Determine Vessel wall thickness and external and internal pressure resistance 2

[erbru]

I am working for a client who has an stainless steel low pressure vertical water tank for which there is no technical data or drawings
It has a useable volume of 33.8 m3 (from nameplate) and has a diameter of approx 3m and height 5m

The nameplate also says:

S 1 bar
PE 1.5 bar (this is France where PE = pression d'epreuve = test pressure)
No mention of whether bar is relative or gauge pressure (we presume gauge) nor design temperature

I guess S is the MAWP but not sure.

Can I use this information (eg assume 1.5 bar test pressure is 0.5 barg) to estimate by back-calculation the possible wall thickness and at what pressure to rate the relief device?
I need also to know the resistance to vacuum.

Or do I have to actually measure the wall thickness (eg by ultrasound) which is a difficult act (empty tank , remove insulation, mounting on a scaffold, etc) to be sure?

The vessel is 21 years old and has undergone temperature cycling (heated once a week from 25 to 85 °C) so may have some fatigue. Not sure how to factor this into the calculation. It needs to last a couple of years more before they replace it.

 

[LittleInch]

This seems to come up quite a lot.

your client does not actually have "a stainless steel low pressure vertical water tank for which there is no technical data or drawings"

what he actually has is a lump of scrap metal which could be used for atmospheric storage.

1 bar / 15 psi is right on most limits of being a pressure vessel at which point all sorts of nasty / pesky regulations and laws start coming into effect.

you need to know a lot of things including material strength, joint type, manufacturing inspection, what was the actual testing pressure etc etc before you can do any back calculation.

I think the effort involved in doing anything which can actually be certified and get you out of any difficulty if / when this thing fails is more than the replacement cost.

Any actual verifiable operating history?
What is the relief system now?
Any in service inspections / pressure tests / wall thickness checks
what shape is it / made from what grade Stl Stl?


If you get good answers to the above then re-test it at 1.5 barg, then you just might get away with it.

As for vacuum - you would need to do some design calcs, but my guess is very small vacuum condition from atmospheric, say 50 mbar.

If you get no answers to the above then walk away.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[JStephen]

Pressure and vacuum ratings, allowable overpressure for venting, set points, procedures for evaluating a change of service, are items that should be covered in the design code used, and will vary depending on that design code. So the first question, what design code was it built to (if any)?

Can you identify the manufacturer, and if so, do they have any information on it?

 

[erbru]

The manufacturer has gone out of business and is uncontactable.

This was built in 1995 in France. They have a design code called CODAP for Vessels but no idea if this was designed to it.

It's pretty frustrating, being asked to set a relief valve for both the vacuum and overpressure scenarios without any data.

I was told by a vessel manufacturer that if I could get the wall thickness one could calculate both external and internal pressure resistance. But searching online the external pressure resistance calculations are complicated and done by Finite Element Anlaysis modelling. The tank is going to be replaced in a few years time by a new one.

I can't just say to the customer don't bother (we are being paid to figure out how to protect his vessel), but I'm tempted to protect it at -10 mbar and +100 mbar and be done with it. But on what basis can I justify this choice of parameters. Apparently some wine storage vessels hold only between -1mbar and +30 mbar.

I could (with some effort on the part of the client) get a set of wall thickness measurements made by a notified body. But I won't do this unless I'm sure it will give us a reasonable degree of certitude on the pressure resistance.

 

[LittleInch]

erbru,

You have some interesting problems...

There are some puzzling things here:
You say there is a nameplate, but it doesn't seem to supply the information you would expect - i.e. code designed to, units of pressure or temperature etc
A design pressure of 1 bara and a test pressure of 1.5 bara makes no sense. If it was only at atmospheric design pressure it would not be tested to 0.5 barg.
You don't give any history of this tank, its operation, its current protection system, nothing.
Did your client buy it as scrap or something??
Or has bought the facility without any documentation at all?? Sounds a bit fishy to me.
So the most likely explanation is that this tank was originally designed to 1 barg and tested to 1.5 barg.

Vessel design is a bit of an art and needs proper analysis. Your only real course of action I think is to take as many dimensions as you can get, send the data to a vessel designer and as them to calculate what the wall thickness should be to some recognised tank / pressure vessel design code in whatever country you're in (France?) using some low strength stainless steel and then see if the thing you have matches up to what you need.

