Basic Principles of Auto AC

Leaks and Contamination

Contamination is a cause for worry. Air in the system will dramatically reduce the cooling efficiency. That is not what we want on a hot day, but shouldn’t do any real damage itself. The problem is that the air always contains some moisture – and moisture will cause mechanical problems. It can react with the oil forming acids and cause corrosion. The acids can attack the moving parts, gum up the orifice tube etc. Corrosion can lead to leaks. The system does come with a small packet of desiccant in the accumulator to absorb and hold any tiny amounts of moisture that get in there during the initial assembly, but that cannot be relied on for subsequent work! Any time the system is opened for more than a few minutes, the accumulator should be changed. Most compressor manufactures require this as a warranty condition.

To remove the moisture and the air, a vacuum pump is used to pull the pressure down to near-absolute zero psi (or about 29.5inches of mercury below atmospheric pressure). At that low pressure, the boiling point of water is below normal room temperature, so it boils off and the pump sucks the vapour out of the system. Again, that can take a bit of time, especially if the water is in solution with the oil, so the vacuum should be held for at least 30 minutes. If the pressure does not rise towards 0 in that time, then you probably don’t have any leaks and can run the pump for another 5 minutes just to suck the moisture out. If the system won’t hold vacuum, then you have some leaks to fix. (note: it is possible for some leaks to seal under vacuum, but leak under pressure or vice-versa, so holding a vacuum is no guarantee. I have seen a crack in an evaporator that would leak a full charge out in less than a day, but hold a perfect vacuum for hours because the vacuum pulled the edges of the crack together.) Vacuum pumps can be rented or borrowed from some Auto parts chains. Harbor Freight also sells a couple of decent models, but stay away from the air-powered venturi pumps. They are virtually worthless and won’t pull a deep enough vacuum for AC purposes. You need at least a single-stage rotary vane pump that seals with vacuum pump oil.

Leaks can be located by partly charging the system and checking the entire system with soap solution (kid’s bubble pipe solution from the dollar store works great for this!), UV dye in the oil or an electronic leak detector. Unfortunately the soap solution doesn’t work for the evaporator, which is hidden deep under the dash. An electronic leak detector is the best way to find those, because you can put the fan on low, close all but one vent and stick the detector in the open vent. Sometimes you can verify an evaporator leak by looking for UV dye dripping out the condensate drain under the car. Unfortunately, while most new cars come with the dye installed at the factory, that was not the case back in the early 90’s when our B-bodies were built. You would have to add the dye yourself. It can usually be purchased in a can with some oil and refrigerant already mixed and just has to be added to the system.

Adding Refrigerant, Oil and Dye

A word on oil:

There’s 4 different kinds. R-12 systems use a mineral oil that does not mix with R-134a. Factory R-134a systems come with a type called “PAG” (Poly Alkylene Glycol). Unfortunately, regular PAG oil just loves to absorb water. An improved version, called “Double End Capped” or DEC PAG oil reduces this tendency somewhat. PAO (Poly Alpha Olefin) oil does not suffer from this problem, however, there are reports that it does not mix as well with R-134a as the manufacturers might otherwise have you believe. This is a Bad Thing. The oil must be mutually soluable with the refrigerant to ensure it is recirculated back to the compressor. That is, the refrigerant vapour must be soluble in the oil and the oil must be miscible with the liquid refrigerant. Otherwise you run the risk of starving the poor compressor of lubrication.

I prefer POE (Polyol Ester) oil myself. Extensive independent studies confirmed it’s compatibility with both R-134a and residual mineral oil in R-12 converted systems. It’s also fully compatible with HC refrigerants (more on that later). I buy the stuff with the UV dye already added.  A dye charge is about $4. An 8 oz bottle of ester oil+dye is $1 more than a bottle of plain ester oil. You do the math.

The oil is the lifeblood of the system. Remember that the oil is distributed throughout the system so if you change a component, you have to replace the missing oil that component held. If you had a leak and lost refrigerant, you probably lost some oil as well. The FSM has a guideline to how much oil each component typically will hold, but it’s also a good idea to drain the old one and see how much comes out. If you are simply recharging the system, get an AC oil analyzer. A small disposable clear plastic can containing a white filter cloth, you depress it over the low pressure port after the system has run for at least 5 minutes to stir the oil/refrigerant. A small spurt of refrigerant  will escape through the filter, hopefully carrying some oil with it. If the filter is completely coated in oil from one end to the other, then the mixture is “rich” enough and you don’t need to add any more oil. If the filter is dry at one end, then you need to add some oil or risk damaging your compressor. Usually ½ oz of oil is adequate for a typical recover/recharge.

Few shadetree mechanics are going to have a 30lb cylinder of refrigerant – and those that do probably aren’t reading this! So I’ll assume you will be buying your refrigerant in those little disposable cans. They often come in kits with a hose and a gauge. Do yourself a favour and throw that stupid thing away. You are saving hundreds by doing this yourself, so make the modest investment and buy a manifold set. The Harbor Freight one is just fine – maybe not heavy duty enough to be banged around a professional shop, but accurate if you take care of it and store it in the nice blow-molded case provided. Why am I so against the kit gauge? Two reasons: first, it is of questionable accuracy and only reads the low pressure side anyway. Second, you can’t purge the hose. Why go to all that trouble to vacuum air and moisture out of the system just to put more back in? That’s one of the reasons the pros call those things “death kits”.

