Introduction to Cascades and Autocascades

July 1st, 2006 at 15:46 under Articles, Phase Cooling.

I thought I was extreme when I first stepped into the realm of subzero cooling. With my paltry peltiers and -10 Celsius I thought I was at the pinnacle of extreme computing till I discovered XtremeSystems.org. From there it all went downhill with stacked peltiers and eventually stepping up to vapor phase cooling. Some people may consider using liquid nitrogen to be the most extreme I really don’t think LN2 is that extreme. Awesome, you pour a really cold liquid into a copper tube, amazing. However my newly discovered passion, autocascading, would definitely qualify as extreme and to some people ludicrous even.

For those of you that are new to phase cooling then jumping headfirst into an autocascade would be very detrimental to your health. Regular phase coolers always have the risk of a leak and a possible explosion if not properly built but autocascades are an entirely different beast. I would compare the two like comparing a 500HP Corvette to a 2000+HP nitro methane injected drag car. Yeah, the Corvette is fast, but the drag car is in a league of its own. If you would like some easier reading and want to learn the basics go check out the articles I wrote on general cooling, watercooling, and basic phase cooling. Now as with all my previous articles on potentially dangerous topics, if you happen to hurt/maim/kill yourself while making something similar, I am not to blame. I will be simply providing research and documentation of what I did, however feel free to ask me questions if you yourself are building a similar machine. Now with that clearly stated, lets dive into some research and designs.

Basic Principles

In basic refrigerators, freezers, and vapor coolers for computers the same properties of evaporation and condensation are used to carry heat. When a liquid evaporates it takes some heat with it, you experience the same thing when you sweat. When a gas condenses it releases some of its heat which you may notice with steam rising and condensing against a lid on your stove. In a single stage unit a compressor raises the pressure of the gas and then a condenser uses fins to remove heat from the gas, causing it to condense to a liquid. This liquid is then forced under high pressure into a large chamber called an evaporator where the pressure drops and causes the liquid to evaporate. When the liquid evaporates it absorbs heat and is then sucked out of the chamber by the compressor and the cycle repeats itself. Simple systems like this are used to cool your refrigerator, freezer, and possibly your dehumidifier. A cascade system takes this process and steps it up by using the evaporator to cool another gas.

Two Stage Cascade

In a basic two stage cascade there are two loops of gas that are separate from each other. The first loop uses a compressor, a condenser, and a heat exchanger in a closed circuit with the purpose of keeping the heat exchanger really cold. The second loop differs quite a bit with there being a compressor, heat exchanger, and evaporator. The gas in the second stage tends to be a gas that doesn’t condense at normal temperatures, hence the subzero cooling from the first loop. After the gas passes through the heat exchanger it is passed into the evaporator where it expands and sucks up an enormous amount of heat at a very low temperature. This system is considerably easier to tune and design than an autocascade but you also have to account for having two large compressors and the electricity they will consume. With two decent sized compressors eating between 1 and 2 kilowatts of power I would never consider this to be a feasible solution for 24/7 cooling so I personally won’t be building one of these.

Two Stage Cascade

Autocascade

Autocascades are nearly opposites to cascades in that they requires a single compressor, are very difficult to build, and even more difficult to tune. Autocascades are considered the most difficult of phase cooling systems to build and tune which make it the perfect project for me to undertake. Besides the difficulty of tuning, autocascades are also finicky and when thrown just slightly out of shape they can quickly approach stratospheric temperatures and pressures. However there exist many devices that can be used to reduce the threat of a rupture which I will definitely be implementing for my safety and the safety of those around me.

To start with, autocascades utilized a mixture of gasses with one of them being the “low” gas and one of them being the “high” gas. Starting at the compressor, this mix of gasses enter into the condenser where the low gas begins to condense into a liquid. Generally the low gas will be an easy to condense gas like propane, r134a, r22, and other similar gasses. After the condenser the mixture of partially liquefied low gas and high gas will enter something called a phase separator. The phase separator creates two paths, one which the denser liquid travels to the bottom while the lighter gas travels to the top. Here you want to separate the different phases, gas and liquid, so that you now have divided the two gasses. Next the two separated gasses reach a heat exchangers where the liquefied low gas expands and absorbs the heat from the high gas and causes the high gas to condense. The heat exchanger will normally be around -40 to -30 Celsius so your high gas should be capable of condensing around these temperatures. Common gasses for the second stage include carbon dioxide and r1150. From the heat exchanger the low gas will travel back to the compressor while the high gas will continue to the evaporator. Within the evaporator the high gas will expand and then it to will travel back to the compressor to complete the loop.

Cascade

Since there are no mechanical separations of the phases you might begin to see how difficult it could be to tune this system. Even when tuned though if the external temperatures fluctuate to greatly you could potentially not condense the low gas or perhaps not enough of the high gas would condense, or a thousand other things could go wrong. This is what poses the challenge of building on of these machines and this is what attracted me to this project.

That is the basic highlights of the difference between an autocascade and a cascade. During the next two to three months I will be building, tuning, testing, and improving my design till I hopefully have a working system. Don’t worry about the material I write here though, while there will be some highly technical jargon thrown into the articles, I will try to keep it readable so that you too can enjoy seeing how I progress. I have created a new tag called “Vapor Cooling” in which I will add all my vapor cooling articles in an attempt to make this place easier to browse and search through.

Trackback/Ping URL

Speak up and be heard!

  1. NoL / n00b 0f l337 Says:

    Not bad, not bad at all.

  2. Chris Morrell Says:

    Thanks man, coming from you that means something. Hopefully other builders will appreciate what I am trying to do.

  3. AdAbsurdum Says:

    In all the articles i’ve read i’ve found your style easy to comprehend which is uncommon. Thanks for sharing your adventures.

  4. Chris Morrell Says:

    That is probably the best compliment/comment I have received so far. I try very hard to keep everything simple and try to lay everything out so that even my little brother can figure out what I’m talking about. Thanks for the comment.

  5. wdrzal Says:

    You Forgot to mention the peressure drop is caused by some type of metering device,…….while the evaporator is low pressure it dosn’t drop there because it’s “a large chamber” as your explanation seems to suggest.

Leave a Reply