I`m intending to build my own Dedion tube rear end due to the distance from suppliers of same & subsequent freight costs,( b4 James tells me to buy a GTS one ). Up till now discussion about these has revolved around round tube but is there any reason I couldn`t use RHS for a dedion tube, maybe 65x65x2.5 or even as big as 75x75x3. It strikes me that whether I go with a straight tube ala Caterham & Fraser or a bent tube als GTS it will be easier to attach brackets etc. A bent welded tube like GTSs would be able to be more effectively strengthened by plating the join & it wouldn`t be necessay to cut thru the rear face of the tube. not sure what the torsional stiffness of square as opposed to round tube is, maybe somepne who can ( Cymtriks?) can do a FEA analysis of this. Its a worry what a guy thinks about on night shifts, & this is the only stuff I`m admitting to!

just my 2p....
from speaking to race car chasis manufacturers, square tube is stiffer for the equivalent weight of a section than round. no info on od size or wall thickness though, but I trust my sources

Ned.

Ned,
That's correct but doesn't it depend on the angle to which the load is applied? What I mean is, if you try to bend a round tube it will have the same resistance to bending regardless of which point of the compass you bend it towards.

However, with square section it will (I think) be weaker if you bend it at 45, 135, 225, 315 degrees compared to bending it at 0, 90, 180 and 270 degrees. The square section will tend to collapse if you bend it diagonally in this manner.

These angles assume that at 0 degrees the flats will be in the 0, 90, 180 and 270 degree planes.

By the way, the GTS axle isn't bent to my knowledge, it is just welded at an angle in the middle.

Cheers,
Craig.

Craig,

Yes, I've only known square tube be used in flat planes as it were. The gts tube is cut at an angle and welded in the centre, mk's early indy with a dedion used a bent tube, but Darren reconed it wanted to straighten itself out which is why he opted for straight and welded in the middle.

Ned.

Just for your info

My dedion tube is square in section ( I can't remember what section but I can measure it later) cut at angles and welded in a similar fasion to the round GTS one.

I decided to use square tube as it easier to measure and cut accurately and easier to fix all the brackets etc accurately. Also I had some in stock!

I am making no claims as to it's suitability - just letting you know what I did.

You can all tell me why it is wrong now.

Phil

Phil,

It isn't wrong, just different.

Craig
A square tube is equal stiff in bending at any angel, at least in theory.

Cheers!

A round De Dion can be bended into shape but once you start to cut it up and weld back together you might as well go for square section.See no reason why not!
All in my VERY humble opinion

square might be stronger due to the higher 'I' figure (whatever that is...!) but i suspect that this is an 'instant' comparison, once it has collapsed via diagonal to flat, it has next to no strength. I suspect (IMHO) that round tube will be more resistant to collapse.

having said that, i would use square anyway, for the two reasons mentioned above- i have loads of 50x50x3mm in the garage, and its easier to get the angles and brackets right.

" but Darren reconed it wanted to straighten itself out which is why he opted for straight and welded in the middle. "
but then darren maybe dosnt work with round section / pipes everyday ???

Leto,
It is the collapse when bending loads are applied across the diagonal that would worry me, and that is why I am intuitively saying that square will be weaker than round when you take this into account. I know that square is stronger when comparing similar wall thickness and size but I don't believe that square will be as strong across the diagonal.

From an aesthetic point of view I think round wins hands down as well.
I've attached a picture of my de-dion arrangment (ex MK Engineering design I believe which I assembled from laser cut parts). I may be biased but I think this looks quite pretty...

Cheers,
Craig.
Rescued attachment De-dion.jpg

With the kind of loads you will put on it with a locost I don't think you have much to worry about.

quote:

---

Originally posted by Peteff
With the kind of loads you will put on it with a locost I don't think you have much to worry about.

---

Live axle tube is only 50-60mm 16g, I would have thought a larger diameter and thinner wall would be the way to go, either round or square

I will be using 2" sq 12 gauge for my DD tube. Someone who was kind enough to donate the metal also provided me with comparisons between other tubing.
Now,this is from information that was supplied to me and I do not want to start a slanging match between the various camps!!
2" Sq 12g is 27% stiffer in torsion/70% stiffer in bending than the tube used in the MK DD and 9% stiffer in torsion/51% stiffer than that of GTS.

