Layered Vessel Built in 1959 to Corps of Engineers Spec

[gvessel09]

I have a layered vessel from 1959 built to the Corps of Engineers Spec. DA-23-086 ENG 59-4 Volume II or III. I have to prove the MAWP of this vessel and evaluate areas of corrosion. All I have to go by is todays ASME Sec. VIII code. Using current calculations I cannot prove the design MAWP for this vessel. My numbers are coming up a lot less than the original design. All I have to go by is a Manufacturer's drawing that is hard to read (ATTACHED). The areas of corrosion are on the layered shell. I am not sure how to address this on a layered vessel.

My vessel was originally stamped for 2800 psi and my calculations from Sec VIII are coming up just over 1650 psi. I have attached the drawing. I need as much pressure as I can get from this vessel. What can I do to get the pressure higher?

 

[stanweld]

Per the NBIC you may not use today's design allowable stresses in your calcs if the vessel were designed to the 1959 Edition of ASME VIII. However, it may be that the NBIC does not apply, assuming the vessel is not under any State Juridiction (still under US Govt.).

I would be most concerned with evaluating the integrity of the vessel through weld examinations. Fortuneately the operating temperature conditions appear somewhat benign. I assume that corrosion wastage is on the ID surface only or has it extended into other layers?

 

[gvessel09]

I am using the material strengths from the original drawing, not today's code. But, I am using today's formula's to apply the values. That is all I have to go by.

The corrosion is on the OD surface and is not very substantial, but I am in the process of collecting NDE measurements on it right now. It seems to be only half way thru the outside layer. I do not have a corrosion allowance to work with and without good MAWP calculations I do not have a Tmin to work with either for evaluation.

 

[rneill]

There was another thread recently talking about where to get old copies of the ASME Code. You may not be able to find the 1959 Corp of Engineers spec but you could track down a copy of the 1959 ASME Code.

Actually, if you contact the Corp of Engineers, you might be able to get your hands on the original spec?

 

[stanweld]

I would be very reluctant to increase the operating pressure based on the use of the current ASME VIII 3.5:1 or former 4:1 design safety factors on the minimum specified tensile strength of the materials or the designs used originally by the Corp. If a crack is present at one of the welds, failure is quite possible at the increased pressure. Catastrophic failures in vessels have occurred more than once when operating pressures were increased. Have you considered the 3:1 VIII. Div.2 saftey factors for comparison?

 

[gvessel09]

Just to be clear. I am not trying to increase the pressure above original design. I am trying to prove original design so I can operate near the original MAWP. My in house regulations require that the orignal calculations be proved if a copy of the originals are not available.

I have researched VIII Div 2, but am not that familiar with the code.

 

[rneill]

ASME Section VIII Division 1 is called "Design by Rules" and Division 2 is called "Design by Analysis". I'm not particularly familiar with Division 2 but in general rather than applying simple rules and formula's as presented in Division 1, it requires much more rigorous analysis but provides the benefit of being less conservative. I believe it also requires more rigorous QA/QC than Division 1 so I'd be reluctant to apply Division 2 analysis to a vessel unless you knew for certain it had been originally constructed to Division 2.

 

[unclesyd]

Your Vessel brought back memories of long ago. As you probably know your vessel is a composite, multi-layer, by A.O. Smith built for the COE. If it was like our it was also designed by A.O. Smith. All the records for these vessels that were held by AOS have been destroyed and according to NASA, Nooter also had copies. NASA has re-certified some of these vessels. As you probably know the Nooter of old is no longer around.
As posted above these are not ASME vessels due to the multi-layers. Even though these vessels are not ASME the NB still might have some documents as our insurance required this.

You might contact The Linda Hall Library for any available information wither through A.O. Smith or The COE numbers.

http://www.lindahall.org/services/document_delivery/

A quick check of the Internet revealed that several design books of that era has some references to Multi-Layer design.
There are several references to ASME publications.

There also was a lot of work on this type construction in Japan.

 

[stanweld]

Multi-hull vessels were also manufactured to ASME VIII. The last multi-hull hydrocracker, manufactured by CBI in the US, installed at the Bayway refinery blew its top head off in about 1968. There were also many multi-hull ammonia reactors made with Carbon-1/2% Mo alloy steels weld cracking was a common problem. These vessels all operated at high temps.

Section VIII, Div.2 does have design requirements for multi-layered vessels.

 

[jamesl]

I don't think Div.2 will help you in this case. There was no Div.2 code until 1968.

No matter which code is used, the formula to calculate thickness or get pressure rating will not make such a difference as you described.

The drawings are not legible. If you read the thickness and diameter correctly from these drawings, and you cannot get the correct pressure rating, the only thing left is the allowable stress used in the calc.

 

[MJCronin]

Some information on analysis of these 50 year old multilayer vessels may be found here:

Structural Analysis and Design of Process Equipment
by: Jawad & Farr
ISBN: 0471624713

These government-owned high pressure vessels made bt AO Smith have been arround for a long time. I worked with a couple of them in an Air-Force Test facility in Tennessee about 15 years ago (are these the same ones !!??)

You might try to contact other owners of these vessels to see what they have done about requalification..... and reuse.

Does LANL have any that they are using ?

They have an extensive (and IMHO superior) mechanical integrity program.

MJC

I was involved

 

[metengr]

gvessel09;
Here is my 2 cents worth of advice in the name of safety and from the perspective of being a current member of an unnamed State Board of Boiler rules where this would be similar to a State Special. If I were you, I would hire an experienced pressure vessel engineering firm that has a licensed PE in pressure vessel design.

The scope of work would be two fold - first I would use ASME B&PV Code, Section VIII, Div 1, Part ULW to validate your current conclusion based on this vessel being designed today except using original material strength allowable stress values. I really don't see how any insurer or operating organization would accept operating this vessel at a higher pressure if the design by rules doesn't support it. This is an incredible amount of risk

Design by analysis could be used as the second approach and this is where an experienced, licensed pressure vessel engineer is required for a proper design review. Included with this approach would be use of API 579 FFS to conduct a proper condition assessment of the existing condition.

The above is what I would expect as a minimum if this were presented as a State Special for an Operating Certificate.

 

[metengr]

By the way, higher pressure in my post is simply going back to original design pressure of this vessel, not the 1650 psi you reached.

 

[rmw]

One thing you should have going for you is that these vessels were designed and calculations done using slide rules so there ought to be generous factors of safety built in, or so it was back when that is the way I worked. We always made it a little more stout than the slip stick said just in case....

rmw

 

[racookpe1978]

Two points:

Can you get any ultrasonic readings through the vessel walls to get "real" thicknesses (of what is left of the "real" steel. If the drawing isn't legible, you CAN'T use the drawing as a source of calculation data for an insurer.

If the ""layered" vessel makes nonsense of the UT reflections, try a wide number of UT shots - to a pattern of reflections vs measureable thickness (like near a manway or opening.) Then take a series of xrays through the most corroded outside and inside points. That should help show if unseen corrosion is present in the "seam" between the two layers.

It would be nice if you could find a nice empty square mile of farmland, and pressure it up to the design (er, desired) MAWP. If it holds, then you have an argument for trying it with real chemicals.

If it fails, you have fresh cow parts for dinner. 8<)