Low temperature refrigeration 5

[c2sco]

Hi,

I need to design a heating / cooling system for a small batch process. Its duty will be to supply alternately heating and cooling to any one of four jacketed vessels, either about 5 kW heat load cooling at -100°C (-148°F), or 10 kW heating at +100°C (212°F), as the batch process progresses, one vessel and one duty at a time. Ideally the waste heat from the system when refrigerating would be rejected to atmosphere.

Can anyone suggest the best technology and who can supply it? The plant will be in the UK.

Thanks

Stuart

 

[MintJulep]

For mechanical refrigeration, you are looking at a cascade system.

With your small load, you might want to consider liquid nitrogen.

 

[c2sco]

Thanks for your reply.

I'd thought of liquid nitrogen and had suggested it to my customer, but he'll need some persuasion that it's the best way. Eliminating alternatives is one part of persuading them, and if it will be a complex and expensive multi-stage machine then no doubt he'll be less enthusiastic about that. They have developed the process that they want to scale up in a lab, where they have a chest freezer that runs at this sort of temperature, so they assume getting a larger version will be easy. I don't know how this freezer works other than it's electro-mechanical.

Does anyone make such cascade machines off the shelf?

Stuart

 

[MintJulep]

I don't know that cascade equipment is ever really off-the-shelf. Almost always it is used for a special purpose, so I would guess it is almost always custom designed.

Start with the manufacturer of your customer's existing chest freezer.

Nitrogen has much lower capitol cost and almost zero maintenance.

 

[Montemayor]

The way to approach this type of problem is always:

1) Find a readily available, inexpensive refrigerant;
2) Refrigerant should have a boiling point at least 5-7 ^oF lower than the stated -100 ^oC;
3) The refrigerant should be "benign" - not toxic, flammable, or explosive;

As MintJulep says, Nitrogen seems to fill the bill. However, for such a small cooling load (5 kW = 17060.7 Btu/hr = 1.42 tons), I seriously doubt you will find interested parties to supply what is normally a complex, cascade refrigeration system. From the data furnished, the simplest and most practical answer seems to be: buy liquid Nitrogen (LIN) as a consumable refrigerant and replace it as needed. You've given a rate of heat transfer, but you haven't mentioned for what length of time this rate continues. It might be for an hour or so; we don't know.

Bear in mind that the LIN is at -195 ^oC and the excessive cold temperature might cause problems, depending on the process.

 

[pipehead]

CF4 has a boiling point of -130 C, condense it with CO2 at a -70 C.

But, I agree, LIN.

 

[c2sco]

Yes, it is rather small. The cooling at this power is only needed for long enough to cool the process vessel, say 1-2 hours, then to maintain the temperature for up to 24 hours, followed by warming up again. It's a rather unusual freeze then distil purification process. Due to the process details, which involves holding the different vessels at -90C, -20C and -10C, then heating to +80C for a distillation step, I'd hoped for something where I could set a temperature controller to the desired value and it would behave like a "normal" chemical plant. I'm not clear how liquid N2 could do this, except perhaps to cool a heat transfer (cool transfer?!)fluid. Actually, that sounds now like the best option. Does anyone specialise in making nitrogen cooled systems?

 

[Montemayor]

I’m not familiar with the compound CF4, but I believe it will be practically impossible to condense it with CO₂ at a temperature of -70 ^oC.

The reason this proposal is a practical impossibility is that CO₂ has a Triple Point (the point at which the substance exists in equilibrium with its 3 states of gas, liquid, and solid) of -56.558 ^oC. This means that at -70 ^oC CO₂ exists as a solid (“Dry Ice”) snow in equilibrium with its vapor at a pressure of 75.124 psia. It is exceedingly difficult to use the snow as a cooling fluid.

Although CO₂ has been used as refrigerant in the past, it is a very inefficient one and not cost-effective. I have used it as a refrigerant in the manufacture of food grade CO₂ and Dry Ice. In that role it proves effective because of its handy availability and cost.

Many people don’t realize that liquid CO₂ doesn’t have an atmospheric boiling point. It only has an atmospheric sublimation point – at -78.5 ^oC. Every technical text book, reference, and data book that I’ve seen has continually made the mistake of labeling CO₂ with an "atmospheric boiling point" of -78.5 ^oC. Even the NGPSA Data Book does this. I make mention of this continuing error because it has, in my opinion, caused engineers to make the erroneous assumption (as in this case) that it can be used as a conventional refrigerant – when it definitely can’t. I’ve used it as a refrigerant successfully, as I’ve mentioned, but at an evaporator temperature of approximately -50 ^oF at 100 psig. To go any lower is to risk invading the solid phase zone and inheriting a lot of problems.

