Should a bypass valve be provided in a City Gate Regulator Station? 2

[TexasPE]

In a natural gas city gas regulator station with two parallel regulators, one primary and one secondary both with backup monitor regulators for overpressure protection, should a bypass be allowed? The station is taking gas from an 800 psig MAOP pipeline, and after regulation down to 60 psig and then after odorization is delivering it to two systems. One has a set of meters for an industrial customer and the second to a stub out for a low pressure polyethylene pipe LP distribution system. There is no relief or high pressure shutdown being proposed on the system so it is strictly relying on the dual regulation to provide overpressure protection. It is my belief that the addition of the bypass should mandate either a front end high pressure shutdown if the back end gets above the MAOP of 60 psig. While it appears the DOT regulations and ASME B31.8 code does not mandate this, it would seem that sound engineering would. Operating with a bypass open would make any overpressure protection by the regulators effectively non-existent. Can I get some discussion on this?

 

[davefitz]

short answer= no. refer to thread815-444133 regarding the sept 2018 gas pipeline failure near boston massachusetts, related to columbia gas co construction upgrade of consumer distribution pipelines.

In general triple redundancy is used to protect consumer pipelines from overpressure if full capacity relief valves are not provided. The capacity of such relief valves would also need to include the negligent opening of the bypass valve if it was provided. Normally series pressure regulators are used plus a pressure switch hardwired to a a fast closing stop valve plus a small leakage releif valve.When you evaluate the safety of the system, you need to assume osama bin operator is deliberately going to operate the bypass at the worst possible time, therefore you should not provide it.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[LittleInch]

This is the conflicting issue that many of these systems face - you need to prevent pressure rising above design in the downstream system, but also need to keep the gas flowing to consumers as a shutdown has huge consequences and costs in getting the supply back on safely.

So for me your system is already not ideal because it apparently has no standalone over pressure protection.

To then compound this by bypassing the protection you have is nothing short of lunacy. My argument in any of these types of discussions is that a bypass eliminates the pressure control. Doesn't matter if the valve is Locked closed or subject to special operating procedures, if it's there then someone can open it and walk away. If it isn't fitted they can't.

I did once see on a city gate type regulator station a set of small bore bypass tubes fitted with orifices / RO which would allow manual control of the pressure by watching the pressure downstream and turning tubes on or off to maintain the right flow, which was their third or fourth line back up plan to prevent the area loosing its only gas supply if all else failed, but it still looked pretty hairy to me. The idea being that there was always some flow so if you had sized your ROs correctly opening one would maintain flow / pressure at min demand, but then you could open more as required. I never found out if they had ever tried it in action, but I think it was there to give them a nice warm feeling that they could do it if ever needed.

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

 

[TexasPE]

I talked with a former manager of a small municipality in Texas that has their own gas system which has gone through much changeout in personnel. His comment was the continuity of service issue. They wanted to be able to continue to deliver gas to customers so they would throttle across the valve and adjust as necessary while they worked on the regulators. The problem is I am now looking at an addition for a large industrial user. Limited distribution pipe for line pack and probably large demand swings as an industrial furnace may turn large burners on and off. My engineering judgment says this should not be allowed in this instance as the pressure rise could overpressure the low pressure system as soon as the demand falls. Add to this an improper design with a bypass on a cross and in line with the inlet. I see this as the recipe for disaster. A picture of an actual application with a single valve bypass in line with the inlet. Now where is any line debris, like black powder, especially if there is no inlet filter to scrub the inlet gas going to deposit. What happens when the valve is initially cracked open and the flow is obstructed. What happened when the accumulated debris finally blows through because the valve is opened wider than it should to let the debris blow through. I think it is fair to say this is unsound engineering and a defective design.

 

[georgeverghese]

Why the bypass? Flow turndown issues with the existing parallel runs? This is often a design error - the solution would be to enable the widest possible turndown within the limits of the 2 existing runs.

 

[TexasPE]

I didn't design and seal the drawings. I am really questioning the design and the engineer that sealed them. Why does New London Texas keep coming to mind? I read the preliminary report on Andover, Ma and the Columbia Gulf issues that LittleInch reported above. Seems there is good cause to be concerned.

