AutoCascade with no phase sep

[Blaster]

Since Stephen's thread was closed i decided to open this topic about Cryo's autoC with no phase sep. This one below:

http://helderfonseca.planetaclix.pt/autocnosep.JPG


Had bunchs of fun building this simple design and charging it over a whole weekend.
Yes, it was a rather ghetto build cause i was only experimenting, ussually i dont show my ghetto experiments but i wanna show it for you guys to see the results.

Cap sizes used:

Cap#1 48” x .026
Cap#2 60” x .026
Cap#3 26” x .026

Gases were supposed to be r123/r22/r23, but since i dont have r123 or r23 i tried with r600a/r22/Co2
Also was supposed to use 3/16'' inside 3/8'' but i used 1/4 inside 1/2''.
Used a 1lt expansion tank as well.

http://helderfonseca.planetaclix.pt/autoc0001.JPG

Insulated it with glass wool, as you can see it was very ugly, but i was just having fun :D

Results:
temp at 1º cap: -12C
temp at 2º cap: -34C
temp at evap: -56.5C

Working pressure: low side: 18 psi
highside: 155 psi

Sorry didnt bring cam when was testing.

So from this results im now trying to find a cheap source of R23 to recharge it, cause with those pressures it was the CO2 evaporating at evap. Couldnt get any colder cause of dry ice forming and blocking (at the time i wasnt understanding why lol).


So, s7e9h3n and Cryo hang in there cause there are ppl that are listening and want to learn :D

[n00b 0f l337]

Haha I was going to be giving that design a shot. Thanks for taking the first plunge.

[s7e9h3n]

Nice work...and this was with a 1/5hp compressor? :clap: Any idea of capacity?

[Blaster]

its a 1/4hp unit

sorry, didnt tested it for capacity, but when work slows down i'll put my load tester on to it

[Revv23]

Nice!!

I really like that. What i want to know is how do you ensure that liquid is what enters the captube and not gas? I would think that the high pressure gas would be the first thing to go into the captube... Any one feel like explaining whats going on here to me?

-Revv23

[[XC] mysticmerlin]

Nice!!

I really like that. What i want to know is how do you ensure that liquid is what enters the captube and not gas? I would think that the high pressure gas would be the first thing to go into the captube... Any one feel like explaining whats going on here to me?

-Revv23

me to, just not getting the autocc thing yet. thanks

[Revv23]

me to, just not getting the autocc thing yet. thanks


Well typically an autocascade works in a way similar to a cascade, you have a high pressure gas and a low pressure one. Only in an autocascade you mix those two together, all in one system.

So the low pressure gas condenses in the condenser, and the high pressure one just keeps going through it, then you enter a phase seperator, and the high pressure and the low pressure gasses get seperated (these can be similar to an oil seperator, the idea is that the liquid stays in the bottom and the gas rises to the top.)

Next you have a typical cascade heat exchanger, the low pressure gas evaporates in it (and goes to the suction) and the high pressure one condenses and goes to the evap.

Thats how our traditional autocascades have worked. This one on the other hand I'm not so sure because it doesnt have a phase seperator and it is three stages long.

[jinu117]

awww... someone beat me to it... gotta love getting work done instead of playing :P Is that loaded temp? than quite good I must say!

[wdrzal]

Stephen those number in cryos avatar don't mean a thing.at least to me

Could you answer our new mods previous question and give the amounts of each gas by weight by weight please?

[Blaster]

Ive done this almost 3 weeks ago Jinu :p , and those are not loaded temps, like i said was just having fun experimenting, at the time i wouldnt understand why could not go any lower than -56,5C.

Cryo calls Partial separation AutoC, so i think theres not a complete separation. He would be the one to better explain this but i think it takes advantage of the fact that in a tube gas and liquid velocity are different, gas tends to have a much higher velocity so it travels through the center of the tube pushing the liquid against the walls. So to work better capillaries need to be as close to the tube as possible.

Charging it was good fun, have absolutely no idea how much quantity of each gas ive put in, did tried to have some sort of method by the pressure gauges, but ended up putting and removing the gasses in small quantities all afternoon testing it.

One thing im sure, it was autocascading, no way r22 would give -56 with low side pressure of 18 psi. At the time decided not to test it any further cause need to finish the 2 stager that im building for my benchmarks.

[Unknown_road]

what happens in here is very simpel, the r600a condenses in the condensor, because it is liquid it is heavier as the gas, so the first cap tube lays in the liquid r600a. the cappilaire effect of the cappilair sucks up the liquid. That's just how cappilaire effect works. gas doesn't flow easy trough such a small tube. the r600a that exits the cappilair boils off making the r22 able to condens etc.

[epion2985]

Can someone draw a better picture or explain as I dont think I understand. All I see is a single stage with a SLHX, and a second cap line going from god knows here to god knows where.

