Hahaha yep deleted em, useless posts and don't add to mytek's thread.Quote:
^^NoL, why did you delete the post? Or was it a Mod?
Printable View
Hahaha yep deleted em, useless posts and don't add to mytek's thread.Quote:
^^NoL, why did you delete the post? Or was it a Mod?
mytek:
i've reading the thread and I see 3 cascade hxes and one aux, but i havn't really figured out what gases you will use.
r123-r22-r23-r14, argon.. as far as i understand, r123 and r22 will be separarted together with oil in the first separator, right?
then we have 3 gases left.
r23, r14 and argon
r23 will be separated after firsta cascade hx in second phase sep, right?
then we have r14 and argon left.. as well as 2 hxes..
argon is an absorber and you normally use a subcooler to get it work proper, right?
r14 will be condensed by the suction gas together with r23 in second cascade hx,
but whats the last hx (cascade hx 3) ment for?
cheers
Tim
Hi Tim :)
I will probably test with at least 2 different blends initially.
Freon Blend: R123-R22-R23-R14
HC/Freon Blend: R600-R170-R14
And then later try adding a small amount of Argon into the mix.
Your observation about there seeming to be one too many HX's would at first appear to be correct, but remember the ability for the system to completely fractionate the individual refrigerants at each stage is far from perfect.
So for example, if we use the HC/Freon charge which has only 3 components. In a perfect world, or should I say a perfect autocascade, you would expect the following:
1st phase sep = condensed R600 (later evaporated in cascade HX 1)
2nd phase sep = condensed R170 (later evaporated in cascade HX 2)
Final Feed = condensed R14 (later evaporated in cold head)
Notice I left out phase sep #3 and cascade HX #3. A system built like this will work. However to achieve the purest form of R14 at the end, it will require excellent heat exchange, and phase separation design. Many times it is simply easier to add another stage, of a less than precise nature, to achieve a final feed composition that is more purified. Of course the more stages that are added, means you will also proportionally lose out on the ultimate heat load that the evaporator can take on for a given system mass flow. So if you want to move lots of heat, minimize the number of stages, or look at boosting the mass flow (bigger compressor).
Also having another stage, will better enable the system to utilize either another condensible refrigerant with a lower boiling point then R14, or one that can work by absorption like argon.
Just remember too, that a subcooler isn't always required in an autocascade when argon is to be employed. This is especially true when the evaporator heat load will be moderate. A subcooler does it's best work when the heat load is comparatively large, since it is isolated from the returning heat of the evaporator.
I hope that answers your question ;)
sounds good michael :)
I'm currently playing a little with my autocascade and I can't use the hcfc's since I'm in europe.
goal:
~-70 -75*C @250w load
gases I have in stock:
r134a, r23, r290, r404a, r410, r507, r600, r600a, argon, co2
gases I aim to use:
r600, r290 and r23,
parts:
14k btu rotary, 2kw condenser
one phase sep (10" height and 1"3/4 diameter, filled with steel wool),
hxes:
12plate hx as cascade hx
and for slhx/aux condenser I have 13ft 1/4" in ½" as slhx/aux condenser due to the lack of 3/16" in stock (purged during brazing ofcourse).
what do you think about that configuration?
what I havn't figured out is captubes I have 0,026", 0,028", 0,031" and 0,042" avaiable at the moment but other diameters is no problems to get.
I was thinking of something like 10ft of 0,042" for first stage (r600/r290 liquid) and about 15ft on cold head. i talked to adam and he mentioned 7ft first and 12ft second, but I've tried this 0,042" and it's really hard to keep discharge up with a decent suction pressure on SS with 0,042" captube and 18cc compressor.. kinda worried if the rotary will be, or not be able to keep it down.
what do you think about this? I have ball valves which I intend to install captube between for easy access to change captubes... a temporary use ofcourse.. when finding the final tune a recovery will be done and ballvalves be removed and then charging pack the refrigerant.. a good idea?
regards
Tim
Hey Tim :)Quote:
Originally Posted by tim-
If they are able to deal with the cold, there is of course no problem.
