[bomondus]

I am told by someone who used to do it often that solid square bar does not reduce in length when twisted.
He used to cold twist no heat.
I assume its got something to do with steel and iron's compressive strength .. anyone give any insights as to whats happening during the process?


Brett

[mikeydesignzinc]

I make jewellery, so all my experience in this area is in gold, silver, and platinum; whenever I twist bars of these metals, I get reduction in length. I always have to allow for the reduction in the length of the starting bar.

I haven't tried it with steel or iron.

[moopogyOvenny]

I make jewellery, so all my experience in this area is in gold, silver, and platinum; whenever I twist bars of these metals, I get reduction in length. I always have to allow for the reduction in the length of the starting bar.

I haven't tried it with steel or iron.

it is all basic induction.

If you want me to pull it all through a draw plate. You can but I'm sure there are others who can explain ductility.

[ceagstuntee]

> I am told by someone who used to do it often that solid square bar does not reduce in length when twisted.
He used to cold twist no heat. I assume its got something to do with steel and iron's compressive strength .. anyone give any insights as to whats happening during the process?

If you twist a bar what you get is a stress state of pure torsion. Torsion is made up of equal amounts of tension and compression. So when twisting there is very little net tension or compression. This isn't something with surface tension.

[bomondus]

Thanks for the reply Moll,
The bar I was referring to it 1" solid mild steel , probably a mid carbon.
If that doesn't reduce in length and from Wocky's experience his precious metals do , is there different things going on internally in the different metals?
Could the sectional sizes have a role ?

[WaydayTew]

Length reduction would suggest that either the cross-sectional area or the density is changing. The density isn't going to do much. What happens to the cross-sectional area when you twist it?

[AdobeCreativeSuite]

I thought they did reduce in length, but just going on dim memory of my materials lectures from eleventy billion years ago.

[AdobeCreativeSuite]

In fact to not reduce in length the edges would have to stretch.

Take a bar square cross section. 1m in length. Each edge is a line 1m long. Now twist it. For the bar to still be 1m In length each edge would now describe a line more than 1m long (because it doesn't go from start to finish in a straight line).

So either the bar stretches along the edges, or it shortens. Either way there is a change in density of the bar, and that change would be dependent on both shape and distance from the centre of the bar"

I think....

[AdobeCreativeSuite]

Length reduction would suggest that either the cross-sectional area or the density is changing. The density isn't going to do much. What happens to the cross-sectional area when you twist it?

If you twist something with a square cross section the cross section must change (otherwise it wouldn't twist), therefore density must also change (unless both crosse tonal areas are the same).

When you twist it the cross section cannot remain exactly a square, because if it was a perfect square then it wouldn't have a twist going up it.

[Patgaepx]

The cross section of a twisted bit of square bar is modified by the flat sides dishing in. will see if I have a piece of square bar and give it a go.


Edit: I have a piece of 10mm bar but it is too small for my 4 jaw chuck, I also have a piece of 1 inch bar but I think I might break something if I tried cold twisting that.

[Ndptbudd]

The cross section of a twisted bit of square bar is modified by the flat sides dishing in. will see if I have a piece of square bar and give it a go.

Or Google image search it. Twisted bar

[SeelaypeKet]

Here's what an elastic finite element analysis says a steel bar will do (10 m long, 100 x 100 cross section):

Longitudinal displacement contours:


Plot of longitudinal displacement along one edge:


End section (displacements x 5):

[SeelaypeKet]

Or Google image search it. Twisted bar

Of course, when you start getting plastic deformation, it's a whole new ball game.

[SeelaypeKet]

Twisting with plastic deformation: