Why do tanks get hot when you fill them from higher pressure tanks?

[jimmyw]

You are a scientific juggernaut.

No I am not. I'm no scientist - just a long term recreational shallow water diver. But, I had asked this question (several times) before and I settled every time for the claim that gas was being compressed and that was the source of the heating. Before coming here to ask the experts I thought about this question a lot and tried to answer it. I didn't know the answer, but I decided it was that claim that lay at the source of my discomfort over the explanations that others gave. I went over it and over it. Does gas expand or compress when you let it out of one tank to fill the volume of two tanks? I could not convince myself that's what a scientist would call compression. It just isn't. The gas ends up expanding to take up more space - two tanks instead of just one. Some gas expands a lot and then gets recompressed a little, but the expansion is always more than the compression - otherwise the gas wouldn't end up taking up twice the space.

 

[RJP]

If I close the fill valve and stop the process of expansion into the "vacuum empty" tank after 10 seconds, the gas will be hotter in that tank. We now have gas in the originally vacuum empty tank, just like the gas in a not-vacuum empty scuba tank.

Reallity keeps getting in the way of your flawed theories...

10 seconds of transfilling is about 9.99 seconds longer than it would actually take to fill a "fully evacuated" scuba tank from 0psi to 14psi (atmospheric pressure) from a donor tank pressurized to 3000psi.

Think about this: Imagine you had a tank at 0 psi (vaccuum) with a stopper that you could simply pull out of the tank opening. If you pulled that stopper out, how long would it take that tank to equalize from 0psi to 14psi. A quarter of a second? Half a second? Now put 3000psi of pressure behind that equalization from 0psi to 14psi. How long would that take? Let's assume it's painfully slow at half a second. The next 9.5 seconds of air going into that tank is being compressed. Follow along with me

One Mississippi...

Two Mississippi...

Three Mississippi...

Four Mississippi...

Five Mississippi...

Six Mississippi...

Seven Mississippi...

Eight Mississippi...

Nine Mississippi...

... and a half.

If you want I can stop at my shop on the way home and transfill an empty tank from a 3,000psi tank for ten seconds and tell you how many PSI of gas I got into that tank. You can then calculate the compression of that gas and tell me the temperature.

---------- Post added February 27th, 2013 at 05:12 PM ----------

I'm no scientist.

Finally, something I think we can all agree upon!



PS - while I do think you're dead wrong here, and can't figure out how someone can be so far off-base, I do appreciate your willingness to tangle on this issue. I'm busting your chops a bit... and you seem like a good sport. Hope you don't take my digs as more than "good clean fun" - Ray

 

[jimmyw]

I still think it's cool that the fundamental reason a filled tank gets hot is different when filled from the compressor directly (where the compressor adds energy to the gas) and when the tank is filled from another tank (where no energy is added).

I'm going to add one more detail to the process that I think is interesting. When the gas in the donor tank is allowed to expand into two tanks, it expands approximately isentropically, meaning it gives up energy. That energy is mostly transferred to the gas departing the donor tank and heading to the receiving tank. When that gas slams into the receiving tank, the kinetic energy of motion of the departing gas (which could have been captured in the transfill whip with a pneumatic motor/generator) is converted to heat, and this raises the entropy level of the gas in the receiving tank. However, we now have a heat differential between the gas in one tank and the gas in the other tank. That heat differential could be sued to power a heat engine. Where did that energy come from? It comes from the expanding gas in the donor tank. Even with the transfill whip process, some of the energy that could have been captured from the expanding gas in the donor tank is always stored in the heat differential. Of course, if we let the two tanks equalize in temperature by letting the gas in the two tanks mix or by connecting them thermally, that stored energy in the heat differential is lost and goes into increasing the entropy of the total system. Until this happens, the total system hasn't achieved "Joule expansion" which requires all the gas to be at the same temperature = the original starting temperature of the gas in the donor tank.

OK, it's just a minor detail, but it explains one factor in the math that was bugging me.

 

[lowviz]

...//... PS - while I do think you're dead wrong here, and can't figure out how someone can be so far off-base, I do appreciate your willingness to tangle on this issue. I'm busting your chops a bit... and you seem like a good sport. Hope you don't take my digs as more than "good clean fun" - Ray

Same. Why not fill out your profile, join us, and stop busting OUR chops.

 

[RJP]

I still think it's cool that the fundamental reason a filled tank gets hot is different when filled from the compressor directly (where the compressor adds energy to the gas) and when the tank is filled from another tank (where no energy is added).

Aarggggghhhhhhhhhhhh!

There IS energy added when a tank is filled from another tank! That energy is stored in the donor tank in the form of 3,000psi of pressure. 1500psi of that energy is "lost" from the donor tank (which cools) and is "added" to the receiving tank... WHICH HEATS UP!

 

[jimmyw]

10 seconds of transfilling is about 9.99 seconds longer than it would actually take to fill a "fully evacuated" scuba tank from 0psi to 14psi (atmospheric pressure) from a donor tank pressurized to 3000psi.

So make it 0.01 seconds, or open the valve a lot less

Think about this: Imagine you had a tank at 0 psi (vaccuum) with a stopper that you could simply pull out of the tank opening. If you pulled that stopper out, how long would it take that tank to equalize from 0psi to 14psi. A quarter of a second? Half a second? Now put 3000psi of pressure behind that equalization from 0psi to 14psi. How long would that take? Let's assume it's painfully slow at half a second. The next 9.5 seconds of air going into that tank is being compressed. Follow along with me

Yeah - good point. I can see you have a good feel for reality.

