Three Stage Cascade

It blew up like a party popper, a white silly string like substance stuck to everything in its path.

CC, good to see you around.

I agree, we will need to hit ~10c beyond condensing temps of the next stage to be safe. With the expansion valves ability to maintain temps under load we dont have to get excessive here to maintain the capacity of the load. With a cap tube I can see where this would be a problem.

We need to hit -100c on the second stage @ ~ 20 psi to get the methane condensing.

A bypass valve may also help on restarts to equilize the pressure and prevent another blown compressor.

We needed a video of last night, two blown compressors (one doa, had start problems on first stage unloaded with R507), A blown cap.

Once running we had no problem getting the first two stages going in no time. They ran for hours and we swapped out the third stage compressor after failure with the first two stages running.

Though i dont know the specs of your plates, judgeing by their looks. I would say they might be a bit undersized for what you are trying to achieve... well with what compressors your ussing.

Lets see if i can make sense here?
plates offer a great deal of suface area. good
plates offer little (if any) restriction. good

Now lets take into acount the raw thermal conductivity of methane in its gas form. right.

Shes a thin little bugger of a gas. So what this means is that though a plate may be rated for 1/2 a ton. it aint gunna pull a half a ton. do you understand what im sayin?

you may need to make that 2nd to third interstage a bit larger bud. :( :(

Oh and you guys should call me next time all of ya hook up. maybe ill swing by and make some :banana::banana::banana::banana: cold with ya :D

Quote:

---

Originally Posted by pc ice

Blew up a capacitor, dont stand over a start cap on a system.. Covered in Oil from the cap,,, More to follow...

me and opp can vouge for that.. :rolleyes: kinda suck's when they start to smoke and whistle before they pop.

---

I dont want to see either of you vogue, strike a pose, bust a move, or even get down on it. thanx

sorry i couldnt help the cheap shot. i cant spell either lol

Actually the plates are five by twelve inches by 10 plates so the surface area is 60 x 10 x 2 -(40 for the two outside plate sides) or 1160 sq inches. almost 10 square feet. Considering the uppr half of the plate was uninsulated and there was no insulation on anything else. I needed to see where we could improve, refrigerant flow is a little differant in plates Unlike tubing both surfaces of the indivigual plate are used for heat sink area. As for the 1/2 ton rating that is at 3 degrees F at 38degF with 10 GPM of water and R22 gas. And as subcollers they are rated at 5 tons with a 25 degF TD. I suspect I blew the compressor a few weeks ago when I tested the pressure limit of the compressor. (Danfoss sc12mlx reaches its pumping ability at 10 psi into 600 psi and the compressor stops pumping and bypasses) DON'T TRY THIS ONE AT HOME PLEASE
The midstage needs more subcolling as the discharge temperature is high. I really need some R14 as we are at the nessassary condensing temperature for that.

i hope you understand my crapy drawing. its just a plain old condenser with three seperate runs for subcooling

So you are using the suction lines to shuffel the heat to the first stage with out using the second two compressors?
Is this what you mean? Like a tube in a tube in a tube in a tube in a tube in a tube. This is what I had in mind after the trial run...
http://www.blairwing.com/images/3stage/4.jpg

ugh your killin me, It isn't anything super inovative. Its just a cool easy way to pipe a desuperheater.

ill explain again.

serpentine condenser.
from the inlet to the outlet, lets say it goes back and fourth 10 times.
in a serpentine fashion.

now cut it so that we make 3 seperate condensers.

in goes in through the inlet.
goes back and fourth in a serpentine fashion.
but then we cut it.
so it only goes back and fourth 3 times
then it goes out to its respective condenser.

lets call it a 3 in 1 pre condenser

someone help me

Sorry cap. Still not making any sense. Sounds to me like it is only a desuperheater. Got some quility Pics? Or are you talking about an air cooled condenser? Where you take the loops of the second and third through the condenser itself? I am installing a small air coil after each sucessive stage also the heat load calls for it also.

this way, cc?

i dont beleive hes having trouble understand the concept but more the ending result.

So why add an air cooled condense rto teh first stage when One is already there?
Add an air coil to teh discharge to the second and third, already have these. I will need to expant the condenser or pre condenser for the second stage substiantially. The subcooler will be made of 6 concentric pipes of 14 inches to 16 inches with y fittings...

awsome chilly1, but where did u get argon?

liquid air or boc gases

NGC gasses affiliate of Aire Liquide

First of, I'm in no way claiming to know better than you guys. I have never even built a phase-change system myself yet.....

