Guidelines/References for Clearing/Blowing Liquid out of a Line with Nitrogen

[Latexman]

We have a tank where one or more different chemicals are charged, mixed, and then fed to the reactors. After the last addition and before mixing, we currently flush the line to the tank with water to get all the active ingredients into the tank.

We want to stop using flush water and use nitrogen to blow the ingredients into the tank.

All the two phase flow information I have found so far deals with continuous flows of liquid and a continuous flow of gas, L/G is fixed. What we will have is a dynamic situation where the quantity of liquid decreases (full pipe to empty pipe) over time. I suspect the resistance to flow of the nitrogen will vary also, from liquid to two phase to gas flow.

I've done a little thinking on how to design this, but I figured someone else has probably done this before, maybe many times before, so . . . does anyone have a good reference, guidelines, or advice on this?

Thanking you in advance for any help you can give me!

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[BigInch]

It will be most affected by profile elevations and gas to liquid flow ratios. I assume you have references that cover the various liquid, bubble, slugging, wavelet, droplet and mist flow regimes. Transient flow can often be broken into a simple series of steps between each of the various regime states. It then only becomes a matter of how fast you make the transitions between one state to another. If you do them slowly the transitions will be relatively smooth, quickly, you may skip a few regimes.

For the most part, if you have enough nitrogen to initiate flow of the liquid column, and sustain it enough to evacuate it in one go, you quickly run out of liquid and on into mist flow, however quickly depends heavily on length and profile.

 

[zdas04]

You could significantly improve your sweep efficiency (and hugely reduce your nitrogen costs) by putting pigging valves in the lines and cleaning the lines with pigs instead of sweeping them with nitrogen directly. You don't say what size lines you are talking about, but pigging valves are available for every size from 2-inch through about 16-inch.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[bimr]

It depends on the line size. Don't think that blowing the lines is practical if the line size is greater than 1.5-Inch.

A larger diameter pipeline when blown may allow the gas to pass over the liquid, particularly if there is a change in elevation.

Have you considered having looped piping systems for the chemicals? If you have a looped system, you can takeoff the chemicals at the point of application using a valve and a short length of discharge pipe.

Are any of these chemicals air pollutants? You may need an air pollution control to collect the blown material.

 

[LittleInch]

Latexman,

You must know the score here, details.....

Does the pipe go up, down, level, top of tank, can you blue at high velocity?, what is your venting rater?

Size, flowrate, viscosity, velocity sill make a huge difference as to whether this is feasible or not.

Give us something to work with here.

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

 

[Latexman]

Sorry folks, it's a small mix tank about 300 gallons and 150 psi MAWP. It operates at 0-2 psig. The vent valve is set at 2 psig. The line to be blown is 1", Sch. 10S, 150 psig SS pipe spec from a drum add station. It's about 75 feet long, a dozen elbows with 2 ups, 2 horizontals, and 1 down into the lower straight side of the tank, which will be below the liquid level by the time it is blown. The drum add tank nozzle could have 5 feet of liquid head on it at the most. The tank vent valve exhausts to the plant's vent system which goes to a relatively huge catalytic oxidizer. All the drum adds have a water-like viscosity with s.g.'s of 0.8-1.05.

I'm sure I forgot something important.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[Latexman]

bimr,

We do have a few dedicated, high volume monomer loop systems, but we can't afford one of those for every monomer. Thus, it's a common drum add station that charges about a dozen different monomers.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[MikeHalloran]

I agree with Zdas04 that pigging sounds like a good idea.
I am also dimly aware that ice pigging technology (really slush pigging) exists, and might be adaptable to your smal lines and elbows.

Mike Halloran
Pembroke Pines, FL, USA

 

[zdas04]

The smallest pig valve that anyone makes is 2-inch. I don't think it would take much to make a 1-inch pig launcher and receiver. With the piping configuration you are describing, I would be shocked if you can flow enough nitrogen to get any kind of descent sweep efficiency.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[IRstuff]

Your concept seems predicated on the fluid leaving no residue to contaminate whatever goes next. Is that valid assumption?

Seems to me that you're not likely to simply push the fluid; a lot of it will evaporate, so the vapor will likely go wherever the nitrogen goes after the tank.

TTFN
I can do absolutely anything. I'm an expert!
homework forum: //

http://www.engineering.com/AskForum/aff/32.aspx

faq731-376 forum1529

 

[BigInch]

75' of 1"

If you got 70' closer, maybe.

