Best AC sealant ?

[350Matt]

Hello all

having replaced every o-ring I can find in the system and it had a new condensor rad about 9 months ago I've still got a leak on my mazda's AC that doens't show up on the vacuum test but does leak enough gas out in 2 days for the AC to stop working


getting a bit bored of it now
so what sealant can anyone reccommend ?

[rigga]

Compressor pump issue ?

When having the vacuum applied to the system, the pump is not running. You seem to be losing gas once you use the system, only thing that's changed is the running pump.

[E-bmw]

rigga said:

only thing that's changed is the running pump.

And the fact that the test is only for 15 mins, to see a significant drop in that time isn't easy with a small leak.

All you can do is to take it somewhere to leak test it properly.

[ColourRestorer]

I absolutely hate the PHers who respond to a question for help with a statement saying "well if I were you I wouldn't be starting from there" , but in this case I wouldn't be trying to use a sealant, of which I'd be concerned about the law of unintended consequences and what a sealant will do to the orifice in the expansion valve, and why it won't all accumulate in the drier and harm that.

Instead I would look at the test regime, because IMHO, there are two significant shortcomings to a vacuum test:
By definition it can only generate a pressure differential of 14.7 psi.
It subjects the seals to pressure in the reverse direction to that which they will experience once in service.

To me it seems that the only way to thoroughly test the system is by pressurising it: obviously not with refrigerant, because of the cost and the environmental concerns, but instead with a gas that will not damage the air con system by way of chemical reaction or introducing moisture.

For that reason we use a nitrogen bottle to pressurise air con systems that we’re struggling to diagnose. If you don't have ready access to a nitrogen bottle, but have a mate who's a welder then you might be able to use argon which is inert and dry. Whatever you do, do not be tempted to use oxygen (dangerous) or compressed air (full of moisture).

[TwinKam]

This^^^.
You need proper, experienced AC engineers rather than a yoof in a fast fit outfit or someone who's just leased a machine and gone on a ½ day course. Besides which, knowingly allowing refrigerant to escape to atmosphere is a criminal offence attracting huge fines, if you've 'lost' your refrigerant more than once and you're still refilling with fingers crossed then I'd say you're drifting into that territory.
Nitrogen with an ultra violet dye is the way to go.