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

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

I think the point of the OP is that overall the net pressure is dropping from the full tank to the empty one, therefore why is there not a net cooling effect.
Yes. That's my point. Overall the net pressure is dropping. You can't find any volume of the original gas in the donor/fill tank that isn't at a lower pressure and larger volume after the fill operation, yet the gas that remained in the donor tank got colder and the gas that moved to the scuba tank got warmer.

---------- Post added February 22nd, 2013 at 09:15 AM ----------

Whatever pressure the gas used to be at is not relevant.
This just can't be right. I want to know what happens to the temp pressure and volume of some gas after it flows into another tank and you say it doesn't matter what temperature or pressure it started at?

Edit: To add a little bit. An individual molecule has no "pressure". Pressure is the sum of all the collisions between the individual gas molecules and the wall of the tank. More collisions per second= higher pressure.
This is true, but it doesn't really address the question. and I don't want to get sidetracked. Pressure also must take into account the mass of the molecules hitting the walls and the speed they hit at. The speed relates to temperature.


Adiabatic cooling by its very definition involves zero transfer of heat or energy. Its caused simply the change in pressure.
And the entire process of filling a tank from another tank is an adiabatic process. We add no heat to the system making up the donor tank plus the scuba tank if we wrap them in an insulator and for simplicity, that's what we should assume.

Has absolutely nothing to do with nozzles and venturi's etc. They only serve to slow down the process slightly and limit the ability for the gains and loses to equalize via the gas through the hose.

I've been working on this all night and I'm now convinced that the orifice/nozzle is key to why one tank gets colder and one gets hotter. I've managed to convince myself that if the orifice is instantly made very large so that the two tanks equalize suddenly, the temperature of the gas wouldn't change.

---------- Post added February 22nd, 2013 at 09:24 AM ----------

He is right. Ignore the source. Pretend you just have a valve on the wall the gas comes out of.
No, he's not right. The gas coming out of the wall has a temperature and a pressure. That temperature and pressure affects the final state of the gas. If the gas comes out cold, it ends up colder. Gas laws apply to the gas, not to the containers for the gas. We can't just say that the pressure in a tank is increasing. We have to look at the gas in the tank and apply the gas laws to that gas.
.

---------- Post added February 22nd, 2013 at 10:00 AM ----------

So why does the high pressure gas that starts in the donor fill tank get hot when it ends up at lower pressure and larger volume in the scuba tank? This is hard. It's very hard. You're missing something that's very fundamental in this whole process. It might help to think of the scuba tank as being welded to the side of the donor/fill tank and then punch a hole at the wall. The total amount of gas in the donor fill tank expands in volume and drops in pressure. By the simple rule it should all cool off.

Or think of the scuba tank as being a cylinder welded to the wall of the donor fill tank that's open to the donor fill tank and has a piston that's pushed against that wall. Then you slowly retract that piston until the cylinder has the volume of the scuba tank. Again, the total volume of gas in the donor/fill tank will have increased and the temperature should decrease. There's something very interesting going on here.
Can you explain in more detail exactly what's happening?
Yes, I can, and it was thinking about this that led me to what I now think is the answer.

I do appreciate all the comments here. They've really focused my thinking on this. It's been bugging me for more years than I want to admit and I finally decided I had to try to get to the bottom of it.

Here's a brief summary of what I think is going on. The gas that remains in the donor/fill tank undergoes reversible expansion and cools adiabatically. The remaining gas adiabatically and reversibly gives up energy in the form of kinetic energy that's transferred to the escaping gas. Basically, the escaping gas is getting pushed along by the compressed gas behind it. The escaped gas in the scuba tank also underwent expansion and cooling, but it was given kinetic energy as well as it was driven by the compressed gas behind it. That kinetic energy is transformed into heat when the escaping gas crashes into the walls of the scuba tank and the other molecules confined therein. That heat is more than enough to overcome the adiabatic cooling of the escaped gas in the scuba tank and it adds enough energy in the form of kinetic energy converted to heat to make the scuba tank hot. The gas in the scuba tank undergoes an irreversible non-adiabatic process.

Here are a couple of other answers to questions I had as people posted here or as I tried to explain my confusion and some answers I found while researching this.

Remember the two 1cc cubes of air in the donor tank I was thinking about? One stayed in the donor tank, expanded and cooled. One moved to the scuba tank, expanded and heated. The heating came from the kinetic energy the 1 cc cube was given as the other 1 cc cube expanded.

In the quote above I imagined the two tanks being tightly bound together so they could transfer heat. In that case, where the the two tanks are tightly bound thermally and all the heat from the hot scuba tank is allowed to heat up the cold donor tank, the final temp of both would be exactly the same as the starting temp. I was able to determine that there's even a name for that type of expansion: "Joule expansion." Wikipedia has an article on it that would make you think the donor tank and scuba tank should remain at room temp. Joule expansion is an irreversible process.

I wondered what would happen if the scuba tank was a cylinder with a piston and I slowly let the piston move away to slowly expand the scuba tank gas. In that case, the gas in both tanks would expand reversibly and both tanks would cool. The difference is that I would have to hold the piston as I moved it to do that, and that would extract energy from the system instead of producing kinetic energy that converts to heat.

I also asked what would happen if I "slowly" moved the 1 cc cube from the donor tank to the scuba tank? To do that, I'd have to resist the pressure differential (a force) while moving a distance. That means the gas in the donor tank would be doing work on me (force times distance) while I made that move. I could store that energy. If I did store that energy, the 1 cc cube of air in the scuba tank would expand adiabatically and reversibly and would cool.

Here's another thought experiment. Put a pneumatic motor between the donor fill tank and the scuba tank and use the pressure difference to turn that motor. connect the motor to a generator to make electricity and store the generated power in a battery. If we did that, the scuba tank and the donor/fill tank gas would both get cold. The pneumatic motor would be capturing all the kinetic energy of the escaping gas. If all those processes were perfectly efficient, the energy stored in the battery would be just enough to run a perfect compressor and put the gas from the scuba tank back into the donor fill tank. The adiabatic heating from that process, like the first direct compressor fill process, would be just enough to bring the donor tank (which got cold as we filled the scuba tank) back to its starting room temperature. By using the motor and capturing the energy we put originally put into the donor/fill tank all the gas undergoes a "reversible" expansion and all the gas cools. Using the compressor to recompress the gas in the scuba tank back into the donor/fill tank is just the reverse of that reversible expansion/cooling process which produces the normal heating by adiabatic compression. that's what "reversible is all about - no change in entropy lets us go back and forth between expansion/cooling and compression/heating.

Alternatively, we could have used that stored battery energy to run a heater. Again assuming everything was perfectly efficient, the energy stored in that battery would be just enough to heat both the donor fill tank and the scuba tank back to room temperature in a non-reversible non-adiabatic process. that's what the wiki on Joule expansion is telling me.

I think that's the answer. Most of the things we were taught in scuba class about gas laws were reversible adiabatic processes and led to cooling during expansion and heating during compression. We sort of got to the mindset that expansion always meant cooling. That's what was driving me crazy. The scuba tank gas is expanding and yet it's getting hot! Every time I asked this question, I got told the scuba tank gas was being compressed, and I knew that was wrong.

The fill from a tank bank process is one process in scuba operations where expanding gas gets hot. Only the fill tank gas was following the usual rule of expansion=cold. The scuba tank gas considered alone is undergoing a non-adiabatic irreversible process and the entire gas system (both tanks considered together ) is undergoing an adiabatic, but irreversible Joule expansion process.

Safe diving for all and thanks for the help.
 
Last edited:
Do you believe that the air going into the tank being filled somehow KNOWS it's being filled from a compressor vs being transfilled from another tank... and should somehow behave differently vis-a-vis heating? Or do you think that the laws of physics apply in either case?

:d

I think the gas "knows" its original state, that that original state of PVT affects the final state) and that the ideal gas laws are enough to predict exactly what happens as it goes from the original state to final state, provided we know how that change occurs (was energy extracted in a reversible process). I also am now confident (pretty confident :) ) that the two situations - filling from a compressor directly versus from another higher-pressure tank are very different processes and that the heating comes about for very different reasons. One is adiabatic compression heating in a reversible process in which no heat is added (the direct compressor fill process). The other is heating by the addition of heat energy (through conversion of kinetic energy) in a non-reversible, non-adiabatic process (fill from a higher pressure tank). One could be reversed. One could not. One is an constant entropy process. One is not. It's this difference that I knew in my gut was there, but couldn't figure out.
 
Last edited:


I also am now confident (pretty confident
icosm14.gif
) that the two situations - filling from a compressor directly versus from another higher-pressure tank are very different processes and that the heating comes about for very different reasons.

This is the major flaw in your logic. The two situations are EXACTLY the same.

