 

bare

[Old Septimus]

Bare's comments are of interest as this is the conclusion I reached over 20 years ago with the help of a very knowlegeable series 1 owner. My series 3, then over 10 years old had a Ford axle without the ghastly plate bracing and had visibly bent with only the power of a standard 1600 Xflow. We got another axle from a scrappy and Brian welded extra turrets below the axle line to take lower radius arms on each side which picked up to the outer Aframe locations. We kept the A frame but all it does is provide lateral location. It's still going strong, hasn't bent and doesn't leak much oil.

[Mike Bees]

The DD tube doesn't get torqued by the drivetrain. The diff mountings in the chassis do.

 

Mike

[Old Septimus]

Our analysis was that it wasn't primarily a "torque" issue but that looking at the axle in plan view under acceleration and (to a lesser extent) braking there was a load applied to the centre of the axle and loads of one-half that sum applied in opposite directions at the outer ends. This caused the axle to bend as a beam loaded in the middle. Can't comment on whether there is an analagous effect for DD tube.

[R2D2]

Looking at the photograph posted at the start of this thread I would think that the failure is due to torsional fatigue.

 

It is quite possible that poor welding has caused a crack to initiate from a highly stressed region but to put a weld into an area such as this just isn't too great and is asking for trouble unless some really good quality control is used.

 

[stevefoster]

I agree Chris. Putting a weld in the that area is one thing, Asking it to take the direct weight of the corner of the car through all the stresses of usage is quite another.

Yes it could be done but I would like to have seen a bigger tag on spreading the load more and thicker metal. Ah the new tube is like that I guess....

 

 

My racing pics, 7 DIY, race prep. Updated often here

Photo's of the year here

Hants (North) and Berkshire Area club site here

 

[Bare]

Old Septimus has begun to understand :-)there is little difference between the DeDion and the live axle setup "force" wise... with the Dedion..the loadings on the ends are further increased by the offset Cat design.. this adds'"leverage" to the forces at the Axle/tube ends.. push it hard enough and the weakest bit fails... not rocket science here.

You can claim weld failures if it makes one feel better.. try gusseting each one, may even work.

Even the late unlamented Alfa GTV6's had a dedion axle.. but not surprisingly they used a 5 link system... and although these cars were just as breakable as a 7.. the was never any dedion /axle end problems even in racing.

Dunno.. but If mine failed 3 times I'd be looking real hard at ANYTHING that might offer a genuine solution..

[Peter Carmichael]

The deeper the a-frame triangle, the weaker the fore-aft loadings on the A-frame. In the fore-aft direction the a-frame merely prevents the de dion from rotating.

 

Agree that the centre loading is double the end loading, but the end loading is the acceleration minus the force transmitted by the radius arm. There is a further small component for the bending caused by the tyre contact patch being outboard of the radius arm pickup point.

 

So the de dion bending load can be made smaller by keeping the radius arm pick-up point close vertically to the centreline of the hub and as far out as possible. There you have it. Moving the radius arm pick up brackets to the front of the de dion will prevent de dion failures.

[Peter Carmichael]

Not sure about old septimus's extra links because if everything isn't exactly parallel the suspension will bind in roll, just like the jeremy Phillips back end of the sylva striker.

 

You have to be careful how many degrees of freedom you constrain in suspension design and you only start with 6.

[Old Septimus]

Yes, we got it all parallel and it all moved freely before we installed the springs. The important point was that the bushes in the new turrets were in the same line vertically and horizontally as the Aframe centre bush and the new radius arms were mounted on extended bolts at the outer Aframe mountings. Accept that in roll the bush compliance is necessary to make it work.

 

The theory is a bit rusty now and it's a long time since I did any physics but our view was that (on a live axle) the torque transmitted through the wheels was forcing the axle to rotate about its own axis and this was balanced by forces applied above the axle line by the existing radius arms and below it by the Aframe. From a side view this all balanced and was fine. In plan you then saw that these forces were applied at different points which caused the bending.

[Graham Perry]

Although mine are bolted through the centre of the tube I didn't like the look of the bottom bush when I replaced my dampers. After removing them completely I realised that as moved the suspension over its full travel, it was twisting backwards the bottom mount and hence my bush was shot. We manufactured a new lower mount with the bolts going across the car parellel with the A frame bushes and radius arm bushes (Sounds like Septimuses) through the dampers, and it now looks, and strangely feels better. I don't think this could be done on a later car through, but that same twisting movement is still there on the later mounting.

[Dave_H]

Graham

 

I found just that when I put my new (well second hand) de-dion tube on. The lower damper mounts seemed to be at the wrong angle and where very difficult to get lined up. I too would have thought that they’d be better mounted at 90 degrees, parallel to the de-dion tube? Or am I talking rubbish

 

Dave

 

[Graham Perry]

Thats what I did. The brackets were made using good quality steel and I used high tenstile bolts

[Peter Carmichael]

Ah. Old Septimus I can now see that the torque reaction you are referring do doesn't occur on the de dion. the bulk accelerative forces though are still partially transmitted by the A-frame.

 

If the brackets are at the wrong angle when fitting, they are relaxed when the suspension is loaded under the car's weight. If they were 90 degrees round they would see flex in roll equal to the roll angle. This might be more than the flexion they see in their current implementation. Obviously with spherical joints none of this matters.

[Bare]

How does one post a Photo here??

Have one of the old Lotus 11 "Lemans" Version that shows how what a simple elegant solution was available and used..before even the Series 1 Seven had it's frame "improved ".

The "Torque Reaction" isn't supposed to happen on the Live Axle either.. If one believes the sales talk/theory.. but the reality is diametrically at variance:-)