How do we know we have the right fitting? 2

[geoffreydavies]

We have experienced weld failure on a number of branch to header connections in thin-wall pipe. The application is 8-inch branches on a 20-inch flare header, where both main run and the laterals are specified as schedule 10S gr. 304 stainless steel.We have had our own problems with welding and quality control in this material but we have a secondary concern as follows. The piping designer has specified "Weldolets" on his MTO, and I am assured that the installed fitting is a proprietary one with an API mark, but I am not convinced that our contractor supplied and installed the correct fitting. What has been installed looks like a simple sleeve placed over the branch pipe and not properly contoured to the main run.

I guess what I'm looking for is a picture of what would regard as a "typical" proprietary reinforcement for the application.

Thanks

Geoff Davies

 

[micjk]

Geoff,
"Weldolets" should not have the appearance of being slipped over the branch piping. This would constitute a socket weld type fitting (which I am not sure exists in that size) and weldolets are butt weld fittings. The fitting should have,along with api mark, material and size 8X20. Which welds are failing, fitting to header or fitting to branch piping. In both cases I would be more prone to believe weld procedure is at fault. If this is a socket weld scenario and the piping service is "hot" maybe the branch piping is being bottomed out in the fitting not allowing for thermal expansion of piping. I've seen this cause cracks on some of our jacketed bayonets. I have also seen steam piping blow apart for the same reason. Roy Gariepy
Maintenance and Reliability Dept.
Bayer Corporation Dorlastan Fibers Div.
Goose Creek, South Carolina USA

 

[hokie96]

You can see what a weldolet looks like on Bonney Forge's website.

http://www.bonneyforge.com

Follow appropriate links: Branch Connections to Weldolet (and its various configurations)

It's interesting to note that Bonney Forge recently acquired WFI, another maker of (similar) proprietary branch connections.

As for your current situation, I'd be a little leary using a weldolet to connect your 8" sch 10S to 20" sch 10S. Weldolets require quite a bit of weld metal for proper installation. Significant distortion and high local stresses can be created when fabricating such a configuration with thin wall pipe.

 

[Thomasjl]

Piping mechanical engineer

geoffreydavies
This contractor aparently dit not use a weldolet. However I do not thnk a weldolet is the right branch reinforcement for 10S wall pipe.
It looks that what the contractor did is the right solution after all, providing he made this reinforcing sleeve from main pipe material (which he probably did as this is the easyest way)

 

[Salzilla]

There has been mention of not using a weldolet on thin pipes. I have seen no mention of this from Bonney F.
The 20" sch. 10 pipe has a wall thk of 1/4".
Is this not sufficient?

I have used weldolets for joining 8" sch. 10 [thk of 0.148"]with 3" sch. 40 pipe with no difficulty.

Should i revise my practice and use a sweepolet in its stead?

 

[DSB123]

Salzilla,
The problem with using Weldolets on thin wall SS piping is that the Fabricator generally does not follow the Weldolet manufacturers requirements as regards amount of weld and more importantly weld profile with the result that there is often an inherent "crack" at the weld to 0'let juncture from which a crack could develop and propagate. I have seen the results of this in the past and am very wary of specifying weldolets for Sched 10s SS piping. As you say a sweepolet is the better option (no possibilities of poor weld geometries). Also as hokie96 says weldolets need a significant amount of weld metal to install correctly which will cause distortion on thin wall SS pipe.

 

[MJCronin]

Salzilla,

Here is my two cents......

There are forged,reinforced branch connections made specifically for Schedule 10S pipe. They are made in 304/316 SS by WFI of Houston TX.

As I understand, the problem with thin-walled pipe is one of maintaining the correct heat during welding. This is particularly difficult when welding thicker parts to 10S.

Both WFI and Bonney Forge offer pictures and diagrams showing what a properly welded fitting should look like.

If failures are occuring in a flare header application, do you really have a vibration (explosion) problem ? The most likely place a thin walled, branched system will fail under shock or vibration will be at the branch connection.....Is the system adequately supported ?

My thoughts only.............


MJC

 

[11echo]

I'm guessing your contactor used half couplings, in stead of a "true" weld-o-let. I'm afraid contactors are notorious for sub-ing materials in construction. Weld-o-lets require a lot more welding and time for installation. IF your experiencing failures on your branch conn.s, you may want to look at expansion problems on the main header. You need to ensure the branch connections have enough flexibility. My 2 cents ...11echo

 

[geoffreydavies]

Thanks, 11echo, I am concerned that we got short changed on the fitting used by the contractor. I have a photo of the installation - I can email it if anyone is willing to let me have their email address and offer opinion on what was installed. I'm not sure if I can attach photos to these threads. my email is geoff_davies@anadarko.com

 

[hacksaw]

use of half couplings is not necessarily substandard pactice. there are cases where they are actually preferred over other means. it all depends on your service conditions, materials, etc.

by the way you can post photos on this site.

 

[geoffreydavies]

Thanks for the idea to post a photo, however my employer's website security does not permit putting pictures onto websites. I can email photos to anyone who would care to pass comment on what we got installed. Sure does not look like any weldolet in the bonney forge catalog
Thanks
geoff_davies@anadarko.com