Weld shrinkage

[Bobfromoh]

When welding a series of carbon steel cylinders together,
is there any "rule of thumb" for the expected weld shrinkage?

 

[metengr]

Bobfromoh;
No. Weld shrinkage is based on several factors (weld size, orientation, deposit sequence, size of the component, filler metal stregnth, preheat, base metal strength......) As you can see predicting weld shrinkage is complex.

 

[SnTMan]

Bobfromoh, a company I used to work for used 1/8" shrinkage per round seam, and 3/16" on a long seam, carbon or stainless, in moddrate thickness, up to say 2" or so. It worked pretty well so far as I ever heard. That is, I never heard it didn't work.

Regards,

Mike

 

[metengr]

SnTMan;
Are you sure about these values? Stainless and carbon steel have different COE, strength values, and I find it hard to believe these numbers.

 

[Bobfromoh]

Thanks for the replys.

BobfromOH

 

[SnTMan]

metengr, yes although these numbers were used only for figuring course lengths and wraps, its not highly accurate work.

My current company doesn't seem to account for shrinkage at all, you just go off nominals.

Its just blacksmithing.

Regards,

Mike

 

[evelrod]

On the structural steel of at least one high rise as I recall, there was no accounting for weld shrinkage, regardless of beam size, on the prints...however, the Ironworkers would spread each weld point by approx 1/8"---seemed to work out. As Mike says, "It's just blacksmithing".

Rod

 

[weldtek]

As metengr points out there are far too many variables involved to establish a rule of thumb for weld shrinkage. In my opinion the suggestions posted above amount to a guess at best and may be misleading in many cases. If the question posed was specific and detailed welding process, material type/grade, joint geometry and thickness etc. a meaningful prediction could be made, otherwise forget it.

 

[SnTMan]

"Measure with a micrometer, mark with chalk, cut with an axe"

Regards,

Mike

 

[desertfox]

Hi Bobfromoh

From a book called "principles of welding technology"

ISBN 0 7131 3402 X


longitudinal shrinkage:-

L1 = Lo * (1-?*?T)

Lo= original length

?= coeff of linear expansion

?T= temperature change after welding (ie melting to solid)



transverse shrinkage:-

transverse shrinkage= k*A/t

k=empirical factor between 0.1 and 1.17

A= cross sectional area of weld

t= thickness of plate

These are for butt welds and are only approximate, maybe if you do some experimental work for your product you can establish some data and combined with the above determine
the "k" factor and hence get a better approximation of the weld shrinkage.

regards

desertfox