 

[1503-44]

"I talked with a former manager of a small municipality in Texas that has their own gas system which has gone through much changeout in personnel. His comment was the continuity of service issue. They wanted to be able to continue to deliver gas to customers so they would throttle across the valve and adjust as necessary while they worked on the regulators."

That's a common, often solely operational, error in contineuity of supply problems. The original and correct design incorporated two parallel regulators so they could work on one while the other remained operating. After operating for 20 years in that manner, the demand has risen to twice the original and the operator says "hey that regulator isn't being used. Turn that one on too." Now they have no way to simultaneously supply while repairing one or both regulators.

Yes it is more than cause for concern. These types of final customer pressure regulation facilities now represent the highest level of danger to the general public and actually to the companies themselves that needlessly expose them to it. $1,000,000,000 could have bought all the equipment they needed to prevent that incident 10,000 times over. No amount of money brings back people killed.

After 50 years of cookbook regulation, the Feds are finally beginning to understand the value of engineering and have recommended that PE's be involved in making gas distribution safe. That is an earth-shaking stand, especially in light of todays antiregulatory gov climate. Please not only make it safe, but make it as safe as you possibly can. Idiot proof (series regulators) is not enough, as idiots bypass, or disconnect their controls. Failsafe only (pressure relief). No needless deaths on the hands of PEs are acceptable. NOT ONE! (If they want an idiot design, they can easily find an iniot to do it.)

 

[georgeverghese]

A HAZOP / process safety LOPA and SIL review audit exercise carried out under the framework of either Company or national guidelines with jurisdiction over these facilities should sort this out. All audit findings should be closed out without compromise before the PE signs off.

 

[1503-44]

georgeverghese That is true in what should be the normal procedure, but it was the lack of even a most basic HAZOP review that went missing in the Lawrence Disaster. They couldn't be bothered to find a drawing of the system. I can easily imagine that was because they knew they did not have any. I once worked for a producer with an 850 well gathering system and 2000 miles of pipelines. I could hardly find a plat with a well location on it, never mind the nearest block valve. As-built of compressor stations, no, not any design drawings. No equipment manuals of anything. I spent most of my years there making as-built drawings in my spare time, which the owner thought were a security risk, because they showed the locations of the wells.

How could I forget to mention this classic operator mistake with direct safety system override implications. Two regulators (ESD valves, or whatever) IN SERIES, designed to provide the "One fails, surely the other won't" safety scheme. The demand doubles and the operator says, "Let's put these in parallel", which almost doubles the flow rate, substantially decreases pressure drop, miraculously solves a number of problems, plus the golden nugget, probably reduces operating cost. Hey, with no engineer, or a HAZOP review, that is exactly what happens.

 

[alchemon]

I have seen, and designed, multiple stations with a hard piped bypass. They are provided for the purpose of continuity of service.

I do not believe it to be as critical in your case as the pressure reduction is quite large. They are more useful when say inlet is 600# and outlet is 400#, as I have seen the inlets drop below the outlets on cold days.

I would suggest to your client that they receive written approval from the PUC. I prefer hard piped bypasses to be above grade, double valved, with gauges and threadolets in the interim space. You could possibly even consider some kind of slam shut device in the bypass as an OPD. You should also consider something other than a ball valve - as they are not ideal for throttling service.

 

[LittleInch]

alchemon,

If you've designed them then what is your view about the fact you are disabling all the built in pressure protection features in the current two streams by installing a hard piped bypass line??

Just because something has built up over time and appears to be commonplace doesn't always make it the right thing to do.

Did you make a typo here ?? "I do not believe it to be as critical in your case " i.e. an extra "not"??

800 psi down to 60 psi and you seriously think a simple bypass with no pressure protection is OK?? - "You could consider..." implies that you think it's ok to do it without said protection.

As I said above, this desire to maintain service seems to cloud peoples judgement as to what is actually safe practice. I do appreciate the cost and issues involved with a loss of gas supply but generally people don't die or buildings explode as they do if you significantly over pressurise your downstream piping.

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

 

[alchemon]

Well - the argument is simple in my mind: this is primarily an Owner/Client decision. I cannot make certain decisions for the Owner, and when I do - I expose myself to undue liability. It is primarily my responsibility to inform the Owner of risks and pros/cons in either direction in this case.

