ASME VII Div.1 Noncircular cross section chamber with openings

[LearningENG-]

Hi all,

Edited: The chamber is roughly 24"X24"X60". It's for insulation of a cryostat but the temperature of the chamber itself is at room temperature. The potential material could be Aluminum 6061 T6 or SS ANSI 304. The reason why I consider Aluminum is that it might be cheaper than a stainless chamber (about which I'm not sure, it would be helpful if anyone can give insight on this)?

Basically I'm trying to design a vacuum chamber of rectangular cross section with flange port on the side panels and end plates. Detailed background is in thread794-463687.

13-4(j) outlined some rules for openings but I think they apply only to internal loading chambers, as they only mentions bending/membrane stress. At this point I'm mostly positive that I would not be able find a published code for my intended design, so I'm thinking about the following procedures to take:

1. calculate a minimum wall thickness as per 13-14, and apply an extra safety factor of 4.0;
2. put openings and reinforcements as per 13-4(j);
3. check the design with FEM software (I'm using SolidWorks) for strength (linear) and instability (non-linear).

Do you think this is a good way to go?

Another small question at last: I'm not quite sure about the meaning of "openings inherent in the construction" in 13-4(j)(1), line 2. Could anyone please clarify?

Any suggestion would be very much appreciated!
XS

 

[LearningENG-]

I know no one has replied it yet but I just want to add that looks like I misunderstood the concept of reinforcement and stiffeners.. I will add also stiffeners around the chamber. Again, I think I won't be able to find any specific codes to my case but is there anything that I can at lease refer to when adding stiffeners?

 

[SnTMan]

LearningENG-, my reading of 13-4(j) does not prohibit openings designed per Code rules, i.e. UG-39. It might be useful to then check such an opening by other means, perhaps FEA. The language about "inherent in the construction" is as far as I know is not formally defined. It applies to openings for which considerations of reinforcement can be neglected due to their small size per, UG-36(c)(3).

In Code-speak, reinforcement for an opening can consist of excess material thickness parts of the connection or opening as well as separate added elements such as pads.

Regarding your proposal 1-3, 1) A safety factor of 4 seems to me uneeded. Doubtless you know Code allowables already posses a design margin of perhaps 3 1/2 (it depends). 2), Sure & 3) Sure if you have sufficient confidence in the software and your analytical abilities.

I'd make the observations that it would help to know if you intend to stamp your unit, general congifuration, if we are talking about a mouse or an elephant, what materials you had in mind, what temperatures it might operate at (I assume quite low), etc, etc.

That's about all I have time for right now.

Regards,

Mike



The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[LearningENG-]

Hi SnTMan,

Thanks a lot for your comments.

Yes I guess another safety factor of 4 does sounds too unnecessary, I might start with what I get from the code plus an extra 1/3 onto the minimal thickness.

I apologize for not having put more design details. I have edited my post so people will have a clearer idea of what I'm trying to design here.

I don't need to stamp the design, it's more of an equipment for a specific physics experiment.The chamber is roughly 24"X24"X60". It's for insulation of a cryostat but the temperature of the chamber itself is at room temperature (plus for Al or SS the mechanical property doesn't really change much even at cryogenic temperatures). The potential material could be Aluminum 6061 T6 or SS ANSI 304. The reason why I consider Aluminum is that it might be cheaper than a stainless chamber, but like I mentioned in the post I could very likely be wrong.

Regards,
XS

 

[SnTMan]

LearningENG-, I've run the Apx 13 calcualtions (details of which I will decline to share) on the vessel dimensions as given. Disregarding closures, openings and other possibly important details, the vessel itself does not appear to be especially difficult.

Much to my surprise, a common plate thickness seems to work for carbon steel, 304 stainless and 6061 aluminum. I'd not have guessed, although I don't have a lot of confidence in the aluminum results due to as-welded properties.

I don't really have the expertise to advise you much further on this, except to say you should consider if cyclic service might be applicable.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[LearningENG-]

Hi SnTMan,

Thank you for your time to run the calculations. Yes I agree this is a not difficult case at all - just my first chamber design thus all the beginner's questions..

Also I forgot to mention that the chamber is split into two equal sized halves by the plane along the 60" direction, which is to be sealed with oring, for easy access to the chamber - not sure if you would have any insight on that but I don't see any thing particularly problematic regarding the design of the flange connection.

Finally if you don't mind me asking, what are the common plate thickness you are referring to, for these 3 materials respectively?

Regards,
XS

 

[SnTMan]

LearningENG-, splitting the chamber as you say sort of changes the problem from straight-forward Apx 13.

OK, using non-current Code rules I'm getting 1/2" for the side plates, each material. Do not rely on this.

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[LearningENG-]

Hi SnTMan,

Thanks for the reply. I have one noob question though, when applying external pressure, do I set the sign of Pe as negative(-)? I supposed so because I think in the case of an externally loaded vessel, the membrane stress should be compression and thus negative sign. Please let me know if I'm right.

Regards,
XS

 

[SnTMan]

LearningENG-, I'd agree it seems membrane stress would be negative. However:

1) Code language in Apx 13 does not state Pe is a negative value. This is consistant with external pressure calculations eleswhere in the Code.

2) The software I used for this exercise requires the external pressure to be a positive value. Else it does not run the unique calculations for external pressure, i.e. those detailed in 13-14. It is a little unsettling (at least to me), but for the membrane, bending and combined stresses of 13-7, signs seem to be the same for both internal and external pressure cases. However, this seems consistant with language in 13-14(a).

Therefore I don't quite have a definitive answer. Draw your own conlusions

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[LearningENG-]

Hi SnTMan,

Thanks for your prompt reply. Actually it starts to make sense combining what you said and the Code:

1)Yes I couldn't find any statement of Pe being negative in the Code, but after a close look at the equations in 13-7(a), the pressure P is always a value you can extract from all the terms of the equation. Since we are comparing the absolute value to SE, adding or not the negative sign does not make a difference.

2)I guess the reason why the software doesn't let you enter a negative Pe is because in equation 13-14(b)(1), if you have a negative Pe then both SmA and SmB would be negative, leading to the left of this equation always smaller than 1.0

However it actually surprised me to see how close the minimal thickness are for Stainless steel and Aluminum. In my calculation I also get 0.4" and 0.55" for SS and Al. I was originally expecting more differences...

Regards,
XS

 

[SnTMan]

Yeh, me too, especially the aluminum. I only checked fractions by eighths, 1/4", 3/8", 1/2". Stuff you can buy

The problem with sloppy work is that the supply FAR EXCEEDS the demand