The filling process from a compressor is nothing more than the compressor creating a small "high pressure tank" and then transfiling the gas from that "tank" into your scuba tank. The only difference is that the compressor does this over and over and over, until the desired pressure in your tank is reached, the pressure in the tank equals the max pressure of the compressor, or your scuba tank explodes... like my head will if I keep reading this thread!

I think the gas "knows" its original state, that that original state of PVT affects the final state) and that the ideal gas laws are enough to predict exactly what happens as it goes from the original state to final state, provided we know how that change occurs (was energy extracted in a reversible process). I also am now confident (pretty confident :) ) that the two situations - filling from a compressor directly versus from another higher-pressure tank are very different processes and that the heating comes about for very different reasons. One is adiabatic compression heating in a reversible process in which no heat is added (the direct compressor fill process). The other is heating by the addition of heat energy (through conversion of kinetic energy) in a non-reversible, non-adiabatic process (fill from a higher pressure tank). One could be reversed. One could not. One is an constant entropy process. One is not. It's this difference that I knew in my gut was there, but couldn't figure out.

This isn't that difficult. Seriously.

Yes, there IS a cooling effect. However, you're looking for it in the wrong place. It exists at one point in place/time in the system. Feel the valve of the donor tank. The air cools there as its pressure drops when leaving the 3000psi tank. This air enters the receiving tank. However, the continued introduction of air from the donor tank increases the pressure in the receiving tank, thus increasing the temperature of the air in that tank.
 
This is the major flaw in your logic. The two situations are EXACTLY the same.
We disagree 100% on this. They are not the same and the heating arises for completely different reasons.

The filling process from a compressor is nothing more than the compressor creating a small "high pressure tank" and then transfiling the gas from that "tank" into your scuba tank.
The small "high pressure tank" in the compressor you are referring to is full of very hot air that was just heated by compression. We expect that air to stay hot as it goes into the scuba tank. As we add more air to the tank, it gets even hotter as it continues to be compressed. The corresponding air in the donor/fill tank is room temperature air that always ends up expanded. It was very hard for me to understand why room temperature high pressure air got hotter when I let it expand into the scuba tank.

The only difference is that the compressor does this over and over and over, until the desired pressure in your tank is reached, the pressure in the tank equals the max pressure of the compressor, or your scuba tank explodes... like my head will if I keep reading this thread!
You only have to put up with reading this thread for a few days. I've had to put up with people telling me I didn't understand this for years. :)

This isn't that difficult. Seriously.
We disagree on this, but I suppose "difficult" is a matter of opinion. I've looked for the answer to this for a long time and wasn't able to find an answer that I was comfortable with. I won't say all the answers I got were wrong, but most were either right, but didn't really explain the "why" of the difference and the others just totally glossed over the difference between the two types of tank filling procedures. They are different.

Yes, there IS a cooling effect. However, you're looking for it in the wrong place. It exists at one point in place/time in the system. Feel the valve of the donor tank. The air cools there as its pressure drops when leaving the 3000psi tank. This air enters the receiving tank. However, the continued introduction of air from the donor tank increases the pressure in the receiving tank, thus increasing the temperature of the air in that tank.

This isn't totally wrong, but it's just confusing the understanding of what's really going on. You're thinking of adiabatic reversible compression heating and adiabatic reversible expansion cooling, but they aren't enough to explain the heating. During the fill process, the first parcels of air that arrive in the scuba tank *are* getting compressed by the later arriving parcels of air. That compression *does* reheat the first parcels of air. I don't disagree with that. The problem is that the early parcels of air are cooled off by a very large expansion and cool by expansion significantly. Even though they are heated by compression later (as the tank fills), the amount those early parcels get reheated by recompression is less than the amount they initially cooled off. They always end up at al ower pressure and larger volume than they started off, so they always end up at a slightly colder temperature.

The only way the air in the scuba tank can get get hotter than it started out, even though it's expanded, is by adding more heat. That's part of why this whole thing is so complex and interesting. If you take heat from the scuba tank and apply it to the valve (which I agree is cold) you still end up with a hot scuba tank and a cold donor/fill tank. You still end up with expanded gas in the donor tank that is colder than it started off and expanded gas in the scuba tank that is hotter than it started off. That needed to be explained, at least it did for me to understand it.

Your explanation continues to try to rely on the idea that compressing air always makes it hotter and expanding air always makes it colder. That's usually true, but not here. the cycle of expansion of air as it initially arrives in the scuba tank, followed by compression of that air from later arriving higher pressure air always involves starting air that expands a lot and compresses less than it expanded. You can't get it hotter from the later compression if the earlier expansion was greater, and that's always the case.

The reason the scuba tank air gets hot is simply that energy is extracted from the donor fill tank (in the form of kinetic energy given up to the escaping gas in a reversible process) and that kinetic energy is converted to heat in the scuba tank. Personally, I find this to be a cool new insight into something that has always bothered me.
 
Personally, I find this to be a cool new insight into something that has always bothered me.

Never let the facts get in the way of a cool story...

You only have to put up with reading this thread for a few days. I've had to put up with people telling me I didn't understand this for years. :)

Yeah, I don't see that changing anytime soon...

Your explanation continues to try to rely on the idea that compressing air always makes it hotter and expanding air always makes it colder. That's usually true, but not here.

Please explain what magical, mystical properties of scuba tanks and valves suspend the laws of physics?
 
Your explanation continues to try to rely on the idea that compressing air always makes it hotter and expanding air always makes it colder. That's usually true, but not here.

Yepper and the air-conditioner in your car also operates on PFM
 
My, my, my... what a load of old cod's wollop. Well, some of it... the question was adequately answered a while back: and then some!

OP, do a google search on the laws of thermodynamics to gain an understanding of the conservation of energy, please.

And the following has made my day...

quote_icon.png
Originally Posted by jimmyw

Your explanation continues to try to rely on the idea that compressing air always makes it hotter and expanding air always makes it colder. That's usually true, but not here.



Which leads me to ask, where is here. Some alternative universe perhaps.

And of course the irony is in this case, that is exactly what's happening. The energy (total and all forms) remains constant throughout the system... mostly. Friction is playing a role too since there is no such thing as an ideal gas... not here... and by here I mean the real world. And there will be radiant and convection carrying off some "energy" from the system.


___________________

OH, i just thought of something.

Let's assume the pressure in the supply cylinders is 300 bar. In the recipient cylinders is is 1 bar. The two vessels are connected with a simple transfer whip. The valve on the recipient cylinder is opened with the supply tank cylinder closed. The pressure in the transfer whip is now about 1 bar... correct. The valve on the supply cylinder is opened... we do not know the flow rate the valve is capable of but what guess would we make for the point in this sealed system where a pressure differential exists? (THIS MIGHT BE WHERE ENERGY IS CONVERTED... perhaps?)
 
Last edited:
Never let the facts get in the way of a cool story...
Please explain what magical, mystical properties of scuba tanks and valves suspend the laws of physics?
You have 8,000 posts here. I've got a dozen in one thread. I'm not going to come into your back yard and pick a fight or try to challenge you. If you don't see why the two situations are different, then you don't. The physics are what they are and I've worked through the physics very carefully. When filling from a compressed tank the air is undergoing Joule expansion except that the gas is divided into two parts, one part in the donor tank and one in the scuba tank, and all the heat that is released ends up in the gas in the scuba tank instead of being equally distributed throughout the gas in both tanks.

If you are really interested in a more complete analysis, and the physics I went through to get to the final answer, I'd be happy to explain further. I considered the internal energy of the gas and the change in entropy, isenthalpic processes and several other aspects of this process before finally deciding I understood what was happening, but as I said, I'm not here to stir things up. I do think it's interesting, but if you don't, I'm comfortable with that. I'm just happy to finally have an answer that satisfies me.

---------- Post added February 22nd, 2013 at 04:17 PM ----------

My, my, my... what a load of old cod's wollop. Well, some of it... the question was adequately answered a while back: and then some!
What was the answer? What did I miss?

OP, do a google search on the laws of thermodynamics to gain an understanding of the conservation of energy, please.
I've done dozens of searches and read parts of several textbooks.

And the following has made my day...

quote_icon.png
Originally Posted by jimmyw

Your explanation continues to try to rely on the idea that compressing air always makes it hotter and expanding air always makes it colder. That's usually true, but not here.



Which leads me to ask, where is here. Some alternative universe perhaps.

Just read the wikipedia link I provided on Joule expansion. That expanding gas doesn't get colder. Or think about the problem for a while. Take any 1 cc cube of air in the donor tank and follow it. If it ends up in the scuba tank it gets hot. If it remains in the donor tank it gets cold. Both end up at lower pressure and larger volume than they start with. I don't see how I can be any clearer than that about what was troubling me. Surely someone here at least understands what was bugging me, even if they don't agree with my answer.

Friction is playing a role too since there is no such thing as an ideal gas... not here... and by here I mean the real world.
I agree friction plays a role in the real world, but we can explain why the scuba tank gets hot without resorting to friction.

---------- Post added February 22nd, 2013 at 04:42 PM ----------

OH, i just thought of something.
Let's assume the pressure in the supply cylinders is 300 bar. In the recipient cylinders is is 1 bar. The two vessels are connected with a simple transfer whip. The valve on the recipient cylinder is opened with the supply tank cylinder closed. The pressure in the transfer whip is now about 1 bar... correct. The valve on the supply cylinder is opened...
I followed you to here and agree.
we do not know the flow rate the valve is capable of but what guess would we make for the point in this sealed system where a pressure differential exists? (THIS MIGHT BE WHERE ENERGY IS CONVERTED... perhaps?)

The air in the whip will have high presure at one end and lower pressure at the other. Two things will happen. 1) As the air flows it will see less and less pressure and expand. 2) It will move faster and faster as it's pushed down the whip. We understand the first part. It's the second part that makes this interesting. The air not only expands and cools, it has energy extracted from the donor tank. As the air in the whip gains speed and kinetic energy it drives down the whip into the scuba tank and begins smashing into the air in that tank and into the tank walls. All that kinetic energy is converted to heat. The air that stayed in the donor tank pretty much stayed in one place and undergoes the normal expansion and cooling that we all agree takes place. It took energy to compress that air originally and as it expanded, that energy was transferred to the air in the whip. If we could prevent the air in the whip from being accelerated, and from gaining energy, that air would expand and cool just like the air in the donor tank. There are several ways we could do that. I mentioned one - running it through a pneumatic motor and doing work with that motor. Or run it through a windmill/fan/turbine and extract the energy. If we did that, the air in the scuba tank would only undergo process #1 above and it woul dbe just like the air in the donor tank - arriving at slow speed in the scuba tank, without any extra energy, and it would be cold, like the air in the donor tank that also was relatively motionless as it expanded.

The easiest way to see what's happening (for me) is to notice that we could use the pressure difference in the whip to run an engine or do work to extract the energy put into the donor tank during its original compression. If we did that, the air leaving the whip would be exiting slowly. OTOH, if we just let the gas rocket through the whip and exit at high speed, it's carrying all the energy we could have, but didn't, extract. That energy is what causes the scuba tank to heat up. I just didn't see the difference between the two cases, and until I did, I couldn't understand why the expanding air in the scuba tank got hot instead of cold.
 
Last edited:
Just read the wikipedia link I provided on Joule expansion. That expanding gas doesn't get colder. Or think about the problem for a while. Take any 1 cc cube of air in the donor tank and follow it. If it ends up in the scuba tank it gets hot. If it remains in the donor tank it gets cold. Both end up at lower pressure and larger volume than they start with. I don't see how I can be any clearer than that.

Another fundamental error. You're confusing the "amount of gas" (how many molecules) with the "amount of space" (how much volume) those molecules occupy. The volume (amount of space that the given quantity of gas molecules occupies) in the two tanks doesn't change. (ie - the tank size has not changed) The lower pressure in the donor tank is actually a function of the fact that a smaller amount of gas now occupies the same volume of space that was previously occupied by twice as much gas. In the receiving tank an enormous amount of gas now occupies the same volume of space that was previously occupied by very little gas.

When thinking about the pressure changes in the two tanks, consider this:

- the pressure in the donor tank has only decreased 50% (3000psi to 1500psi)
- the pressure in the receiving tank has increased >1000% (from 14psi to 1500psi)

---------- Post added February 22nd, 2013 at 06:22 PM ----------

OH, i just thought of something.

Let's assume the pressure in the supply cylinders is 300 bar. In the recipient cylinders is is 1 bar. The two vessels are connected with a simple transfer whip. The valve on the recipient cylinder is opened with the supply tank cylinder closed. The pressure in the transfer whip is now about 1 bar... correct. The valve on the supply cylinder is opened... we do not know the flow rate the valve is capable of but what guess would we make for the point in this sealed system where a pressure differential exists? (THIS MIGHT BE WHERE ENERGY IS CONVERTED... perhaps?)

Yup. The OP wants to follow a cc of gas from one tank to the other? He should do that... and think of the pressure/volume/temperature changes that occur at each and every step along the way. I think he believes that each cc of gas goes from being 1cc of gas at 3000psi and X° in one tank, directly and immediately to being 1cc of gas at 1500psi and Y° in the second tank.
 
Last edited:
https://www.shearwater.com/products/perdix-ai/

Back
Top Bottom