Air Conditioning - Vacuum Leak Test - New System

I just installed two Fujitsu 16.5 SEER split systems. I ran the copper myself (four sets of lines) since wanted lines inside walls, not outside. I have a vacuum pump and Digital Vacuum Gage. Before we release the gas, how much leakage is acceptable? What is the best way to check for leakage?

Bob

The right way to evacuate the system, & important when working with R-410a, is to pull down to 500 microns & break the vacuum with dry nitrogen. Repeat this twice more & evacuate again. On the final evacuation, valve off the pump & allow to sit with the micron gauge on the system for at least 15 minutes. The vacuum should not rise over 1,000 microns. Check the installation manual for exact evacuation numbers. In general these are good but some manufacturers have different specs. I hope you used a Rigid or equivalent flaring tool. Flares are CRITICAL on these systems. The Rigid tool is expensive but if you are going to do these systems on any kind of regular basis, worth the money.

Grady - Thanks for the prompt and concise reply. Looks like my system has a few leaks. Friday afternoon I evacuated the lines for an hour, let the system set for 5 minues, then 15 minutes more. The lowest number at the end of evacuation was, for system 1A, 850 microns. After 5 minutes, reading increased to 1000 microns. Another 15 minutes, increased to 1300 microns.

The same test on system 1B gave 610/910/1250 micron.

Systems 2A and 2B gave identical results of 680/1150/1450. The gas valves for 2A and 2B are open since we, my contractor and I, opened the gas valves last week. The contractor evacuated the lines for 15 minutes on each set of lines, and felt the system was secure. He read vacuum from the -30 hg meter on the manifold. System 2A did not work for long. The contractor blames my bad flare fittings.

Originally I used alkylbenzene oil to seal the flares, as recommended by Fujitsu. My contractor advised me to use Blue Block sealant when we strarted on this project but I didn't listen. Before Friday's leak test I re-did the flares that I did (one end of the tube was pre-flared) and used sealant on all connections.

I have a Rothenberger flare tool. The mistake I made with the tool, as discoved toward the end of my re-do effort, was not having the adjustment screw tight enough to center the flare cone. Some of the flares are off center.

I checked my pump alone with one length of Yellow Jacket vacuum hose between the pump and Micron Meter. The best I could get was 500 microns. I changed pump oil and the number dropped to 490 microns. The pump is a new JB DB-4A, capable of 25 microns.

Comments?

You may have a bit of moisture in the system. If the the vacuum climbs from say 500 to around 1500 and stops it is a indication of moisture and not a leak, continue vacuuming. If the reading continues to rise you have a leak. Another problem you can run into concerns the quality of your hoses, all hoses have some porosity a plus 2 hose as manufactured by yellow jacket or a equal is generally recommended for pulling deep vacuums consistently. A good way to leak check a new system is to pressurize the system to max factory test pressure (probably 400 psig) with a Nitrogen refrigerant mix ( a few ounces of refrigerant is sufficient) now with a good quality electronic sniffer you can pinpoint any leaks easily. The higher pressure will shoot the gas mixture out thru the smallest pinhole. Quick and efficient. Only problem is you have to have the proper equipment. Most people do not invest 2k to gear up to do their own system but if you can borrow it or have the extra cash laying around heck man do it yourself. Sounds like you have a good start. Note: If your lines were unsealed for a long period of time in humid weather you may never be able to completely remove all the moisture, be sure to install a quality sporlan or equal filter drier in the liquid line.

I have found a lot of electronic vacuum gauges will not give an accurate reading when essentially connected directly to the pump. Thru some experimentation with two gauges & three pumps I found by putting some volume (I used a recovery tank) between the pump & gauge I got good readings. The line set & evaporator will serve the same purpose.

Hoses, as Mr. Fixit mentioned, can make all the difference. I use only the heavy wall black hoses for evacuation. Some guys use the braided stainless hoses.

Flares are machined seals. DO NOT use any kind of "sealant". A drop of oil is good but no Blue Block, Leak Lock, or any other kind of sealant. When you break the vacuum with nitrogen, pressurize to 10-15 psig & allow it to sit for about 15 minutes. The nitrogen will help to absorb any moisture.

If your contractor only uses the vacuum gauge on his manifold, he's a hack. Fire him. :madhell: