"diamond heated pipe bending" 1

[Hansac]

Dear all.

It is now 2020 and ASME PCC-2 2018 has been published with this paragraph:

"307-4.3.2 Cold bending refers to bending below Tcrit. The pipe may be bent cold if permitted by the applicable construction code or post-construction code. Cold bending may require heating below the critical temperature; heating is usually on the side that needs to be shortened and is applied in a diamond pattern."

My question is simple. What is "diamond pattern heating"? Can anyone give a pictorial example? Google does not yet have it.

Much thank you.

thread378-205206

 

[LittleInch]

I can only assume that this refers to heat via a torch and applied in a criss cross fashion to spread the heat out and prevent any hotspots developing.

All sounds rather dodgy to me like everyone said in the other post you referred to.

I'm not surprised that no one is admitting to doing it this way.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[TBP]

Diamond heat on carbon steel pipe. I've seen it done years ago, typically to straighten-out something like a pipe that had some o-lets welded onto one side in close proximity to one another, to make a header. After welding, the pipe had a visible "banana" shape to it, with the bowed-out side opposite the o-lets. The fitter rolled it in the pipe stands, so that all the connections we pointed down. He then applied heat from the oxy-acetylene torch to the back-side of the pipe, opposite the o-lets. He applied this heat evenly, moving the torch slowly in a diamond pattern - with the longer points oriented to the ends of the pipe, and the shorter points of the diamond to the side. My recollection is that he did this opposite each o-let individually. Good practice or not aside, it worked - the pipe was straight.

Here's a link to a discussion on another forum:

https://app.aws.org/forum/topic_show.pl?tid=2066

 

[moltenmetal]

This is an old pipefitter's method of straightening pipe or making minor adjustments in angular tolerance on flanged sections of pipe. Pipe is heated, then shrinks back when cooled. At the moment I can't remember if you heat on the side that needs to be longer or the one that needs to be shorter when you're done, but it definitely does work if you remember and do the right side!

Straightening pipe distorted by welding O-lets with heavy welds to fill the full bevel of the fitting (especially on thin walled pipe, where welding out the full fitting bevel is often done but is BAD practice) is a common example, which makes me suspect that TBP is right- you heat the side that is too long and it shrinks back when it cools.

I've also seen it done on vessel nozzles where angular tolerances were close but weren't quite met on the nozzle neck to flange joint, for parts that had critical fit-up tolerances- i.e. a probe has to be inserted through the nozzle and the flange to nozzle and nozzle to shell tolerances determine where the probe actually ends up located inside the vessel etc. It's less invasive surgery than chopping a flange off and re-welding it.

The diamond pattern is as TBP mentioned- a diamond shaped area of pipe taken generally to red heat, with the long ends of the diamond along the axis of the pipe, and the short ends across the diameter.

 

[robyengIT]

quote "At the moment I can't remember if you heat on the side that needs to be longer or the one that needs to be shorter when you're done, but it definitely does work if you remember and do the right side!"
You have to heat keeping in mind the cooling process, so that you have to heat the side that has to shorten

 

[LittleInch]

The issue for me isn't so much that it works - we know that - but what level of stress does it build into the pipe and is there a risk of material change in terms of hardness, hard spots, brittle sections etc.

About 25 years ago I saw a ship builder using this technique on large plates to form complex shape forms using precise strips of heating in various locations and this flat pate became curves in three dimensions.

The idea that someone with a flame gun heating to red hot can do this to the level required to not change material properties is where I have serious concerns.

I think yes you can do it, but then you really need to heat treat the piece of pipe you've just messed around with. I.e. basically like you do for induction bends.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[berkshire]

This technique is also used on steel tube aircraft fuselages, where a crossmember is welded in, the tee shape tends to shrink on the welded side so heat is applied on the opposite side to straighten things up.
B.E.

You are judged not by what you know, but by what you can do.