Relief Valve Capacity Derated by Inlet Piping 3

[Feedyourhead]

So we have engineering software that is capable of calculating the effect that the piping has on a relief valve. But how is it doing it? What is engineering theory that would provide an analytical solution to how much a relief valve is affected by pressure loss?

The typical relief valve situation for my application is as follows:

A section of pipe T's off from the system to be protected and has a relief valve at the end. The piping from the T to the relief valve is considered the inlet piping. There is also outlet piping after the relief valve typically straight up like a chimney.

From what I understand is that engineering code recommends no more than 3% pressure loss on the inlet to prevent chattering, and no more than 10% pressure loss on the outlet to prevent too much backpressure. But what about derating the actual capacity of the relief valve with the pressure losses? How is it calculated? How is my software calculating it? Say for example I am able to calculate that there will be a pressure loss of 5 psi in the inlet piping, and that the relief valve is able to pass 400,000 cubic feet per hour at full capacity. What is the new capacity now that there is a 5 psi pressure loss in the inlet piping because it is 100 feet long?

 

[LittleInch]

I'm not sure I'm following your argument here.

Is the 5psi < or > 3% of set pressure?

Which software?

My guess is that it doesn't impact the capacity or if it does it calculates the flow with the differential pressure across the relief valve.

The type of relief valve also makes a big difference between simple spring, balanced bellows and pilot operated

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

 

[The Obturator]

Inlet pressure loss ultimately affects Pressure-relief Valve capacity, and if unavoidable/fixable, should be considered. Most major PRV manufacturers sizing programmes allow you to input the pressure loss which is then considered for the rated capacity calculation against the PRV type selected. Some can even calculate pressure loss should you be able to input the necessary data.



*** Per ISO-4126, the generic term
'Safety Valve' is used regardless of application or design ***

*** 'Pressure-relief Valve' is the equivalent ASME/API term ***

 

[shvet]

The piping from the T to the relief valve is considered the inlet piping.

Incorrect as this segment is a branch, not inlet.
The piping from the relief source to the relief valve is considered the inlet piping.

More info see here

https://www.eng-tips.com/viewthread.cfm?qid=502295

From what I understand is that engineering code recommends no more than 3% pressure loss on the inlet to prevent chattering, and no more than 10% pressure loss on the outlet to prevent too much backpressure.

Valid for conventional valves. There are other designs suitable for >3% and >10%.

From what I understand is that engineering code recommends no more than ... 10% pressure loss on the outlet to prevent too much backpressure.

Incorrect.
From what I understand is that engineering code recommends no more than ... 10% pressure loss on the outlet to prevent too much overpressure.

Note that (a) backpressure and overpressure are interrelated, (b) allowable overpressure for piping is much higher than that for vessels.

How is it calculated?

Derate pressure drop and fluid properties when calculate mass flux.

 

[Feedyourhead]

For this theoretical discussion lets say 5 psi more more than 3% of set pressure.

Typical relief valve designs for my application are pilot operated backpressure regulators.

The capacity of the relief valve is affected by the inlet pressure at the relief valve, the relief valve set point, and lets say the outlet pressure is close to atmosphere. But the program I have derates the relief capacity if there is too much inlet piping as a separate consideration to the differential pressure across the relief(set pressure).

Interesting thought, and definitely intuitive, that the inlet piping is from the overpressure source. Say maybe you have 20 feet from the source to the T.

Allowable ovetpressure for piping may be more than for pressure vessels, but it still isn't much. Maybe like 6 psi over relief valve set point.

 

[shvet]

Allowable ovetpressure for piping ... still isn't much.

302.2.4.f.1.a ... it is permissible to exceed the pressure rating or the allowable stress for pressure design at the temperature of the increased condition by not more than 33% for no more than 10 h at any one time and no more than 100 h/y ...

What PSV protects? What is piping design code and pressure rating? What is relieving temperature?

 

[MortenA]

@shvet,

Why do you say that the 10% loss on the outlet is for overpressore (as opposed to backpressure)? For the case where we have choked flow (ca > 15 psig) flow is independent of downstream pressure. However, a high back pressure can cause unstable flow?

--- Best regards, Morten Andersen

 

[shvet]

@MortenA

5.3.3.1.3 In a conventional PRV application, the allowable built-up backpressure is equal to the allowable overpressure.

See table 3 for illustration. In case of a conventional valve allowable built-up backpressure depends on relief case and design code.

