Using a flat plate instead of an end cap on a pipe 2

[borobam182]

Hello!

First of all – let me say that this is a great forum for pipeline and piping issues. I visit this forum on a daily basis to read about different topics and issues that other engineers have to try and learn more. The discussion and advice is always very informative and helpful.

I now have an issue of my own that I would like some help with. I’ve had a look through the relevant design codes and around this forum and I’m unable to find any information to help me.
I am designing a high pressure anti ice pipework system for an air intake filter house. The design code is ASME B31.3, design temperature is 410 oC, design pressure is 16 bara, pipe material is A312 Gr TP316L and pipe size is 3”.

I need to provide end caps for the 3” riser pipes. I have specified an ASME B16.9 end cap in A403 Gr WP316L material. My understanding was that you would need to use a ‘dished end' end cap to distribute the stresses correctly and also to allow for the full penetration butt weld.

However, our manufacturing team have proposed an alternative and I’m having difficulty to find the relevant clauses in the code to rule out their approach.

They would like to weld on a square flat plate over the ‘hole’ in the pipe instead of use an ASME B16.9 end cap. They will prep the end of the pipe and leave a weld gap between the end of the pipe and the plate to provide some penetration, but I don’t feel comfortable using a flat plate when I would usually specify a ‘dished end’ end cap. A pressure vessel had domed ends - and in my opinion, a piping system is no different. A flat plate would be subject to large bending stresses due to the internal pressure and a 'dished end' end cap can manage these stresses better due to its shape and construction.

Do you agree that the flat plate is not a safe and recognised approach?

Any advice would be very welcome.

Thank you in advance.

Craig.

 

[LittleInch]

You need to look at section 302.2 This basically states that you can use "listed components" i.e. listed in other standards like B 16.9 without needing to do anything else to establish the pressure and temperature rating.

A flat plate is a bad idea and is not a listed component.

To use an unlisted component is possible but you will need to follow 304.7.2

The costs to do this work for a 3" 16 bar end would be many many times what a 3" dome end cap costs.

Hope this is what you're after.

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

 

[borobam182]

Hi LittleInch,

Thank you very much for your response.

I feel happy that you agree with me that its a bad idea.

I will review 304.7.2 of ASME B31.3 to understand what is required when looking at unlisted components.

Thanks again.

 

[KevinNZ]

As above you would have do some design to use a flat plate. Plate thickness and weld detail will need to follow code requirements. We have done this many times. We have also found that a cost effect source for the plate is a blind flange. Better than getting a disk cut from boiler plate stock.

 

[LittleInch]

Still don't understand why the manufacturing division want to use a flat plate. End caps cost is negligible.

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

 

[trashcanman]

Have whoever came up with the flat plate sign and seal the drawings.

 

[borobam182]

haha - I wish it was that simple @trashcanman - they have sent a deviation into my department and are trying to get me to approve it! all roads would lead back to me - not them sadly!

thank you for the message @KevinNZ - I'm finding that the plate thickness would be large (much larger than they are proposing to use). a blind flange would ease my worries a little as this is a listed item to a specific ASME code. But I feel that this suggestion would be more expensive that the flat plate.

And like @LittleInch highlighted, the cost of an end cap is low when compared to the alternatives. An end cap would be off the shelf - a plate would require a certain amount of preparation.

Thanks guys.

Craig.

 

[LittleInch]

Just write back to the deviation stating that a flat plat is not a listed component in ASME B 31.3 and therefore cannot be used without considerable analysis. The cost of this analysis will be many time the cost of using a listed component such as an end cap or a weld neck flange and blind flange to ASME B 16.9 / 16.5 respectively.

Job done.

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

 

[blacksmith37]

Years ago I was on the ASTM A1 committee; A user reported he got plate when he ordered a blind flange forging . most of the committee was upset and wrote requirements into the forging specs to be certain plate flanges were prohibited. ( I don't remember the exact problem ).

 

[borobam182]

@LittleInch - I 100% agree - that is the most sensible way forward. The time and cost associated to verify the alternative far outweighs any possibly small cost saving for using a plate. And the analysis could prove that the plate would have to be very thick and need machining and hence, more expensive that originally expected. Thank you.

@blacksmith37 - thank you for your feedback on this issue. This adds more weight to my argument against using plate.

 

[DekDee]

borobam182,
No matter what the industry you are working in you will always find the "manufacturing team, fabrication team, construction team, production team" will have a short cut they think is a great idea.
Their job is to get the job completed as soon as possible.
My job (I am in quality) is to ensure it complies with the relevant codes / standards.
Your job as an engineer is to ensure it complies with the design requirements.
This is not what the "manufacturing team, fabrication team, construction team, production team" want to hear and that is why there are always conflicts between the various disciplines,
Good luck with your battles,
DD

 

[maxdistortion]

As per 31.3 sec. 302.2.3 Unlisted Components, if components are not listed in Table 326.1 of 31.3, which is where B16.5 fittings are listed and rated, a designer has the option of using or "creating" a component, as long as he/she is satisfied with the mechanical properties and integrity. This allows you to fabricate plate blinds. There are other caveats to this clause but you can essentially make a plate blind. The "satisfaction of design" and extent will vary depending on the owner and their criteria. It is mostly more expensive to perform extensive analysis than to purchase a rated fitting.

We use home made plate blinds all the time. We made charts and created weld details to use them up to a specified pressure rating, but we do so only for hydrotesting purposes. In this case its all about thickness, grade and compatibility in welding to base metal than it is for a design of operation, as these would be subjected to temperature-pressure limits, longevity and corrosion.

I hope this helps

 

[borobam182]

@DekDee - you're right - the manufacturing team always have an idea which is cheaper or quicker to order than the way that I would prefer to do things. one of my daily battles. it is obviously important that the final design isn't too expensive and can be built in a competitive timescale - but the integrity of the design should never be compromised.

@maxdistortion - thank you for your input on this issue. I agree that the time spent and money spent performing this analysis would be much more than the cost of the rated fitting. my manufacturing teams thought is that if we can design a cheaper solution once, then we can re-use this idea on future projects. but I still feel that a rated component would be the better and safer option.

 

[maxdistortion]

@borobam182 I think there's not much money to be made by making your own components, I am of the opinion that reinventing the wheel is not very efficient. And if you guys create a new type of fitting, what is to say that a client will accept it, you may have spent lots of money registering and jumping through hoops with the codes authority in your area. Where we do work, our clients often issue pipe specs with procurement, just so vendors and contractors are using similar fittings and branch tables.

Where I have seen lots of home made fittings is in the non-registered piping. For example, one large vendor that sells water treatment packages makes their own flanges, in their case, the size of B16.47 flanges are hard to get, and they don't require to be rated to high pressures, so they make their flanges out of plate and just weld it onto rolled plate duct.
This approach work well for them.

What type of fittings do the manufacturing team want to make? maybe they want to go into the business of making fittings and selling them by the thousands.

And as far as safety, we designed caps out of plate for our hydrotests, we use them all the time, they are cheaper, we make them faster to whatever size we need, and we do it safely. Some of these spools get tested to 20'000 kpa. But we can only do it on our property, and once the pipe goes to the client's property, we can no longer test with plate and have to use rated fittings from a qualified manufacturer to perform our tests (with a few exceptions).