Morphing Air Conditioner into Autocascade System

[mytekcontrols]

The new project is coming together, and now pressurized for leak check. Insulation of the HXC stack should occur possibly later today.

As can be seen in the photos, there are quite a few valves involved. Here is a breakdown of their use:

1. Sporlan A3F1 - Buffer Valve
2. Sporlan A3F1 - Defrost Valve
3. Sporlan A3F1 - HXC ByPass Valve
4. Danfoss BML10 - Comp Suction Valve (hiding under solenoid valves)

The "HXC ByPass Valve" is activated during the defrost cycle at the same time as the "Defrost Valve". It's purpose is to reroute the returning hot gases from the evaporator directly to the compressor suction. This prevents heating up the HXC Stack while defrosting (allows for faster cool-down when defrost is terminated, and the unit is once again put back into cooling mode).

The "Buffer Valve" is used to dump gases from the 1st phase separator directly into the combo Buffer/Expansion Tank whenever compressor discharge pressures exceed a preset amount. These gases are then metered back into the suction via a very small captube.

The "Comp Suction Valve" allows for doing pump downs to verify the integrity of the compressor, and also to suck in a charge from a pre-mix tank. It has an extended shaft routed out to where the front panel will eventually be, to allow access.

AC3 Modes of Operation:

1. Compressor ON = COOL MODE
2. Compressor ON w/Defrost Valve Open = DEFROST MODE
3. Compressor OFF when Defrost has completed = STANDBY MODE

There is no independent Cool Control Solenoid Valve. The ever so higher flow of the defrost circuit, effectively prevents any flow from the final evaporator captube, and thereby prevents cooling while defrosting. When defrost is done, as determined by a temperature switch monitoring the evaporator return, the compressor power is terminated. This puts the system effectively into "STANDBY MODE". Reactivating the compressor puts us back in "COOL MODE".

Eliminating the Cool Valve allows for standard refrigeration rated components to be used, and thereby keeps the cost reasonable.

Target Evaporator Temperature: -155C

[yngndrw]

That looks really awesome, great job. Probably the neatest setup with that complexity that I've ever seen on here. I love the attention to detail such as the flares coming out of the HZ box so that maintenance is easier at a later date - And the little labels on everything.

Question: Why have both a Defrost valve and a HXC ByPass valve instead of just routing the Defrost valve straight to the suction line ? (I hope I haven't missed the answer to this earlier in the thread.)

[mytekcontrols]

Question: Why have both a Defrost valve and a HXC ByPass valve instead of just routing the Defrost valve straight to the suction line ?

Because the normal return path from the evaporator goes through the HXC Stack (passes through all the cascades and the auxiliary condenser). And the hot gas defrost is teed into the evaporator feed connection. So if I didn't bypass the returning gases around the HXC Stack during defrost, the stack would be warmed up as well.

Does that make sense? Or do I need to show a diagram?

I love the attention to detail such as the flares coming out of the HZ box so that maintenance is easier at a later date

Yep I can't stand having to open up a unit to verify pressure readings against a gage set, or to recharge it. It is amazing how many commercially built units completely overlook this. Especially considering the minimal cost to incorporate this added accessability.

Note: the 2 aluminum pieces holding up the manifold are temporary, they'll get removed when the front panel is installed which has the appropriate mounting holes.

[[XC] gomeler]

Michael, you never fail to impress. I'm slightly concerned about how the little blue condenser will hold up but I don't even know how much heat you are dumping, so besides pump-down it should be sufficient given a sufficiently small load. I eagerly await your analysis of the unit once it is operational, especially if you can at least share a little information on the gas blend you'll be utilizing. I ask this as you are hitting temps I'm eyeballing as I mentioned to you earlier albeit with two compressors.

I'm incredibly jealous, can't wait to see the finished product.

[Marvin]

Michael, that is a lovely job.
What kind a fan are you using with this condenser ?
regards

[yngndrw]

I still don't quite understand the use of two valves. Here is a diagram of how I see it, but I've probably got the whole thing mixed up:
http://yngndrw.hostilezone.net/uploads/AC-2.png (Sorry about the poor diagram.)

When I said HZ box before I meant HX box (Typo) - The internal flares coming out from where the HX stack is.

[[XC] gomeler]

He wants to also dump hot gas into the evap to warm up the evap. However since the HX stack feeds off the suction gases of the evap, he has the 2nd valve to bypass the HX stack to keep it chilled so it won't require a complete system restart and pulldown when he just wants to defrost the evap.

[yngndrw]

Ah I get it now, ignore my drawing I totally messed that one up - No idea why I had it coming from the liquid line something went (Badly) wrong in my head.

Sorry about that.

[aenigma]

Beautiful as always mytekcontrols, what an awesome job you have!

[teyber]

that 80/20 sure is expensive stuff... props to you for going all out

[mytekcontrols]

When I said HZ box before I meant HX box (Typo) - The internal flares coming out from where the HX stack is.

Yep did that for several reasons. ease of maintenance, maintain good seal for keeping water out of insulation foam, and able to eventually create modular drop-in replacement HXC stack.

I'm slightly concerned about how the little blue condenser will hold up but I don't even know how much heat you are dumping, so besides pump-down it should be sufficient given a sufficiently small load.

What kind a fan are you using with this condenser ?

Yes my load will be small, probably 15-25 watts. So a large condenser shouldn't be necessary. Also keep in mind that because of my "Regenerative HXC" on the compressor discharge (for creating the hot gas defrost from the colder 1st phase sep gases), I have effectively extended the length of the air cooled condenser.

As for fans; I am using the stock fan that comes with the Little Blue condenser, but I also intend to supplement this with an additional 120CFM fan directly behind the compressor (will be mounted to outside of the future rear panel).

I eagerly await your analysis of the unit once it is operational, especially if you can at least share a little information on the gas blend you'll be utilizing.

Chris - I'll be using R600,R170,R14,Argon in this puppy, but unfortunately can not disclose the exact amounts of each (this unit is aimed at becoming a commercial product).

[plasmatique]

wonderful work

We can soon buy....

[Jor3lBR]

Chris - I'll be using R600,R170,R14,Argon in this puppy, but unfortunately can not disclose the exact amounts of each (this unit is aimed at becoming a commercial product).

So is there any Company backing you up? Will we see some of these units on the market?

Outstanding craftsmanship

[mytekcontrols]

So is there any Company backing you up? Will we see some of these units on the market?

Nobody is backing me at present. But I do have a company in mind for my first potential customer. As for seeing these units on the market; it wont be anything to do with PC cooling. Instead it would be something that might show up in a vacuum systems trade journal.

We can soon buy....

Sorry Plasmatique, but as I said this is not aimed at cooling a CPU (25 watts tops). Also the market price for the final version would be at least $10,000.

[plasmatique]

I had definitely understood that this systeme was not to cool down a central processing unit of treatment but principle could etre adapted... To future let us wish

You say apparatus for space, pump moleculaire?

[sjg0]

Mind if I ask what this unit would be used for, if not cooling a CPU?
Or is that a trade secret?

[mytekcontrols]

The application (at least the intended initial one) is for cryopumping water out of an inline hard disk platter coating machine. Some of these machines use up to 24 tiny cold heads coupled with just as many turbomolecular pumps for this purpose. Each one is creating a mini environment suitable for coating the magnetic material onto the glass or aluminum disk.

One such company building these machines is Intevac.

[ultralo1]

Let me know when you need a service rep for the Southeast

[Jack]

I see you lost quite some valuable posts
I just wanted to say again that this build is awesome!

And thanks for replying my question.

Can you braze in such a tight space without setting the other parts on fire?

Well actually there is a tendency for the residual foam insulation to burn, but once that's gone, no problem. Of course this wouldn't be the case for the area around the compressor and valves, so it does get tricky, and takes a lot of wet rags draped over things to keep the fire in check. I also fashion heat shields out of SST sheet that can be placed in-between things.

[[XC] gomeler]

Oh jesus This thread took a huge beating in the data loss.

[EvoCarlos]

Oh jesus This thread took a huge beating in the data loss.

lets hope the losses were just Nol's 717 posts ha ha ha

[Jack]

lets hope the losses were just Nol's 717 posts ha ha ha

Well as far as I can remember, we're missing posts about Mytek swapping the condensor and changing the condensor fan.
Also he did some work on the discharge and liquid line piping, and removed the complete HX coil.

Unfortunately I did not save the email notifications

[mytekcontrols]

Don't worry, I'll restore what got deleted in the next week or so. Although it'll be a bit different then it was, because this server is not the only thing that suffers from memory loss

Edit: 8/28/2009 -- Sorry that I haven't updated as promised. Been kinda busy, and also haven't been feeling good for a while due to a sleep apnea problem, and ineffective treatment. But here is a list of what has been happening.

1. Larger fan installed.
2. Larger fan was not able to keep up with higher load due to increased Argon additions. Changed to water cooled condenser (10 plate -- flat plate HXC).
3. Built Bong Cooling Tower for water source.
4. Abandoned first HXC Stack design due to inability to go much below -100 C.
5. Built new stack based on more conventional design with 2 full phase separators.
6. Had cross-feed problem with stack, but was able to fix it.
7. Ran out of R600 for experiments, so temporarily used HFE-7000 & R290 instead.
8. Best Cold Head temperature to date is -137 C.
9. Found new source of R600, and its on the way.

Anticipate even better temperatures with R600, R1150, R14, Argon blend. I will hopefully have results in a few weeks. Stay Tuned.

[[XC] gomeler]

Just a bump on this thread to expose people to the awesomeness contained within. Going to sift through it in my sparetime and link the important posts into the sticky guides for everyone's future reference.

[[XC] gomeler]

Turned this into a sticky, I have the next 2 days off so I'm hoping to compile it all into a novel in my spare time. Make no promises though =\