Set pressure of safety valve 1

[ekschwab]

On a non-fired boiler steam line... has anyone ever seen the design pressure of B31.1 piping being LOWER than its relief valve setting?

There are multiple relief valves in the line, each at an incremental pressure than the previous. The first valve will lift at the design pressure of the pipe. Subsequent valves that lift to accommodate the capacity will be at higher pressures (above the design pressure of the piping).

I did not design the system. I am observing and questioning what is already being considered. It does not sound right to me and I do not see B31.1 clearly permitting this.

 

[whammett]

Never. That sounds very odd to me. However, your wording seems to be conflicting. In your first statement you say that the piping pressure specs are lower than the relief valve setting. (Worst case scenario, piping giving way before relief valve lifts). But in your second paragraph you say it lifts AT the piping design pressure.

Granted, I've never seen this either way.

"Scientists dream about doing great things. Engineers do them." -James Michener

 

[ekschwab]

Thanks whammett. Me too - never.

You are correct, it does sound conflicting the way I presented it. I am basically saying that once past the first valve lift, the piping pressure will continue to rise above the design pressure until the next set pressure is reached. At this point this is almost the same as having a single releif valve whose set pressure is higher than the piping design pressure. I don't see how one can "design this in" this way.

 

[BigInch]

Then B31.1 is different than B31.3, 4 and 8 in that it doesn't give any allowance over design pressure for what may be short term pressures that fall into the category of transient pressure loads, correct? I don't know the B31.1 code.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.co

 

[rneill]

If you look at Figure 15 of API 520 Part 1, it indicates that the maximum allowalbe set pressure for the first or for single PSV's is 100% of the MAWP. This figure indicates it is permissible for additional valves to be set at higher pressures. Secondary process PSV's can be set up to 105%, and PSV's for fire exposure can be set at 110%.

I know this is not the B31.1 Code itself and I don't know whether B31.1 allows the leniency that API 520 does in setting the set points. There is probably an interpretation on this out there somewhere if you have the time to look.

From a practical perspective, I would say that if you have multiple valves, you would not generally want them at the same set point since they will fight with each other. It is much better to use staggered set points so that the tendency for the valves to fight will be reduced.

I have designed one B31.3 system with three process relief valves where the first was at 95%, the second at 100% and the third at 105%. Note that I set the first below MAWP in order to provide a 5% margin between them and to ensure that the last still complied with API 520.

 

[BigInch]

B31.3 references BPV Section VIII, Division 1, for maximum relief settings. API 520 is not referenced.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[rneill]

Ahh, but B31.3 does permit temporary pressure excursions over the MAWP and with 2 PSV's at or below the MAWP, I am very confident that I will not exceed the permitted time (per Clause 302.2.4) at pressures in excess of 100% MAWP.

Given that ASME B31.3 allows excursions above the MAWP, I fall back to API 520 for guidance; although, API 520 is actually more conservative than B31.3 which would allow pressure excursions far above 105% of the MAWP.

In regards to Clause 322.6.3, it does say that "relief set pressure shall be in accordance with Section VIII Division 1, with the exceptions stated in alternatives 1 and 2 below."

Those exceptions allow set pressures to be higher than specified in Section VIII provided it is approved by the Owner and in accordance with Clause 302.2.4(f).

 

[ekschwab]

I understand that B31.3 permits temporary pressure excursions over the MAWP. And I believe B31.1 allows this too. But I am still having a problem with all but 1 of the 6 PSV's being set above the design pressure of the B31.1 piping.

 

[gmax137]

Do I understand the situation: the lowest set safety valve is set at the design pressure, and there are additional safety valves set at higher and higher set pressures?

That's the situation I have seen (many times) in the main steam piping in nuclear units (B31.1 power piping).

The Overpressure Protection Report requires that the peak pressure remain below 110% of the design pressure.

 

[rneill]

Clause 102.2.4 of B31.1 allows for short term occasional operation above the MAWP; up to 20% greater than the MAWP.

However, Clause 107.8.3 refers to ASME Section VIII Div 1 for relief devices and does not specifically indicate that exceptions are permitted based on the occasional variations allowed by Clause 102.2.4 (which is clearly specified in B31.3).

Based on this, it is not entirely clear that B31.1 intends to permit the set point of a relief device to be higher than MAWP based on Clause 102.2.4. However, if you were not ever permitted to set a set point higher than MAWP, I'm not sure how you could actually take advantage of the allowance for occasional operation?

Consequently, my interpretation is that you are permitted to provide set points above MAWP provided you are within the limits specified in Clause 102.2.4. That said, I think it is generally good practice NOT to routinely use these allowances in normal design situations. I would typically only consider these allowances when there was no other practical way to achieve a required design objective.

I did review all the B31.1 Interpretations and did not find anything that specifically addressed this situation.

 

[BigInch]

If you're doing B31.1 pipe design, wouldn't you in that case refer to BPV III only for the specifics of the design not given in B31.1 Otherwise its kinda' like you're building a BPV, no??? I doubt either code intended to cover parts of the other's scope, hence if PSV set pressures are given in B31.1 as MAWP + 20%, using lesser MAWPs of BPV III doesn't really make sense, don't you think?

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[sbnz]

Very interesting discussion. Thanks ekschaw for sharing your observation.

gmax137 - you said you have seen similar design (pipe design pressure lower than (at least one) safety valve in the line). Would you please share further which line (steam?) and in what type of plants have you seen that? Just for curiosity and educating ourselves.

As a piping design engineer, in my (limited) experience of direct/ indirect design of steam systems in various power stations in different countries I have never seen and would not have conceived this could be an acceptable practice. Not saying though this is conclusively wrong - may be my ignorance - but need strong reasons/ precedences to be convinced this could be an acceptable practice.

