21 replies to this topic

[latexman]

"My thinking is that, as long as the superimposed backpressure does not increase the actual lifting pressure of the valve above the allowed accumulation (assumed to be 10 percent), and that the total backpressure when the valve is relieving is less than 10 percent of the setpoint, then I am ok."

No, you are not okay. The PSV on the 350 and 357 psig vessels will open at a pressure > MAWP, which is a violation of Code. The other two vessels will be fine.

Compensation for superimposed back pressure which is CONSTANT may be provided by reducing the set pressure of a conventional relief valve. A balanced bellows is recommended when built-up back pressure is expected to exceed 10% of set pressure.

When superimposed back pressure is VARIABLE a balanced bellows is preferred.

Discussion on a pilot-operated relief valve was omitted.

You'd need to back the setpressure down 10 psig on the PSVs on the 350 and 357 psig vessels. If built-up back pressure is 10% or less, conventional PSVs would be fine.

[Nirav]

hello,

To me, for any kind of combination in the discharge, you need to analyze mainly 3 aspects.

[1] What's the maximum load in common relief? This is based on the chance(s) of more than one PSV discharging at the same time or single largest discharge.

[2] What's the maximum back pressure on PSV at the time of relief? (superimposed + built up)

Notice that the first point is necessary because, it may vary the "built up" pressure if you consider more than one PSV discharging at the same time. (When PSV under consideration is part of simultaneous discharge).

[3] The calculated back pressure should not go above recommended limits of type of PSV. e.g. 10% of set pressure for 'conventional'.

If above points are met, I don't think you need to bother anything else as far as discharge pressure profile is concerned. Does anyone have more point(s)? In case, if i'm missing..

now, I do NOT agree with following statement from Latexman

No, you are not okay. The PSV on the 350 and 357 psig vessels will open at a pressure > MAWP, which is a violation of Code.

It is not violation of code. Refer to the attached file. It is from API-RP521 and based on ASME Code Sec.VIII.
Set pressure of PSV CAN be equal to MAWP. It can NOT be more than MAWP as per Code.
PSV can operate > MAWP during its discharge because of accumulations. This is allowed within limits as shown in attached file.

Thank you.

[pleckner]

Nirav,

You have misinterpreted what Latexman said. He said the PSV cannot open at a pressure greater than MAWP by Code and this is correct. He did not say it cannot operate at a pressure greater than MAWP.

I would add to Latexman's response that theoretically, there are times a built up back pressure greater than 10% of PSV set pressure for a conventional PSV is acceptable. A conventional PSV is nothing more than a differential relief device being held opened by a spring. As the pressure against that spring increases, it will tend to force the spring to close. The only thing to keep it opened would be the opposing force of the pressure in the vessel. So, if the over pressure were to be 21% (fire), then yes, the built up back pressue will also be allowed to increase to 21%. This is acceptable practice as stated in API RP520.

Can the conventional PSV still function if this "rule" is violated? Yes to a point. The valve would then have to be derated by a factor. Unfortunately, as the built up back pressure as a percentage of set pressure exceeds the over pressure, a conventional style PSV capacity falls off dramatically. So the only way to stay out of potential trouble would be to head the recommendation to use only balanced bellows PSVs in these situations. This is also why I would never specify a conventional PSV that is tied into a header system or manifolded with other relief devices. There is nothing wrong with going back to the vendor and requesting balanced bellows PSVs.

And just to second Latexman, you can't exceed the MAWP of the vessel, no matter what, period and end-of-story. So if the superimposed back pressure will be variable and the PSV is set at MAWP, then I would not even say a balanced bellows PSV is preferred, I would say it is mandatory.

[latexman]

pleckner,

I had to study your second paragraph for a while because I have the 10% overpressure number engrained into my brain due to my company having standardized evaluating inlet and outlet losses at the rated flow at 10% overpressure, even in fire cases. After re-reading your second paragraph 2 or 3 times, it made more sense to me why they did this. Thanks.

[Nirav]

Phil / Latexman,

Thank you for that clarification on my interpretation. I just misunderstood little on "open" & "operate" part of PSV in Latexman's first response.

In case of conventional valve, due to variation in superimposed back pressure, it MAY "open" at higher pressure than its initial set pressure. Which means set pressure is raised by itself by some amount during actual practice which is definitely NOT permitted. Therefore, in such cases, balanced bellow is must when Set Pr.=MAWP.


Thank you again,

[pleckner]

There are two issues. The first is the set pressure and the second is the affect of built up back pressure.

Taking the first issue, we alll agree that the set pressure shall not be any higher than the MAWP of the vessel and this is clearly spelled out in ASME. We also agree that for a conventional PSV, if the superimposed back pressure is greater than what the spring was designed for (probably atmospheric in this case), then the valve won't open at set pressure but at set pressure + superimposed back pressure. And I believe your argument Mvancleave is that the latter is OK as long the allowable over pressure within the vessel is never exceeded before, during and after the valve opens.

