high pressure reactor problem 2

[dramarc]

Hi!
I´m currently calculation of the dimensions of a small (inside diameter 100 mm, height 200 mm)) cylindrical lab reactor with a flat head closure (like a blind flange with bolts). THe working pressure is 120 bars.

The flat head blind fange type of closure have three approx 10 mm round threaded NTP holes for inlet, outlets and pressure transmitter.

My problem is that I don´t know how to calculate the desired flange thickness. COuld you give me some ideas how to to this and in idea of the desired amounts of bolts and dimension of these?

best regards

Andy

 

[jte]

Andy-

You didn't mention the material of construction or temperature, but the easiest/cheapest way to go is to get a pipe with a (probably) CL-900 raised face weld neck flange on each end with matching blind flanges. Standard stuff per ASME B16.5. Just be sure to match your actual material to the pressure/temperature ratings. You may have to move to a heavier flange class.

jt

 

[PAN]

jte,
For weld neck flange ANSI 900# of NPS 4 in, The length (flange face to neck) is 121 mm. For both ends, the length will be 242 mm (greater than 200 mm). This may not be conformed to the requirement of Andy.

 

[dramarc]

THanks for the answers!

although I wasn't clear enough. The head of the vessel will be buildt according to my specs. The head of the vessel will be flat and bolted to the cylindrical vessel. This flat head will have four 1/8 or 1/4 " threaded inlets/outlets for the connnection of pressure transmitter, processmedium in/out etc.

THe reactor will be working at maximum 100 BArs but have to withstand 200 Bars (test pressure)

I can calculate the desired shell thickness and the desired flat head thickness (without drilled holes).

My problem is that I don't know how this holes in the flat head will affect the construction and how much thicker the head have to be in order to withstand the working pressure.

Is there an approx empirical formula for this or do you have to use finite elements calculations?

This reactor is for lab test pruposes only.

best regards

Andy

 

[codeeng]

First calculate the blind thickness per UG-34 formula (2). Then use any excess thickness/provide a repad for reinforcement.

 

[jte]

PAN-

Yes, that's just the engineer in me. Trying to take a proposal and suggesting a way to perhaps achieve the same goal at substantially less cost. Perhaps the laboratory calculations could be changed for less cost to match off the shelf components. If, however, the dimensions are critical, then of course the shell must be custom fabricated along with the blind flange.

Andy-

It would be interesting to get quotes from a fabricator for both approaches. Let us know what the difference in cost is.

This is all assuming that although excempt from the scope per U-1(c)(2)(i) that a design per ASME VIII Div. 1 is acceptible. For lab use I would tend to order the vessel "designed per ASME VIII-1 but not stamped."

You said that you already know how to calc the head thickness (VIII-1 UG-34(c) formula (2) as codeeng pointed out) and are now stuck on the opening calc's. Take a look at VIII-1 UG-39. The easiest way to provide reinforcement if you are having the blind custom machined is to add some excess thickness to the flange thickness and take credit for it as reinforcement. The easiest way to provide reinforcement if you are buying a B16.5 flange is to use something like a Thredolet (

http://www.bonneyforge.com).

jt

 

[Larry617]

The calculation of multiple openings in either a standard or customer fabricated Blind can be appracohed in the same way as normal openings in a vessel.

ASME B16.5 does actually allow you to tap holes upto a stated size into a blind without considering reinforcement.
However, this assumes only one hole per blind and that the hole is located in the true center of the blind. In this case we have 3 connections (assumed to be all in one blind). Therefore the ASME B-16.5 exempts do not apply.

The best appraoch is to apply DIV. 1 calculations per para's UG-36 thru UG-39. Be carefule of your limtis of reinformcent as these should not over lap or exceed the RF the RF dia of the blind. A commonly over looked problem. Also since we have 3 connections we need to be careful of the UG-36(d) cluster conditions/restriction that norally no do apply to exempt nozzle openings.

You may claim surplus materials in your blind. This can be done be perfroming the design calcuation to determind Thickness required for pressure. I generally do not claim this value however as most B16.5 flanges we claculated actually failt to meet the required thickness. But are accepted by the industry due to service experience.



I have performed similar calc's for DIV.1 U-Stamped vessels and have had few preoblems obtaing Canadian Registration Numbers (CRN) approvals.

 

[dramarc]

thanks for all the answers!

I live in sweden and I don´t have the ASME. Although I think I get the idea with the reinforcement and I´ll see if I can find anything similar.

