Why Would Relief Valves be Required on a Metering Run?

Is your function to produce a bid of the proposed system, or to analyze the proposed system?

This answer lies with you and the organization requesting the the ITB. If it is not an RFP (request for proposal) your path seems clear to me.

We are currently analyzing the proposed system to get a flier when we get the ITB. The drawings are of low quality with numerous other errors. I think the inclusion of PSVs is a mistakeand I want to get it into the heads of the bean counters that a lot of rework is likely to produce a good engineered solution. But I dont want to hold up the PSVs as an issue if there is a good reason for them.

This ideally needs to be done before we get the ITB.

Without an isometric piping diagram, assumptions have to be made. Most likely the PSV is there to protect the metering station from the shock/back pressure pulses/waves that occur when downstream emergency valves slam shut.

The extremes in the elevations of Colombia, when coupled with the length of the NG pipelines, can lead to rapid pressure pulses that may travel throughout the system. Where the pipeline deviates from a relatively straight line, those pulses can cause huge mechanical forces at the transition points.

My guess is that you'll find PSVs in many locations where there are significant elevation changes. They're probably there to act as snubbers or shock absorbers, but only looking at well annotated drawings will reveal their true purpose. Contact the pipeline owner for the most accurate info.

The maximum elevation of the pipeline is 1800m. Currently the PSVs are only at the cutody metering on the beach and at the tie in to the National Transmision System (near Cali in the middle of the country). It may be that they have not yet considered the various significant changes of elevation.

I am not sure that even an ESDV would slam shut in 30" piping I would expect a 15-20sec closing time. Are the back pressure pulses that severe in gases? I am used to seeing water hammer in liquid services but I thought that the the main driver of that phenomenen was the incompressiility of liquids.

There is no current pipeline and certainly no well annotated drawings.

I will need to go away and dig a bit more into this area thanks

I can't help with most of your queries, but some things aren't clear.

Is the 1800m the difference in elevation along the pipe?

Is the 100bar the pressure of the liquid or the gas after vaporisation? I assume there is a pressure drop in vaporisation. And the 30" diameter?

Heat will be absorbed as it vapourises, where does this come from?

What type of pump is it?

1800m is the maximum elevation of the pipeline about 2/3 the way along the route. The final elevation is about 1400m. There are about 6 peaks and troughs along the route. (Final route has not yet been selected).

100bar is the discharge pressure of the pump. The vaporizers are low pressure drop typically less than 1bar.

I am guessing you are asking if 30" is OD or ID. In this size of piping standard pipe dimensions are for the OD. ID is likely to be 29".

Vaporizers have not been selected yet as it depends on the balance of land facility or FSRU (Floating Storage and regassification Unit ie a converted LNG Carrier).

Options would be on land an ORV (Open Rack Vaporizer) that uses seawater.

On FSRU and indirect heater that uses flue gas froma burner to vaporize propane that that condenses on tubes vaporizing the LN.

Pumps will be centifugal - typically a low head (7-8bar diff) submerged pump in the LNG tanks and then a larger standard cryogenic pump (probably a canned type) for the rest of the duty.

OK thanks. As I said it's not my field, and I wouldn't presume to try to answer your query, just thought it might help to clarify.

Pipe diameter - I meant is it 30" handling liquid or gas? I assume dia is much larger after vaporisation. (I know it's OD, 14" and above, nominal dia is OD).

The 30" piping is the Nat GAs sideof the vaporizers.

Thanks CM

Thought so, 30" pipe would carry a lot of liquid LNG!

Aren't there established ways in the LNG industry for designing these systems? You wouldn't want to take any chances with large quantities of LNG. If it's never been done before you would need to work from first principles (after giving it some serious thought!) but that seems unlikely.

It is worth keeping the PSV(s) in the scheme at the moment so that when "value engineering" starts, its/their removal can take place.

This brought a smile to my face this morning

Thanks it will be in the back pocke for cost savings

This seems too obvious to me so I am probably missing something since this is outside my field of expertise. At the same time gas flow and pressure does follow electric theory.

Your source for the gas is a liquefied gas thus over pressurization is a possibility. I know that the vaporization system is regulated to maintain a proper pressure but this regulation might fail. Probably the specified meters are the devices with the lowest safe pressure level and thus near them is a PSV that releases gas at a safe level. Then there are the possible scenarios of back pressure generation at the flow meter. This back pressure can be generated by anything from the proposed back waves from valve closings to improper devices being attached at the dispensing end.

Thanks redfred you've set some thoughts off here.

The whole pipeline would have to be designed for a common design pressure so that it can be tested although you are right that the piping is expected to be good for the limits of the pipe class. I was thinking that setting this DP at the limits of the pump would be adequate but you are quite right that the expansion of the gas in vaporization would increase pressures in a blocked system. I would normally put the PSVs for this at the vaporizers but it needs more thought

Do you have the freedom to change the design?

Eventually.

During the pre ITB and ITB we can ask for clarifications.

We must submit a bid that is compliant with the ITB (and subsequent clarifications). Normally the client will issue composite clarifications during the ITB - this will summarize all the clarifications they have agreed from all the bidders so that we are all working to the same target.

We could also submit a bid to a different design but the client could choose to ignore that. If you want to submit an alternative design you almost always have to submit one that is compliant.

e.g. the isolation shown is inadequate (single block instead of double blocks and bleed) and the kicker line is too small. We will submit these as clarifications and price the bid on the improved isolation and larger kicker line. Although what we do if the client thinks the isolation and kicker line are ok I am not sure!

If we then win the FEED (Front End Engineering & Design) we can then change the design

With a cryo system, you need to protect for phase change "water hammer" conditions, which can generate massive short duration pressure pulses through the system. In this case, you could potentially have a gas filled line and collapse it with a slug of subcooled liquid phase. In that case, the line pressure goes to nearly pure vacuum and ingests more liquid at velocity. In the worst case, the liquid finds an elbow and knocks the pipe off the hangers. More likely the liquid will flash to gas phase as the liquid transits the pipe and forms a bubble ahead of the liquid which is then compressed by the mass and velocity of the following liquid and turns into a compressed high pressure gas pocket. The relief valve vents the high pressure gas at pressure and allows the pocket to act as a cushion in the system to decelerate the liquid phase in a controllable manner.

Hi Jpfalt

I just need to clarify a few things as I an unclear on the mechanism of this phenomenom

I think this requires some upset in the vaporizers (pump failure etc, there are endless options) that lets the the cryo LNG into warm high pressure line.

In this operation the gas is actually dense phase at about 5C so I am not sure that it can be condensed to produce the partial vacuum you mention.

At lower pressures there would be a balance between heating up the sub cooled liquid and cooling the superheated gas. But to then condense the gas the liquid has to vaporize and as they are the same material that would just be a one to one swap.

I think I am missing something here can you point me in the direction of a web site or article.

I think I can see how this would work if the gas was saturated and you introduced a subcooled liquid. In that case latent heat of condensation would go against sensible heating of the sub cooled liquid but I am not sure that case applies in my situation.

But I do know I need to look into this more so thank you and if you have any references I would be grateful

I had a few more thoughts and played around with a simulation program (HYSYS) and I wonder if the issue is the dynamic interface. I can only use HYSYS in steady state but at 40bar (ie below the triple point) when I mix the sub cooled liquid and superheated gas I just get more gas until I get to a liquid rate about 60% of the gas flow. However in this case we have a slug of liquid pressing against the gas so the mass balance will be in favor of the the liquids. This is going to create local vacuums and then depending on the behaviour in the liquid if the warmed liquid is recirculated away from the interface this could keep going pulling the liquid through as you noted.

Sorry if this is all a bit rambling I have been thinking of new things as I composed this post.

Once again thanks for the post its certainly been food for thought

You can start here: https://www.tlv.com/global/TI/steam-theory/waterhammer-steam-distribution-lines.html

The situation occurs I any two phase line. As long as the gas can condense, then a subcooled liquid can collapse the gas, generate a reduced pressure and accelerate a slug of condensed liquid. The geometry of the piping system will determine whether a high pressure gas pocket or a mechanical impact will occur.