[Xeon th MG Pony]

Very nice test set up there mate looking forward to see how this progresses


edited by runmc :)

[jinu117]

Well... I never thought complete separation of gases is the way to go or would happen in practical manner. Both for capacity and involved circuits. It could be possible and probably get colder temp but would need bigger displacement in essence than what could be done with partial separation.
Possibly reason why we had so much problem with load on autocascade (other than not so optimal gases used).
as for charging from my slhx enabled single stage experiment (thing held -50ish load at 300w toward end of experiment... don't ask discharge pressure though lol)... I used refrigerant scale and went by oz per each gas until decently than by .1 oz bleh. What i did was make t with 3 1/4 flare connection, 2 connected to 2 different gases (well may need cross for 3 refrigerant), put refrigerant on scale and end of t connected to 2 refrigerant and low side, and alternating charge carefully and write down the amount used in weight.
Will do same when I do auto experiment (Single stage SLHX was dropped once I found out it was doing partial auto cascading and I had to make SLHX to some exact size to accomodate varying pressure temp condition to work within pretty small margin of error to make it not go too high pressure...)
Way too much tuning needed for extra 10-15c I was looking for.

[Unknown_road]

wdrzal feel threatened? He doesn't sell anything, he doesn't get money for the info he gives, he's just helping people out here and now he is curious how the system works as lots of people here. And with works I mean how do you easely charge an autocascade, how do you prevent refrigerants migrating to the cold etc. those diagrams I have seen thousands of times years ago.

@jinu: I'm working on an autocascade design in which complete seperation takes place. When it works I'll post everything I've got on it.

edited by runmc ;)

[Blaster]

found another pic lol

the solenoide was there for no reason as it was already part of the unit

http://helderfonseca.planetaclix.pt/autoc0002.JPG

[wdrzal]

But to quote to my dad, you don't know if your ass hole was bored or punched !!!:D

edited by runmc :wasntme:

[wdrzal]

quote by blaster "i think it takes advantage of the fact that in a tube gas and liquid velocity are different, gas tends to have a much higher velocity so it travels through the center of the tube pushing the liquid against the walls. So to work better capillaries need to be as close to the tube as possible."

I think you may be geting on the right track, study how a vortex tube works.

[ozzimark]

guys, drop the philosophical humanistic discussion and focus on figuring out how to build one of these things. i was honestly pissed when i saw what stephen's thread had come to, and i'm annoyed at the attitude showing up in here too. is it really too much to ask to just put things aside and work towards trying to build a good auto unit? :toast:

[wdrzal]

I have always challanged posts that that I beleive are either unsafe,factually wrong or misleading. Like I said before if you hold up a dog and claim you cloned it, you better be able to explain how.


edited by runmc:D

[n00b 0f l337]

Blaster is your flow up through the HX or down through the HX?
If your trying to condense in the tubular design you want the most low stage to make it to the evap capillary. Therefore wouldnt you want to flow downwards and hope the most of the liquid makes it to the evap?

[Blaster]

Blaster is your flow up through the HX or down through the HX?
If your trying to condense in the tubular design you want the most low stage to make it to the evap capillary. Therefore wouldnt you want to flow downwards and hope the most of the liquid makes it to the evap?

thats a very good question,

discharge is going upwards and suction is going down, when i was building i was thinking about it

if it was reverse my thought is that it would be a little bit better for discharge line but the suction woud tend to flood a bit too, dont know if that would affect velocity of liquids and vapours... guess someone will need to test it :)

[wdrzal]

Is a few pictures and few creditable paragraphs too much to ask for ???

If you think a auto cascade is just suction line hx and these basic phase seperators,I think your seriously mistaken.

here a diagram of a coal liquafaction plant (state of the art) looks simple

edited by runmc :toast:

[Unknown_road]

the gas speed is too low for the liquid to be pushed to the sides of the tube I think. Gravity will win this fight I think :D

[wdrzal]

why not just seperate the gas into 2 parts, hot gas and cold gas

[jinu117]

thats a very good question,

discharge is going upwards and suction is going down, when i was building i was thinking about it

if it was reverse my thought is that it would be a little bit better for discharge line but the suction woud tend to flood a bit too, dont know if that would affect velocity of liquids and vapours... guess someone will need to test it :)

Hmm is it possible that discharge going up is actually limiting your flow? I mean on plate hx i always put metering device to bottom port -> top suction while discharge on another stage comes from top->bottom. In fact, this is the recommended way to plumb for hx. I don't see why it would be different in our case with autocascade since we still want steady stream of liquid from hx as well as evaporated (vapour) gas going up hx. Just some thought... :)

[Blaster]

Hmm is it possible that discharge going up is actually limiting your flow? I mean on plate hx i always put metering device to bottom port -> top suction while discharge on another stage comes from top->bottom. In fact, this is the recommended way to plumb for hx. I don't see why it would be different in our case with autocascade since we still want steady stream of liquid from hx as well as evaporated (vapour) gas going up hx. Just some thought... :)

yes, like i said, i was thinking about it when i was building it but it was already brazed so my try was with suction going down... humm the slope is not that much :D

question: so who will try it the other way around? and with r23 cause CO2 is limited to -56 or so... im trying to get R23

right now with the litle time i have left need to finish this (http://www.teampuss.com/forums/viewtopic.php?t=1251)

[epion2985]

never mind my question above, I just understood it.

[jinu117]

I would be... but using separation :P
gimme a week :) Probably will use r23 or r116 both which I do have handy... (my wallet isn't that handy anymore though)

[n00b 0f l337]

Grrr i need low stage gases. :(