Drainage of a 10 in LNG line by displacing with NG? 2

[JustSomeRoark]

Dear colleagues,

A colleague and I have been tasked with coming up with a procedure to drain some 500 m of a 10 inch pipe initially totally full of LNG. We must empty the line into a vessel which is elevated by some 8 m, and the line is on the ground. Unfortunately there are a couple of liquid pockets (i.e. culverts with a height of 2-3 m) which are giving most of the headaches.

The idea is to push the LNG back into the tank by displacing it with NG injected from the far end of the pipe. We hope that injecting the NG will "push" the liquid, which will flow like a piston, back into the tank. We have reservations on whether the LNG will be adequately pushed by the NG in the vertical runs of pipe or if the gas may find its way into the LNG as bubbles which would "break" the piston flow

Do you think it sounds plausible or should we abandon the idea completely?

Regards.

 

[georgeverghese]

LNG will pool back into the culvert low points initially.
But since compressed BOG is much warmer than LNG, if you continue to blow this warm BOG through, the LNG in the culvert sections should vaporise away into the BOG.

 

[shvet]

Gas flowrate should be high enough to avoid stratifying and further backflow of liquid. As per my experience such rate is hard to achieve, especially in a long distance piping having high hydraulic resistance. High-capacity high-pressure gas source is a major issues.

 

[1503-44]

LNG is at -161 C. Injecting NG into LNG will liquify the NG.
Normally one would let the 10" pipe warm slowly and take the vapor out through a vent, possibly using a compressor to move it into a tank as CNG.

I would not try to move liquid with hot gas. At the interface you are at the critical point between phases, LNG wants to vaporise and NG wants to liquify. If LNG vaporises, it's volume increases 600X and pressure rises rapidly. If you don't vent it, pressure Increases and any vapor may return to liquid phase as its volume collapses by 600X and pressures rapidly drop. The volume changes can be accompanied by extreme shock waves and rapid variations in flows. It can be difficult to control at higher temperatures even when moving normal NGL, Propane and butane mixtures. Operating well above critical point, or well below, are preferred.

If you do get it moving, then you still have to deal with the two phase flow problems, bubbles, gas and liquid slugging and potential vapor locks.
This is what happens when the venting is inhibited. Adding more pressure to move the LNG wouldn't appear to be a great idea.

https://www.eng-tips.com/viewthread.cfm?qid=495842

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

What sort of velocity are you talking about?

How much capacity does the receiving tank have for the sudden release of 500m of gas at whatever pressure you are working at?

Doing this historically damages the tank....

My guess is that you'll displace about 90% as liquid LNG if you're doing about 2 to 3 m/sec, then the rest as a mixture of gas and liquid.

The safer method is just to isolate it at one end and let it vent slowly into the tank as it warms up.

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

 

[MJCronin]

I agree with LI ... as you know, LNG is not like most other liquids

It has been my experience that the most spectacular accidents are the result of operating a well-known and established process "far away" from it's original design ... Like, ummmmm... say.... forcing drainage of a very large amount of liquid into a container not designed for a huge surge.

Despite what your prancing boss may tell you, the forced clearing of LNG pools within the main pipe by an uncontrolled gaseous method.... Although forced gas discharge makes your job easier, it will be very, very dangerous and forces unknown loadings on the tank (container) and the massive relief devices

This seems like it would be a commonly encountered problem in your industry ... What do other people do ???

Can anyone else help out here ?

(BTW, IMHO I believe that eng-tips should have a seperate LNG Forum where these questions can be discussed..

MJCronin
Sr. Process Engineer

 

[GBTorpenhow]

In small scale LNG liquefaction and vaporization at least, when a process run is complete leaving a line liquid full, appropriate valves are left open and the liquid product is allowed to safely and slowly boil off back to the tank and boiloff gas handling equipment.

Tank details? With an atmospheric tank there is the obvious risk of overpressuring the tank but 8m sounds like it might be bullet tanks. Even with a pressurized tank as soon as the gas has a path across the liquid traps you will just equalize the tank vapor space with the motive pressure and be done moving liquid. You would have to displace all that volume out of the tank too so boiloff handling equipment might be easily overwhelmed. Trying to force liquid back into a tank with gas pressure in the 10" line is probably not advisable.

You could maybe do essentially the same thing with a small cryogenic pressure vessel at/near the low point(s). Drain some LNG from the line into the vessel, then use gas pressure to push it back into the tank in batches. With a small enough 'trip tank' vessel and a plethora of controls and lockouts you could probably get this to pass a PHA. Might be pretty expensive compared to just boiling off the product in the line.

 

[LittleInch]

You will need to give this service full HAZOP like review, but some more details would help.

As MJC says, without knowing the design of your tank its not ready to say what will happen when you get gas blow by, which will happen.

I've seen huge storage tanks get lifted off the ground when gas gets in the wrong place.

This has all the prospect of being featured in the engineering disasters forum at the moment...

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

 

[Caloooomi]

25 m3 of LNG? What's at the other end of the pipe?

 

[georgeverghese]

LNG is a low viscosity liquid, so plug/piston type flow with high backpressure buildup upstream of the 8m riser is unlikely. Keep BOG flows low, and keep an eye on pressure upstream of the culvert sections.

 

[pierreick]

Hi,
What about pushing back the liquid with hand pump similar to the one used for pressure test?
Safety is key.
my 2 cents
Pierre

 

[georgeverghese]

Oops, just saw this line is full of LNG, and not just the culverts, so it will be a large plug of LNG going into tank behind which will be high pressure gas with a pressure of at least 50kpag. What provisions do you have at the moment for draining out most of this line ? And what is the process design pressure of this downstream tank ?

 

[georgeverghese]

2 solutions come to mind, there may be more :
a) Install heat tracing in the culverts. Thermal input duty / LNG vaporisation mass rate of all tracing circuits to not exceed sum of (i) BOG gas compression unit design feedrate (ii)any excess BOG to flare venting capacity. Tracing heating method may be electric or warm BOG or warm N2 (but not steam).
(b) Pressure drain out LNG in the reverse direction towards BOG compressors, and divert to an existing cold drain vessel through a new jumper line (teed off this pipeline from downstream of BOG compressors) installed with a restriction orifice, PIC-PCV and a drain drum PSH. Pressurise with warm BOG.