View Full Version : co2 liquifier

29-11-2006, 10:05 AM
Dear all,

I am now working on designing a co2 liquifier, some parameters:
duty:450 kW
co2 side: flow: 9.4 ton/h
inlet: 40 C, 30 bar, gas
outlet: -10 C, 29.5 bar, liquid

I am new to this project and I look for many informations about this ,but found no useful information on the co2 condenser/R22 evaporator designing.
Could someone help me especially on calculating the co2 condensation heat transfer coefficients in-tube? and more information on the shell and tube exchanger design?

Maybe it's a simple problem for you seasoned guys, but for me it's huge. Need your help! Please!

many thanks!


Andy P
30-11-2006, 12:10 AM
Hi Cygbob,

I don't remember the coefficients that we used to use for these plants - maybe someone else can help with that, or I can raid the archive at work. However one issue worth considering is that most systems I'm familiar with have the CO2 outside the tubes and refrigerant inside. This gives some advantages: you can purge non-condensing gases more easily from the CO2 if it is on the shell side and the condenser still works well when the load is light, whereas with CO2 in the tubes they tend to fill with liquid when the flowrate is low. It doesn't need to be a dry expansion system: many of the older ones were gravity fed with refrigerant in the tubes (kind of like a thermosyphon oil cooler). It will be interesting to hear some other views on whether the CO2 should be inside or outside the tubes


Andy P

30-11-2006, 12:31 AM




30-11-2006, 01:52 AM
Hi,Andy P
Thanks for the advice, we also considered it that way, because in the co2 liquid the impurity gas (N2,H2) is less than 0.3-0.5ppm, but because the co2 pressure is much higher than R22, to be safe we put the co2 in tube.

to bruceboldy:
thank you for the help!


30-11-2006, 11:03 PM

Hi, as promised

we worked up a model for your requirements.

identification if you want to talk to the company
iso-therm.com attn. zahid

model to use is ZX 1812 DEi
18 in diameter by 12 foot overall lengh
the unit is all carbon steel
the r-22 is dx in the tubes
the co2 is in the shell
base the r-22 suction on 4 degrees F
Price is based on usa dollars ( 23,719.00 )
Delivery is based on order entry usually 10 weeks
Performance guaranteed

hope this is useful in your planning, I look forward to visiting on other co2 applications as needed, or just refrigeration thoings that come up. This is a good site for good refrig info.


US Iceman
30-11-2006, 11:42 PM
Andy_P raised some good points that I will add to. One of the things I always looked at is which side the fluids should be on.

I thing there are several considerations. The first two big items I can thing of are pressure and fouling. For pressure it is generally easier (and cheaper) to design the exchanger with the highest pressure fluid in the tubes, rather than the shell.

CO2 will certainly have the higher pressure. Having said that, the condensing process could take precedent for fluid location. The condensation (of CO2) on the shell-side is probably better, since you do not have to worry about accumulation of the condensate, which you would if the CO2 was in the tubes.

Gravity drainage would be the only concern in this case from the shell. In this exchanger configuration you would have a thermosyphon chiller (flooded tube-side), which responds well to varying loads, much as a flooded shell-side chiller will.

However, I think the thermosyphon design will have a lower refrigerant charge (than a flooded shell & tube design if the CO2 is in the tubes).

Another possibility is to use a single pass tubeside exchanger so any condensate would run out of the exit end of the tubes. This is very common in dehumidification streams in air separation plants. In this case, the exchanger is simply a flooded shell & tube.

I doubt fouling is of much concern with CO2. It's not a nasty gas like chlorine condensing, in which case, the chlorine would be on the tube-side for easier cleaning.

I don't think I would recommend a multi-pass tube side for condensing the CO2 due to the drainage problem potential.

TXiceman might have some interesting thoughts on this also.:cool:

Cleaning the shell-side is next to impossible, so this is why I mentioned fouling as a context for fluid placement in the exchanger.

I will go on the record as saying I'm not too fond of DX for any process related cooling. DX has limits in turndown ability and you need slightly more heat exchange surface to produce the superheat for the TXV bulb & control.

Flooded shell-side or thermosyphon tube-side would be my recommendations.;)

The thermosyphon tube-side will probably have the best heat transfer coefficients due to forced convection boiling in the tubes, rather than pool boiling in the shell. Just a thought,

01-12-2006, 02:43 AM
Dumb question, but how are we aquiring the co2 and why are we liquifying it?

01-12-2006, 03:03 AM
To bruceboldy,
Thanks for the information, it is a choise for us.
Another point is that could you give me some materitals for calculating? as a new engineer , I'm eager to learn how to get it. Many thanks!

To US_Iceman,
Thanks for the analysis, it helps me lot!

To Dan,
it seems a different problem:) I don't konw how we get too but I can tell you the co2 liquid is for brewage and drink.:)

01-12-2006, 03:53 AM
Are you using a blower to push the CO2 vapor? Or is it by natural convection? It has been a few years but on a small CO2 tank, we used a "knock-back Condenser" to keep the tank pressure in range. I believe it was a vertical design flooded on the shell side and use a gravity feed drum to keep the shell flooded.

Also seen a X-changer Inc, unit which was DX fed setting on a tank operating like a knock-back condenser.

As for the heat transfer characteristics, best thing to do is get B-Jac or join HTRI.