I do not feel that we are removing all fail safes in the design. The regulation set up in the beginning seems a little risky anyhow (with no relief), a little piece of trash hitting the secondary will void the entire system - but that was not the question raised. Rather, we are merely introducing a human element where manual operation is required/allowed. Kind of like being able to turn off the autopilot on an airplane (I’m sure Boeing wishes they would have made this easier on their latest 737)... Poor maintenance, inadequate monitoring, and a whole host of risks are reasons that this arrangement could fail.

The point I was making in not being as critical was the I would advise the Owner that having a manual high pressure bypass is not as critical/important here (in this particular case) due to the larger pressure drop across the station. I highly doubt that the inlet pressure ever drops below the outlet due to the large difference. I have experienced the transmission inlet falling below the outlet, and if this occurs - there is not much you can do other than run the station on bypass (course station needs to be designed so that in this case gas is still oxidized).

My example shows that if a differential is not as great, the bypass can be incredibly useful. The question is always: what is the operational reliability of the upstream pressure? Most firm delivery contracts do not specify, or the minimum inlet pressure is something very low. Therefore, I do not believe it is wise to holistically discount manual bypasses as a general design philosophy.

Say the delivery contract on this gate station states that the pressure will never fall below 300# unless force majeure: your regulators may be designed to provide adequate flow at 600#, but are they designed to do so at 300#? If the transmission pipeline routinely falls to say 300# on a cold day - you may have problems. Course with a redundant set, the combined flow may be adequate, but then you are back to not true full redundancy.

It is up to the engineer of record on this project to make a judgment decision on whether the hard piped bypass is necessary and useful to possibly include a slam shut or other OPD, and then advise the Owner. My opinion with limited information is likely not necessary, but if the Owner wants it - there is nothing wrong with the arrangement, and the EOR should provide a reasonable standard of care against over pressurization.

 

[1503-44]

Maybe I totally misunderstand, but I can't make sense of a lot of that. First, I am not sure what you mean by,

"I have experienced the transmission inlet falling below the outlet, and if this occurs - there is not much you can do other than run the station on bypass (course station needs to be designed so that in this case gas is still oxidized)." How do you get forward flow in that condition? When inlet pressure is lower than outlet, reverse flow always occurs, unless there is an active block or check valve, in which case you get no flow. When inlet pressure is lower than outlet, you would get backflow through any open bypass. You could probably increase outlet pressure via a bypass, due to a higher pressure drop within the bypassed facility than in the bypass piping, but the inlet pressure to the bypass piping would still be greater than the outlet pressure, or backflow occurs. So using a bypass to increase delivery in those circumstances would defeat all value of having the bypassed facility. It makes no sense. And when pressure falls below minimum operational system pressure, it is time to shut it down, not open bypasses. Continuing to flow at minimal pressures can reduce pipeline inlet pressure and downstream pressures even further, which can result in downstream regulators going full open, possibly, eventually extinguishing pilot lights and incorrect sales meter readings, as well.

"Say the delivery contract on this gate station states that the pressure will never fall below 300# unless force majeure: your regulators may be designed to provide adequate flow at 600#, but are they designed to do so at 300#?"
Of course they are. Every delivery contract I have seen firstly specifies the flowrate. It might specify pressure, but often does not. Where pressure is an important requirement for delivery, it will specify pressure too. An end client normally does not care about delivery pressure so much, because if he is getting the required volume, he's happy and his use is almost always at relatively low pressures anyway. If he had a good engineer review his contract, it would specify pressure too, no matter how minimal it was. A downstream pipeline "client" cares about delivery pressure because it won't get into the downstream system at all, if pressure falls below that minimum. If you are a producer delivering to a downstream pipeline, you typically care, because the downstream pipeline often simply won't purchase your stuff at all, if your pressure falls below minimum, or they will buy only what quantity you can deliver at lower pressures, if any and if still possible, and just maybe you might be allowed to make up volumes not delivered to them at some later date, but it is also possible that you will simply have to pay for not delivering the volumes specified in the contract. Or maybe you won't get paid for anything not delivered. That's usually how it works. And pipeline companies can't afford to put a guy at every facility to open some bypass whenever system pressures get low. That's why they have a minimum pressure specified in the contract. You might be able to sustain an operational flowrate with a redundant bypass open, but that was not the purpose of having the redundant bypass. Redundant bypasses are reserved for maintenance purposes only; they not for operational use. A redundant "regulator" valve to be used for continued delivery using "hand" pressure control, if provided and could be used safely considering the downstream, would probably be a plug valve, not two PCVs. Possibly in an extreme emergency bypass operation could be considered, but I can't think of what kind of emergency would have to exist to do it, nor have I ever seen such thing written in a delivery contract.

