Hi,
As stated in title. I put in 2 cans of 134A (it's been converted) runs cold for 2~3 days then stops working and clutch won't engage due to low pressure switch so I suspect I have a leak. Now, I was thinking about getting a Tracerline leak detection kit since I have lots of leaky cars to contend with BUT, I read that I might not want to try detecting an A/C leak like this since it can't show a gas-only leak, only if the oil's leaking out too and that the dye may or may not mix well with the oil. Should I just bring this in to a place that can detect leaks with an electronic device? Are there sections of this system that are more prone to leaks than others. Assuming a compressor leak and with probably getting a new one here for about 200. What would be a reasonble guesstimate for labor to swap out the compressor, flush the system, swap the expansion valve (I have a new one) recharge and test the system? From grunge near where the hoses lead to the condenser I am thinking there might be a leak there since the hose was changed to a barrier type. I'm afraid to futz with it since it's pretty awkward to get a couple of wrenches in there to test for tight. Any advise is appreciated. Thanks.

I will be shocked if a leak of this magnitude can't be located with UV dye.

I have a tracerline system. The light works great, but the dye application stuff was much less than impressive. Since it's 134, it's easy to get a can of UV dye and just charge with it. I saw a UV light at O'Reilly's the other day for $35 or so.

I got a system from the local auto a/c shop that had an injector that put in 1 oz. of oil with dye. It works great.

I would look for the leak and repair it, replacing all the parts you mentioned is most likely unnecessary if your compressor is running okay. Since it's 134 finding the leak should not be a ridiculously expensive or time consuming project. Regardless of what else you do, you MUST fix the leak so do that and then reevaluate.

Make sure you have a way to evacuate before charging.

My $0.02,