Thermal Relief - Pipe Protection 2

I haven't looked into the code requirements of certified/non-certified in ammonia service but for thermal relief unless you are some longs runs of large pipe, the relief requirements versus the capacity of just about any PSV are orders of magnitude different. As long as the PSV is going to relieve at set pressure, I wouldn't likely be too concerned. Now, if I was protecting a pressure vessel in ammonia service, that's a different question.

That doesn't answer your question though. It will be interesting to see the rest of the replies.

Bill, in the latest version of API 520 Pt 1 (8th ed), the section numbers are different and the section you cited is worded a little differently. I don’t have the version you’re looking at, but to find the latest explanation, which appears to be worded a little more clearly, refer to section 5.9 and specifically 5.9.1.

The term “capacity certification” simply means that the valve has undergone capacity testing, using the prescribed ASME testing protocol, according to National Board requirements. In other words, the valve has a certified liquid flow coefficient, and it's published in the National Board “Red Book”. To fully understand the text that you cite, you need to understand a little history. Prior to ~1985 there was no such thing as a capacity certified (“ASME code certified”) liquid PSV. Up until then, ASME capacity certification was limited to just vapor PSVs. That is, there was no certified testing protocol for liquid PSVs. In 1985, ASME established the certification requirements for liquid service PSVs, and they were added to the red book. One of the ASME certification requirements stated that the valve’s certified capacity was to be determined at 10% overpressure. Prior to that, almost all liquid valves didn’t reach their capacity until 25% overpressure. So, manufacturers went back and designed liquid trim valves which reach capacity at 10% overpressure.

The first sentence you cite is simply saying that ASME now requires liquid service PSVs to be capacity certified. In other words, if you want your design to be compliant with ASME, then you have to use a capacity certified (“code certified”) PSV – one that has a flow coefficient listed in the red book. The second sentence you cite is effectively saying, “Before 1985, liquid PSVs were sized using the following equations....” And, “This is the sizing method you can use if: (1) your PSV doesn’t have an ASME liquid flow coefficient (isn’t listed in the red book), or (2) your installation doesn’t have to comply with ASME.”

The bottom line answer to your question is this..... If this piping needs a PSV, and you want it to comply with ASME B31.3, then use an ASME certified liquid PSV. Check the vendor catalog to determine whether or not the valve is ASME certified.

I will aggree with TD2K.

Ref: API 521: Section 5.14 Hydraulic expansion

5.14.2 Sizing and set pressure
" The required relieving rate is not easy to determine. Since every application is for a relieving liquid, the
required relieving rate is small; specifying an oversized device is, therefore, reasonable. A DN 20 × DN 25
(NPS ¾ × NPS 1) relief valve is commonly used. If there is reason to believe that this size is not adequate, the
procedure in 5.14.3 can be applied. If the liquid being relieved is expected to flash or form solids while it
passes through the relieving device, the procedure in 5.21.2 is recommended."

E.g. dont let a detailed engineering contract engineer start to create a calculation sheet for the sizing case.. it is typically not required..

I also agree with TD2K. but why are you looking at ASME Section VIII for ammonia piping? why not look at OSHA NH3 handling requirements?


"1910.111(b)(9)(ix) A hydrostatic relief valve shall be installed between each pair of valves in the liquid ammonia piping or hose where liquid may be trapped so as to relieve into the atmosphere at a safe location."

don't read too much into the above. it does NOT say "An ASME certified hydrostatic relief valve shall be installed..."

TD2K is correct.

First, thanks to all for the input. Ben, the facility I am helping chooses to apply ASME B31.3 and Section VIII to all of its piping systems, hence the tie to the those standards.
In fact, during the initial review of the plant, the NH3 storage standard was dismissed since this facility is an ammonia producer.

My interpretation, while conservative, is to apply the prinicples as outlined by Don. I does seem a bit strange to replace valves where a fractional gpm relief rate is involved, but I can find no basis for maintaining these Rego (non-certified) valves on the basis of saying that the valves are so oversized that they certainly ok.

