a/c recharge interval?
 

I know there are a very larger number of discussions on the poor air conditioning in the pre-1990 911's as well as many. many posts on fixes, upgrades, etc.

Well, here's a question: What is your experience on how often the a/c needs to be recharged to maintain reasonable cooling?

My 1987 black/black coupe has been converted to R134a (by the previous owner) and has Griffiths barrier hoses installed (by me), a new expansion valve and a rebuilt OEM compressor (again, by me with help from Griffiths). Oh, and I installed a pwm a/c fan motor controller. The condensers and evaporator are stock (in the near future this may change).

I'm finding I need to have a complete evacuation and recharge once per year to maintain some degree of cooling. Is this consistent with others' experience?

My other cars (admittedly much newer) have lasted over 10 years and the a/c is still blowing plenty cold.

Time for a leak test.

I heard Griffiths, one of their reps anyway, is a member of this forum.

Perhaps they can offer additional help.

Sorry - I forgot to mention that I've hunted around with my UV light and didn't find any leaks and when I vacuum test the system it holds vacuum well.

So if there is a leak it is either very small, in a location I can't inspect, or maybe in the compressor itself.

The last time I found a leak it was the charging port itself - not something that shows as a leak during standard "vacuum hold" tests.

Definitely a leak somewhere. No need for a “routine maintenance” refrigerant/oil change on any A/C system car or home. There are some other things, like keeping them condenser coils clean, that do need periodic attention,

The system in my just sold daily driver (Honda CRX Si) went from 1991 to 2019 (when the compressor failed) with it’s original charge of R-12 and worked fine up to the time the compressor died. I know this is exceptional in the extreme but the point is if an A/C stops cooling a leak and low refrigerant is the culprit at least 75% of the time.

Back in 2008, at 124,000 miles I did the Mr. Cool project from Griffiths. All the parts that hold pressure were replaced, except the front and rear condensers, and my original compressor. That old original compressor died about a year later and I had to replace it, so of course a recharge.

I never drive with the windows down and have only opened my sunroof three times in the 29 years of owning my car, the AC or heat is running if I am driving. When driving in the rain for extended periods, the windows start to fog up, and the heater can defog the windshield, but then it gets hot and humid in the car, so I run the AC to cool off the car and dehumidify the interior. The AC system runs a lot in my car living in Oklahoma.

The second compressor died just two years ago. I went for over a decade and many thousand miles before the second compressor developed a nose seal leak. I replaced that compressor two years ago, and the system is untouched since. I am at 196,000 miles now.

My 1986 El Camino has fantastic AC when it is working right. It has the GM designed R4 compressor. It is noisy and does not last long. I have to replace it every few years. I need to break down and convert it to use a different compressor, but new R4 compressors are cheap.

See this recent post. It explains really well why you may think your system is tight because it holds a vacuum, while in fact it isn't. Don't forget that an AC system is designed to work under pressure, not under vacuum.

Also, if you want comparative data, my car AC has gone through a very similar process as yours. I changed all lines for dual barrier from Griffiths, changed the TXV and the receiver drier, flushed the two condensors and the evaporator and changed the compressor. I charged it more than 3 years ago and it still blows cold. Not freezing cold, because that's the Porsche design from the eighties, but it blows as cold as 3 years ago. No leak.

That is a really good point. The only reason why I have been testing using a vacuum is that is the standard procedure for vacuum & charging.

I'll be sure to research how to pressure test the system. One complexity is finding a source for very dry gas because I do not want to introduce all the moisture contained in standard dry air.

(Interesting aside: Very early in my career I worked on industrial air driers for use in telecommunications systems, and later on extremely sensitive leak detectors for gas laser gyroscope fill systems. Oh how I wish I had access to that equipment these days...)

Here’s my experience:
I’ve had cars sit unused for months if not years and the ac would function reasonably well, other cars would fail completely if the system wasn’t activated. Most 911’s and classics are seasonally used and that might be part of the problem. There’s no telling why some seals last longer than others. My Fords lasting a decade on original charge, but the BMW was years.

Minute leaks can come from seals o-rings not being lubricated during the cycling, I’ve found my BMW to benefit with a low side recharge with a can of AC Pro after it sat for two winters.

