There are many ways to do leak checking, and the best way depends on who you ask.
For my experiments, I have used bubble blowing soap, bubble bath liquid, scrubbing bubbles bathroom cleaner, and softsoap hand cleaner gel. Anything that will stick to the suspect surface well (more than a minute or two without running off) and form long-lasting bubbles or suds will work. Dead reckoning approach: bubbles sighted confirms a leak.
If this condition slowly occurred, the best approach is to look for signs of oil that leaked out with the refrigerant. This is best done with an ultraviolet (black) light source. If you smeared oil or some kind of sealer on your fittings during assembly, it may or may not look the same as fresh oil. UV dye can be injected into the system easily, and is available at supply houses everywhere. This makes detection much easier, as even a tiny amount of oil will shine brightly under the UV light. I have used this method in automotive systems, both to spot refrigerant and engine oil leaks.
Another thing to consider is this: when refrigerant leaks out of the system, conditions inside the loop become favorable for contamination. If enough charge is lost, the low side will operate at less that atmospheric pressure, encouraging ingress of non-condensible gases and moisture. If only SOME refrigerant leaks out, compressor discharge temperature can rise to levels that will "crack" the oil into smaller molecules. One of these molecules is water. If more than a few PPM of moisture gets into the system (especially with R-410), this is cause enough to evacuate, drain the oil, and start over with brand new refrigerant and oil.
I have had better luck with these "caveman/luddite" methods than with modern, electronic sniffers. I have heard both ways about them: some swear by them, some say they will only detect larger leaks reliably.
|