Effects of altitude on liquid CO2 utilization

[MrFreezeMiser]

I’m a contractor using liquid CO2 (LCO2) to freeze water pipes (for valve maintenance) while the water service remains at full static pressure (i.e., water is not flowing).

I do the work in two locations: (1) near Denver, CO, where the altitude is 5,300 feet and (2) near Detroit, MI where the altitude is 1,000ft. Interestingly, I’ve found it takes WAY MORE LCO2 to do the same work in Denver than Detroit. At this point, I don’t have any data collected (i.e., water temp, ambient temp, etc.), so initially I’m looking at things theoretically with anecdotal observations.

My question here is: Assuming exact same situations (i.e., same copper pipe material, same size, same potable water temperature, etc.), how could one calculate and/or quantify the effects of altitude on LCO2 utilization? In other words, given the exact same work but at two materially-different altitudes, how many fewer pipes would be expected to be frozen with the same volume of LCO2 (e.g., standard 20-lb cylinder) in Denver vs. Detroit?

The last freeze in Detroit was a 1-inch type-M copper pipe, and it took 10-minutes (note: consistent with manufacturer’s specifications ) and used approximately 3-lbs of LCO2. The manufacturer of the freeze equipment does not provide adjustments for freeze times with higher altitudes, but I can state in Denver it seems to take a lot more time and use more LCO2.

Thank you in advance. -Mr. Freeze Miser

 

[russ_watters]

Air pressure in Denver is about 85% what it is in Detroit. What is the pressure of the liquid CO2? Is it pressurized or atmospheric pressure(Can CO2 even be liquid at atmospheric pressure?)? Do you have a table of boiling point vs pressure? If the system is a pressurized cylinder eventually discharging to atmosphere can you put a valve at the end to throttle it and increase the pressure?

 

[DaveE]

My guess is that if what you say is true, then it must be that the CO₂ evaporates (sublimates?) more rapidly with lower ambient pressure. But I don't know how your machine works and how that matches up with the CO₂ phase diagram. If it's truly liquid for any significant length of time then it's pressure must be much greater than 1 atm.

 

[Bystander]

full static pressure

Denver vs. Detroit? How much difference? You're looking at two different public utilities.

 

[Lnewqban]

Copied from
https://www.plumbermag.com/how-to-a...out_shutting_down_the_whole_building_sc_01ck7

"The system freezes all types of liquids in steel, copper, cast iron, aluminum and plastic pipes ranging from 1/8 to 2 inches. The freeze kit uses carbon dioxide (CO2) from a dip-tube tank available at all welding supply houses. The cold liquid CO2 is minus 110 degrees so it can freeze the water in just minutes. A 1/2-inch copper pipe freezes in just five minutes, or just three minutes in cast iron. The ice plug is so strong it can withstand up to 7,000 psi."

 

[Lnewqban]

The last freeze in Detroit was a 1-inch type-M copper pipe, and it took 10-minutes (note: consistent with manufacturer’s specifications ) and used approximately 3-lbs of LCO2. The manufacturer of the freeze equipment does not provide adjustments for freeze times with higher altitudes, but I can state in Denver it seems to take a lot more time and use more LCO2.

Please, see:
https://www.engineeringtoolbox.com/CO2-carbon-dioxide-properties-d_2017.html

https://homesteady.com/how-6514409-calculate-flow-based-differential-pressure.html

"The change in freezing point (water inside pipe if system is opened to atmosphere) at different altitudes is much smaller than the change in the boiling point (liquid CO₂ expanding to atmospheric pressure).
The freezing point increases very slightly at higher altitudes, due to the air pressure. Because ice takes up more space than water, a lower air pressure will cause water to freeze at a slightly higher temperature."

 

[berkeman]

(Can CO2 even be liquid at atmospheric pressure?)

Yes, but not for long. Please don't ask me how I know this...

Spoiler

Broken LCO2 connection from a dewar that I had to deal with in the lab -- crazy.

https://newbestsm.live/product_details/47728720.html

 

[MrFreezeMiser]

Air pressure in Denver is about 85% what it is in Detroit. What is the pressure of the liquid CO2? Is it pressurized or atmospheric pressure(Can CO2 even be liquid at atmospheric pressure?)? Do you have a table of boiling point vs pressure? If the system is a pressurized cylinder eventually discharging to atmosphere can you put a valve at the end to throttle it and increase the pressure?

Thank you, Russ! I’m not certain exactly what the pressure is. It’s not high but not low either. Open the valve to atmosphere and it’s going to be loud and sporty. The LCO2 tank has a dip tube so the LCO2 is forced out vs. gas. As the tank empties you still get CO2 but it’s no longer liquid. The device sprays LCO2 literally (but lightly) on the pipe. Eventually it freeze a plug inside the pipe and about a 3/4” thick x 3/4” wide ice ring on the outside. You keep it running the entire time you’re doing work, then shut it off and about 5-mins after, flow naturally restores itself. One other note is the tubing coming from the tank connection is like 1/8” OD. You’d think that would freeze solid and snap with LC02 going through it! But it stays springy/coiled (stretchable to about 10ft). At the tank, you connect the manufacturer’s adapter and flow of LCO2 is ”regulated” by a sintered brass filter device. There is no regulator; only flow control is the tank’s valve (on/off). Lnewqban (thanks!) posted a link to the ColdShot which is what I’m using.

 

[MrFreezeMiser]

My guess is that if what you say is true, then it must be that the CO₂ evaporates (sublimates?) more rapidly with lower ambient pressure. But I don't know how your machine works and how that matches up with the CO₂ phase diagram. If it's truly liquid for any significant length of time then it's pressure must be much greater than 1 atm.

View attachment 338510

Thanks, Dave. I think you’re on to something here. Sublimation with LCO2 is totally new to me, so I will get familiar with it. I bought this setup based upon manufacturer’s representations. It says freezing results will vary…but I didn’t expect such dramatic difference. Unfortunately, there’s no mention specifically of altitude effects. Needless to say, in Denver I BLOW THROUGH LCO2 vs. doing exact same work in Detroit. I’m really fascinated by all this vs. have any hard feelings with the manufacturer. I will say high school and undergrad physics never prepared me for this! Hopefully the brains of this forum can help explain what’s really going on and perhaps share this most interesting, real-world work scenario with world.

 

[MrFreezeMiser]

Copied from
https://www.plumbermag.com/how-to-a...out_shutting_down_the_whole_building_sc_01ck7

"The system freezes all types of liquids in steel, copper, cast iron, aluminum and plastic pipes ranging from 1/8 to 2 inches. The freeze kit uses carbon dioxide (CO2) from a dip-tube tank available at all welding supply houses. The cold liquid CO2 is minus 110 degrees so it can freeze the water in just minutes. A 1/2-inch copper pipe freezes in just five minutes, or just three minutes in cast iron. The ice plug is so strong it can withstand up to 7,000 psi."

Thanks, Lnewqban. Yes, this is the system I’m using. I cross-sectioned an ice plug with 3/4” CPVC and was super impressed by the very dense ice…extremely robust.

 

[MrFreezeMiser]

Denver vs. Detroit? How much difference? You're looking at two different public utilities.

Denver vs. Detroit? How much difference? You're looking at two different public utilities.

According to Russ, it appears pressure is approx 15% less in Denver. I do private, commercial work vs. work for public utilities.

 

[MrFreezeMiser]

On an interesting side note, when I asked my Denver LCO2 gas supplier why utilization here was falling off the charts, the sales manager said he couldn’t explain why but said altitude has a huge effect. Then he shared this example, and brace yourself this get weird. It supplied the liquid nitrogen to keep some frozen guy frozen (yes, it’s a creepy tourist attraction!) while in transport being relocated from Nederland, CO to Estes Park, CO. The planners said/ordered 6 large liquid N tanks. Needless to say there was serious panic because they made emergency requests for more and more tanks of LN. I think he said they ended up using 20 tanks instead of 6. This is consistent with my LCO2 experience.

 

[MrFreezeMiser]

Yes, but not for long. Please don't ask me how I know this...

Spoiler

Broken LCO2 connection from a dewar that I had to deal with in the lab -- crazy.