-JTE I´ll let you know about the costs when it´s finished.

best regards

andy

 

[dramarc]

by the way, the reactor is for in house lab use only.

 

[dramarc]

could anyone give me an apporox thickness increase of the flat head when adding three 1/4 " tappings (holes) to it?

 

[jte]

dramarc-

A temperature and material of construction is necessary data which you need to provide. For carbon steel no higher than say 200°C your req'd thickness would be around 65 mm. Assuming the openings are several diameters away from each other and the gasket surface, the additional thickness for reinforcement would be around 33 mm for a total of around 100 mm. These are "talking numbers", not final, engineered numbers!

jt

 

[GenB]

Actually you do not need any special requirements on this flange,you can have as large as 1- 50mm opening w/o any compensation, so 4 small 1/8 and 1/4 openings comply with ASME Code,
note that thread'olets will do nothing to reinforce a flat plate on this condition, the opening is still there,
a pad! same condition, only added thickness or a buid-up pad can make a difference,
ER

 

[dramarc]

The working temperature is 20 to 40 degrees C (room temp.)
MAterial i stainless steel SS2346.

A 65 mm thick lid seems really thick! Accroding to my calcs (minimum values) a lid 2 cm thick is required when the shell thickness is 1,5 cm. this is a lid without the four 1/4 " holes.
HOw do you reason to come up with 6,5 mm thick lid?

 

[jte]

genralboiler-

Ummm... I think you are referring to the reinforcement exception in UG-37(c)(3). Unfortunately, per UG-37(c)(2) that paragraph is only applicable to cylindrical or conical sections or formed heads. Unless I missed something, there is no exception in VIII-1 for reinforcing openings in flat heads.

Not sure about your claim of a Thredolet being useless. Can you elaborate?

dramarc-

It all depends on what formulas you are using and what allowable stresses you are using. I made the calc based on ASME VIII-1 using 20000 psi as an allowable stress. On the other hand, I may have miscalculated...

jt

 

[prex]

dramarc, assuming your SS2346 may be taken as equivalent to a SS304 (it is not anyway allowed under ASME VIII, if you can use a forging per ASME II it would be better) and using a 4" class 900 flange+blind (a very good advice given above by jte, it requires 8 bolts 1 1/8" dia.), the working pressure could be as high as a little more than 140 bar (at less than 50 °C temp., and don't worry about the 200 bar test pressure, it is OK for that).
Now the standard thickness of the class 900 blind is 44.5 mm (plus the thickness of any raised face). I'm pretty sure that this thickness will take much more than your pressure, especially if you use an O-ring gasket (for such small flanges the required thickness may greatly vary with bolt and gasket materials).
Concerning the openings, GenB is correct: if their centerlines are spaced apart more than the sum of their diameters (when taken 2 by 2), threaded openings as small as 1/8" or 1/4" do not require any reinforcement : in other words the cover thickness stays with the thickness calculated without openings.

prex

http://www.xcalcs.com

Online tools for structural design

 

[GenB]

jte,
I am talking about excempted openings in flat heads and flanges, the flat plate will be only weakened if ligament efficiency is affected.
In this case if the small openings are in the 2in perimeter or far apart as per not affecting efficiency is also okay,
I do not have the Code with me but 2 diameters apart is the excemption.

as per the thred'olets being useless
it is only my engineering thinking; they may have some reinforcing properties but the Code will not recognize that.
GB

 

[dramarc]

HI all and thanks for all answers!

I´ve another crazy idea. THe reactor is still a shell with a flat plate closure. THe internal diameter of the shell is 10cm and the outer diameter is 17 cm. In order to se throuth the reactor I´ll sight glasses. The sightglasses are mounted in a stainless steel cylinder with a 2" NTP mal tread coonection.
I´ll attach 2 sightglasses of this kind directly into the shell approx 5 cm height from the bottom. Do you have any formulas to use in order to calculate the max strain/stress in the material due to the addition of these sightglasses?

best regards
Andy

 

[dramarc]

the sightglasses will be connected on "opposite" sides of the cylinder in order to be able to look through the reactor.

 

[prex]

If you mean that the shell is 35 mm thick, then I see no concern with opening reinforcement, just drill a hole and tap it. As I assume the bolt holes for the caps are also drilled into that thickness, you only need to be sufficiently far apart from those holes: would suggest the solid material between the closest points of the finished holes to be at least equal to the minimum required shell thickness.

prex

http://www.xcalcs.com

Online tools for structural design

 

[hacksaw]

Great way to design a pressure vessle, who needs outsourcing!