 

[sterl]

Systems of this capacity and scale, done with multicomponent refrigerants, often termed Autocascade, can be "broken into" at varous temperatures but without some really sophisticated control you wouldn't want to be relying on the cracking performance of the various heat exchangers to provide a variety of different temperatures all at the same time.

Trick would be to apply a brine (nothing salt based, it would have to be something like benzene or White Gas or delimone) and mix the heated with the cooled.

Obviously, making this thing also heat would be pretty complex as well, though if the highest VP fluid was transcritical at the heat-of-rejection condition, it might all work...

I believe overall that 3-stage conventional Cascade with the lowest stage R508 will get you the temperature. Its still going to involve some pretty exotic machinery and materials but the fluids will do it.

http://www.hvacindia.org.in/journals/2000july/article05.html

a presentation by Rudy Stegmann on the alternatives, interesting reading.

Harris Cryostar makes chests for -150Deg. C. Autocascade, backed up by LN2 in most cases.

 

[dcasto]

We used methane as a refrigerant at -240 F, its cascaded with ethylene and propane.

 

[pipehead]

Thanks Montemayor, I forgot about that part. I had just finished running a similation on replacing freons with CO2 and was caught up. Actually, I was trying to point out there is always an option with wierd stuff, CF4, Freon 14 is real and I've never seamn it used as such, is also known as Halon 14. I'd still look at LIN system that uses a control loop that cools the vessels, maybe a closed loop of Freon with a thermal syphon system.

 

[unclesyd]

Here is a cryogenic refrigerator that I recently saw in use. This one had a Ln2 backup though they offer a LCO2 system This one replaced a Harris box.

http://www.coleparmer.com/catalog/product_view.asp?sku=4414125

Here is the refrigerant used in the CP system.

http://refrigerants.dupont.com/Suva/en_US/products/suva95.html

Montemayor,
Next time down on the coast if you get a chance checkout some of the fish and shrimp freezers. Quite number of them use LCO2 as the refrigerant. They let the Dry Ice formed on the product sublime to continue coolong.

 

[Montemayor]

Pipehead:
I’m glad you brought out the subject of possibly using CO₂ as a refrigerant because it allows us all to analyze its merits with other refrigerants on a level playing field. I started to envision using it as a universal refrigerant when I started out in the Compressed Gas industry, right out of college in 1960. When I saw it being used as the refrigerant in producing liquefied CO₂ product at 250 psig and -8.5 ^oF at the Harrison, NJ Liquid Carbonic CO₂ plant, I thought I had the universal solution. When I was later was sent abroad to operate and later install other CO₂ and dry ice facilities I tried to incorporate it in the Dry Ice production cycle. That’s when I was forced to get into the phase equilibria of CO₂ and its thermophysical properties. To my dismay, I found out the bitter truth – there was a tradeoff in applying CO₂. The Harrison plant was successful in its application because it seized on a convenient property of CO₂: it was readily and cheaply available and it fitted in the type of compressors employed.

Harrison plant used the conventional, 3-stage reciprocating compressors employed at that time to produce what was then called “high pressure liquid CO₂” and distributed in high pressure steel cylinders. However, the industry saw the multiple advantages of radically changing the type of product and switched to “low pressure CO₂”, which was saturated at 250 psig. This allowed the used of standard, ASME section VII vessels (called “bullets” in the LPG business) fabricated out of ASTM A-212 steel. This allowed the vessels to be designed for -20 ^oF. To produce the new level of product (called "LiquiFlow"), Liquid Carbonic designed the Harrison plant with the conventional 3-stage compression cycle – but with a major process change: the second stage, which conventionally took the gas from 65 to 250 psig was used as the last stage of compression in the production line. The third stage was converted to serve as a closed cycle, refrigeration stage while using CO₂ in a closed cycle. The third stage took the vaporized CO₂ from the main CO₂ condenser at approximately 200 psig and recompressed it up to the critical pressure of 1,100 psig so that it could be readily condensed at 85 ^oF with cooling water. The high pressure liquid CO₂ was stored in a liquid reservoir and served as the refrigerant in the CO₂ condenser, completing the cycle.

When I tried to employ a CO₂ refrigeration cycle in the production of Dry Ice I ran smack dab into the Triple Point and I couldn’t go any further. That ended that effort - real fast.

The above shows that indeed CO₂ can be used as a refrigerant. However, the caveat is that it is limited by the type of compressor used and the compression suction pressure it has to withstand. A suction pressure of 200 psig on a compressor is a special case – and consequently will be reflected in pricing. Also, as I recall, the Coefficient of Performance (COP) for CO₂ is terrible as compared with the conventional refrigerants – especially Ammonia (which I consider the best, practical refrigerant).