There's nothing wrong with a bypass arrangement, until somebody leaves it open too long, or starts texting his girlfriend on a Saturday night callout. Operating by following procedures for conducting maintenance is not a good road to follow.

 

[alchemon]

First comment: it is a city gate station. They are pulling gas off a transmission (midstream) pipeline and going to a distribution (downstream) system. The gate station’s purpose is to reduce the pressure from the midstream pressure and odorize (depending on where you are in the country - some states require midstream to odorize).

When inlet is less than outlet - you bypass the station and the REGULATORS. The flow will equalize and eventually you have flow from midstream to downstream. Hopefully, the design is smart enough to employ check valves to prevent reverse flow.

The bypass is to bypass the REGULATORS of the gate station. Do the regulators work, have they been maintained properly, do they have adequate capacity. There is no right or wrong opinion here, each company must decide how they wish to proceed. And many, choose to have a manual bypass.

The bypass at the gate station is an “oh crap” moment. Doesn’t cost much money to set a guy up for one night to manually bypass the station on a cold winter night (as compared to relighting countless services that would be lost). I have witnessed it first hand.

The differential between the normal and minimum is often below what the downstream party can handle operationally. Let me know how it goes for you if you consistently deliver 600# to your customer, but then start cutting it back to 300# when peak demand sinks in. You may say - “well we met our contract”. But trust me, if this occurs over a number of years, your customer will go to a different supplier and build a new pipeline if that is what it takes to remain operation. This is happening as we speak to one of the largest midstreams in my corner of the world. They have over sold the pipeline, everyone knows it, and they cut the inlet pressures to dramatically low levels on cold days (when peak demand occurs).

A redundant bypass is not, in my opinion, only for scheduled maintenance purposes (depending on how critical the load is). It should primarily be designed to serve as a true back up.

As a design rule, I would say, if the MAOP of the downstream system is higher than the minimum upstream pressure specified by the contract - then you should strongly consider a manual bypass.

 

[1503-44]

If delivery pressure is at minimum specified by the contract, normally extra pressure would be available, so just raise the delivery pressure until it is sufficiently higher than downstream to flow again. If no additional pressure is available for some emerrgency situation, bypassing won't help as the same inlet pressure - station pressure drop is still less than downstream; no forward flow.

Might as well have an additioal parallel regulator rather than a manual valve. The cost benefit ratio is almost equal, or a great many times higher, plus you have complete redundancy for delivery during maintenance. You can adjust the regulator pressure higher to suit your scenario above if needed, up to MAOP of the downstreams system, but then again, why would it not have been set there already, unless of course there was a very good reason that it wasn't. In which case a manual bypass again makes no sense. They are good for black starts, not a routine operatioinal mode, and if care is used, but that's about it. If you need one to supply flow during anything but an extreme emergency, not sure what that might be, in my PE opinion, it's just not a good design. Full stoop. Period.

 

[davefitz]

Unless you are installing this station in a completely undeveloped country, there are laws and statutes that require the design meet specified codes and standards, which in turn do not allow for personal opinion on whether or not one provides automatic pressure control on such a natural gas pressure reducing station supplying gas to the public. The state and federal gov'ts require the use of these design codes, and so does the relevant insurance companies. Such a bypass would invalidate the function of the automatic pressure control .

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[alchemon]

Usage in an extreme emergency - that is exactly what they are there for .

When the emergency hits: it is too late to cut the manifold and install it. STATION lifetime is say 50 years (?). If you have one or two emergencies in 50 years - the bypass will pay for itself and then some after the first time.

