AC to plateHX Chiller Conversion...56K Warning!

[Naja002]

NASA......we have lift off!

Finally have 99% of the stuff in-house. Picked up the AC unit today and started working on it. Took a lot of pix and 1 video in hopes of helping those that are timid about de-casing an AC for a ghetto chiller. So, experienced folks--please bear with me on the initial pix....

The main intent in this project is simply to convert a standard 5K btu AC into a plateHX chiller that can be controlled by the onboard AC controller or sent sub-zero at the flick of a switch. Wish me luck! ()

The AC is in-house:

[Naja002]

Ok, easiest step--yeah, just remove the shipping block:




In order to remove the front panel/grill remove the window screen slide attachments on both sides:




remove this screw:





Go ahead and pull the filter and remove this screw on the other side:



Next you will need to sort out the plastic tabs holding the cover/grill on. Not a big deal. If you don't want/need the cover....then don't be afraid to break the tabs.

Once the cover was lose:



I needed to remove the 4 screws holding the control board:



Takes just a few minutes:




Next, this AC is not a slide-out, so the entire housing cover needs to be removed. 10 screws total...not a big deal:

3 screws on this side:



4 out of 6 screws on the rear (the bottom 2 hold the condenser):



2 screws on the top:



and 1 screw on the other side:



The top will now slide right off and here ya go:

[Naja002]

Next is getting the evap fan out of there. The styrofoam housing is split--it's 2 pieces....a top and bottom:




You will have to pay attention and check your particular unit, but the 2 that I have dealt with were basically the same. Foam on the side:



I just "loosened" half of the small piece and sliced the long piece:



If you are going to make a ghetto chiller then taking care of the styrofoam is not a concern. However, I will be keeping and using it, so I need to take care of it.

Once free, the top will slide right out:



In my limited experience, the evap fan has been held on with a simple nut/washer setup:



Just get the right size wrench, mount it and spin the fan and the nut will come right off.

In order to get the fan out, you're going to need to unscrew the evap:



and slide it out just a bit.

Once the fan is out you'll be left with the shaft and the bottom styrofoam will be very easy to remove:




While researching around I found a very cool common sense tip for removing the fan shaft---let the motor do the work! DuH!

Here's a 4:59 min video of the entire shaft removal process:



Took a file with the fan motor running and cleaned up the little nipple left over:




End of ghetto mod assistance.....

[Naja002]

My first question is: What is this gooey stuff? Just anti-vibration....? This unit only had what is seen in the pic below. My ghetto chiller had a whole bunch of this stuff:



The electrical diagram shows both a "Room temp sensor" and a "Room tube sensor". I am assuming that this thing is the room temp sensor...? :








Compressor sticker:




Condenser connections:












Evap connections:







Fan motor sticker. Not really relevent, but here anyway:








I am going to need to study this unit a bit in order to determine where to use the piercing valve. I want to do as little brazing as possible in order to get this unit converted to the plate HX.

The only thing I'm waiting on now is the N2 reg. But I still have to deal with the piercing and recovery. Hopefully, the reg will be here very soon.

[Naja002]

A little progress made today.....

I setup and did the recovery. Best I can tell....things went well. We'll see when I go to get the refrigerant back out of the tank!



I put the piercing valve on the high-side where it's still vapor.



I put a sight-glass on the recovery unit. Confused me for a bit, because I didn't see any liquid. Finally I realized that I was recovering vapor. Everything behaved normally otherwise. Like I would actually know what "normally" is..... But, honestly everything with the gauges made sense...

I managed to get the wiring and control box out of the way and flipped back over the condenser:



So, I went ahead and removed the evap:





While I was still feeling spunky, I went ahead and removed the piercing hole:




Cut back the tubing where the piercing hole was and taped everything off:




Now I'm still waiting on the N2 reg. Going to start getting pissed if the guy doesn't have it here today or tomorrow. Can't do any purging until I have a reg to work with. I could possibly try to set a CO2 tank at a slow leak, but I would rather have the reg.

So, right now I'm at a holding point. Not sure what more I can do until I'm setup to purge....

[[XC] Hicks121]

Woot! I was wondering what was going on with this. Nice work as always brother!!

Down here on Florida ATM. Lets's see what ou get done with the long weekend.

[nascasho]

Whoa, most hardcore thing I've ever seen.

[n00b 0f l337]

Nicely and PROPERLY done!
*tips hat*

[Naja002]

Whoa, most hardcore thing I've ever seen.

Thanx! 1) check out my other projects in my sig and 2) keep surfing the chilled and phase forums.....I'm just a beginner here....it get's a whole lot more hardcore than this!

Nicely and PROPERLY done!
*tips hat*

Thanx, Nol! Everyone started back-peddling as soon as I started actually buying stuff, so it looks like I'm on my own here. But it is reassuring to have someone with experience chime in with some positive words! Thank You!

[Naja002]

Ok, more progress made.....


Got the pipe work sorted out and brazed up the plate HX:







I'm not over-joyed with the stainless to copper brazing. I think they will be Ok, but I will have to wait for the pressure test. Certainly not the prettiest joints.

[Naja002]

So, I carried on with the pipe work:





















Started pulling a vacuum since I cannot currently pressure test. Think I have a leak, but it's really hard to tell what's going on. So, I'm checking out different things with the gauges/pump/etc. Really wish I had picked up some Hose-end ball-valves!





Welcome to My Nightmare:

[Naja002]

Ok, looks like I've got it holding a vacuum. Yes, it's only been ~20 minutes and I know: Stick my thumb in my nose and just wait! Got it! Check!


Anyway, seemed like there was something wonky going on: It would hold a vacuum for a few minutes and then start increasing fairly rapidly, but pretty much stop at around 5-10psi vacuum. Just didn't seem right the way it was behaving. I tried tightening the hose fittings. Realized at one point that I had the damn things on backwards! The valve pin pushers were up at the manifold. Could swear I checked that a few days ago. So, I got that straightened out, but it was still acting strangely. I thought maybe it had something to do with either the hoses or the vacuum pump. So, I pulled another vacuum, closed the manifold valves, removed the vacuum hose and capped the manifold off with an access valve cap, opened the manifold valves again.....and it's been holding a vacuum for ~30 mintues now....


Capped off:



So, now it's wait time.......

[yngndrw]

Hey, I'm just wondering what is the black bottle in this image:
http://i50.photobucket.com/albums/f3...rk-7-23-09.jpg

As well as what pressure testing setup will you be using ?

[n00b 0f l337]

Those joints don't look proper at the HX.
You need to make sure the steels clean, the flux is applied heavily enough, and the copper and steel are both hot enough. Looks like you melted the rods on.

[[XC] gomeler]

^^ Yup. HX joints don't look correct.

[Naja002]

Hey, I'm just wondering what is the black bottle in this image:
http://i50.photobucket.com/albums/f3...rk-7-23-09.jpg

As well as what pressure testing setup will you be using ?

That's a 9oz C02 paintball bottle/tank. Thought I needed it to block that port, but have realized that I do not if I open both red-handle ball valves 100% and just tweak the C02 with the main valve.

I will be using N2, but still waiting on the reg. Filed a dispute with paypal earlier today.

Those joints don't look proper at the HX.
You need to make sure the steels clean, the flux is applied heavily enough, and the copper and steel are both hot enough. Looks like you melted the rods on.

^^ Yup. HX joints don't look correct.

I agree that it doesn't look correct on the outside. It looks like a cold braze. Not being argumentative here, so please don't take it that way. This is just FYI. I seem to be a bit of a "Blobber". I add more than is absolutely necessary. I like the joints to look what? "Full and complete"? I don't know exactly how to say it. The fittings where definitely hot and I think the "inside" of the joints are pretty good actually, but I did start running into problems later as I was trying to "form" the "cap". However, after 10.5 housr it does appear that I've lost 1 inch of vacuum. So, I have to assume it's a leak in the AC fittings. First guess--the HX.

Also, I cleaned those HX joints up with a wire brush and a couple of needle files, so I could get a better look at them. But, Yes, we all agree that they don't look right. Guess I'll break out the mapp/oxy.

On your HX joints, you did heat the fittings and then apply the braze, you did just get the brazing liquid and put it on there.

I dosent look like you had the fittings hot enough. They should be a dull to bright cherry red before you put the braze rod to them. The fittings should be hot enough to melt the braze, it should not be the flame that melts it.

When pressure testing with vaccum you only are using 14.7psi to test with. Vaccum testing should not take the place of pressure testing.

You also need a micron gauge to do the vacuum correctly. Not a must have but pretty close IMO.

Looks pretty good so far.

I did not know that you went to the chiller section, I dont usually cruise the wet side.

As above, I think the fittings were hot enough during the initial braze, but the HX was sucking the heat while I was trying to "form the cap".

I'm still waiting on my N2 reg to arrive. Without it--I can't pressure test. So, that's the only reason that I am doing a vacuum test--trying to make progress with what I have atm. If I could currently pressure test--I would. It would help me pin-point a leak. But I can't, so vacuuming was an option. Yes, I will be pressure testing.

Yes, I understand the value of pressure testing v. pulling a vacuum. And I would definitely like to have a supco vg64, but they are ~$200. I was hoping to get past this project before investing--considering all the other money that I've already spent. But I may go ahead and pick one up. We'll see..... Right now, it looks like I need to purchase another N2 reg....



I'll work on those 2 joints and see if I can't improve them...... Thanx Everybody!

[ultralo1]

On your HX joints, you did heat the fittings and then apply the braze, you did just get the brazing liquid and put it on there.

I dosent look like you had the fittings hot enough. They should be a dull to bright cherry red before you put the braze rod to them. The fittings should be hot enough to melt the braze, it should not be the flame that melts it.

When pressure testing with vaccum you only are using 14.7psi to test with. Vaccum testing should not take the place of pressure testing.

You also need a micron gauge to do the vacuum correctly. Not a must have but pretty close IMO.

Looks pretty good so far.

I did not know that you went to the chiller section, I dont usually cruise the wet side.

[n00b 0f l337]

As above, I think the fittings were hot enough during the initial braze, but the HX was sucking the heat while I was trying to "form the cap".

The HX needs to be hot enough as well

[Naja002]

The HX needs to be hot enough as well

Not sure what You mean.... Do I need to pick up a camping stove? And use that to "preheat" along with the torch?

[n00b 0f l337]

The main trick for copper/steel, is high contact between the two. Placing heat directly on the Steel will often times burn it. So heat has to be transferred, and towards the end maybe a little direct flame at the joint. The steel should be red hot, just the same as copper to copper brazes, the torch shouldn't melt the rod.

[Ozzfest05]

To add to what NOL is saying basically you should make sure the steel is red hot once that is achieved continue to heat the copper once both are hot enough remove the flame from both and add your brazing alloy it melts very quickly when done this way so make sure to cover those joints nicely if you keep applying heat you will just melt away all the brazing material you just built up its tricky but you will get the hang of it.

[n00b 0f l337]

The same applies for brass, but you can more readily heat it over steel.

[Naja002]

The main trick for copper/steel, is high contact between the two. Placing heat directly on the Steel will often times burn it. So heat has to be transferred, and towards the end maybe a little direct flame at the joint. The steel should be red hot, just the same as copper to copper brazes, the torch shouldn't melt the rod.

Ok, I understand that the torch shouldn't melt the rod....just like soldering and sweating. However, trying to transfer the heat via the copper is a trick I'm not sure I can master.... The stainless was red hot. I think that the inside of the braze is a lot better than the outside, but there's no way to be sure one way or the other. The main problem came in while trying to form the "cap". I think maybe the purging gas was cooling things just enough to cause problems. Maybe a bit too much gas. Not really sure.

To add to what NOL is saying basically you should make sure the steel is red hot once that is achieved continue to heat the copper once both are hot enough remove the flame from both and add your brazing alloy it melts very quickly when done this way so make sure to cover those joints nicely if you keep applying heat you will just melt away all the brazing material you just built up its tricky but you will get the hang of it.

See? This is were I get lost. It's not supposed to be orange, but I cannot get these 15% rods to melt otherwise. So, I am doing something wrong. Maybe just need a bit more patience? Keep heating even after I think it's ready? I try to apply the rod opposite of the flame--in that space where there really is little or no flame. I've got a pretty busy week ahead, so maybe instead of working on the HX I'll just take a little time and try to sort some things out--copper-to-copper--with the brazing. What I've been doing probably isn't as bad as it may sound, but I don't think it's quite ideal either.

The same applies for brass, but you can more readily heat it over steel.

I picked up a propane stove from wally-world. I know people use them for phase evaps....is there any reason why I should not use it on the HX? Set it right on it, get things cooking well and braze it up.... Propane for the stove and either mapp or mapp/oxy for the torch....

[n00b 0f l337]

15%?
Won't do steel.
You need to be using around 45% and separate flux.
Don't throw it on the stove though, last thing you want to do is mess up the brazed plates.

[Naja002]

Apologies, You're right--it is 45% stay-silv and the black paste flux. The 15% is for the copper to copper, no flux.

I've been re-watching some online brazing videos. I think I understand a bit better. I'll have to do some work on it when I get a chance, so progress may get a bit slow for a few days...