PS - while I do think you're dead wrong here, and can't figure out how someone can be so far off-base, I do appreciate your willingness to tangle on this issue. I'm busting your chops a bit... and you seem like a good sport. Hope you don't take my digs as more than "good clean fun" - Ray

Ray, thanks for this. I think you're wrong, too, but no more wrong than I was, and you seem like a decent guy, too. I do think I understand it now, but I also recognize that being out on the limb as I am I need to be sure of my footing. I'm not trying to convince others, nor stir up trouble. I really do think that gas leaving the donor tank expands and I really do think it gets hot. It's the "why" I needed to work out. If anyone is able to knock me off this branch, they are going to have to do it with some convincing arguments, and each time I go to the text books or the math or Wiki, I find that the branch gets just a bit wider. It would just be nice to find someone who understands the arguments and can refute them or explain with decent support why they are wrong.

---------- Post added February 27th, 2013 at 04:34 PM ----------

Aarggggghhhhhhhhhhhh!

There IS energy added when a tank is filled from another tank! That energy is stored in the donor tank in the form of 3,000psi of pressure. 1500psi of that energy is "lost" from the donor tank (which cools) and is "added" to the receiving tank... WHICH HEATS UP!

Yes, I agree. The gas in the receiving tank gets energy and that energy is the source of the heat.. But the total gas in the system - the original gas in the donor tank, gets no additional energy. The fill process separates out the expanding gas into two parts - the gas expanding in the donor tank, from which energy is extracted, and the gas expanding out of the donor tank into the receiving tank, which picks up that additional energy. The expanding gas in the donor tank is doing work and cooling down. The expanding gas going to the receiving tank is not doing work and stays at constant temperature by "free expansion = Joule expansion" and carries the energy released by the expanding gas in the donor, ultimately resulting in heated expanded gas above its original temperature. I think it's interesting (cool! ) that expanding gas going to the receiving tank gets hotter when no energy is added to the entire system of gas and the system is able to separate itself into two expanding parts - one getting hotter and one getting colder.

The difference is this. When direct filling from a compressor, you start with 40 cf of air at STP and compress it into a mere 11 liters of the tank to end up at about 1500 psi. You have to add energy (plug in the compressor) and the compression results in heat. In the case of transfilling, you start with 11 liters of air at room temp and 3000 psi and let it expand out into 22 liters of volume at about 1500 psi. No energy is needed - you just open the valve and release previously stored energy. One is compression heating and one expansion with a system that divides itself into hot expanded gas and cold expanded gas. In one you don't need to have ever heard of Joule expansion. To explain the other, you need to understand Joule expansion - it's key to how it's possible for expanding gas to heat up.

 

[rhwestfall]

My intuition is that you live in a cold place... cold in a way that keeps you from diving......and thus doing this...... Please, please please either find a way to go diving, or find another hobby to occupy your non-diving time....

 

[steveharriss]

awap...you must be right

 

[rjack321]

The difference is this. When direct filling from a compressor, you start with 40 cf of air at STP and compress it into a mere 11 liters of the tank to end up at about 1500 psi. You have to add energy (plug in the compressor) and the compression results in heat. In the case of transfilling, you start with 11 liters of air at room temp and 3000 psi and let it expand out into 22 liters of volume at about 1500 psi. No energy is needed - you just open the valve and release previously stored energy. One is compression heating and one expansion with a system that divides itself into hot expanded gas and cold expanded gas. In one you don't need to have ever heard of Joule expansion. To explain the other, you need to understand Joule expansion - it's key to how it's possible for expanding gas to heat up.

So in your universe there's: 1) heat of compression and 2) heat of expansion? Have I got that right? Is there compression cooling? How does that work?

 

[RJP]

In the case of transfilling, you start with 11 liters of air at room temp and 3000 psi and let it expand out into 22 liters of volume at about 1500 psi.

No. You start with 22l of gas. 11l in the donor tank at 3,000psi and 11l in the receiving tank at 14psi. You keep conveniently failing to deal with this tidbit of reality in all your theories and calculations.

No energy is needed - you just open the valve and release previously stored energy.

No. That's the point. Energy is needed. The "previously stored energy" in the 3,000psi is no different than the "previously stored energy" that comes from the wall to the compressor in the form of electricity, which may have come from energy that was previously stored in an atom, a fossil, the weight of water falling through a turbine in a damn, etc, etc. You seem to think that the air in the tank knows or cares how it was compressed and should - or should want - to behave differently depending on where the energy of compression came from.

The difference is this. When direct filling from a compressor, you start with 40 cf of air at STP and compress it into a mere 11 liters of the tank to end up at about 1500 psi. You have to add energy (plug in the compressor) and the compression results in heat. In the case of transfilling, you start with 11 liters of air at room temp and 3000 psi and let it expand out into 22 liters of volume at about 1500 psi. No energy is needed - you just open the valve and release previously stored energy. One is compression heating and one expansion with a system that divides itself into hot expanded gas and cold expanded gas .

No. There is no expansion. You start with 22l of gas. 11l in the donor tank at 3,000psi and 11l in the receiving tank at 14psi. You keep conveniently failing to deal with this tidbit of reality in all your theories and calculations.