derSchwamm

iTrader: (1)

I have a 94 touring with a MANA system (I thought all tourings had Denso? whatever) and the AC hasn't worked since I bought the car. I've spent a lot of time trying to find the leak but just can't find it.

What I did, in order, since buying it:
- Filled the system with ES12a w/ dye. It all leaked out within a couple weeks. Never found a joint or spot with dye on it
- Replaced all O-rings while my engine was out
- Pulled vacuum after o-ring install: 26in peak, dropped to 0 in 15-20 minutes
- Pulled the evaporator and checked for leaks in a bucket of water (blowing into a hose, plugging the outlet)
- Plugged the lines to to the evaporator with nitrile gloves and vacuum hoes, then pulled vacuum again. 26in peak, dropped to 0 in >10 minutes like before. So I think my evaporator isn't the problem

Without being able to spot the leak with dye, I am getting frustrated. I'd like my AC to work but I can't find this leak to save my life. What else can I do to find it since dye isn't working and I can't reach half the joints with soap and water?

DaleClark

iTrader: (55)

That can be the hard part, just finding the leak. I recommend using Big Blu -

https://www.refrigtech.com/big-blu/

I think you can get it from Amazon and other places. It's like supercharged soapy water - even the tiniest leak will make it bubble and foam like mad.

Some common places are the smaller high pressure lines around the turbos, I've seen (and had happen on my car) those lines rub up against something and the wall get super thin so it makes pinhole leaks.

I would also thoroughly check the compressor, many spots there that can leak.

Also if you are only pulling 26" of vacuum that's a sign too if you are using a proper vacuum pump. You should get to 29" with a good vacuum pump. The cheap pumps that use compressed air won't get that low and don't do a good enough job FYI.

Keep at it!

Dale

alexdimen

iTrader: (13)

Some other places to check are the high/low pressure line schrader valves. If you have a 134 quick connect adapter on there I have had nothing but headaches with those things leaking in the past. I just used the original R12 ports to charge/service on mine.

derSchwamm

iTrader: (1)

Good point, I have those adapters as well. I don't know how to pull vacuum on the system if I remove them though... I'll need to dig up r12 adapters

derSchwamm

iTrader: (1)

I noticed some evidence of rubbing... I'll pull the hose and check more closely. Thanks!

So... the vacuum pump is an autozone rental, electric. On my protege that just got a new compressor it pulled about 29" so in this case the vacuum pump isn't the problem! I have a good sized leak

DaleClark

iTrader: (55)

Good, you're using the correct vacuum pump.

There are 134a adapters out there that are better than others. I've bought them at O'Reilly's in the past with the help of a buddy who works for them and knew which was the good one. They were behind the counter in a cardboard box, not out front.

May want to google around and find different adapters. Also you want adapters that require you to remove the original Schrader valve and have their own new Schrader valve. These may be more "pro" adapters than the consumer grade ones sold at the front of the parts store.

Dale

nmoffatt

iTrader: (15)

The Napa adapters have worked really well for me. No issues. The part numbers are 409901 and 409903.

Neil

t-von

iTrader: (8)

I find my A/C leaks the easy way. Rig up a ball valve with pressure gauge and adapter to connect to your lowside fitting and connect all this to an air compressor. Use an air compressor and run about 100psi and close the valve. This will pressurize the system so you can listen for leaks and spray areas with soapy water. Just make sure you drain your compressor of any moisture before doing this.

Edit: You can also rig up a vacuum pump using an old Refrigerator compressor. They're quiet and pull pretty good vacuum.

mikejokich

iTrader: (1)

The two most common places for leaks that are hard to find are the dryer connections, which was the cause of two of my previous leaks over the last 25 years and the one that is the most difficult to find is in the compressor itself. The seals over time give out. I also found that the original Mazda O-rings last much longer and seal better, particularly on the dryer. I have changed mine three times in 25 years and the aftermarket ones failed in only two years. If the dryer is not the problem and you can't find the leak with the ideas described in the posts from others, I would pull the compressor to check it. They did make a rebuild kit for it years ago, but I don't know if it is still available. I still have a rebuild catalog for compressors if you need a number for the rebuild kit.
Mike

jza80

iTrader: (32)

You can also use a R12/R134a leak detector. The pro models are a couple hundred $$, but thanks to Chinese innovation they can also be found for much less:

https://www.ebay.com/itm/Leak-Detect...-/221901247299

Maybe seems like a waste for a one time job, but if it can detect where you are having the leak, and you can actually fix it, it may be worth the expense...

KompressorLOgic

iTrader: (6)

I couldn't find any leaks with dye, so I suspected the compressor. I replaced it with a new one and when taking the clutch and pulley off it was all yellow with dye in there, the shaft seal has been leaking (slow leak, took a few months to leak out)

derSchwamm

iTrader: (1)

Maybe it’s worth pulling it to check. I can’t see any dye on it from up top but it’s hard to see.

Many other suggestions require positive pressure in the system but my car isn’t running right now. I had a heater core die and I’m installing BNR twins, so the dash is out, turbos are out, and many of the AC parts are exposed that are normally buried so I figured I’d look. My focus is on the turbos and heater core/dash first and I’ll keep working on the AC later, now that I’ve ruled out the evaporator and hoses on that side of the engine.

DaleClark

iTrader: (55)

You can pressurize the high and low sides of the system with refrigerant to check for leaks. Compressor doesn't have to be running.

