What exactly are you wanting to calculate?

The reinforcement check needed for the manhole. API650 gives tables with needed pipe size and reinforcement size, EN14015 gives a formula based on the area replacement method, like EN13445 does. But the area replacement method formulas are always based on round cut-outs, not on square ones.
I assume the local stresses around a square cut-out are different from those on a round cut-out. That's why I'm looking for some code-formula's to check the reinforcement needed for a square manhole (or cut-out in general).

Depending in the size of your tank you could reference API12F.  That is the only API code that I know of that allows the use of rectangular manways.  This would only be suitable for tanks with low shell stresses (The maximum liquid head that 12F applies is 25')  API 12F is not a design guide, it doesn't have any reinforcement calculations, but it does give you an idea of what situations rectangular manways have been used.

The stresses around rectangular manways can be large due to the discontinuity at the corners.  I have seen many oval manways used to reduce these stresses.

Thanks for your reply.
I have seen rectangular manholes in roofs, but not in shells indeed, I think for the reason you mention (high stresses at the corners).
If an oval manhole is used can I then use the area replacement methods used in the different codes, using the small diameter for reinforcement in one direction and the large for the other direction?

Maybe another option is a rectangular manhole with rounded (half cirle) ends. I assume I can use the same area replacement formula using the diameter (which is equal to the cross section of the cut) and add a reinforcement around the manhole?

Is there any code adressing one of these options?

Thanks in advance for your reply.

The reinforcing calculations for manways don't make any allowance for stress concentration around an opening.  It's just a matter of replacing the area you cut out.  In that sense, it would be identical for a rectangular opening.  Typically, there isn't any calculation actually done on an API tank, you just use the standard-sized reinforcing plate and you're done.

Using a square manway, you have stress concentrations you don't have with a round manway, and the potential for cracking in or at the corners.  You also don't have the past experience that you do with conventional nozzles or manways, so you're on your own.  If the manway is square, it would still be preferred to round the corners, or make it an obround manway or something.

thanks for your reply.
I was already afraid I am on my own in this issue.
If the client really wants a square manhole and the normal shell stresses are high I think a FEA will be the only option then.
If normal shell stresses are low I'm not really afraid about the stress concentrations.
 

ASME Sect VIII UG-34-c-2 gives some guidance on thickness requirements for non-circular openings.  You could look to see if they give anything on reinforcement as well.

Even if you use FEA, you won't be able to model the crack initiation of a square cut, so you will have to assume some radius.  If you can pick a reasonable radius like 3 inches then you might have a good chance at the model being a good predictor.  Still, you will have to specify that the cut edges be ground smooth etc.  See the treatment in API 650 for flush clean outs for some additional ideas.

"If an oval manhole is used can I then use the area replacement methods used in the different codes, using the small diameter for reinforcement in one direction and the large for the other direction?"

Normally, there wouldn't be any check in the vertical direction.

What about a 36" or 38" round manway ?

Aside from preference, why must the manway be square ?

Manways are used to move "things" in and out of the tank.

What reason does your client give when you ask why must it be rectangular ?

 

   

The rectangular shape is just for easy entrance in the tank. I know a larger diameter round manway could also be used, but client insists to use a square one.

Eventually the problem is solved much easier than I could expect.
After some questioning the client told me he has already a couple of the same tanks with the same manhole in use for many years. So instead of making a calculation we have now decided to use the "copy from the past" method. It's in use for years and no issues have been found. Both internal and external inspection bodies agreed.

Thanks all for your replies on this matter.

API 650 basic design crude tanks are witnessed to have sludge gates/D-doors at Side Shell/Bottom Peripheral area with dimensions  occasionally the same& bottom edge nearly flush with tank bottom,whereas top slightly rounded like letter "D".

In all fairness the respective formula should be applicable in your case most presumably indeed!

Best Regards
Qalander(Chem)