Big_Sam's R23 cascade... need help!

[Big_Sam]

Greetings!
Thats my first post. sorry fo my bad english.

After making couple of simple phase-change systems, i finally decided to go cascade.
Parts:
2 Aspera compressors- 2178Gk and 2168GK
P404a for the first stage and R23 for the second
Lytron LL520 heat exchanger ( http://www.lytron.com/standard/he_liquid_drawings.htm )





When i started building this cascade i didn't know what freons
will be used, so first i tested it with R404a on the first stage and
R410 on the second.



Then i got R23 in the second stage and sinse then i've been having problems with load.
Without loading it looks so :



My main problem seems to be the capillary tube in the second stage.
I think i shoud cut some of the lenth. My goal is to get this cascade stable at 220-230W...Certainly I shall pick up by practical consideration length of cappilary tube, but on it many time and work will leave.
I would be very greatful if someone could help me with it...

[chilly1]

YOur capillary tube is too restrictive. R23 boils at 0 psig at -82C so you are in deep vacume if you go below this your capacity will also drop. Ideally you need a captube length that will give you 0 to -.1 bar with a static charge of around 12 to 14 bar.

[Big_Sam]

I was hoping to find R1150 so i put 2.8m cap tube 0.8mm diameter. But when i got R23 i thought i'll give it a try the way it is. Static charge was around 13.5bar, and with no load i got 7.5bar on the high side and almost vacuum on the evap. Thats when temperature dropped to -101C.
To get the system working under load i charged some more R23:
static charge to 17bar. With no load pressures went to 9bar on the high side and 0.4bar on the suction line.Temperature stabilized at -93C.
To emulate heat load i use 90w peltier. So with this setup system holds at -82C.
After that I tried cascade on a real system: P4E @ 5Ghz, when idle temperature stays @ -70C but under load it goes up very fast, all I could bench was SuperPi 1Mb: http://images.people.overclockers.ru/22208.gif
Theoretically cap tube lenth shold be around 85-86cm (0.8mm diam), but i think its too short...

[Unseen]

nice project! You r at -100c club!
I tried 503 to my cascade and under load (P4 3,6@4700) gone -80. -93 at idle. I can play too with some r23. I want to pass -100.

[runmc]

Very nice Sam!! Welcome to the -100C Club. You have a very powerful looking cascade.

[hatemi]

Try with around 2m of 0.8 captube. It should get you much better holding temps.

Oh and BTW that thing looks just marvelous

[cold_ice]

These plate HXs are nice, maybe I'm gonna test one soon. Keep up the good work

[Big_Sam]

Thank you guys, i put a lot of time into this progect, hoping to make it work just right.
Here's some pics of the whole setup:




I put all devices under the table, more convinient for me...
I'm afraid that 2m of cap tube is still going to be too long, Berkut wrote once that 2m is good for CO2, but R23 has lower pressure compared to CO2.

[hatemi]

I'm using 185cm of 0.78mm cap in the second stage of my cascade. still testing with CO2 but If I get the first stage realy working it will see some ethylene very soon If you want some major capacity you might want to go for even less than 2m but that should be a good starting point. If it still isn't holding decent loaded temps then cut 40-50cm of the cap tube of and it should hanlde anything you can throw at it with ease

[Big_Sam]

While continuing to work on my cascade i changed some things.
First i put a small condenser in the secong stage-desuperheater
and then i shortened the cap tube to 1.2meter. Calculations show
that i should have 10bar discharge pressure and 0.8bar suction pressure
(-30C condensing temperature)
And as Chilly said- 14bar static pressure.

But after this was done i ran into even bigger problems. With the second
stage looking like this my first stage could not keep up.So i tried
finetuning the first stage by installing TXV but the problem is still there.

And now i really need help, anybody has ideas or solutions?

[Redwolf]

What is your first stage doing? CST SC SST SH

----small tweeks---
Use a piece of copper sheet to cover the thermo bulb and suction line, then use a stainless steel hose clamp to hold them together.
Reroute the liquid line to bend down before the thermo bulb. So the bulb doesn't get warm readings from the liquid line.
Insulate the Suction line, pipe to pipe exchanger, liquid line from the exchanger to the txv, the txv and it's outlet pipe, and the plate exchanger too. Basicly anything thats going to get cold.

[Big_Sam]

Room temperature- +25C
So when both stages are working this is what i get:
R404a -31C entering heat exchanger, -27C leaving heat exchanger,
1.8bar suction line pressure, i dont have manifold on the discharge line but
temperature next to the filter is +25C, 10cm to TXV +18C.
The first stage works while i have 8bar static charge in the second stage.
That was the first stage.
Here's what i have in the second stage:
discharge-9.6bar, suction line- 0.8bar, evaporator temperature -84C...
No heat load.