Option 1 - PSV protects a pressure vessel, ASME VIII, firecase
built-up backpressure < 0.21 * MAWP

Option 2 - PSV protects a pressure vessel, ASME VIII, overpressure case, staggered set pressure (2 valves in parralel)
built-up backpressure < 0.16 * MAWP

Option 3 - PSV protects a pressure vessel, ASME I, overpressure case
built-up backpressure < 0.06 * MAWP

Option 4 - PSV protects piping, ASME B31.3, overpressure case
built-up backpressure < 0.33 * pressure rating

Option 5 - PSV protects a pressure vessel, PED, overpressure case
built-up backpressure < 0.10 * design pressure

and so on

In topicstarter's case (pilot valve) total backpressure (superimposed + built-up) may be as high as 98% of set pressure. Capacity derating is the only issue.

 

[Latexman]

A well labelled and dimensioned sketch/drawing/P&ID from relief source/protected vessel to PSV outlet would be most welcome.

Good Luck,
Latexman

 

[Feedyourhead]

"What PSV protects? What is piping design code and pressure rating? What is relieving temperature?"

It protects a natural gas distribution system. I think we have multiple at various operating pressures. 40 or 60 psi maop. Allowable ovrrpressure by federal code is 6 psi over maop for piping at 12-60 maop.

"A well labelled and dimensioned sketch/drawing/P&ID from relief source/protected vessel to PSV outlet would be most welcome"

I don't really feel like sharing internal information on a public forum, I meant to ask for a theoretical approach to determine if there is x pressure loss in the inlet/outlet piping then it derates the relief valve capacity by y. If anyone has heard of any analytical approaches as to how that might be calculated. Rather than relying on a computer program.

We can determine the flow rate by determining what the maximum flow the upstream valves can pass is and what the upstream pressure is. So the relief valve must be able to pass that amount of flow (hundreds of thousands of cubic feet per hour) in the event that there is a mechanical failure.

 

[shvet]

engineering code recommends no more than 3% pressure loss on the inlet to prevent chattering

It protects a natural gas distribution system.

Make sure this "engineering code" does contain such recommendation. For example - ASME B31.3 does not. Your case looks like an overkill as "3% rule" relates to vessels, not piping.
More info can be found by the link above.

 

[GBTorpenhow]

But the program I have derates the relief capacity if there is too much inlet piping as a separate consideration to the differential pressure across the relief(set pressure).

I assume what is happening in the program is say you have 5 psi of inlet drop, an allowable overpressure of 6 psi, and a pilot operated PSV that needs 5% overpressure over the set point to be at full capacity. If the design pressure/MAOP is low, with 6 psi allowable overpressure that sounds like say 60 psig or so, e.g. a PSV set at 60 psig wont be at full capacity until 63 psig at the device inlet. With 5 psig inlet drop you have 68 psig at the piping protected by the PSV, which is more than the allowed overpressure. So you either have to derate the PSV because you can't allow it to use its full capacity or you have to reduce the set point which also reduces the capacity.

ETA - Saw your post while I was typing this, to do this yourself in a rough way you can get the valve coefficient and orifice area from the PSV manufacturer's literature usually but if its close IMO its preferable to have the vendor do the sizing calcs with their software when quoting the valve.

 

[Feedyourhead]

I kind of thought that the pressure buildup would be less than 5% of set point since it is pilot operated. But now that I think of it pilot operated regulators can't hold their setpoint at high flows. So 5% could be possible. Do you really expect 5%? Because that definitely hurts, that's already 1/3 to 1/2 of the allowable pressure buildup. But that could always be remedied by lowering the setpoints a bit.

But lets consider an ideal situation where the pressure loss and pressure buildup are perfectly uniform.

The required capacity is 300,000 cubic feet per hour. At a set point of 60 psi with a buildup of 3 psi the rated relief capacity is 301,000 cubic feet per hour. The inlet piping has a pressure drop of 2.9 psi. So now the overpressure in the system is 65.9, 0.1 less than the limit. The relief capacity is only 1000 more than what is required. Again, this is an idealized scenario so it can work. Or does the inlet piping being so long that it causes a 2.9 psi pressure drop derate the relief valve so now it only has a capacity of 250,000 cubic feet per hour instead of 301,000?

 

[georgeverghese]

"Typical relief valve designs for my application are pilot operated backpressure regulators."

Pls note such devices are only used as process control devices to bleed away excess pressure in the upstream system; they are NOT Coded safety relief valves.