View attachment 338528
https://newbestsm.live/product_details/47728720.html

lol…we don’t need to know! I’ve done plumbing work for companies the use a lot of LCO2 and CO2…never pleasant when things go sideways. Thankfully, I haven’t triggered any of the alarm systems yet!

 

[256bits]

One other note is the tubing coming from the tank connection is like 1/8” OD. You’d think that would freeze solid and snap with LC02 going through it!

A misunderstanding.
Why should the tube 'freeze'?
The liquid CO2 in the bottle is at room temperature.
So is the liquid CO2 passing through the tube.

 

[256bits]

I do hope you wear safety goggles for your eyes.

 

[Chestermiller]

Copied from
https://www.plumbermag.com/how-to-a...out_shutting_down_the_whole_building_sc_01ck7

"The system freezes all types of liquids in steel, copper, cast iron, aluminum and plastic pipes ranging from 1/8 to 2 inches. The freeze kit uses carbon dioxide (CO2) from a dip-tube tank available at all welding supply houses. The cold liquid CO2 is minus 110 degrees so it can freeze the water in just minutes. A 1/2-inch copper pipe freezes in just five minutes, or just three minutes in cast iron. The ice plug is so strong it can withstand up to 7,000 psi."

According to the phase diagram for CO2 presented in post #6, at -110 C, CO2 is a solid. The temperature must be >-55 C and the pressure in the tank must be above 5 bars absolute (4 bars gauge, 60 psig) for there to be liquid CO2 in the tank.

Here is a pressure-enthalpy diagram for CO2.

At -40 C (=-40 F), the pressure in the tank is going to be about 140 psi (125 psig), and the CO2 exiting that valve and tubing will be a mixture of about 45% CO2 vapor and 55% CO2 solid (assuming constant enthalpy through the valve and tube), down to 15 psia. The temperature of this mixture will easily be <-40C, and could easily approach -110 C. The final pressure (1 bar vs 0.85 bar) could effect the final mixture temperature by about 10C (say -120 C vs -110 C, according to the diagram). This should not substantially affect the freezing time of the pipe.

 

[MrFreezeMiser]

A misunderstanding.
Why should the tube 'freeze'?
The liquid CO2 in the bottle is at room temperature.
So is the liquid CO2 passing through the tube.

Yes, I agree it shouldn’t freeze. It’s just me and kind of like my response to a magic trick…it should, but it doesn’t.

 

[Lnewqban]

... The final pressure (1 bar vs 0.85 bar) could effect the final mixture temperature by about 10C (say -120 C vs -110 C, according to the diagram). This should not substantially affect the freezing time of the pipe.

Could the difference in freezing time of the water plugs described in the OP be due to a reduced heat of sublimation of the gas at lower atmospheric pressure?

 

[Bystander]

According to Russ, it appears pressure is approx 15% less in Denver. I do private, commercial work vs. work for public utilities.

Water pressure; how hard a freeze to stop the flow? Two different public utilities, two different hydrant pressures.

 

[256bits]

This should not substantially affect the freezing time of the

I wouldn't think so either since the heat of vapourization of CO2 would be the dominant criteria for heat removal and stabilization.

Ideal case would be if the camber has a liquid-gaseous quality, with liquid mass flow input matching gaseous outflow,
Too too little liquid inflow and the quality within the chamber surrounding the water pipe would drop to 100% vapour. The benefit of the heat transfer due to heat of vapourization of CO2 would move from the chamber towards the direction of the tank along the piping.
Too too much liquid inflow and the quality would become saturated liquid. Liquid would exit the chamber and turn to the gaseous state outside the assembly, completely negating any cooling effect to the chamber.

I propose that the lower back pressure of the atmosphere inclines the liquid input into the chamber to increase, moving the quality of the CO2 towards a direction of more saturated liquid state. so that more CO2 liquid is escaping from the chamber and not contributing anything to the cooling effect of the chamber. It is throwing away liquid CO2. This can account for the increase in amount of CO2 usage.

The increased inflow of CO2 adds another burden for cooling according to the specific heat capacity of CO2 liquid, which has to be additionally cooled from the input temperature, say of the tank, to the freezing point of water and below.