Everything is isolated, pictures above are only to see the structure.



I will move TXV bulb close to the heatexchanger....but i dont thik this will
help much, its got to be a different problem, i'm sure i need more radical
changes than just a buld.

[Redwolf]

I'm assuming your condensing tempature is about 30C. Basicly the drier temp + 5C. However dont think condensing is your problem. The system looks to be running good.

So you need to drop the Suction pressure atleast a bar. What kind of txv do you have on there? R404a with ?? orifice.

Have you tried turning the superheat adjustment crew in?

[Big_Sam]

TXV - for R404A... Danfoss TS-2 / 00
At pressure in 1.8 bar R404A boils at -33C... I have -31C
At pressure in 0.8 bar R404A boils at -50C !!! It's not real for my device !!!
Naturally I tried to turn ТХV and now at me the best variant from possible ...

[hatemi]

You might be having the same problems I had wiht Danfoss SC18C in my Autocascade project. I was fighting and fighting to get it to nice hx temps. I tried with TXV ont the HX and with captubes. Then I got frustrated and tore the ting apart and assembled it as regular single stage direc die cooler. With that compressor and R290 -35c was the best it could do without a load. It just wasnt able to keep the suction preasures in vacume so the compressor was busted Too bad since it would have been very nice compressor.

[chilly1]

WEll teh TS - 2 is fine for a valve But I would go to a 01 oriface (1/3 Ton) the 00 oriface is 1/5 ton.
Sounds Like your not getting enough gas in the second stage also. I would add a 0.5 cutf chamber connected to teh suction of the second stage with a 5 ft 0.028 captube this will be a good expansion tank the captube will limit start current. YOu high stage should run in teh .1 to .5 bar range positive numbers. If you get too low the temperature at ehend of teh metering device will flock up and slow refrigerant flow.

[Redwolf]

If your valve isn't hunting.... Hunting is where the pressure and superheat go up and down constantly. Which means either drastic load changes, poor bulb placement, or to large of a valve.

Then you could possibly have a leaky discharge valve or blowby in the cylinder.

Try closing the txv all the way off and watch the suction pressure. Or if you have a suction or liquid shut-off valve close it and watch the suction pressure.
Any refrigeration compressor should be able to pull a 20" Hg vaccum with ease.

[Big_Sam]

Thanks Chilly. I've found that 01 oriface and one of these days I'll try it. But I dont understand how it's going to help. Compressor capacity stays the same, i think it's only going to increase temperature an pressure in the heat exchanger
About chamber with cap tube I have understood the idea, but I have only 0.031 tube, and I have not understood what chamber volume is necessary. How many it will be in litres?
I'll have Parker constant pressure expantion valve in a couple of weeks.. Chilly do you think that if i install it in the second stage i will be able to tune the system without expantion tank?

[Big_Sam]

In the end i installed 01 TXV oriface, reciever for the first stage and
Parker PEV in the second stage... There were no problems at startup, everything
worked flawlessly...






But then i tried to experiment with the first stage gas, i mixed R410a and R404a...
It was running fine for about 10 minutes but then i noticed this:



so I got a new condenser (bigger one) and looking for a receiver...



...to be continued...

[Tonic]

OMG It looks very dangerously

[chilly1]

In the end i installed 01 TXV oriface, reciever for the first stage and
Parker PEV in the second stage... There were no problems at startup, everything
worked flawlessly...






But then i tried to experiment with the first stage gas, i mixed R410a and R404a...
It was running fine for about 10 minutes but then i noticed this:



so I got a new condenser (bigger one) and looking for a receiver...



...to be continued...

Cut that off of there and add a pipe cap.

[berkut]

Make it 2 pipe caps, one on the bottom and one on the top

dont you love those condensers ? compact and powerfull

*edit*

like WTF ? didnt that pipe expand on the top ? like a baloon ?

[Unknown_road]

get a smaller diameter pipe man, the bigger the pipe the lesser amount of pressure it can handle And indeed use pipe caps. You can't use this pipe, it will explode eventually!!!!

btw a 00 orifice (danfoss) R404a is rated at 700watts and a 01 is rated at 1600watts(almost 1/2 ton).

[Big_Sam]

Cut that off of there and add a pipe cap.

Make it 2 pipe caps, one on the bottom and one on the top

I am afraid to leave this pipe. Metal after heating becomes soft. I shall make a new reciever of a steel pipe.

dont you love those condensers ? compact and powerfull

Yes... I constantly use them in the systems.

And indeed use pipe caps

It is a greater rarity for me...

[skate2snow]

Sam, currently, whats the lowest temp you achieved?