 

[Feedyourhead]

That's interesting. My role is to scrutinize existing designs. These backpressure regulators were stamped by a PE and approved by regulators. I guess its possible they didn't realise?

 

[georgeverghese]

Non coded back pressure regulators are some times used as relief valves, but only in innocous applications such as water. Certainly not for flammable or toxic or high pressure applications that can harm operators.
Perhaps the P&ID has erroneously shown these devices with the standard safety relief valve symbol ?

 

[Feedyourhead]

Ok let me clarify. The application is a model of regulator (the same that might be used to step down the pressure from 100 to 40 psi) but configured with a pilot to operate as a relief valve. Not a model of backpressure regulators.

 

[shvet]

The application is a model of regulator but configured with a pilot to operate as a relief valve.

Regulator is not a Pressure Relief Device. This is a regulator, a kind of control valve. Yes, self-acting but still a control one, a part of a control system, not a PRD and a safety system. Control valves has another design, response time, codes, practices etc. Chattering and "3% rule" are not applicable to control valves.

Anyway see

The verbal forms used to express the provisions in this document are as follows.
...
Should: As used in a standard, “should” denotes a recommendation or that which is advised but not required in order to conform to the standard.
...
7.3.4 PRV Inlet Pressure Loss Criteria
The total nonrecoverable pressure loss between the protected equipment and the pressure-relief valve should not exceed 3 % of the PRV set pressure, except as noted below:
...
— remotely sensed pilot-operated relief valves (see 7.3.9);
— an engineering analysis is performed for the specific installation (see 7.3.6).
...
7.3.9 Inlet Loss Criteria for Remote Sensing for Pilot-Operated Pressure-Relief Valves
Remote sensing permits the pilot to sense system pressure at a location that most accurately reflects the actual pressure of the protected system. Remote sensing will mitigate the effect of excessive inlet pressure losses due to the inlet piping configuration... The addition of a remote sensing line allows the pilot to correctly sense system pressure and to keep the valve from rapid cycling or chattering due to high inlet piping pressure losses.

Pay attention what you are doing (all together):
- discussing a PSV in a pipeline forum
- mixing doubtful/unproved practice with widely accepted practice
- mixing an advise with requirements
- mixing equipment overpressure with piping overpressure
- mixing a petroleum industry code with a NG distribution one
- mixing safety system with control system

What is a plan and a final goal? These are not obvious - your questions and comments contradicts each other. Plan is to confirm NG is distributed in a safely manner? What you are trying to avoid - undercapacity caused by chattering? In this case a way you chose is hard to name "proper", more likely "wierd".

 

[GBTorpenhow]

Tough crowd.

There is nothing in US pipeline code 49 CFR 192 (just a guess on my part that OP is in the US under DOT based on use of freedom units) that I can remember that requires pressure relief to be provided by ASME coded relief valves when providing overpressure protection to a pipeline or distribution main.

Diaphragm type self-operated control valves marketed primarily for use in pressure reducing applications, known colloquially as 'regulators', can be and often are configured as pressure relief devices in this application.

The 5% overpressure I quoted is typical of the pilot operated relief valves marketed to the natural gas industry for this application to help meet the DOT/192 overpressure limits for lower pressure applications. The one I have used is the Anderson Greenwood "High Gain" modulating pilot on Series 400 relief valves.

At a set point of 60 psi with a buildup of 3 psi the rated relief capacity is 301,000 cubic feet per hour. The inlet piping has a pressure drop of 2.9 psi. So now the overpressure in the system is 65.9, 0.1 less than the limit. The relief capacity is only 1000 more than what is required. Again, this is an idealized scenario so it can work. Or does the inlet piping being so long that it causes a 2.9 psi pressure drop derate the relief valve so now it only has a capacity of 250,000 cubic feet per hour instead of 301,000?

No, the device cannot 'see' the inlet drop, so the stated capacity of 301 Mscf applies as long as there is 63 psig or more at the inlet to the device.

 

[Feedyourhead]

What you are trying to avoid - undercapacity caused by chattering? In this case a way you chose is hard to name "proper", more likely "wierd".

Yes trying to avoid undercapacity caused by excessive inlet piping.

The addition of a remote sensing line allows the pilot to correctly sense system pressure and to keep the valve from rapid cycling or chattering due to high inlet piping pressure losses.

Yes that would be ideal, but it should be noted that laying out long sense lines could be dangerous and increases the risk of third party damage. if someone breaks your sense line that is a big problem.