• This topic has 25 replies, 13 voices, and was last updated 5 years ago by  J-R.
Viewing 26 posts - 1 through 26 (of 26 total)
• Fluid dynamics and liquified gas question.
• Imagine a gas cylinder with a slightly rounded top with a fill hole in the centre.
Now, if you were to tilt the cylinder by about 30 degrees and fill it with water, you would end up with an air space in that part of the cylinder above the fill hole. You can see this effect by holding an open clear plastic water under water at an angle and letting it fill. The bottle need to be near upright to fill to the brim.

What if you connected two gas cylinders top to top though, assuming it’s a liquified gas under pressure like butane?
Would you get the same effect,with a small volume of gas at the top, or would it fill with liquid?

The air gets trapped above the water because it is less dense. It can’t go upwards, because of the container and it can’t go downwards because of the water.
My suspicion is that the trapped gas will be compressed by the liquid gas flowing in to the cylinder, until it too becomes liquid.
Am I right?

By the way, there’s no pump involved, the liquid is flowing from the upper to lower cylinder entirely by gravity.

Stoner
Member

from my uneducated view, Id say you could replace “liquid gas” and “gas” for water and air. i.e. the gas cannister would behave like the water filled bottles…assuming that as you say, the incoming liquid is not under pressure.

Although you do say earlier: “What if you connected two gas cylinders top to top though, assuming it’s a liquified gas under pressure like butane?”

SO lets assume you mean you have balanced pressure between the upper and lower cannisters.

The gravitationally created pressure of the liquid wont be enough to condense the vapour in the lower cannister.

boblo
Member

Just a small point; there isn’t any air in the part empty canister at the bottom. When gas bu

gonefishin
Member

Assuming that what you have in the two bottle is the pure gas (butane in this example) rather than a liquid and an uncondensable gas (your water air mixture) then the pressures in the two cylinders will equalise and you will be able to fill the tank (the gaseous Butane will just condense into liquid). I’m assuming that there is sufficent elevation difference between the two tanks to achieve this, although it may take some time.

Stoner
Member

(the gaseous Butane will just condense into liquid)

will it though? Doesnt that depend on the mass/pressure of liquid in the upper can?

hot_fiat
Subscriber

You see, this is why I hated thermo. It’s a compete head****!

As drummed into me by the last person whips tried to teach me this: “It all depends where you draw the boundary.”

teacake
Member

I don’t fully understand your setup but if the cylinders are connected together by a hose the pressure inside the whole system will be approximately the same. The only thing which will affect the pressure at any given point will be it’s height relative to another point. As we’re only talking a metre difference in height – it’s safe to assume this has little effect relative to the larg pressure (several bar) inside the system.

Does that help?

nickjb
Member

The trapped gas will ‘compress’ to a liquid state until the pressure is too low for that to happen. Propane certainly does and at a pretty low pressure. If you want better gas transfer warming the doner tank with warm water helps a lot. We do it regularly.

Some tanks have a dip tube so you take liquid propane. Others don’t so will transfer faster if you turn them upside down to get to the liquid but you run the risk if transferring crud and rust.

gonefishin
Member

will it though? Doesnt that depend on the mass/pressure of liquid in the upper can?

Not really. Butane in a cylinder (like a stove cylinder) is a saturated liquid and so is always at its boiling pressure (assuming there is sufficent present to actually form the liquid) with it’s temperature being defined by the surroundings. Adding any additional liquid butane to the lower cylinder will reduce the volume of the vapour causing the pressure to rise, however this rise in pressure can’t be maintained at the cylinder temperature so some of the gas has to condense (think of steam at 100C and atmospheric pressure, if you squished by reducing the volume of the steam some of it would condense).

This is true for things like Butane and Propane at ambient temperatures, but it can get a bit more complicated for other “gasses” due to different critical temperatures.

A bad illustration to, er, illustrate the question.

Both cylinders are filled with liquid to the dotted line and gas above.
The pressure within the whole system will be equal.
My guess is that the weight of the liquid in the top cylinder will compress the gas in the bottom one, allowing some of the liquid in the top one to flow through until the bottom one is entirely full of liquid.

Stoner
Member

If I understand gonefishin correctly it’s not really a “compression” of the lower gas as much as it readily condenses.?

gonefishin
Member

The compression causes the condensation, yeah. From that sketch you will end up with one (rather oddly shaped) cylinder with a single gas “cap”.

I may be using the wrong terminology, but yes, the gas will condense under the weight of the liquid in the upper cylinder.

Another way to look at it is that in my underwater bottle example above, the air is trapped between the water and the bottle. That air has got to go somewhere before that space can be filled with water.
With the gas cylinders, it’s all gas. The gaseous gas doesn’t have to go anywhere, it just turns in to liquified gas.

thisisnotaspoon
Subscriber

Depends on the pressure and temperature, a condensing gas kicks out a lot of heat*. The static head of a few cm of liquid won’t be enough to substantialy increace the pressure in the lower bottle. So assuming you’re not adding or taking away energy from the bottles you won’t get any condensation/evaporation, and therefore it will behave just like the water/air mixure.

This all falls appart if you’re working arroud the critical point of the fluid, at the critical point the latent heat of evaporation drops to nothing, so there is not liquid and gas, just supercritical fluid. Then you could get evaporating/condensing occouring with very little heat transfer into or out of the system.

You can fill a tank with liquid from a line containg compressed gas, but only via the Joules Thompson effect, which is what you observe when you let the pressure out of a tyre quickly and the valve freezes.

