SS warping

Date 04-11-2007 17:07

Anything you do is going to be of limited success. It is just the nature of SS material (low thermal conductivity-high thermal expansion). What you are seeing is typical. You can attach chill blocks as heat sinks (though in my opinion more trouble than its worth), use pulsing (not an option really, if you don't already have the equipment) or use low parameters (as you have already surmised). But in the end its going to happen anyway. And in general it does nothing to interfere with the service life of the component. It just looks bad.
Keep in mind, consistent with other threads in the forum, that you may be adding more weld metal than needed. 16 gage, half couplings, this would be my guess.

By js55

Date 04-11-2007 17:12

There may be one other method that could help if you are using full pen welds and autogenous welding is not prohibited. Are your welds full pen?

By piperdolan (*)

Date 04-11-2007 18:16

Full pen ( full penetration)??. I'm using .030 wire, its about 1/16 fillet mostly less. no penetration on this weld. I figured that the chill blocks would do little to stop the material from moving. havent tried pulse, I do have that option.

By js55

Date 04-11-2007 18:49

I was thinkin GTAW. Sorry bout that.
I've seen full pens on half couplings. Even with material that thin. For whatever reason.
The chill blocks wouldn't keep the material from moving, just assist getting heat out of the material quicker.
Pulsing may help some, or since its wire, short circuiting (if thats not what you are already doing), but in the end its gonna happen.

By piperdolan (*)

Date 04-12-2007 10:30

this is GTAW, thanks for the response I'll keep trying till something is more satisfactory.

By js55

Date 04-12-2007 13:46

If you really need to minimize it there is one other thing you could try, but its a lot of trouble. An ID spider jack. Spider the ID with a strip of pipe the same diameter as the one you're welding to support the pipe under the O-let. Set the strip underneath the weld point and press it against the ID with a jack. Then relieve the jack to remove the spider.
The pipe will still move but the sink may not be as readily visible.

By aevald

Date 04-12-2007 14:46

Hello js55, I can see the validity of your suggestion for the spider/jacking approach. However, I believe in this case that could be a real problem as I am pretty sure this is a very small bore application, he's only drilling a 1/4" hole for the pass through on the half coupling. Just an observation. Regards, aevald

By js55

Date 04-12-2007 14:49

Thanks Allen.
For some reason I didn't have it pictured that way. Oh well, perhaps in another application.
Spider jacking would not be a solution to a small bore application. Not sure they make jacks that small. :>)

By aevald

Date 04-12-2007 15:52

Hello js55, I understood where you were coming from. Others who have read this thread will come to appreciate your suggestion for this in an application where the size allows it to be used. Nice recommendation. Regards, Allan

By aevald

Date 04-12-2007 14:50

Hello piperdolan, one thing did come to mind here as I was reading the original question again. You said that you were using a half coupling to weld onto the tube. Is this half coupling fitted to the radius of the tube or are you just filling the gap? Having a tightly fitted half coupling that matches the OD of the tube would help to somewhat minimize distortion also. Just a thought. Regards, aevald

By welder5354 (**)

Date 04-13-2007 03:54

Hi Piperdolan, would it be possible to drill a hole in the coupling, insert the tubing and then braze it
with silver solder (easyflow 45).

By piperdolan (*)

Date 04-16-2007 10:58

Thanks for responding, the half couplings are not fitted. I measured the gap and its 1/32 both sides. It looks like proper fitting, pulse arc maybe the way to go for me.THe half coupling is for a measuring sample line and catalytic substrate is pressed into the tubing, so any distortion from the half coupling causes problems in pressing that substrate in.thanks again for the response