Ian

Colin Chapman would go for 25 x 25 x 1.5,
drilled along the entire length

Ian,
I'd be interested to see the working for those figures if you are still in contact with your mate. I must admit I find it hard to believe but this is by no means an area of mechanical engineering that I am comfortable with. If it is so much stronger than the MK DD setup (which is what I have built) then I'd have to question if it is in fact overkill because my own axle seems over-engineered already. Also, do these figures take into account the seam in ERW square section? Can you get seamless square section?

Cheers,
Craig.

Craig

The bending load at wish permanent deformation occurs is the same regardless of angel. Total collapse is much harder to predict. In real life, I believe, small dent and other imperfections will affect the outcome more then the orientation of the tube.

Cheers!

Okay, but let me ask another question...

Why are virtually all commercially available de-dion systems made out of round tube then?

I've never actually seen or heard of a square one although there is bound to be one out there. Surely the designers of at least some of the commercial axles must have known what they were doing and they ended up using round tube. Why didn't they use square if it is potentially easier/better?

While I have been building my Locost I have either researched a given topic from first principles and designed a part myself or copied an established design. In the case of the de-dion axle I copied an existing design by MK engineering which used round tube. I don't see any good reason to go against the grain unless there is a distinct advantage in doing so and no bad side-effects. Sorry but I don't think that is the case here.

Cheers,
Craig.

My thoughts on structure....

comparing a round tube to a square tube of equal weight in the same material, assuming a sensible design of geometry the following can be true.

The axial stiffness will be the same for both.
The bending stiffness of the tube will be higher.
The torsional stiffness of the tube will be higher. (alot)
The bending stresses in the rhs will be higher.
The stress due to torsion in the rhs will be higher.

The component WILL NOT be designed for buckling failures !

Best practice is going to be straight and round for a tube, this is beacuse any offset from the wheel centreline (simpest description) will mean the tube is subjected to multiplane bending, shear torsion and axial foces. The behaviour of a round tube is understood for these loading condition wheras square requires more development proving. If the member is not straight then it will not behave in a linear fashion, i.e as loads get higher the deflection and stresses will not be proportional to load.

As per usual stick with the proven approaches. Take a look at a caterham tube.

Having re-checked what I used last night it was 50mm x 3mm wall - Overkill if anything.

Yes there is a reason why everyone (nearly) uses round but I think that for ease of construction a heavier square version is a reasonable compromise.

I think that any forces sufficient to damage my axle will be causing damage elsewhere anyway. Ok it isn't light but then again the V6 at the other end isn't either - mind you doesn't do a lot for unsprung weight.

It's all compromise - we'll see how it works.

Phil

quote:

---

Originally posted by ned
just my 2p....
from speaking to race car chasis manufacturers, square tube is stiffer for the equivalent weight of a section than round. no info on od size or wall thickness though, but I trust my sources

Ned.

---

quote:

---

Originally posted by leto
Craig
A square tube is equal stiff in bending at any angel, at least in theory.

Cheers!

---

Jeffers,
What about collapse of the square section. Won't it tend to happen more easily if the bending loads are applied at 45degrees to the "square" axis?

Cheers,
Craig.

quote:

---

Originally posted by craig1410
Why are virtually all commercially available de-dion systems made out of round tube then?

---

Thanks to Craig (cheers again mate), I am making my own De Dion. The tube on my Mini will be straight as I have a chain driven Radical Diff unit that's sits far enough forward to miss the tube. So I am making mine out of 1 3/4" 14G T45 hi strenght tubeing. I hope Colin would be pleased.....

FYI

Mk will make you a de dion tube to any dimension you want for a price that makes it a bit silly to make your own.

Cheers

Chris

Not when you live 12,000miles from them which is where I started this thread from. Several posters have echoed my point re easier fabrication, attachment of brackets etc, & while the MK style as used by Craig, ( Been following your progress Craig, looks ab fab to me, hope I can come close to yours oneday) & the GTS one look really smart, from the point of someone doing their own fabrication by & large ( me) I`m trying to use standard sections & avoid tricky things like bending which I`m sure to take several goes to get right or have to pay someone for. To date I haven`t heard anything to dissuade me from using square tube. Still open to further ideas though & will have to see what my approving engineer says. Cheers, Steve.