Unclesyd:

Nice to hear from you again. Yes, the preservation and transport of seafood has been a traditional application of direct, liquid CO₂ refrigeration by expanding the refrigerant directly onto the material to be cooled. My old company, Liquid Carbonic pioneered this type of application many years ago – in the 1930’s, I believe. The method is nothing more than the same effect of using a CO₂ fire extinguisher. The free, adiabatic expansion of liquid CO₂ produces a 2-phase mixture of dry ice snow and cold CO₂ vapor – both at approximately -109 ^oF. This effect was tried on both interstate trucking applications and rail cars. It was quickly found that direct application of dry ice on the consumable products produced a “burning” of the product and ruined it. The method worked, but not by direct contact; rather, the cold resulting vapors from the expansion and from sublimation were circulated about the consumable product and used to do the resultant cooling through a difficult gaseous film coefficient. The refrigeration and preservation of Ice Cream, however, was a different and more effective application. It is still the principal alternative method to preserving Ice Cream other than mechanical refrigeration.

I’m not familiar with any liquid CO₂ or Dry Ice applications on fish or shrimp here in the Texas Gulf Coast. Although I was born, raised and educated in Galveston, Texas (the traditional home of Gulf of Mexico Shrimp and fishing industries) I never saw or heard of it being used there while I grew up in the 40’s and 50’s. However, when I was in Peru in 1963, I furnished a lot of Dry Ice and even some liquid CO2 for refrigerating fish in transport when the Peruvians had the largest fish industry in the world, based on their catches of anchoveta. Again, the system worked there, but it wasn’t cost competitive in most cases because of the cost of the CO₂. Dry Ice is used today in such applications as those of hunters and fishermen who have to transport relatively small quantities of preserved meat product. I seriously doubt if one can allow the solid dry ice formed to come in contact with the fish or shrimp in order to cool it. The result, in my experience, would be “cold burns” which would make the product un-salable. I believe if it were to be still used today (which is likely in some situations) the liquid CO₂ would have to be expanded in such a way that the solid dry ice were kept away from the product shrimp or fish. The cold, -109 ^oF vapors would be still used as the effective cooling medium alone. However, I could be wrong since I’ve been away from the CO₂ industry for some time.

 

[TXiceman]

As small as this system is, you will be better off with a liquid nitrogen system. The cost of installing a low temp cascade unit, plus the operation will not be feasible.

Included with the design using the lower temperature refigerants, you will be faced with oil recovery problems and the lower temp refrigant pressures at ambient conditions. The settling out temperature will put your design pressure too high unless you let all of the refrigerant fade-out to vapor and reduce the pressure to a level that equipment is deigned to operate.

Ken

Ken
KE5DFR

 

[EdStainless]

I have run a low temp system that was purchased as a package. It was small and used LN2. We simply vented the gas off. They are commonly built for low temp heat treatment. You could set a temp (-75F) and it would regulate LN2 flow to maintain.
The nice thing is no rotating equipment.
Yes, you have ongoing LN2 costs, but for a small system it is an attractive option.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[c2sco]

Thanks a lot everyone. I'm convinced! But so is the customer - and so he's gone away to sulk in his lab and consider other ways of purifying his raw materials!!

Never mind, he'll be back with a new proposal soon I guess. If he can't find one, LN2 must be the way to go at this scale. Maybe next time he upgrades his plant we might consider a mechanical solution.

 

[sterl]

Montemayor:

A lot of interest in using CO2 as Industrial Refrigerant in cascade with NH3. Partially this because most larger food plants have use for Low Temp and Medium Temp and have been running multistage compression systems on NH3 for many years. BUT as production rates got higher and direct flash CO2 to atmosphere got grayer, demand for systems in the minus 60 to 70 neighbourhood led to very large NH3 Boosters (because of the specific volume so low at 20" Hg SSP); some really big pipes; and some very specialized lubricants to maintain transport and distribution at such temperatures. Even screw compressors at these conditions have some pretty specialized starting arrangements, oil cooling arrangements, seal arrangements and so on. And the maintenance issues are quite a burden, to a processor who thinks of refrigeration as utility.

The big pipe topic alone makes over-the-roof installations pretty costly overall: One installation went cascade and reduced multi pipe run load on 34-foot tall flat roof structure by almost 140 lbs per running foot. CO2 also avoids a lot of the occupancy/ leak detection scenarios that lead to some very detailed ventilation/evacuation/general alarm arrangements, required even where NH3 is in such a deep vacuum....

R717 is a great refrigerant, and we don't see transcritical being realistic in a large scale sprawling system, but CO2 installations are becoming very popular, especially for those who package equipment at the 75 to 150 TR capacity levels....