I hate spending wasted $! Thanks again.

ASME B31.3 is very appropriate for Ammonia piping. However, as BenThayer wrote, ASME Sect VIII is only for Pressure Vessels; and if you are in the USA, the OSHA requirements [restrictions] are mandatory.

You need to IMMEDIATLY buy CGA [Compressed Gas Association] G-2.1 -- it has superceded the ANSI K61.1 spec that OSHA cites. Those requirements, through OSHA, have 'force of law' in the USA. Violate them, and somebody may end up in jail.

Further, if handling liquid ammonia, be advised that you have an Auto-Refridgeration potential -- NH3 boils at -29°F. Standard piping materials, SA-53 & SA-106 pipe, and SA-105 flanges, are only certified down to -20°F. To be Code compliant, you will have to use SA-333 Gr. 1 or 6 piping -- it is certified to -50°F; or stainless steel - good to << -250°F.

And yes, you have to replace the non-code PRV's. Non-Code is non-compliant; ASME B31.3 andCGA G-2.1 are mandatory, and OSHA will hand out massive fines and possible jail time for nowingly & wilfully violating these mandatory Codes. Buy some 3/4" [I find 1/2" too fragile, in "the Real World"] PRV's, with a standare small "D" orifice. Then you can make the statement that "Per good Engineering and Code practices, these PRV's will relieve greatly in excess of the credible scenarios' requirements"

Done.

Duwe6,

OSHA 1910.111 specifically exempts ammonia producers, which the OP said they were.

Interestingly, the *main* intent of OSHA 1910.111 was to make it difficult for illegal drug dealers/makers to steal ammonia from farmers, co-ops, low tech distribution facilities, etc. As such, it also addresses security, pin-hole orifices under pressure guages and transmitters, detectors, alarm systems, etc. OSHA figured an ammonia producer would have good security anyway, thus the exemption. The ammonia producer lobbyists saw to that.

Good luck,
Latexman

IRRC, Latexman is correct, 'producers' are exempt from the OSHA statute. However, liquid is guaranteed exert thermal expansion. If your valves hold, your gasket(s) inisolated lines will blow out at about 2K psig. Thermal exp. relief is still necessary. Nowhere in ASME [including B31.3] are non-Code PRV's allowed.*

Non-Code is Non-Code: gotta have an ASME [or API, if you can find one} acceptance stamp on all primary PSV's [presafety devices; fusible plugs, PSV's, rupture disks, etc.] to have a Code-compliant system.

Ask your Insurance carrier what the financial risk is, if you continue forward with your Non-Code PSV's. It will be in the millions if anyone gets hurt. Makes a few off-the-shelf PRV's or R/D's look pretty cheap, risk/benefit vs. price.





They can be used as a back-up to a Code PRV, or set to pop at a lower pressure than your Code PRV.

Great comments! You have confirmed that we are on the correct path - i.e. to replace with certified valves. Thanks.

i respectfully disagree with the "code" vs. "non-code" when you are complying with the OSHA standards on piping. for pressure vessels, absolutely go with code certified.

i worked at an ammonia producer (1500 TPD nh3 plant and typical downstream urea, hno3, nh4no3, etc.) and 60,000 tons of atmospheric storage and we routinely used the small Rego hydrostatic PSV's in liquid piping systems. our insurance carrier and OSHA inspectors (VPP site) never said "boo" about our PSV selections.

if an event happens, regardless of if it was a code or non-code valve, you are pretty much toast either way. that is, whenever anyone gets hurt it is because there was some type of non-compliance.

regarding the auto-refrigeration properties of ammonia, you are fine with A-106 and do not need A-333 or stainless piping for piping that is normally at ambient temperatures, etc.

when an ammonia line is depressured, it will cool but then there is essentially no pressure on the system. a quick look at B31.3 and coincident temperatures and pressures and you will see that ammonia at essentially no pressure (e.g., <4 psig) is acceptable.

that stated, i prefer SCH 10 stainless depending on the situation because too many contract pipefiters will think they are "replacing in kind" carbon steel for carbon steel and you can lose control of piping systems that were originally installed as A333. also, try and find a 3" A333 elbow on short notice vs. 304 SS for repairs/modifications.