The pressure-temperature excursion allowance in B 31.1 is very commonly considered while designing for occassional loadings (seismic, wind, transient pressure/ temperature spikes etc) but I could not have imagined to use that to compromise the relief setting of the protection device in the line.
In fact, trying to recall - I have (probably) not come across any piping system which had multiple protection device set at different pressures (multiple safety valve or rupture disks for capacity are common, but are they "always" not at same setting for the same section of the pipeline?
Would follow the thread for more interesting opinions/ experience from others

cheers,

 

[rneill]

sbnz,

Since both B31.1 (Clause 107.8.3) and B31.3 (Clause 322.6.3) generally refer to ASME Section VIII Division 1 for specifics around relief devices (presumably to prevent reinventing the wheel), I thought I would actually look at Section VIII Div 1. Now it turns out that this particular issue is covered in UG-134 which is included in the B31.3 reference but is not included in the B31.1 reference to ASME VIII? Anyway, UG-134 says:

"When a single pressure relief device is used, the set pressure marked on the device shall not exceed the MAWP of the vessel. When the required capacity is provided in more than one pressure relief device, only one pressure relief device need be set at or below the MAWP, and the additional pressure relief devices may be set to open at higher pressures but in nor case at a pressure higher than 105% of the MAWP, except as provided below."

It turns out that this is consistent with API 520 (which upon review states that it's requirements are based on ASME Section VIII).

Apart from the specifics of the Codes, I believe it is considered poor practice to set multiple relief devices at the same set pressure since you can set up a dynamic where the pressure will bounce around quite severely and exacerbate the effect of system upsets.

One of the Major International Oil and Gas Operators that I worked for previously specifically required that when multiple relief devices were installed, they were to be installed with staggered set points. The first being at 100% of the MAWP and the subsequent being at 105% (or 103% in the case of Section 1 boilers). They also required that where feasible, the valves should be different sizes and that the first to relieve should have a smaller valve opening. The stated reason for this was that for smaller routine upsets you would only have the smaller valve opening minimizing the effect of the upset, reducing relieved volumes and minimizing the potential for valve damage. During larger (less common) events, the second relief device would relieve.

Anyway, clearly with B31.3 piping it is absolutely permissible to have subsequent relief devices with set points in excess of the MAWP (backed up by the language in both B31.3 and ASME Section VIII). In the case of B31.1, it does seem a bit muddy since the language is not explicit and there is no specific reference to UG-134. This leaves me with the thoughts I expressed earlier ...

Myself, I have only ever installed multiple relief devices at the same set point when they were redundant valves with only 1 in service at a time. When multiple devices were intended to be in service simultaneously, I have always installed them with staggered set points.

 

[BigInch]

I can agree exactly with you rneil; my experience exactly mirroring yours for B31.3, and also for B31.4 & 8. However I don't know specifically how that is handled in B31.1. Left to my own devices, I would tend to reason that B31.1 takes precedence over the others, when designing a B31.1 piping system.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[rneill]

I should note that when practical, I use staggered set points at or below the MAWP rather than above the MAWP.

I agree that B31.1 takes precedence over other related documents for B31.1 systems but it does seem to be silent on the issue ... if I get some time, I might write up a code interpretation request for submittal to the B31.1 committee.

 

[ekschwab]

This has been an interesting discussion. Thanks to all who contributed.

Since my original post, I determined that the highest setting on the last (6th) PSV is <5% above the B31.1 piping MAWP which is within Section VIII's UG-134. (I was originally told 10% over the MAWP.) I am feeling better about this, but would have preferred to see the staggered setpoints below MAWP as rneil states. Unfortunatley it is not my system, do not have design authority, and I cannot cite a clear code violation.

rneil - A code interpretation would certainly be beneficial for the future.

 

[BigInch]

I certainly can't see any logic for setting PSVs below MAWP. Think about it. What's the point in setting your design pressure so high, if you're not going to be able to use it. No. I could use it, if it was held at a steady state flow. As soon as it raises just a tiny little bit, but still below MAWP, the PSV pops and drops it below my original steady state pressure. The PSV created a negative transient. And for sure somebody set the pump to hold design pressure and for sure the instrument engineer didn't tell anyone that the PSV is set to trip at less than design pressure. That will surprize somebody late at night. How logical did that sound? Sure seems strange to have a MAWP that you can't operate at. What's MAWP mean then?

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[ekschwab]

BigInch: If we had a single relief valve I would totally agree. With Sect VIII UG-134 in mind I would still agree for almost every case. But 31.1 is not clear on this and that is the code being used for the design.

Also, in this piping system the possibility to get close to MAWP is only during startup and unit trip (expected every 1-2 years). So there is no concern for repeated popping/closing of the PSV.

At normal operation we are several hundred psi below the piping design pressure. So having setpoints slightly below MAWP for PSV #1-5 (staggered) and #6 set at MAWP wouldn't bother me. The PSVs will be set at or above MAWP by design.

(This isn't a pumping system.)

 

[rneill]

ekschwab picked up on what I was getting at. When you normal operating pressures are far below the piping design pressure (for plant piping often governed by flanges and piping that is set to match flange ratings - but that is a whole different discussion) then what is the harm in having the first PSV set at 95% of the system rating and the second at 100% of the system rating.

If the system operating pressure by necessity needed to be "near" to the MAWP then I wouldn't consider it "practical" to set any of my PSV's below MAWP (as is the case described by Biginch).

 

[BigInch]

You mean today you're not running at design pressure. In the future you will be. That I can pretty much assure you is only a matter of time... usually a short time.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv