Recharging the AC with hydrocarbon refrigerant

Update:

I finally got the new condenser installed after a long struggle with removing the old one. The bolt that held it to the lower left radiator bracket was so rusted that the head got stripped, so I had to cut open the frame of the old condenser to get enough room to remove the rusty bolt with pliers and penetrating oil. Took me about a day. This is what 15 years of exposure to road salt does to fasteners.

The good news is there are no more large leaks in the system and the AC compressor still works. After installing the new condenser (including new o-rings, oil and dye) I pulled a vacuum on the system for 30 minutes and made sure it could hold a vacuum for 45 minutes. I then pulled a vacuum for 22 hours to remove moisture, and charged a can of Enviro-Safe Industrial as a liquid into the vacuum. After that I let the car sit for two days to check for leaks with a combustible gas detector and a UV flashlight. The gas detector did not pick up anything (and I've confirmed that the device actually works), but I could see small amounts of dye around some of the crimp fittings on the rubber hoses and a tiny bit where the suction hose connects to the compressor. Not sure if this is my dye or leftover dye from when it was last charged by an AC shop, or (less likely) spillage from the large leak at the condenser. The system hasn't had any refrigerant in it for two years, and I guess it's impossible for dye to slowly leak out without pressure, so I figure there must have been a slow leak back when it did have refrigerant. I'm not really comfortable with any size leak in combination with flammable refrigerant, although in this case it's probably too small to hold a flame if it were to ignite.

Anyway, I manned up just enough to start the car with the doors and windows open, put the AC on full blast, and it worked. The manifold gauge needles moved from 57/57 PSI (static pressure, I've learned) to approximately 30 PSI low side and 100 PSI high side at an ambient temperature of 14 C (57 degrees Fahrenheit). Cold air was coming through the vents and I could hear the compressor working - a bit noisy maybe? I then shut the car off, removed the LP and HP quick couplers and, here comes the bad news: As I went to purge the manifold set, I expected a small hiss, but instead I got a large amount of liquid refrigerant, oil, and dye spraying all over the engine bay, windshield, headlights, etc. Greenish yellow goo was bubbling out of the fitting, streaming down the hose. The high pressure gauge was still showing 100 PSI when I cracked open the screw, so perhaps I should have seen it coming. It was a mess. And I can forget about tracing any leaks with the UV dye now because it's now everywhere under the hood.
However, my main concern now is the oil that was lost. Judging by the amount I picked up with paper towels, and the oily stains everywhere, I think a significant amount was trapped in the high pressure hose. I really need some qualified advice now as to how much oil I should add back into the system.

 

Thank you for the feedback. I'm not sure how this would work on my gauge set. It's this one:

EDIT: It appears I'm not allowed to post links, so I'll just attach a picture of the gauge set instead.

It does have a push valve right above the fitting where you screw on the yellow charging hose, but it's my understanding that this valve is only for bleeding air out of the charging hose so you don't push a pocket of air into the system together with the refrigerant. When I was done charging the Enviro-Safe and tried to disconnect the charging hose from the can tap, a considerable amount of liquid refrigerant sprayed out, so not sure there's an automatic shut-off valve, or maybe I did something wrong.

Any thoughts on the oil? I'm assuming some of the goo that sprayed out was compressor oil. As you can see from the first picture, the air filter box, coolant hoses, etc. are coated with an oily residue, and the stains you see on the plastic trim and windshield wipers is the same stuff. Should I add maybe 5-10 ml of POE oil?
EDIT #2: To be clear, this mishap with the hose removal was days after I had charged the AC with refrigerant. It happened right after I had hooked up the gauges to check the pressures with the car running. I did not put in any refrigerant with the compressor on. The only oil I've added is the 45 ml I poured into the condenser inlet.

The Enviro-Safe I used is the "Industrial" version without dye or any other additives. This stuff must be charged as a liquid. I believe that's because it's a blend of hydrocarbon gasses (propane and isobutane) with different volatility - so charging it as a vapour would give you the wrong mix if you don't use the entire can. I tried to charge it by weight holding the can upside down on a kitchen scale, but that proved almost impossible, so I gave that up and just winged it. The can had 226 grams of propane/isobutane which supposedly equals 598 grams of R134A. I disconnected the can when it was nearly empty and vented whatever was left in the hoses into the atmosphere.

 

Thanks for the advice. I'll give the car a good wash when I'm done with the AC work. I might even clean the engine compartment with a mild degreaser to get rid of the dye stains. Never heard of Super Clean, I doubt it's available here. I usually use a water soluble degreaser called Bilt Hamber Surfex HD. When that doesn't work, I resort to a petroleum based degreaser.
Apparently the maker of this dye also makes a "dye remover" but it costs $50 for a 1 liter bottle. I might as well pour my money down the drain.

Yes, rust is a massive problem here in Norway. I've done so many rust repairs on this car. Three years ago I replaced almost the entire front and rear suspension. I even took the crossmember out, sandblasted it and painted it with 2K epoxy primer. When I don't have the time to remove the rust, I use Fluid Film.
If you look at the second picture I posted, you'll notice that the upper radiator brackets look nice and black. That's not how they looked when I removed the condenser. They were heavily corroded. I cleaned them up, submerged them in citric acid overnight, then painted them with several coats of 2K epoxy primer and one coat of alkyd paint.

 

Thanks. So if I open the high and low valves, air won't be sucked in through the yellow charging hose?

EDIT: Just so there are no misunderstandings, I did not have a refrigerant can attached to the charging hose when this happened. I had simply hooked up the gauges to read the pressures, I didn't touch the knobs. In fact I had disconnected the yellow hose and put the red dust cap that came with the manifold set onto the charging fitting.

 

I would have to check. I looked at the instructions that came with the manifold and it says to unscrew this cap (they refer to it as a "nut") and press the valve behind it with a screwdriver to bleed air out of the charging hose. I'll see if the other end of the yellow hose connector can be screwed onto it
Is there another way to transfer most of the high pressure liquid back into the system without messing with the manifold? What if I let it sit for a while with both quick couplers connected to the car (and fully open), would the pressure gradually drop? I have the time to do this if that's an option.

 

The AC is still working pretty good with the hydrocarbon refrigerant. Thanks to the dye I found another leak at the suction hose to compressor connection which I've now (hopefully) fixed. There was significant buildup inside the fitting, some hard yellow substance that was even underneath the o-ring. I scraped it off and installed a new o-ring.

Anyway, I have another issue now: There seems to be a problem with the air conditioner ECU. I hooked up my laptop with Techstream (latest version, legit subscription, Tactrix OpenPort cable), and every system is responding except "Air conditioner." It tries to connect for a while and then gives me the error "the selected system is not responding." I've tried maybe 15 times. I did get a semi-successful connection one time, but it only displayed the compressor target speed before communication was lost. The common solar sensor DLC (B1421?) is present, but I assume that's because I'm doing the test inside the garage. Anyone familiar with this issue?