*heat as in the temperature effect of enthalpy, not as in it gets hot.

boblo
Member

The way they do it us to cool the receiving canister and slightly warm the donor.

here

And just to clarify, this isn’t something I’m trying to do, I just wondered if it’s as dangerous as I think it is.
If you were trying to drain the dregs of several gas cylinders in to one using the technique in my drawing above, then I think you would end up with the bottom cylinder 100% full. Tilting it like that won’t help to create any ullage like it would if you were just pouring water in.
If you fill a cylinder to 100% in the shade, then take it out in the sun, the pressure will increase, possibly beyond the design limits of the cylinder.

thisisnotaspoon
Subscriber

No, tiliting it will create a vapour space. You can compress this vapour as the pressure increaces, but you can’t condense into this space without removing energy form the cylinder.

If you want to be safe, weigh a new cylinder from the shop and make sure the filled weight is less than that. The pressure in the cylinder will be equal to the vapour pressure of butane and the temperature, so as long as you don’t fill the cylinder so much that the normal expansion of the butane would fill the can then you only need ‘some’ vapour space. Having more vapour space won’t decreace the pressure, that’s why an almost empty container still provides almost the same pressure as a full container.

Don’t do it with propane, and don’t do it with propane/butane mix, as propane is much higher pressure, and you don’t know the concentration of proapen in the drgs of the cans (it should be less than the starting figure though).

Sundayjumper
Subscriber

TINAS is correct. It’ll behave just like your water / air example.

gonefishin
Member

You can compress this vapour as the pressure increaces, but you can’t condense into this space without removing energy form the cylinder.

Natural heat loss to the environment will achieve this as the system will more closely resemble isothermal than adiabatic, although the time to achieve it may well be considerable.

In reality though there will be a small amount of incondensibles in the vapour space that will mean that it will never clear completely. It is however a dumb ass idea as the container will empty on it’s own as you use it so I’m not really seeing what you would achieve by trying it.

If you fill a cylinder to 100% in the shade, then take it out in the sun, the pressure will increase, possibly beyond the design limits of the cylinder.

Believe it or not we do actually consider things like maximum temperature when designing pressure systems like these. ðŸ˜‰

pjt201
Member

the biggest problem i have with this is (if you are trying to drain a number of partially empty cannisters into another empty one to make a full canister) that as soon as the receiving canister has a pressure (volume of gas in this case) higher than the filling canister won’t it just put gas back into the filling canister in order to equalise the pressure?

boblo
Member

That’s why you cool the recipient and warm the donor…

tomd
Subscriber

Believe it or not we do actually consider things like maximum temperature when designing pressure systems like these.

He’s talking about a liquid full system, not designing for the maximum vapour pressure at the maximum ambient / solar gain temperature. I’m not totally sure what happens to these kind of cylinders if you do fill them completely and then heat them up, I assume that they are designed to leak at the top rather than fail completely (Glasgow Airport comedy car bombing suggests this is the case).

Gonefishin, I’m aware properly designed systems incorporate safety features to prevent overfilling.
The LPG tanks on my Land Rover came with instructions to ensure that they are fitted at the correct angle so that the internal float valve would shut off the supply at 85% full.

I’m still not convinced that a gas pocket would remain above the liquid in the lower cylinder.
Going back to the water & air example, the weight of the water in the whole system would pressurise the air in the lower cylinder slightly above the pressure of the air in the upper cylinder.
Wouldn’t the weight of the liquid gas do exactly the same to the gaseous gas, turning it to liquid?

The gas in a sealed system like this must, by design, be at exactly the pressure where it condenses to liquid.
It would only take the smallest increase in pressure from the weight of liquid above to condense the gas in the lower cylinder, while a similar amount of liquid in the top cylinder evaporates.

boblo
Member

Right necessity being the mother of invention:

I made an adapter out of brass tube about 25mm long. I slotted the ends so they wouldn’t seal against the inside of the valves and gas/liquid could transfer. I then sleeved the brass with a length of silicone tubing to form a seal around the outside of the lindal valve when compressed:

I’ve just transferred 50g of fuel from one canister to another. One now weighs 100g the other 150g. The max for these is 190g when full. This was at room temperature and I’m not dead but expect to be at some point in the next 40 years.

Subscriber

Its not pressure increase which condenses gas, its removal of heat energy by cooling.

Due to the pressure differential tho, between the upper and lower vapour pockets, the lower pocket would slowly condense (the liquid itself cooling the pocket), whilst the upper pocket would get larger due to the dropping liquid level which would cause the pressure to drop, but ofcourse any pressure drop here will cause some evaporation ensuring the pressure stays the same.

J-R
Member

Graham, if I have understood your diagram properly I believe:

1) Some liquid will flow out of the top container into the bottom container due to the effects of gravity (we call it liquid head)

2) As this happens the vapour in the lower container will be squeezed and it’s pressure will increase. Some of the vapour will condense, and the heat of condensation will warm the remaining gas, so it is now gas at a higher pressure.

3) In parallel the opposite will happen to the vapour in the top container – it will expand, cool and and up at a lower pressure.

4) Soon the pressure difference between the two vapour pockets will be equal and opposite to the gravity effect and the liquid will stop flowing.

5) In the longer term the flow of liquid will continue very slowly from the top container to the bottom one, as the differences in temperature tend even out and the pressures in both vapour spaces try to return to their original values. Depending on the size of the original vapour spaces and the properties of the liquid, the original vapour space could disappear entirely.

Please do not try this at home with any sort of pressurised flammable liquids.

Viewing 26 posts - 1 through 26 (of 26 total)

The topic ‘Fluid dynamics and liquified gas question.’ is closed to new replies.