The mass flow of the liquid CO2 is related to the pressure difference between the tank and the atmosphere, although not that large in quantity, does seem to be playing hacoc with the amount of CO2 usage at higher elevations.

Assumption of course is that there is not choked flow to limit and regulate CO2 liquid, gaseous or mixture through an exit orifice.

 

[256bits]

Water pressure; how hard a freeze to stop the flow? Two different public utilities, two different hydrant pressures.

Flow is stopped before when using this machine, ie some valve(s) are closed to prevent water flow into the section one is working on.
The ice plug(s) of ice are there so that whole of the pipe line does not have to be drained.

 

[MrFreezeMiser]

Water pressure; how hard a freeze to stop the flow? Two different public utilities, two different hydrant pressures.

It’s designed for static pressure only. There can be very little—preferably no flow.

 

[MrFreezeMiser]

I wouldn't think so either since the heat of vapourization of CO2 would be the dominant criteria for heat removal and stabilization.

Ideal case would be if the camber has a liquid-gaseous quality, with liquid mass flow input matching gaseous outflow,
Too too little liquid inflow and the quality within the chamber surrounding the water pipe would drop to 100% vapour. The benefit of the heat transfer due to heat of vapourization of CO2 would move from the chamber towards the direction of the tank along the piping.
Too too much liquid inflow and the quality would become saturated liquid. Liquid would exit the chamber and turn to the gaseous state outside the assembly, completely negating any cooling effect to the chamber.

I propose that the lower back pressure of the atmosphere inclines the liquid input into the chamber to increase, moving the quality of the CO2 towards a direction of more saturated liquid state. so that more CO2 liquid is escaping from the chamber and not contributing anything to the cooling effect of the chamber. It is throwing away liquid CO2. This can account for the increase in amount of CO2 usage.

The increased inflow of CO2 adds another burden for cooling according to the specific heat capacity of CO2 liquid, which has to be additionally cooled from the input temperature, say of the tank, to the freezing point of water and below.

The mass flow of the liquid CO2 is related to the pressure difference between the tank and the atmosphere, although not that large in quantity, does seem to be playing hacoc with the amount of CO2 usage at higher elevations.

Assumption of course is that there is not choked flow to limit and regulate CO2 liquid, gaseous or mixture through an exit orifice.

Unscientifically…something like this is what I feel. The manual says a freeze of 1/2” copper pipe should take 5-mins AND you should get 30-freezes for 20-lb. cylinder. In Denver, it’s taking 5-mins, but I’m only getting 3 (maybe 4) freezes per cylinder. Now, I haven’t done 1/2” copper yet in Detroit, but I did do 1” copper, and it took 10-minutes and used 3-lbs using a refrigerant scale. So extrapolating this, I should get at least 6 freezes per cylinder. The manual suggests yield should be 15 freezes. Anyway, in Denver it just feels like LCO2 somehow just disappears.

 

[Chestermiller]

From the p-H diagram I presented in my previous post, if the temperature of the tank contents is ~70 F initially, it will be at 800 psig. What pressure does the regulator show initially? What is the volume of the tank. Is the amount of liquid CO2 initially in the tank really 20 lb?

I intend to do some serious calculations on this operation, including heat transfer and cooling/freezing of the water in the pipe.

At 1 bar,
saturation temperature -109 F
saturated solid density = 1.562 gm/cc
saturated vapor volume = 363 L/kg

At 20 C
Pressure = 5.729 bars = 830 psia
saturated liquid density =1.294 gm/cc
saturated vapor volume = 5.149 L/kg

 

[Chestermiller]

Based on equilibrium vapor pressure data, the temperature of the solid-vapor mixture coming out of the tubing should be about 8 F cooler for Denver than for Detroit (say, -117 F vs -109 F). The streams should have about the same mass fractions solid (pretty low). Based solely on these temperature of the streams, the water freezing at Denver should be more rapid than the water freezing at Detroit.

We know that the vapor flow rate affects the heat transfer. How do you know that the mass flow rates were approximately the same. given that, as the tank empties, the pressure in the tank decreases so that, at a given valve opening, there is decreasing flow as the tank empties.?

 

[256bits]

given that, as the tank empties, the pressure in the tank decreases so that, at a given valve opening, there is decreasing flow as the tank empties.?