A yearly recharge isn’t normal for a car that if driven on a regular basis with occasional operating the ac and or defroster use even in winter months. Once the seals go bad, no amount of sealing properties in the refrigerant will fix them.
Hose crimps deteriorate and the hose itself from engine heat, age and movement - like taking the compressor off to get it out of the way during other maintenance or repairs.

So your AC cools then?

1986-1989 evap's are prone to leakage.

A vacuum test for leakage is only for a 'gross' leak' check before charging; perfect vacuum at sea level is approx equivalent to 14.7346 psi. R134a running pressures can range from 150 up to 350 (depending up on ambients an charge levels, above 300 is not nice) on high side,and static can run from 6 at 0f to 199 at 130f



Dye leak tests are only show up leaks where you can see the leak.
Hence, use a elecrtonic refrigerant sniffer
https://www.fieldpiece.com/news-articles/better-than-bubbles-the-ins-outs-of-electronic-leak-detectors/#:~:text=The%20four%20main%20types%20of,best%20fit %20for%20your%20toolkit.

Can't say changing the refrigerant oil 'annually' is common practice unless you have residual contamination from 'before' such as debris, mixed oil or moisture you failed to remove.

Most common leaks are:
O-rings (old or new crushed or sheared).
Schrader valves.
Compressor manifold, nose seals or case o-rings.
Forgetting to put in a o-ring or snug up the fitting.
Evap years mentioned.

short answers, time to go make the donuts.....

Another very good point. I looked back and the car's a/c was not used for approximately 12 months and the car had months long periods of sitting unused. Also - testing the oil using the quick (and largely non-quantative) Duracool oil tester showed that the system might have been low on oil, which of course could add to seal issues.

Certainly good advice, but not entirely practical from an economic perspective. Assuming one vacuums and charges their own system, and assuming the price of R134a does not change significantly, one could fully charge the system once a year for (10) years for the cost of one of the recommended electronic sniffers. Taking 10 years to have a positive ROI doesn't look all that good.

Notably, it is my understanding (and experience from using industrial gas detectors for He and other gasses) that the cheap electronic gas leak sniffers are garbage. I would of course be very happy if someone was to prove me wrong about this.

ASSUMING !

1) No body recommended a brand of sniffer. Someone (:) provided a link to the description of common types of sniffers. Your understanding of gas detectors must be limited. I have a very old Tif 5650A we bought 25 years ago. It still works. It is still sensitive enough to find just about any rate of leak.

2) If you want to talk bout ROI's, you are in the wrong forum.

3) When a client contacts us because their shop can't find a leak or failed to find a leak, well tell them to go on Ebay, find a decent used brand Tif, Mastercool, Yellow Jacket, etc, in a case, with a money back guarantee. Buy, use it. When they are done put it back on Ebay and resell it. Pass it on. Or, rent one.

4) Put your money on the table.

^^^More fun gadgets and toys ... oil testers, freon sniffers ... to add to the long list :)

+1 on the "money back guarantee" recommendation, the last ultrasound sniffer I bought on ebay, Snap-on to boot, was a dud.

Same experience here with the DDs.

What's with these O-rings failing without being "oiled" regularly ? The "oil" is mainly for the compressor, is it not ? Or are we simply not using the right material, surely that exists.

Thanks for the recommendations. My knowledge of gas detectors is wide, but focused on gas lasers (He-Ne), medical-use gasses (O2, N2, CO2, He, etc.) and process gasses (Ar & others previously listed) and detecting leak rates as low as 10E-8 atm-cc/sec (and even lower - so much so that I used a quadrapole mass spec), but limited when it comes to Halogens.

A question: In the absence of a calibrated leak standard (preferably a NIST-traceable one) how does one verify the sensitivity and functionality of a Ebay-purchased gas leak detector? If it doesn't detect a leak, is it because it is isn't working or because there is no leak? Same for false positives.

With used electrical and mechanical test and measurement equipment, verifying performance is typically quite straightforward, but with gas sniffers perhaps not so much.

I was always skeptical of gas leak detectors. If one has a leaky area, then replaces the O ring to stop the leak, how long does the refrigerant from the previous leak remain? How does one clean the old refrigerant leak smell the detector is finding?