But I'm really intrigued by reaching really low temps, and I have read most of the amazing book: Cryogenic Engineering by Thomas M Flynn. It has taught me a lot on the physics involved in reaching these temperatures.

One thing that puzzles me about all the approaches you have chosen is that you don't seem particularly concerned about reusing the cold you have worked so hard to achieve. What I mean is that what you do with the gas from the third stage in you setup chilly1 is to use a fraction of the remaining cooling potential in it to subcool the incoming fluid/gas mix to the evap. But after that it is sent directly to the compressor! What a waist! If that gas is leaving the evap at -100C (if your goal of -150C evap temp that is even a bit high?) it still have quite a lot of cooling capacity (and that at a really low temperature even!).
What is important here is at what temps your heat loads is at. You all know that lifting 1 W at -150C takes WAY much more effort and energy than lifting 1 W at +20C. This implies that it is also les energy demanding to cool a gas at reasonably high temps (in the 0C range) compared to cool it to really low temps, and even possible to do it "for free" if the temp is above room temp.

This can be translated in to the following:
If you want to reach low temps you will gain efficiency by cooling the gas in the third stage of your cascade in several steps. What you want is to use an as efficient cooling as possible. First of you should implement an air (or rather water) cooled HX to get rid of excess heat above room temp (water temp) introduced by the comp. This will releave the second (and first....) stage of having to lift the heat introduced by the third stage comp. Then you would possibly want the next cooling stage to be done by an as efficient system as possible meaning the first stage. This stage is possible to make really efficient compared to the cascade the first and second stage is made up of. This way you will not have to use the not so efficient cascade to cool the gas from around room temp to a reasonably low temp (lets say about the evap temp of stage one….), thus using les energy to cool the gas, thus being able to lift more heat at the low temp.
Then there is the possible cooling potential of the gas leaving the CPU evap. after it leaves the subcooler. Remember that this is low temp, and therefore precious heat lifting capability that should be put to use. Remember that each heat lifting Watt here has gone through 3 stages of cascade and used a lot of energy to get at that low temp. This heat lift potential could be used to cool the gas in stage three before it enters stage 2/3 HX. Each W lifted from the gas before it enters the HX gives one more W heat lifting capacity or less energy used to cool stage 3.

To sum it all up, by using cooling processes that work in the appropriate temp region and to split it where the temperature range is to big is the key! The energy needed can fast reach 100 W used to lift each W at -150C! So you want an as efficient system as possible if you don’t want a HUGE electric bill…… and a hot computer room :)

It also seems to me that you should be careful to make sure the HX’ have good recuperative capabilities…. hm… not sure if that is the correct word. What I mean is that it is important that they are counter flow, and that the temp of the cooled gas leaving is close to the temp of the cooling gas entering. Plate HX’ with a wide footprint (meaning they are fairly wide compared to the length) isn’t as good in that respect, and tend to spread the heat to well if you understand what I mean.

In reaching these temps the HX efficiency is also really important, but the most important factor is in fact what is happening at the top of the system! Flynn had one comment this. Most designers seem to forget that all the work is done on the top (in and around the compressors where the temperature is high) and 99.999% of all refrigeration research has historically been done on the cold end. I think we should take this in to account and start by thinking again at what ways we have of improving things in the system.
Precooling is a major efficiency increasing factor. An example in the book describes the way precooling with LN2 to make LH2 liquefaction increases the capacity significantly. And the amount of used LN2 isn’t even more than about the same as the amount of LH2. And since LN2 is a LOT les energy demanding the net result is a huge saving on the energy used.

To get a better idea of what I’m talking about take a look at the Linde Liquefaction apparatus and how it operates. The key here is the heat exchanger that preserves the cold at witch the cooling takes place and makes the apparatus able to cool it to so low temperatures. If the HX’ efficiency is to low it will limit the lowest temperature it is able to reach.
The problem is that if the HX isn’t able to transfer enough energy needed to preserve the cold the energy the temperature reachable by the process is limited by this. And at the temperatures here the efficiency of the HX’ need to be better than 90-95%! So that is another place I would start looking. I’m sure you guys know how to measure and calculate it.

