Are we outside the limits of UG-39? 3

[heaterguy]

Hello,

I have found this BBS to be helpful in the past. I hope to have the same outcome. Will someone else calculate the following:

Design Pressure is 3200 psig
Design Temperature is 225 'F
Corrosion Allowance is 0.125"
Flange size is 12"
Flange rating is 1500#
Flange material is SA-105
Gasket is RTJ soft iron
Bolts are SA-193-B7
We are drilling (84) 0.484" holes with a 0.75" pitch

UG-39 calculates the flange thickness to be 9.5"! Is it possible that these calculations were never meant for such extreme conditions? Is there another method for doing this calculation?

Regards,

Heater Guy

 

[waskillywabbit]

Why not use a B16.5 standard class 1500 flange? (According to the Taylor Forge flange rating tables that will work) Why so many bolt holes?

Brian

 

[heaterguy]

We are making a "tubesheet" for electric heating elements. The 84 holes are for the heater rods. These rods are welded to the flange / tubesheet.

 

[metengr]

heaterguy;
If you are making a tubesheet for a heat exchanger, you would be best served using ASME Section VIII, Div 1 Part UHX for your tubesheet design calculations.

 

[heaterguy]

Thanks Metengr, but ASME has replied that UHX does not apply to our heaters and that we should us UG-39.

 

[unclesyd]

Take a look at SIFCO ASME insert flanges. You are almost there with their Class 1500 insert flange. The Class 2500 would be an overkill. If it is possible to use an insert flange that means you would only have to machine the hub. I don’t have the referenced codes to do the calculations if the hub is solid and how it would affect the pressure rating. They make specials so a slightly thicker ring might fall right in line. If this is possible you would only have to replace the hub not the ring.
Down load their catalogue and look at table 1 and the go to the Class 1500 and 2500 insert flanges.

http://www.insertflanges.com/

 

[prex]

Agree with your 9.5" value using UG-39(b) (perhaps a bit less, but over 9", however didn't make a detailed check).
You should get a slightly better result by using UG-39(e), though still over 8".
A more difficult way you could try to follow is by the use of UG-42(c): in fact if you encircle all your openings in a single opening you obtain a ring flange that might be calculated per App.2 (though this is not mentioned in UG-42). However then you will need to calculate the holed portion as a flat head: the diameter will be less and there is no bolt moment, however I guess that the thickness would be still in the 8-9" range after accounting for the holes.
By the way I don't think that the required thicknes could be anything less than 8" or so: if you were not bound by ASME, what kind of calculation would you intend to use (without collaboration by the rods and their welds of course) to determine the required thickness? And what thickness are you thinking of?

prex

http://www.xcalcs.com

Online tools for structural design

 

[heaterguy]

Unclesyd, thank you for that suggestion. We will look into it. We are also looking at Reflange:

http://www.oceaneering.com/reflange/Reflange.htm.

Prex, thank you for verifying our calculations and offering other sections of ASME to review. Electric process heaters have historically used ANSI blind flanges and have never had a problem due to over pressure. About 5 years ago, ASME recommended these flanges be code stamped and the industry now offers a partial code stamp on the heater. However, the 1500# and 2500# flanges are getting too thick and costly. We are looking for a way to comply with ASME and reduce the thickness required. Again, thank you for those recommendations and we would welcome any further thoughts.

Regards,

Craig

 

[unclesyd]

Have you looked at gasket seating stresses to get the flange thickness reduced. Here are two that I’ve had very good luck with.

http://www.helicoflex.com/catalogs/

http://www.selcoseal.com/index.htm

Thinking out of the box and under the hood. This is a nice picture if this flange assembly is amendable to your design and application. There are several people in the business with similar type flanges.

Checkout the picture.

http://www.vectorint.com/mainSections/Section02_SPO/2_products_intro.asp

Literature of an American Supplier.

http://www.ltsusa.com/Library.asp

 

[GenB]

I seen your other post, why another post?

I see no way out of the thick flange for the pressure involved, you could use different materials and make your own flange and thin it on the way of the higher strenght material.
I see no way out with std flanges because of the material you are taking out for the holes, one thing the holes may
not be affecting the strength of the flange because they are so small you do not even need to calculate for but for ligaments only. If you are using the heaters on Sect I boiler you may have to comply with Sec I Code and not Sec VIII
GB

 

[heaterguy]

GenB,

The other post had resolved itself. This post is to see if we are outside of the intention of UG-39. Should we pose the question to ASME to see if we only need to calculate for ligaments? Would this make the flange thinner?

Regards,

Craig

 

[GenB]

If the Code allows ligament exemption in your design
you may use the flange T as it is at max. pressure and temp,
The Commitee will not do your design, they will only give you a ruling.
I notice you are using Sec VIII-1. Is your heater for an unfired PV or for a steam gen for sec I construction?
the rules may be different.
GB

 

[heaterguy]

GenB,

You have my attention. I am not an ASME expert. I have done some simple calculations using canned software, but rarely get into the actual code.

Where should I look for ligament exemptions? Do you think 0.484" holes are small enough for an exemption?

These heaters are going into unfired pressure vessels, not boilers. Most of our applications are heating flowing fluids like N2, air, water, crude oil, natural gas, etc. Other applications are for heating the liquids that sit in the bottom of ASME vessels to prevent freezing or thickening.