After going through API RP520 and 521 again for the umpteenth time, I almost came to agree with you until I went through ASME Section VIII, Div 1 again for the umpteenth time. This is what you missed (by the way, I think you may have an older edition of ASME because my copy is 2006 and the wording you quote is not quite the same).

UG-134 PRESSURE SETTING OF PRESSURE RELIEF DEVICES

(a) When a single pressure relief device is used, the set pressure (60) marked on the device shall not exceed
the maximum allowable working pressure of the vessel. ......

The (60) above is really a superscript that refers to a note. And this note reads:

(60) The set pressure is the value of increasing inlet static pressure at
which a pressure relief device displays one of the operational characteristics
as defined by opening pressure, popping pressure, start-to-leak
pressure, burst pressure, or breaking pressure. (The applicable operating
characteristic for a specific device design is specified by the device
Manufacturer.)

Obviously, this note pertains to many types of relief devices, not just to PSVs.

What we see here is that ASME defines what they mean by "set pressure" and the intent is to define the "set pressure" as an opening pressure. It does not appear that the intent of ASME is to separate the concept of set pressure from opening pressure as you seem to be interpreting the readings. Therefore, if you accept this defiition for set pressure, then you must accept the fact that the PSV on the vessel shall not open or begin to open at a pressure greater than MAWP. And, if you manifold the discharges of your conventional PSVs, you will most certainly have the potential to violate the Code.

I won't argue that API RP520 seems to allow this condition to exist but in my opinion, they are allowing you to violate the spirit of ASME if not the letter and this is unaccpetable. And remember, API RP520 is NOT CODE, just a recommended practice. And another thing, the definitions given for "Accumulation" and "Over Pressure" refer to dynamic states, i.e. the PSV is in relief, not static where it is waiting to open. So again, I don't think that if you want to stay within the spirit of ASME, you can adapt the argument of allowing accumulaiton or over pressure to build up before the valve actually starts to relieve.

One last thing on this particular issue, you quote a section from the Crosby Engineeing Handbook. This was lifted from API RP520 and you really should go to the source standard and not rely on a vendor's hand book.

3.3.2 Effects of Superimposed Back Pressure on Pressure Relief Valve Opening

3.3.2.2 .... Balanced spring-loaded or pilot-operated pressure relief valves should
be considered if the superimposed back pressure is variable.
However, if the amount of variable superimposed back pressure
is small, a conventional valve could be used provided:
a. The set pressure has been compensated for any superimposed
back pressure normally present; and
b. The maximum pressure during relief does not exceed the
Code-allowed limits for accumulation in the equipment being
protected.

API uses the word "small" and Crosby seems to use the word "low". I'll go with API and perhaps it has to do with the version they lifted the text from. But I ask you, who defines "small" or "low"? API certainly doesn't! Crosby certainly doesn't! And I don't see it in ASME.

Bottom line, no, you should not design your system the way you intend. Perhaps you can decouple the vessels so that you won't violate the Code for the vessels in question and only manifold the vessels that won't be affected by this.

I don't have time right now to go into issue two but I think I addressed it in my previous post.

[latexman]

Phil,

Your explanation agrees with my experience and my companies training and practices. Set pressure is the opening pressure with no tailpipe to atmosphere. Opening pressure is when the device opens installed on the vessel and in the worst credible scenario relative to raising the set pressure.

[benoyjohn]

Phil,

Your explanation agrees with my experience and my companies training and practices. Set pressure is the opening pressure with no tailpipe to atmosphere. Opening pressure is when the device opens installed on the vessel and in the worst credible scenario relative to raising the set pressure.

Mike,

I am wondering why you were limiting the manifold backpressure to 7 psig.
You can consider upto 20 psig i.e 10% of (341+14.7) psia and still use a conventional valve?

Regards
Benoy

[pleckner]

Latexman, I'm not sure I fully understand what you mean in your last sentence, "Opening pressure is when the device opens installed on the vessel and in the worst credible scenario relative to raising the set pressure."

In their definiton of set / opening pressure, it doesn't appear that ASME differentiates between a PSV with a tail pipe or not or, discharging to atmosphere or to a header or installed on the user or on a test stand. Also, I wouldn't restrict the definition to just the worst credible scenario (if I am interpreting your sentence correctly). The PSV needs to open at set pressure even if the cause is a scenario with a lower relieving rate. It will just mean the valve will probably chatter for a time because it will be too big for that particular relief scenario.

Benoy, always work in gage pressure when talking about relief set pressures, superimposed and built up back pressures and percentages of set pressure.

And lastely, I think everyone agrees that in general, if you can keep the variable back pressure (define it anyway you want) to within 10% (gage pressure) of the PSV set pressure, then you can use a conventional valve in the system. This addresses issue #2 of this thread.

[pleckner]

Mike,

You said, "I still think that the BPVC is referring to the stamped set pressure, determined by the valve manufacturer, rather than the installed lifting pressure." First, the valve manufacturer does not set the PSV set pressure, you do. You tell them what you want and that is what they must design to. They put this value onto the stamp, nothing more, and nothing less. And I will go back to what I've said based on ASME definition, the PSV set pressure and the lifting pressure are one and the same. I can't see how one can interpret this any other way. Definition is definition. And, there is nothing that you quoted from PTC 25 that is any different from what ASME states.