The vapour volume within the tank increases as liquid is discharged, so there should be some boiling and temperature drop, unless the heat transfer from atmosphere to tank is able to keep up.

 

[Chestermiller]

The vapour volume within the tank increases as liquid is discharged, so there should be some boiling and temperature drop, unless the heat transfer from atmosphere to tank is able to keep up.

The temperature within the tank is indeed dropping, but there is only saturated vapor entering the exit tubing at the enthalpy of the saturated vapor. The enthalpy of this saturated vapor does not change in throttling though the valve and exit tubing, and, according to the diagram I presented in post #16, the temperature of the vapor leaving the tubing will be cooler in the tank due to Joule Thomson cooling. The highest the exit enthalpy from the tubing will be is 140 BTU/lb, and the highest the exit temperature will be is about -80 F for this maximum saturation enthalpy. Toward the end, when the pressure in the tank has dropped to about 1 bar, the exit temperature from the tubing will be about -109 F.

 

[Dullard]

I think that 256 nailed it in post #20.
The flow regulation in this setup is a function of differential pressure, where the low end is atmosphere. you're flowing more CO2 than you can effectively use (in Denver). I suspect (very strongly) that the addition of a 'throttle' ball valve (partially closed in Denver) will make 'Mile-High' performance very close to that of 'Motown'. There are some other differences, but flow regulation is the 'gorilla.'

 

[MrFreezeMiser]

From the p-H diagram I presented in my previous post, if the temperature of the tank contents is ~70 F initially, it will be at 800 psig. What pressure does the regulator show initially? What is the volume of the tank. Is the amount of liquid CO2 initially in the tank really 20 lb?

I intend to do some serious calculations on this operation, including heat transfer and cooling/freezing of the water in the pipe.

At 1 bar,
saturation temperature -109 F
saturated solid density = 1.562 gm/cc
saturated vapor volume = 363 L/kg

At 20 C
Pressure = 5.729 bars = 830 psia
saturated liquid density =1.294 gm/cc
saturated vapor volume = 5.149 L/kg

Unfortunately, the manufacturer doesn’t include a pressure gauge, but maybe I can get the data or see if the supplier can tell us. I have refrigeration pressure gauges that go to 800-psi, but now that you‘re suggesting it might be 800-psi or more, I might have to buy a higher limit gauge for testing. I have two tanks to refill in Denver, so I could get pressure at temperature from these close-to-empty tanks. Then when I get replacement tanks (supplier just swaps them), I could gather that data as well and post it.

Interestingly, today I tried to freeze 3/4” CPVC when the ambient temperature was approx 20ºF. I brought a full tank that was kept on the truck on it’s been like -5ºF here, so I bet the tank was somewhere around this temperature. As ironic as it sounds, yes, I was freezing pipes in freezing cold weather. And here I wanted to see what would happen in ambient conditions. My guess was just gas would come out based upon what I’ve learned from folks here. So after hooking up everything, yep, LCO2 would not flow. Only gas hissed like you hear when the LCO2 tank is close to empty. So I took one of our portable heaters and warmed up the tank really good [read: hot], and like magic, LCO2 began to flow like normal.

Typically, we freeze 3/4” CPVC for 20-minutes in Denver (no data for Detroit). Today, I did 25-mins—just to be certain—and it worked perfectly with pre-heated tank and maintaining heat throughout the process. But I bet I used a lot of that brand new 20-lb cylinder. These jobs are critical and consuming, so I forget to gather data to stay focused. I need to do better with this.

Supposedly cylinders are full with 20-lbs of LCO2 from the supplier. When I get replacements, I’ll get actual net weight(s). My guess is it’ll be 20-lbs or slightly over.

 

[MrFreezeMiser]

I think that 256 nailed it in post #20.
The flow regulation in this setup is a function of differential pressure, where the low end is atmosphere. you're flowing more CO2 than you can effectively use (in Denver). I suspect (very strongly) that the addition of a 'throttle' ball valve (partially closed in Denver) will make 'Mile-High' performance very close to that of 'Motown'. There are some other differences, but flow regulation is the 'gorilla.'