My company requires us to buy UV stamped reliefs for U stamped vessels. That's all. On B31.3 pipe, they do not require a code relief. That said, there are an awful lot of other factors to think about. Availability. Quality. Spare parts on the plant books. Familiararity in the shop. Readily available nozzle areas and flow coefficients. Do they answer the phone at 3 in the morning. Do they come in and train the folks in the shop. Do they have an annual golf outing. Etc., etc., etc.

Good luck,
Latexman

one thing i neglected to mention is to pay attention to PSV capacity and blowdowns, etc. one nice thing about the small Rego PSV is that even if they fail or if they work because they need to, they release a much smaller volume of ammonia than a certified "D" orifice psv.

you might want to talk to the enviromental folks and see if the opening of a larger certified "D" PSV results in an ammonia cloud off-site impacting the public.

sometimes the thermal expansion case is because of a hot day. and if you are in nebraska, etc., in the summer time, it is not unlikely that piping in the direct sun can exceed 100°F. and a PSV popping with a long line of saturated 100°F ammonia liquid behind it might stay open longer than was originally anticipated.

A common misunderstanding among engineers is that "code valve" means an API 526 valve (D-T orifice valve). API 526 valves are simply a subset of code valves, but there are 1000's of code valves that are not API 526 valves.

All "code valve" means is that it has undergone capacity certification according to the ASME testing protocol.

As Ben states, one ought not automatically select a D orifice valve for liquid thermal expansion due to ambient heating. In the vast majority of such cases, a much smaller valve is better and safer. A small liquid service valve is a far better selection for those applications rather than a API 526 pop valve, and plenty of those smaller valves are listed in the Red Book (code certified).

We have used Farris series 2700 relief valves with "C" orifice for similar applications.

Try and have the set pressure equal to the MAWP of the pipe you are protecting.

Once these valves relieve they can be a nusiance (don't reseat easily).

We usually route the discharge into some type of separator pot. Recently more of the pots have the vapor line connected to a flare header.

OSHA / EPA is getting more strict on NH3 venting etc.

Good luck.

Great posts! I guess we all will never be in complete agreement regarding code vs. non-code, so to err on the conservative side, I will be replacing the Regos with code valves. And yes, there are plenty of Code liquid relief valves with smaller than D orifices, and that is what I plan to use.
Bill

For a blocked-in liquid experiencing thermal expansion, the relief quantity will be miniscule. To drop the [liquid] line pressure from 500 psig down to about 250# will only take about 1 ml/5 drops per about 500-ft of small-bore pipe, or 200-ft of big-bore pipe. Ammonia liquid, like all normal liquids, is essentially incompressible. The increase in pressure due to the expansion is being offset by expanding the steel pipe, blowing it up like a very strong, tough balloon.

The liquid didn't expand much, but the spring rate of the pipe is so 'fast' that a tiny expansion gives a huge increase in pressure. When your PRV 'burps' out a teaspoon 5ml or so, the thermal expansion has been accomodated and the pipe relaxes back to a MUCH lower pressure. Should be the vapor pressure of NH3 at the skin temperature of the pipe, after the liquid and pipe have cooled some, due to a little autorefrigeration - latent heat of vaporization of 5ml of NH3.

And if all the flanges were a prudent Class 300#, giving a room-temperature MAWP of 740 psig, and the PRV set at 350-650 psig, that PRV will slam shut after after one 'burp'.

Locked-in Liquid Expansion [in a rigid container] is waaay different than relieving an overpressured gas. One 'burp', and you are done.