US Iceman
01-12-2006, 04:36 PM
I don't know how we get too but I can tell you the co2 liquid is for brewage and drink.:)

Is this being installed in a brewery to recover the CO2 off of a fermentation tank? As the CO2 releases from the beer, it is collected and condensed by a separate refrigeration system. The CO2 is later injected into the beer prior to capping the bottles or sealing the cans.

If the CO2 is not a recovery operation off of a beer fermenting process, is your refrigeration system to be used in an air separation plant where CO2 is generated and collected?

02-12-2006, 02:47 AM
"knock-back Condenser"

Iceman, was there a name you gave to this that I asked about in another thread? TX, what is a "Knock-back" condenser?

US Iceman
02-12-2006, 03:07 PM
Iceman, was there a name you gave to this that I asked about in another thread?

A name for what? I'm confused.:confused:

03-12-2006, 02:37 AM
A name for what? I'm confused.

I can't remember the name. It was a reference you made to a tank that could contain the released gas from an over pressured system. Like "Knock back" yours was a slang that I never heard of before. I am tempted to invite a thread for slang... a glossary of sorts. It would be fun to do if I could find the time.

US Iceman
03-12-2006, 03:58 PM
Ahhh, I remember now. The item in question was a "fade-out drum". It's an old school term for an expansion tank used on cascade systems.

This vessel simply allowed the refrigerant to expand into gas (during shutdown periods). By providing sufficient internal volume in the refrigeration system, the gas would expand to fill this volume with no extraordinary increase in system pressure.

I have heard of the term TXiceman used. My definition for a "knock-back condenser" is any heat exchange device that simply recondenses the vapor back to a liquid.

In either case, the main purpose of these two methods is to prevent the system pressure from increasing beyond the pressure relief device release pressure.

In some cases, a knock-back condenser could also be used to condense a specific fluid out of a mixed gas stream.

The terms knock-back condenser and fade-out drum are not slang. They are highly technical terms used to describe a specific function.:D

04-12-2006, 03:57 AM
Here is an excellent source of condensing dat from Wolverine...http://www.wlv.com/products/databook/ch3_3.pdf


05-12-2006, 04:46 AM
Here is an excellent source of condensing dat from Wolverine...http://www.wlv.com/products/databook/ch3_3.pdf

thanks a lot!
i've looked it and it helps.
but i'm not sure the formula suit for the normal refrigerants such as R22,134a,717, whether it suit for the co2?


US Iceman
05-12-2006, 05:31 PM
You have to look at the properties and the equations. Most of the condensing equations are based on specific refrigerants. The equations sometimes have constants that are refrigerant specific.

If the equations are not refrigerant specific, then you can assume they can be used for general purpose, as long as you use good judgement.

Find the CO2 properties at the operating conditions (or an average value at the operating conditions). Plug these values into the equations and see what the resulting coefficient value is.

Try it also with some other refrigerants. Try to analyze how much difference there actually is between the condensing film coefficients of different fluids.

Some equations are refrigerant specific and those should state the applicable range or tolerance. You also have to be careful of the the basis of the equations. Some may be suitable for full regime flow (laminar, transistion, or turbulent). Others may only apply to a specific flow regime.

Part of developing the heat transfer for a new application is something of an art, rather than a science.

You have to see what the range of values are and their impact on the surface area required. The same applies for fouling factors also. High fouling factors can significantly increase the surface area requirements.

The one positive beneft you have with this exchanger is that you have phase change on both sides of the tube, which is the best case.

Andy P
10-12-2006, 11:46 PM
Hi Cygbob,

If the CO2 is coming from a brewing process then it will need to go through a fob tank and wash column before it gets to the compressors (to keep the beery bits out of the compressor) - there's no doubt that these will leave plenty of non-condensibles in the gas stream, which is why you need to worry about purging. The gas coming off the fermenters is at atmospheric pressure - the higher you can get the pressure the better for liquefying, hence the compressor. Typically the CO2 discharge was about 20 bar, which is roughly -20C.

There was always a great smell around these plants. As Frank would say "Mmmmmmmmmm...."


Andy P

11-12-2006, 01:42 AM
Fob tank
Knock-back condenser
Fade-out drum

Adding those to other terms I have seen, I wonder if we should open a thread called: Glossary of terms. Hmmm... this is for the moderators to consider. I will post this there, too.

12-12-2006, 03:16 PM
Dan, there is definately a difference in some of the terms used between what is termed commerical refrigeration and industrial or process refrigeration. Somet6imes, one item may have several different names.

Chiller barrel-evaporator
accumulator-knock out drum
head tank-Thermosiphon drum
pumper drum-liquid recirculator
vapor condenser-knock back condenser
fade out drum-volume bottle

Then there is a whole lost of items that can be really confusing to new comers in the industry.

Gotta' love this industry.


Andy P
17-12-2006, 11:34 PM
Good point Dan,

I only disagree with one of TXIceman's equivalents: a knock-back condenser is what I would have called a "reflux condenser" - one used to re-condense the vapours coming off the condensed liquid as it warms up so that the pressure is kept down to a reasonable level. This is more than just a vapour condenser - it might also be referred to as a re-condenser, if "reflux" gives you indigestion;)


Andy P

18-12-2006, 05:01 AM
I agree, reflux condenser is another name often used.