All this costs money, so I would write report listing all the data you have, what the issues are ( no design code, no documentation etc) and then list some options with cost and schedule.
1) Use it as an atmospheric tank +/- 10mbar
2) Use it as a low pressure tank (API 620 say) with your -10 / + 100 (low risk and all you need to do is test to 150 or 200 mbar)
3) Full design cod check and multiple wall thickness check and material strength test to prove it to the 1 barg / - whatever the design code check says - cost X,000, schedule 4 months

Job done




Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[jte]

LittleInch-

Good post. I disagree with your recommendation #1, however, unless we modify it a bit. erbu mentioned that it has been thermally cycled weekly for decades(?). No mention of any inspection besides that it would be, you know, extra effort to actually check wall thickness. The UT would be a good start. But given the thermal cycling a PT would be wise prior to any continued use, pressurized or not.

erbu-

No offense, but only a process engineer would look at the pressures you mention and question whether they are absolute or gage. Any vessel designer would look at 1 bar design and 1.5 bar test and immediately conclude that they are "normal" units - meaning gage. We don't typically use absolute pressures in the design of vessels except when we have to translate to chemical engineers. Seriously, get somebody who knows how to evaluate this lump of steel. Mechanical design is not within your area of competency (just as I do not pretend to be competent with process engineering), and people are at risk of getting hurt. This forum is not a substitute for a competent paid engineer taking a close look at the situation and evaluating it.

 

[erbru]

It was bought second hand and previously used for milk storage as far as we know. There is an unused heating element inside the vessel. It's now used to store water at atmospheric pressure and temperature. It is stainless steel and currently protected at 100 mbarg by a (very probably undersized) relief valve and not protected at all for vacuum.

I think you are right about the bar being barg. 'g' doesn't mean anything in french as the word for gauge pressure is 'pression relative'. So I think they often write just bar. The problem is that it's an assumption.

There are units of measure on the name plate (bar, litres) but no indication of vessel design code or temperature of service. But back in 1995 the rules for low pressure vessels maybe did not require this information in France I think.

I was going to get a notified body involved to help quantify matters but all that costs money.
Your plan sounds like a good one.

 

[erbru]

No offense taken. I should have known really that pressures are always in gauge. It's just that I've been doing ideal gas law calculations with absolute units today (and I obviously don't have the pressure vessel engineer's reflex).
I am well aware this is dangerous and I will make sure the RV we put on will be properly sized, I just needed to make sure that we didn't risk being on the wrong side of the design or MAWP pressure.
I had also been told by a relief valve supplier that wine vessels are sometimes protected at 30 mbar and this is an ex-dairy vessel so I was concerned it could be similar. Also I didn't understand the rationale of 1 barg rather than atmospheric. But could be down to the heating of milk in it...

Still, should we go ahead with all the analysis or just protect it at 100 mbarg as currently. It has been thermally cycled for decades. All these tests, including PT are expensive and not necessary if we just protect it well within the suspected range. (eg -10 mbar to + 100 mbar)

 

[jte]

All these tests, including PT are expensive and not necessary if we just protect it well within the suspected range. (eg -10 mbar to + 100 mbar)

So... are you suggesting that failure cannot occur if it has existing cracks which have initiated and propagated to an unknown extent under previous thermal and hydrostatic loadings if going forward you continue to cycle the liquid level and have eliminated the thermal cycling?
Allow me to try an analogy: If the tank previously stored milk, and we emptied it out and didn't look in to it or attempt to clean it, and now will use it to store deionized water - or better still WFI (water for [human] injection) used in the pharmaceutical industry, would we think that the cleaning process and subsequent inspection to verify sanitization is too expensive and unnecessary since milk is ok for humans to drink?

It comes down to risk. If failure has no direct consequences (perhaps this tank is out in the middle of a field somewhere, nothing and nobody around it) and no indirect consequences (no economic loss to production or otherwise), then I'm probably ok with it. I might use this tank to store washing water at my camp site. But I would not use it in a circumstance where the failure of the tank would cause harm (is it in an attic above occupied spaces?) or if losing the tank would necessitate the shutdown of production and related lost profit opportunities. Finally, will the failure of this tank impact your personal liberty or finances?

 

[carthago]

In France , for pressure vessels the CODAP or the EN is in use
For storage tanks ( up to 60 mbar ) the CODRES ( Code de reservoir cylindriques VERTICAU et atmospheriques, a bit similar to EN 14015 is in use.
Besides you have to satisfy the several national Laws called Decrets which stipulate WInd, Seismic, inspection, Approval.
Your Service des Mines will advise you ..in this regard.

Please state if your tank is vertical or horizontal on 2 saddles ( berceaux) ? my opinion and watching the pressure the Codap will do.
From your pressure , this would be a thin wall item.

If it is a non cylindrical vessel it should be designed according Good practice. there is an UL standard for it ...etc


take care and thanks