 

[LittleInch]

lateexman,

I think this depends on what you're really trying to achieve. Is it simply to get ~90% of the liquid in the line into the tank before mixing OR strip the line of nearly 100% of the liquid to avoid contamination of the system the next time?

If the former the key velocity will be the one that maintains a virtually solid interface between gas and liquid in the vertical section. My estimate is that would need to be in 10 m/sec range, but that might generate some significant pressure drop and if the control is such that you would just stop injecting nitrogen after you hear it enter the tank, then you could generate a lot of vapour.

Much slower than the range I gave above and you risk getting a significant blow by of the gas on the vertical up and horizontal legs. Suspect given your piping layout this will be a matter of trial and error to see how much of the pipe contents make it into the pipe.

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

 

[racookpe1978]

To bad there are bends and instruments and valves in the line. If not, you could "freeze" a short piece of the product into a slug small enough to act as a pig, then blow it through the line to push out the liquid product with the nitrogen.

 

[Latexman]

The main reason the plant wants to do this is winterization. It’s near Chicago, and we have problems with the water freezing (no shock there!), even though it is well traced and insulated! The issue is we use a bulk monomer as a carrier to always have about the same total quantity and same flow rate. This makes control a snap. However, in winter the bulk monomer gets as low as -5^oC. With a little bit of water and a lot of cold monomer, the water crystallizes and plugs strainers, control valves, and nozzles. We don’t have time for the tracing to warm up the tank. We could heat the monomer coming in, but that rings all kind of alarm bells with safety folks when it is not done in a proper Reactor. We think the ultimate solution is to have a flushless design; no water. No water also progresses us to a long term goal of higher solid products.

Zdas04 and MikeHalloran – I agree, if we could "swine the line", that would be an excellent solution. But, winter is on the way! I don’t think we have time to help develop the 1” pigging technology at this time. Plus, it would be relatively expensive IF blowing with nitrogen works good enough. If I see it will not work good enough, I may have to revisit pigging.

Racookepe1978 – those dang elbows ruined your brilliant idea! Thanks though!

LittleInch - We don’t have to get 100% of the liquid out of the line. I’d like to see the line 95% empty. These monomers are not incompatible from a safety stand point. We just want to keep the contamination low enough that quality is not affected. We may have to pay some attention to which product follows which product though. 90% empty would be about the minimum we could stand. I did a little Googling yesterday on this line clearing subject and came up completely empty. Today, I’m going to search some more and see what I can find. Then, I’m going to take what I find and your advice and run some numbers.

More later!


Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[zdas04]

I think Racookpe has a brilliant idea. If you have had water in the line in the past and the process can tolerate a bit of water, freeze a 6" section and blast it with 20 psig N2. The ice plug will get torn up, but who cares. Sounds like something you can do for the price of a bit of tubing and a few pounds of nitrogen and you don't have to recover the pig.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[racookpe1978]

Elbows, eh?

What you need is something that will go through the bends while not twisting or rolling over: The axial length of the total effective "plug" needs to be longer than the ID of the pipe (minus the maximum expected tolerances and any weld inclusions obviously)

Freeze two shorter plugs with a compatible cord between them.
Get fancy and make a two-part wooden mold for the plug like a pair of bar bells (two spheres on a central smaller diameter section supporting both ends)

 

[bimr]

Ice pigging is typically used to clean pipes. Ice is pumped through the pipes. I doubt it would be of much value here when you are concerned with water freezing.

Have you considered pneumatic transport of the chemicals? You can be completely water free.

 

[Latexman]

Currently, I'm researching pneumatically conveying the last full pipe of liquid into the tank.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[LittleInch]

Seems it has been looked at before - Note the reference to 10m/sec!

http://www.google.com/patents/US7950403

or inject a bit of this stuff and then blow.

http://aubingroup.com/products/pipelines/pipeline-gels/

Or use these things!

http://www.ricoservices.co.uk/products/?by=cat&id=19&reset=1

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

 

[zdas04]

LittleInch,
I've used the pipeline gels for cleaning pipe, and they do that very very well. The first time I tried it, I took the manufacturer's name (gel pig) literally and just poured the goop into the launcher chem injection port and opened the gas. What a mess that turned out to be. It really needs a pig in front of it to keep it pushed to the top of the pipe and one behind it to keep it moving (the video omits the leading pig, but I found that it was hard to get much gel to the top of the pipe without it). With two poly pigs (not spheres, spheres made a horrible mess because they could roll over the gel) it is fantastic, but I don't think it really applies here.

The pellets look interesting if there is a way to get them into and out of the line.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist