Internal manhole Flange

[victorpbr]

Hi,

I'm designing a horizontal vessel for refrigerated CO2 and the client wants me to use an internal manhole flange, as the one shown in the attached file, he claims to be using it for several years never had any problem. I've never designed or even specified this type of flange before.

Any hints on the aspects I should analyze to see if its code compliant?

My first impression was that the number of bolts was too low, but given that the pressure inside the vessel is "helping" the sealing, maybe it is appropriate. Anyway, I'd like to be able to calculate its components to see if it is allowable.

Gasket is a teflon elastometer

 

[JStephen]

The concept is very similar to an elliptical manway. Is this a commercially-available unit, or something that needs to be custom-fabricated?

 

[victorpbr]

I'm not familiar with elliptical manways, I'll give a look into it right now.

This particular piece is not commercially-available, it is custom fabricated

 

[weeeds]

Make sure there is no vacuum condition in your design as these types of closures will not seal under any level of vacuum.

 

[JStephen]

http://www.cmforming.com/pdfs/asme-code-elliptical-handholes-manways.pdf

 

[victorpbr]

Thank you for the reference JStephen, found some material on UG-34 and some procedures at Moss book. I'll give a look into them, meanwhile if you have any more technical references on the design of it it would be very appreciated.

Regards,

Victor

 

[victorpbr]

Ok, after some reading here are my thoughts

Based on the Procedure 3-8 from Moss 4th ed, I see it is simply using the Eq from UG-34(c)(3)(3). However since my geometry is a bit different I'd assume that using Eq from UG-34 (c)(2)(2) would be more appropriate as I`m using a flat plate.

However since my bolts are on the inside part of the flange, is it correct to change the formula from t= d*sqrt(CP/SE + 1.9Whg/SEd^3) to t= d*sqrt(CP/SE - 1.9Whg/SEd^3) ? This way I assume it would account for the opposite bending moment? The closest sketch similar to my configuration I found on figure UG-34 was item (k), so can I assume C has the value of 0.3 or there is more suitable way to determine its value?