Dale

derSchwamm

iTrader: (1)

Interesting, I thought the compressor had to be running to suck it in. So I can just hook up the can to each side and get enough pressure to check? I've got two more cans of es12a I can use so I don't waste real r12

DaleClark

iTrader: (55)

Yep. I would pull down to a vacuum then charge both high and low sides. You shouldn't need a ton in there, just 30-40 psi should be more than enough.

Normally when you charge a system with the compressor on you only charge to the low side. You may need to put the can of refrigerant in a warm water bath to help it all go into the system.

Dale

jsemerica

iTrader: (1)

I got my compressor tested for leaks before recharging my AC and was found to be leaking on one of the main body seals and the shaft seal. If there's a place nearby that'll test it I'd suggest starting off with that.

derSchwamm

iTrader: (1)

I finally fixed all my leaks (after 6 years of ownership) and wanted to share how I did it. I invested in a vacuum pump, high quality gauge set, a refrigerant leak detector, and Big Blu AC leak detector spray. To avoid wasting R12, I also hooked up my welding gas to my manifold gauges so I could pressurize the system with inert argon/c02 and still get the refrigerant detector to work without damaging the system or wasting R12. That was a neat little trick.

After replacing the compressor, dryer, and all O rings I found 4 separate leaks using a combination of Big Blu spray and the leak detector. Both soft lines in/out of the compressor leaked in the rubber, my dryer connection leaked, and a condenser connection leaked. After fixing all of this I was able to hold vacuum for days. I would have not have found all the leaks with a single tool alone, and I probably would have wasted a lot of refrigerant in the process.

Now that the system holds pressure, I have to figure out why it still doesn't work, but at least it doesn't leak!!

DaleClark

iTrader: (55)

Thanks for the update!

It's more of an art than a science to find leaks, that's for sure.

What's the next step - if you are charging it, does it charge up but the compressor isn't kicking on?

Dale

derSchwamm

iTrader: (1)

Yeah... I started a new thread with details: https://www.rx7club.com/3rd-generati...stuck-1161763/

Pete_89T2

iTrader: (4)

^Are you absolutely sure the system holds normal operating pressures and isn't leaking when it's running? Reason I ask is it sounds like you're basing that statement on the fact that it held a vacuum for days - keep in mind a vacuum test doesn't really stress the seals too much since the differential pressure acting against them is only barometric pressure (~14.7psi at sea level). If you can pressurize the system to say ~300+psi with an inert gas (or with refrigerant & compressor running), with the manifold gauges on it and don't see it drop pressure over time, that's a much better test to rule out leaks.

My FC actually had similar symptoms after the pressure relief valve on the compressor failed. It could hold a vacuum for days but once it was fully charged with refrigerant and running, it would leak down to nothing in no time. On the FC, the pressure relief valve installs on the high pressure side of the manifold piece that bolts onto the Denso compressor - it's supposed to open up & relieve pressure if it ever gets excessively high, like 350+ psi for the old R12 system. I found that problem by evacuating the system, partially charging it with R152A (computer duster gas), and hunting for the leak with a cheap Harbor Freight electronic leak sniffer with the compressor running. Found it right away once the compressor was running on a partial charge. You might have a similar problem, but possibly from a different location.

derSchwamm

iTrader: (1)

That's a good point. Before pulling vacuum I left the welding gas in the system for like 2 weeks but it was a pretty low pressure, like 40psi. I could try that again, but unfortunately that means wasting the R12 in the system (which might be unavoidable at this point)

Pete_89T2

iTrader: (4)

If you have R12 in there now, first thing to do is just slap the manifold gauges back on there and check the static pressure (system not running). I think there's an ambient temperature dependent spec for static pressure somewhere in the FSM, but low & high sides should be the same, and fall in the ballpark of 70~80psi at about 70*F or so IIRC. That will give you some idea how much R12 you still have left in there, and if it's not completely leaked out, go ahead & run the A/C to fire up the compressor and check the your low & high side operating pressures. If there's enough in there for the compressor to run long enough without the pressure switch shutting down the compressor, have your electronic sniff detector handy to start hunting for leaks.

derSchwamm

iTrader: (1)

After sitting for 24 hours, the low pressure side is 80psi and the high pressure side is 88psi. It's weird to me that they haven't equalized yet, but that also doesn't seem indicative of a huge leak. I put in right about 19oz and not the full 21oz anyway

Pete_89T2

iTrader: (4)

I wouldn't worry too much about an 8psi difference between L/H sides, though they should be equal after 24 hours of non-use. Unless you're using professional grade high $$$ gauges, or measuring each separately with a single gauge, that's only about a 10% error tolerance between the two different gauges - about what you would expect from a set of manifold gauges most of us have.

With that much of a charge, and the compressor operating, you should be feeling some cooling out of the vents. After verifying that the compressor runs & cycles as it should, and the electrical control end is all good, the only thing I can think of that could prevent it from cooling is perhaps moisture in the system or a buggered up expansion valve or dryer perhaps. On the moisture end, did you bleed out the yellow charge hose before charging the system with R12? If you didn't, your system would have a hose full of air laden with whatever moisture was in it, in addition to the R12 you charged it with, and that can cause all kinds of weird things to happen.

derSchwamm

iTrader: (1)

Well as of this morning I am at 75psi low, 80psi high, so perhaps I do have a small leak still, at least at higher pressures. I'll see if I can't troubleshoot that some more while it still has strong pressure.

Even so, I feel like this is enough pressure for the compressor to kick on. I'll follow the steps in the service manual for electrical troubleshooting next and see what I find.

I am not sure what you mean about bleeding the yellow charge hose. I did pull vacuum for a long time but had to switch the yellow hose over the R12 can, so I assume there was some air in that hose just from swapping it over. I don't really see any way to avoid that?