I’m going to look into flow control. When I first starting doing this work, I was just happy it froze and my company didn’t cause an international incident! There’s no Blowout Preventer on this deal, so you have to be sure your work is as cool as the ice plug. Fascinating stuff! I so much appreciate everyone’s involvement and learning what’s really going on here. Plus, it will help me do smarter this work. “Mile HighMotown”…still

 

[Lnewqban]

... So I took one of our portable heaters and warmed up the tank really good [read: hot], and like magic, LCO2 began to flow like normal.

Caution with heating pressurized tanks is advised:
https://www.linde-gas.com/en/images/LMB_Safety Advice_01_66881_tcm17-165650.pdf

A bath of hot water is preferred, never a torch or local heating.

 

[MrFreezeMiser]

Caution with heating pressurized tanks is advised:
https://www.linde-gas.com/en/images/LMB_Safety Advice_01_66881_tcm17-165650.pdf

A bath of hot water is preferred, never a torch or local heating.

Thank you. This is great information. It’s definitely not recommended. I was extremely careful. The initial salvo I did in 140º water. A 2-gallon bucket, most of the water went all over the ground, and it didn’t do much being 7ºF. So I let the warm air passively heat it and turned the tank often to get even, rotisserie heating. I’d put my hand in front of the propane forced air heater to make sure it wasn’t beyond a certain temp and used Fluke thermal gun to monitor tank as well. It’s sort of like EOD/bomb disposal…DO NOT try this at home, but someone has to do it.

 

[MrFreezeMiser]

I’m going to look into flow control. When I first starting doing this work, I was just happy it froze and my company didn’t cause an international incident! There’s no Blowout Preventer on this deal, so you have to be sure your work is as cool as the ice plug. Fascinating stuff! I so much appreciate everyone’s involvement and learning what’s really going on here. Plus, it will help me do smarter this work. “Mile HighMotown”…still

I wouldn't think so either since the heat of vapourization of CO2 would be the dominant criteria for heat removal and stabilization.

Ideal case would be if the camber has a liquid-gaseous quality, with liquid mass flow input matching gaseous outflow,
Too too little liquid inflow and the quality within the chamber surrounding the water pipe would drop to 100% vapour. The benefit of the heat transfer due to heat of vapourization of CO2 would move from the chamber towards the direction of the tank along the piping.
Too too much liquid inflow and the quality would become saturated liquid. Liquid would exit the chamber and turn to the gaseous state outside the assembly, completely negating any cooling effect to the chamber.

I propose that the lower back pressure of the atmosphere inclines the liquid input into the chamber to increase, moving the quality of the CO2 towards a direction of more saturated liquid state. so that more CO2 liquid is escaping from the chamber and not contributing anything to the cooling effect of the chamber. It is throwing away liquid CO2. This can account for the increase in amount of CO2 usage.

The increased inflow of CO2 adds another burden for cooling according to the specific heat capacity of CO2 liquid, which has to be additionally cooled from the input temperature, say of the tank, to the freezing point of water and below.

The mass flow of the liquid CO2 is related to the pressure difference between the tank and the atmosphere, although not that large in quantity, does seem to be playing hacoc with the amount of CO2 usage at higher elevations.

Assumption of course is that there is not choked flow to limit and regulate CO2 liquid, gaseous or mixture through an exit orifice.

 

[MrFreezeMiser]

Weighed the two new 20-lb cylinders. First one was spot on 20-lbs net LCO2. Second was short by 1.7 ozs (only posted pics of 1st cylinder).

 

[MrFreezeMiser]

Some other notes: Local supplier didn’t have any ideas/recommendations for a LCO2 flow regulator. I told him a pressure regulator wouldn’t work anyway because we still need high pressure just at a lower, adjustable flow rate. Harris makes some flow regulators—use them in various welding processes—but I will have to contact it to be sure it would fit this tank [valve] style and is designed to be used safely with LCO2.

Supplier didn’t offer a blanket heater to keep tank warm, but pointed me to the web. Said the tank is rated for 1,800 PSI. He didn’t know the blow-off relief pressure, but said last summer a 5-lb released sitting in the sun on a hot day. And once the relief pops, it empties.