Going further, you said, "I do not think that the spirit or letter of the code requires that the actual lifting pressure of the installed relief valve be less than or equal to MAWP, as evidenced by the fact that an allowable tolerance for relief valve set pressures is stated in UG-126(d)." Allowable tolerances are there to take into account that anything built by man, woman or thing is not perfect and therefore cannot be expected to act perfectly. The tolerances are granted by ASME to help manufacturing and to recognize there will be some slop in the function of the relief valve. You are never to use these tolerances within the basis for your design. They are to be considered to be "unmentioned", kept in your back pocket, etc.. Why do you think ASME specifies the testing pressures that they do (130% of design pressure, or MAWP)?

When we talk about "spirit" of the Code rather than the 'letter" of the Code, we now enter the world of interpretation. Then I would say someone needs to write to the ASME committee and ask them how this whole concept should be interpreted. Let them make the decision, not us poor souls trying to do what's right. BUT, they did put something into the Code that is quite explicit and that is the definition I keep going back to. That to me turns this from "spirit" (interpretation) to "letter".

And in your very last post, the answer to your question is yes, we are talking about built up back pressure; I should not have been so general in my response.

[latexman]

pleckner,

Set pressure is the pressure the PSV opens on the test bench (no tailpipe and discharge to atmosphere).

The pressure at a PSV discharge flange before the device opens is called superimposed back pressure. An evaluation of the PSV being designed and all other relief devices discharging into the same discharge header must be done to determine the worst credible superimposed back pressure (WCSBP).

Opening pressure = set pressure + WCSBP

That's what I meant.

[pleckner]

Latexman, thanks for clarifying that for me.

Mike, you are welcome and I really enjoyed being part of this discussion! Many more to come I hope.

[pleckner]

Mike,

A whole lot of words that basically mean the same thing.

Ref: Your 5:57 P.M. post.

The only thing 3.3.2.1 talks about is the spring setting for a conventional style PSV should be compensated for the superimposed back pressure. That is exactly what the vendor does to ensure your PSV will open at the designated set pressure. This is also why the PSV will not open at the designated set pressure if the superimposed back pressure is anything other than what you tell the vendor it will be.

3.2.2.2 goes into why you would want to use a bellows or pilot PSV instead of a conventional PSV for this reason. The statement about superimposed back pressure being "small" is one of my problems. Please define "small"? Also, condition 'a' still wants you to compensate the set pressure for superimposed back pressure, doesn't it?

Then there is paragraph 3.2.2.3. where they contradict their own statements / requirements in the previous paragraphs! Based on this, I might as well just design the PSV to open at MAWP + accumulation.

Ref: Your 7:50 P.M. post.

First, these are interpretations from the 1982 edition of the Codes. I don't have these on hand and even if I did, they would no longer be valid. Things are removed, added and changed over the years so I will only go with what I see now. There is no UG-135(g) and UG-135(f) is somewhat different. Even still, it only refers to sizing the discharge piping so that the PSV is not adversely affected. I take that to mean it doesn't affect its opening or ability to protect the vessel.

I have to remind you that API RP520/521 is not Code, just recommended practices (guides) and they are not perfect by any sense of the imagination. So I stick with my original conclusion based on the definition of set pressure given by ASME Section VIII, Division 1, which is Code. And by the way, here is the definition of set pressure from ASME B31.3, CODE for process piping, "Set pressure is the pressure at which the device begins to relieve, e.g., lift pressure of a spring-actuated relief valve, bursting pressure of a rupture disk, or breaking pressure of a breaking pin device." Two different CODE documents from the same organization, but B31.3 says it even better. And, B31.3 refers back to ASME Section VIII, Div 1 for PSVs.

My best suggestion would be to write the ASME committee explaining your situation in detail and be very speicific. Don't use generalities. Let them define/redefine set pressure and see if what you want to do is legal? You may also call your insurance company and ask them. You could call your local municipality officials and ask them, afterall, they are really the final word. Maybe we've all been interpreting "set pressure" wrong all these years. Hey, we're human too, just like the people who write the API RPs. (One problem with the ASME committee, you probably wont' hear back from them until almost retirement; OK, I exaggerate but you can get my point.)

Mike, by your own words, "While I still think there is (arguably) a tiny bit of wiggle room in the letter of the code (the words "marked on the device" in UG-125), I will concede that it is quite small, and engineers shouln't be in the business of taking advantage of possible wiggle room." You're trying real hard to "wiggle". I give you an A+ for trying.

[pleckner]

Mike,

First, happy holidays to you and your family as well (and to all of our subscribers).

If possible, can you post the letter you are writing to the committee? At the very least, please post their response when you get it. Remember, you want to ask them if the set pressure is the same as the required opening pressure and if that shall not be greater than MAWP...or something along those lines. I guess we'll all learn together and that is exactly what this forum is all about.