Butt weld or fillet weld: where do you switch?

[XL83NL]

In my every day work I frequently encounter the discussion when one chooses for (full pen.) butt or (non full pen.) fillet welds in the design of pressure retaining welds (like encountered in pipelines, vessels, etc).

Now, before replying, dont get me wrong; all code vessels and pipelines thats been worked on (usually skids) have full pen. butt welds, where possible.

However, for a simple small reactor (sometimes with capacities less than 50 ml), containing a 6x1 mm tube as in- or outlet, it's difficult to weld that tube to the treactor with a full penetration butt weld.
And there's no design code the reactor has to meet.

If this reactor would operate at, say 30 bar and 300 deg C, of course youll want to make sure youll make full pen butt welds.

But what if, assuming no design code has to be met, the design pressure is 0.5 bar, and operating temperature is say 150 deg C? Or 100, or 300 deg C?

Non full pen. fillet welds are often way more practical, but .. where do you draw the line?

 

[stanweld]

It's the designer/manufacturer's decision, based on service safety, joint efficiency and cost. Refer to Joint Type No. (4),(5)& 6)in Table UW-12, Section VIII, Div.1 as examples.

 

[XL83NL]

Stanweld, thanks for your reply, but thats not what Im looking for, prlease readm y first agaian. I was looking for 'criteria' baded on temperature and pressure. Not specifically from design code (I can imagine they dont hand these), but perhaps more from industry practice or experience from people here at the board.

 

[Duwe6]

Since your detail is not "legal" under the ASME Code, there is no guidance. How about you install a socket-weld coupling on the reactor? Then the connection of the tube will be a socket-weld, which is a wrap-around fillet weld.

 

[XL83NL]

Hi Duwe6, thanks for joining the topic and providing help.
Socket welds are neither full pen welds, and are a probable source for crack/split corrosion. Hence not preferred for the higher operation regions.

As I mentioned in my first, I prefer not to have such weldments for a certain range of pressure and temperature. (which is to be determined)
Now, Im looking for suggestions when it's a good idea to switch to (the more cheaper/easier) fillet/socket weld, and when to make sure to do full pen butt welds.

Lets take example: say 150 deg C and 0.5 bar design conditions. hydrogen service; full pen butt welds or fillets?
Keep in mind that Im not only talking about 1" or 2" Sch 40 pipe, but also 6x1 mm tubing. See practical execution, NDO, etc plays a role too.

 

[moltenmetal]

You're only going to get an opinion answer to this question. Here's mine:

300# class and below (obviously 2" and below): socket welds, up to 1000 F.

600# class and above: butt welds- unless one or more of the following criteria are met:

- corrosion conditions are modest (i.e. especially cracking promotion mechanisms), and
- temperatures and temperature cycling are modest (rule of thumb? ~400 F max), or
- project design life is short (i.e. a couple of years at most)

If these conditions are met, we go well beyond 600# class with SW without too many sleepless nights.

Stress intensifications on these assemblies can be bad news. Flexibility analysis is not usually done on lines this small, but since we build skids and things are tighter than usual we tend to do it for anything beyond 400 F- it helps reduce the major problem locations.

 

[SnTMan]

XL83NL, for hydrogen service full penetration welds are much preferred, but if not possible, unvented spaces must be avoided.

Regards,

Mike

 

[XL83NL]

@ moltenmetal; thats exactly the kind of discussion Im after. I understand there's no exact answer to the question I asked in my first, and thats whaty I mentioned in my forst post ...
second; are you talking about design pressure corresponding to 300 lbs range, or about explicitly about 300 lbs flanges?

@SnTMan; the before posted message was just an example. hydrogen service came to my mind first, which is (in my course of applications) probably the most severe service encountered for which the design of of the device doesnt have to meet any specific design code.
e.g., a 50 degree 3 bar vent or nitrogen pipe is more towards the opposite. in such a case, I wouldnt be far less afraid to use socket welds.

all in all, its difficult to mark the line where you go from one to another, but im interested in others opinions.
moltenmetal's reply is good example, e.g. about other aspects (like corrosion) to think of.


last but not least; product design for the business I work for (kinda R&D) is very short, sometimes just a few months or only cycles.

 

[XL83NL]

what I meant was; product design life for the business I work for (kinda R&D) is very short, sometimes just a few months or only cycles.

 

[moltenmetal]

When I say 300#, I mean MAWP/T within the limits of ASME B16.5 flanges of that class. Otherwise you got the basic idea.

Total judgment call. We've used SW up to 1500# class under the caveats noted above.

 

[XL83NL]

Hi moltenmetal,

Thanks you for reply, and sorry bout my late response. I appreciate your input.
When I was going through some standards a while ago, I noticed they usually approximately apply the same limits (if any) for non full pen. fillet or full pen. butt welds. With all the input aobve, I think I have enough information on where to draw the line for each instance and application/service.

Regards