But to me the temperatures we are talking about here seem to call for other methods of cooling. And you chilly1 seems to have most parts of a G-M refrigerator capable of really low cryogenic temps. But the costs to operate it alone (both the ultra pure He and the absorbers in the filters) seems rather high to me. And then there is the efficiency problem. You have seen the compressors needed for this system, and they are as you have seen LARGE (wasn’t it 5 Hp or something???)! That is because the efficiency of this process is magnitudes (as in possibly 1/10 ) as efficient as the more efficient methods (phase-change, stirling, turbo-bryton and others). The reason for it being used in the vacuum chambers as you probably already know is the durability and the ease at witch cascading is accomplished.

My (over a very long time though) plan is to try to make a stirling cooler. I have a CNC-mill (oh well, soon the z-axis is converted too) in my basement, and I’m going to use it to make both stirling coolers and motors. My plan is to make a not so high-tech version of the wisper-tech stirling motors.

Hmmmmm…. sorry for making this such a long rant, but it’s kind of hard to explain these things in short terms in my own language, and even harder in a language not my mother-language.

And again, I’m NOT saying I know more than you guys. On the contrary…. I’m just trying to understand how real life implementations are done of the engineerisch (my second mother-language too :) ) language discusions in that book. BTW I really recommend this book to those of you really interested in reaching low temps (and understand engineerisch…..)! It has whatever you want to know of cryogenic properties of both fluids (gasses at normal temperatures) and solids. A lot to be learned on heat conduction in there! And there is even a whole chapter on insulating that is really informative and useful for our purpose! It’s expensive though, so a library visit is probably the best way to have a look, as I did.

akb1212 going back to lurking mode..... :cool:

Here is a drawing of my propased changes to the three stage as a result of findings from the first run. And yes I am trying to simplify the three cascade as well as make it more effecient. At the same time I wish to stay inside the relm of what I can build in my garage or purchase off the shelf and also power with house current. I do have economizerizers on this cascade but there is way more heat being generated by the second stage than I antipiciated and I have a solution for that also. There is a limit to how much superheat you can add to the gas entering a compressor and still be able to compress it. SO itf you heat the suction line too much you will end up too high a suction pressure and you will lose effeciency. The Subcooling HX in the drawing should shuffel the heat from stage to stage bypassing the compressors. This may help... http://www.blairwing.com/images/3stage/3stage.jpg

We got the compressors to cool down considerably. The methand was condensing in teh third stage but it does not have the capacity to work in this system. R14 will as teh mid stage condensing temp (R503) is -85C first stage was-30C and the final temp with Ethlyene was -114C, Looking for R14 for this rig now. sent it over to fuggers shop for bench testing. Its been a learning experience, We need to find some better gasses for the third stage or add a fourth either an autocascade forth and fifth or other method to condense Argon or nitrogen.
Well here are the final pictures I took here...

http://www.blairwing.com/images/3stage/5.jpg

http://www.blairwing.com/images/3stage/6.jpg

http://www.blairwing.com/images/3stage/7.jpg

http://www.blairwing.com/images/3stage/8.jpg

http://www.blairwing.com/images/3stage/9.jpg

holly sheat....u r crazy...!

We ended the night with a blown circuit breaker and didnt have the camera ready to take pics to the temp meters. With proper insulation and finer tuning I expect to hit closer to -120c

I also have the capability to subcool the entire unit 20 to 30 degrees colder than Cilly1's garage.

The third stage compressor ran nice and cool, it had plenty of cooling getting back.

The R503 is pretty nice, it drops to temp very quickly and has huge capacity. I have never seen the R1150 drop so fast. Watching the meter -19,..-25,..-46,...-72,..-95,... After we pulled the methane with first and second stage running we shot in some R1150 and fired it up.

We did condense methane. We reached -60c or so before making the call to switch back to R1150.

Add another Aspera 3/4HP compressor to the dead list.

It refuses to start with bypass wide open, it even tripped the breaker without a load.

Going with tripple danfoss SC12MLX

man u guys are lunatics!

600 posts, wheee!

man i cant belevie what i see that is a work of art guys. :slobber: :slobber: :toast:
well done cant wait to see some more

you added fans on the oil cooler/desuperheater????

After the first run the temperature and test results pointed at different areas to remove heat from and not add it to the interstage hx's This made it possiable to condense methane, which BTY has little heat capacity looking for alternatives now.... R14?

think its a good idea in a regular cascade?