Greets
I think right after cap tube maybe a bad idea, however post the flexible line, like Bazx has done will cause less problems.
3 grand for some R-14, don't think I'll be getting any!
Ya for what, like 10 lbs? I payed a crapload for my cylinder of vapor :(
$500 for R-507 and 480 for 13.6Kg R-124
Mr. Pony; using HC's R1150 (ethylene -104C) with a small amount of R50 (methane -161C) might be just the ticket to substitute for the R14 (-128C). I don't know if this would be cheaper, and/or easier for you to get in your area, but it would probably work just fine. The amount of R50 needed would be very small (5%-10%), and it would definitely require at least 3 Cascade Condensers with 3 Phase Separators.Quote:
3 grand for some R-14, don't think I'll be getting any!
Yep thats something I want to try once I get a bit more cash on hand.
Michael, sorry for high jacking the thread, but do you think that you could use the mix of ethene and methane you metioned for the third stage of a conventional cascade? That would be awesome since R50 is cheap, and ethene is anyways needed for the second stage.
This means that ethene and methane go into solution like R14 and argon?
No hydrocarbons don't necesarily work as solute and solution. They can actually make mixes in between that fully condense. Look up non-polar solutions for more.
Wait for Kayl's project. Two stage cascade with twists.
Thanks Michael I was rather disapointed upon hearing that price! R-124 thankfully is within reach albeit pricy. and I all ready have a SGHX ready and waiting ;) Now it is just saving up and geting the gases and prefabed HXs
Quote:
Originally Posted by n00b 0f l337
;) Just have to collect some more parts before i start :up:
I suspect that it would, but maybe not to the extent that it would in an autocascade, since we effectively have 2 HX's and a phase separator working with the R1150 and the R50 part of the mix (Cascade Condenser 2 and 3, and Phase Sep 3). This gives us a thermodynamic advantage over the basic 3rd stage of a conventional Cascade. Of course you could always make the final stage of your Cascade into a simple autocascade (Aux and Cascade Condenser with a single Phase Sep). Now wouldn't that be a sweet solution :up:Quote:
Michael, sorry for high jacking the thread, but do you think that you could use the mix of ethene and methane you mentioned for the third stage of a conventional cascade?
This still allows you to make a fairly easy to tune system, since the first 2 stages use only one refrigerant, and the 3rd stage utilizes a binary mix, which is a lot easier to tweak, than a full blown autocascade with 3 or more refrigerants.
Yes it should, and a combination of R14, R50 (methane), and Argon has been used in one of Polycold's smaller units to help bridge the gap, and yield even better load handling at low temps.Quote:
This means that ethene and methane go into solution like R14 and argon?
No problem :) , but don't you mean R-123 instead of R-124?Quote:
Thanks Michael I was rather disapointed upon hearing that price! R-124 thankfully is within reach albeit pricy.
No, I meant R-124 as you stated it would work best with water cooled units, and it is the only one that is reasonably affordable in the quantity range that is of use for me.
Any and all units I make will be water cooled any way so it will work nicely.
Yes that would be a good choice for your application, almost the same as R-600a. The only downside is that it will require POE oil.Quote:
No, I meant R-124 as you stated it would work best with water cooled units, and it is the only one that is reasonably affordable in the quantity range that is of use for me.
It seems to be the only oil around now days in these parts any way! Speaking of which need to pick up another can of the stuff.
Mytec, may I add my thanks and also admiration to the collection of thanks that youve been offered here. Damn nice work in a field thats clearly black magic. Ive read this whole thread over the last few days and the fact that youre prepared to share your hard won experience with the board is very generous.
I did have a couple of questions too.
When I first read about autocascades on this board, I got the impression that they only worked properly at one specific load, and that if the load strayed from this (either higher or lower) the performance would drop off rapidley. But some of the things youve written seem to suggest that that is not true. Could you comment on that? For example, if this system you are now creating is tuned to do say 100w at -130C, what sort of temp would you expect at say 75w, 50w, 25w and 0w?
The other thing is that this thread has me wondering how low you can take this technique. Two things come to mind for me when I see this thread and contemplate following your design to make a copy - one is to act as a first stage for a stirling cooler, JT cooler or pulse tube cooler, with the aim of liquifying helium (I would consider this the holy grail of cooling no?), and the other is to make my own liquid nitrogen which would be useful for some of the vacuum work I like to do sometimes (electron tubes and vacuum stuff).
So how about going down to say -180C - I realise that you might not have done this before but in your opinion do you think its possible?
Thanks again for sharing so much information!
Pete
You can get to a methane stage and beyond yes, so liquifing nitrogen could be possible, large amounts is incredibly hard though.
Peterpion let me see if I can answer all your questions...
Yes an autocascade is not limited to a narrow range when it comes to load. Although just like any refrigeration system, there will always be a preferred, or should I say "as-designed sweet spot" load point where it will work at its best.
Many of the Polycold units I have worked on, have quite a diverse range of loads as used in the various customer applications. The biggest unit (PFC-1102HC) will handle as much as 3600 watts of heat load, while still producing an average evaporator temperature just below -100C. But it can just as easily (and comfortably) run at 150 watts forever, with an evaporator temperature of -140C.
However reducing the load even more (10 watts, 50 watts, ect.) wont produce much colder temperatures, since the proportions of the refrigerants in use have pretty much done what they can.
As Nol pointed out, changing the make-up of the charge will allow even colder temperatures, but at ever decreasing efficiency. As an example; the same system with additional Argon, and less R14, can be coaxed into even colder temperatures (-165C @ 100 watts, -145C @800 watts), but it can no longer handle loads much beyond 800 watts.
And yes -180C or colder is possible, and was done in a few prototypes when I worked at Polycold. It took additional stages, and additional refrigerants such as methane, nitrogen, and neon. But in order to get any useful cooling capacity at these temperatures, the system was gigantic with a 25 HP compressor.
The idea of combining an autocascade with other refrigeration methods as you suggested (stirling cooler), might produce some interesting results, but I can't say that I have ever tried such a thing. Although in theory it would seem like a good approach.
Good question, and I'll see what it does when I get to the testing stage. Which if all goes well (unfortunately it never does), I should be conducting the first load tests late next month.Quote:
if this system you are now creating is tuned to do say 100w at -130C, what sort of temp would you expect at say 75w, 50w, 25w and 0w?
Here is the state of the AC-2 Project as of today:
- Heat Exchanger Thermocouples have been installed
- Refrigerant Line has been Insulated
- Heat Exchanger Stack has been Foamed
Next will be trimming off the excess foam insulation. If you are wondering why there is so much extra on the top, it is because #1) I wanted to fill all the voids and required that the foam rise with a lot of pressure, and #2) because it is just plain difficult to judge how much will be required.
I also need to stretch out the line sleeving, glue it down, and cap off the end.
Looks like I am getting pretty close to firing this baby up, just need to do the wiring, and add an access fitting for evacuating the expansion tank. Unfortunately I wont be able to initially chart more than 8 thermocouples due to limitations in our test software. So I think I'll chart out the following which should give a good overview of what is happening:
- Compressor Suction
- Compressor Discharge
- Strainer#1-4
- Eaporator Inlet
- Evaporator Outlet
The last photo shows AC-2's Big Brother in the background. Interestingly enough, AC-2's Heat Exchanger Stack is very similar in geometry. However Big Brother has 2 stacks in parallel.
Impeccable assembly mate!
Looks great! I love the huge autocascade in the background. "Say hello to my little friend". Can't wait to see you fire it up, really interested in it's performance.