Just talk to the owner and ask what they want. It is not that hard.

For this specific, original example - the bypass is dangerous and not useful (in my opinion). If the owner really wants one here, try to talk them out of it. If they persist, put on a slam shut or relief and go home. You have more than met your duty to the public and to the client.

And RE comments about codes and the like - please point me to anything in CFR 192 that states you can’t have a manual high pressure bypass.

What is automatic pressure control anyway. A human sets the regulator - if they don’t do that correctly and then leave the station - downstream pressure can still build and create an overpressurization. Gauges and SCADA can fail.

There is no substitute for a trained technician that can go out and troubleshoot the station and verify without doubt that the regulator is operating correctly.

I realize certain states may have different language, but there is nothing in the federal code that prohibits them. Nor can we in this industry remove the final human oversight element.

 

[1503-44]

"And RE comments about codes and the like - please point me to anything in CFR 192 that states you can’t have a manual high pressure bypass."

Codes never tell you what you cannot do, and neither should well-written specifications. They tell you what is required. So, with that in mind, show me where the CFRs require manual operation of anything.

And the CFRs do indeed define specifically what equipment is required for pressure reduction in the operating scenario. It permits three options, none of which are by manual pressure control by the way. All options work well under typical conditions,but they are still insufficient to adequately protect the public as EVIDENCE of the Lawrence Mass.diisaster plainly shows. That is my problem with those regulations, as it is with much of the public, some state legislatures and a US Senator- presidential candidate that I have communicated with on this matter, that are now aware that additional requirements for positive pressure protection in the form of relief valves must be considered in future regulations, due to the recent history of the magnitude of the disasters that can occur when such is not required. If I'm not mistaken, gas distribution has a worse record than nuclear power mishaps, maybe or not in death tools, but certainly in property damages.

 

[LittleInch]

please point me to anything in CFR 192 that states you can’t have a manual high pressure bypass.

Well 192.195 seems to cover this pretty well.

192.195 Protection against accidental overpressuring.
(a) General requirements. Except as provided in § 192.197, each pipeline that is connected to a gas source so that the maximum allowable operating pressure could be exceeded as the result of pressure control failure or of some other type of failure, must have pressure relieving or pressure limiting devices that meet the requirements of §§ 192.199 and 192.201.

(b) Additional requirements for distribution systems. Each distribution system that is supplied from a source of gas that is at a higher pressure than the maximum allowable operating pressure for the system must -

(1) Have pressure regulation devices capable of meeting the pressure, load, and other service conditions that will be experienced in normal operation of the system, and that could be activated in the event of failure of some portion of the system; and

(2) Be designed so as to prevent accidental overpressuring.

Now sure it doesn't say "no manual bypass, but please tell me how you are going to ensure that the previous provisions are possible to be met by use of a manual bypass with no other devices. By designing one to be installed you would, in my opinion, violate the b) 2) part of the code.

Your latest post now say, correctly, put on a slamshut or a relief (valve). Fine, but that's not a "manual bypass" as most people understand it.

I do truly understand the reason why they exist and i the rare times they are used have managed not to blow too many people up due to the skill and professionalism of the operators and technicians invovled.

But that and procedures to prevent accidental opening are not enough. IMHO.

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

 

[1503-44]

"Now sure it doesn't say "no manual bypass, but please tell me how you are going to ensure that the previous provisions are possible to be met by use of a manual bypass with no other devices. By designing one to be installed you would, in my opinion, violate the b) 2) part of the code."

Exactly so LittleInch, manual pressure regulation while in any operational mode is not listed in any of the options for controlling pressure to protect downstream piping from overpressure, so thereby not permitted to be used for that purpose. Manual pressure regulation via bypass, or not, is not prohibited and therefore can be used for purposes, such as operation or maintenance affecting downstream piping that is protected with a listed option, thereby not endangered in any manner from overpressure, or other methods, such as maintenance when lower pressure piping is appropriately disconnected, blocked or blinded, or for controlling flow where pressure reduction is not at all necessary to protect downstream piping from overpressure, such as black start of facilities whith >= MAOP of upstream piping. Scenarios requiring pressure reduction for lower MAOP pipe, no.