My First DIY AC Recharge Story

[hasoan]

This is more of a story than instructions, so if you want that look somewhere else. I do not reccommend any procedures that follow, although in the end I was able to finish the job. This post is not intended to express an opinion on what you or anyone else does, or what is right or wrong. I only thought it could be of use to someone to possibly avoid some mistakes I made.

It started with me replacing my heater core on my 96 Explorer XLT. To do that, I had to disconnect the evaporator line. Before doing that, I took my car to a shop and asked them to evacuate and recover the refrigerant (R134a.) They said something like "wait, you just want us to empty it? not recharge it?" I explained to them what I was doing but I'm guessing this doesn't happen a lot, if at all there. I paid almost 50 dollars just for this. Definitely a rip-off, but I was already there so I just did it.

I then replaced my heater core, but that's another story.

About a month later (June) I decided I had better recharge the AC. I had originally planned to take it back to the shop, but I was on a roll with car repairs, so I decided to do it myself. I'm sure many of you probably can relate to that. I mean, It's just refilling it. There shouldn't be any problem.

I bought an R134a manifold gauge set, a refrigeration vaccum pump, a accumulator/dryer, some o-rings, a can tap, 3 (12 oz) cans of R134a, and some PAG 46.

Project day. Here is where I made my first mistake. (Well, on this procedure.) I installed the new accumulator/dryer first. It never occurred to me to test the system before doing that.

I then hooked up the gauges and the pump, and let it run. The system would not hold a vaccum. I was running the thing for up to an hour at a time, repeatedly. The whole time the vapor mist was coming out of the pump. When I shut off the pump, vaccum quickly dropped to around 10 inHg, then slowly even lower. so I wasnt really removing all the moisture because somewhere air was getting back in. I changed nearly every o-ring in the system... twice... even ones I didn't touch before... and over a couple days I could not find any obvious leaks. I don't know at this point if that accumulator/dryer is good anymore.

I found a place that rents Nitrogen tanks, but I would have to buy a regulator for like 75 bucks. So I came up with a plan to use CO2 instead from a paintball bottle to pressurize the system and try to get UV dye in there. I didn't run the compressor with the CO2 in there (because it's probably not a good idea) so the dye didn't distribute very well. BUT, the system held the CO2 at a constant pressure overnight (about 60 psi I believe)

The system holds pressure, but not vaccum. I read somewhere this could be from a "bad" compressor shaft seal. In other words, it SHOULD hold a vaccum, but is really more designed to hold pressure. The same could possibly be true of o-rings, but that would suggest that the car has a design flaw. I didn't see any posts on here to suggest that.

So at this point I decided to "evacuate" it again and put in just enough R134a to kick on the compressor a few times (somewhere around one 12oz bottle I think). Now the UV dye circulated. I checked for glowing leaks with a UV lamp: None. I checked again the next day: None.

This is where my chemistry classes started kicking in (although it's physics, but I didn't write the syllubus...) Because I added the R134a, I could not test for leaks by using the manifold gauge, because there was liquid R134a in the system at that point. In that case, the pressure should always read at the vapor pressure of R134a, even with a small leak. (if some leaks, some vaporizes and takes its place) Another result of this is that the temperature reading on the gauge should match the actual temperature. It did not. The pressure was about 10 psi too high for the temperature outside.

Without going into too much detail, Dalton's Law of partial pressures was what I was looking at, and I refreshed my mind with Wikipedia. Basically the total gas pressure of a system is equal to the sum of the partial pressures of each gas. That means I had air in my A/C, which was not entirely surprising. Also, even with a low charge I could see the high side gauge was getting too high compared to the low, and that concerned me.

So now how do I get that air out of there? Starting over won't help because neither me or a professional will be able to pull a proper vaccum without replacing more parts. And it WORKED BEFORE anyway. I came up with a plan. I think you could call it "fractional purification." If I vent a small amount of the refrigerant, air will come out with it. Then the refrigerant will boil and displace the air that came out. Repeated a few times, and my pressure gauge was within 1-2 psi of where it was expected to be (after letting it warm back up to ambient temperature of course.) It would be impractical try to use this method beyond that point because of deminishing returns. Keep in mind that this method is also ILLEGAL.

After taking care of that I added two more cans of R134a, and got both the low and high gauges well within the proper range. It's been nice and cool all summer.

The moral of the story here is that I would have spent the same or LESS (on tools) by taking it to a shop. I also would not have had my neighbors looking at me strangely while I deliberately vent R134a. They probably now think I caused global warming and am killing their grandchildren. And on that note, I am not happy that I did it either. But I basically forced myself into that situation. "Buy the ticket, take the ride."

So there you have it.

[Update: 2014-05-18] The AC is still working great! I don't know what the deal with the vaccum leak was.