So, assemble your manifold set with the blue hose on the side of the blue valve and the blue connector on the blue hose. That will be our low side line. Assemble the red stuff on the red side, and the high side is ready. Connect the yellow line to the centre bottom connector – the one without a Schrader valve. Turn the manifold valves clockwise to close. The port valves on the ends of the hoses turn the opposite direction to close! That’s right, turn them both COUNTER-clockwise to close. This seems stupid until you realize the reason for that is when they open, they not only open the hose end, they also extend a presser to open the Schrader valve on the system port. Check that the gauges read exactly 0. If they don’t, the gauge “glass” (plastic actually) can be unscrewed and the little silver screw on the face of the gauge is then turned to adjust the gauge needle to exactly 0. Connect the blue valve to the low side port and the red to the high side port. Put the port caps in a safe place because if you leave them lying on the front clip or airbox, it’s guaranteed you will knock them off and lose them before you are done.

Now, if you were replacing things, you would connect your vacuum pump to the yellow hose, start the pump, open both manifold valves, then open both port valves and pull a vacuum as described above before the next step. Close both manifold valves before turning off the pump.  

If you are working with a pressurized system then with the manifold valves closed, open the port valves on the ends of the hoses. Crack one manifold valve slightly until you get a puff of refrigerant out of the centre bottom port then close it. Do the same with the other. You have now purged the air from the red and blue hoses.

Wind the can tap valve all the way counter-clockwise, connect your can to the can tap, connect the can tap to the yellow hose. Wind the can valve fully clockwise – this will extend the piercing needle and close the valve at the same time. Make sure the manifold valves are closed. With the valve on top of the can, open the can valve (by rotating counter clockwise). Depress the Schrader valve core stem in the centre front manifold connection until there is a puff of refrigerant. You have now purged the yellow hose. All moist air should now be out of the lines. Now would be a good time to weigh the can with the hose attached if you have an accurate postal scale, good to ½ oz or better. Otherwise, take note of the can’s net weight and start counting cans. It won’t be completely accurate because some refrigerant will always be left behind, but it’s better than nothing.

If you are filling a system under vacuum

Congratulations, this is the best starting point because you have an accurate knowledge of how much refrigerant is already in the system: none. You will need four 8oz cans if you are doing a complete fill on a B-body. Invert the can so the valve is at the bottom. This way you will be filling the system with liquid not vapour. Open both red and blue manifold valves so the vacuum can suck as much liquid in as possible. You should be able to see the liquid rushing in past the manifold sight glass. Close both valves when the can is empty. If there is still vacuum in the system, you might even get some more in the same way. However, it is more likely that the system is now under pressure, so go to the next step: filling under pressure.

If you are filling a system under pressure

You should now be looking at two gauges with equal pressure. The gauge will have several different scales. One will be labelled with pressure and others will be labelled with temperatures, each temperature scale being for a different refrigerant. If there is any liquid refrigerant in the system and you are using the HF manifold set with R-134, then the needle should be pointing to the ambient temperature on the outermost scale. If it is below the ambient temperature, then you don’t have any liquid in the system, just vapour.

Have the can upright, so the valve is on the top. This way you will be filling with vapour, not liquid. (Note: if you are adding an oil charge can follow the instructions on the can. Probably something like “shake well and fill with the valve down”. If filling valve down -- ie: liquid -- fill slowly!) Put the can in some warm water to help force more of the refrigerant into the system.  Make sure the hoses are clear of any moving or hot bits. Start the engine, turn the AC on and set the temperature to the coldest setting (with Automatic Climate Control, this means turn the temperature down to the minimum possible and push “Auto”). Open the can valve. Open the blue manifold valve ONLY. Do NOT open the red valve. If the low pressure is above 40 psi, the compressor should kick on. The low pressure should drop and the high pressure should rise. If the low pressure drops to about 23 PSI, the compressor clutch should disengage, allowing the pressure to rise again. The cycling will be quite rapid at low charge levels and/or low ambient air temperatures.

If you don’t have at least 40 psi in the system, the compressor will not engage. As long as the can is above 45°F, refrigerant will be entering the system, since the vapour pressure of R-134a at 45°F is 40 psi. so be patient. Once the compressor is cycling regularly, you can increase the engine rpm to 1,500 or 2,000 rpm to speed the fill process. Be patient, this will take a while. When the can is empty, try to time the closing of the valve with the compressor cycling – close it just as the compressor reaches it’s lowest temperature. It doesn’t really matter which one you close first: the can valve or the manifold valve, so long as you remember to close BOTH! Then crack the hose connection at the manifold to bleed the pressure from the hose. Disconnect the hose and point it away from you  or leave it cracked and stand back while opening the can valve to bleed the residual pressure from the can. The refrigerant can be cold enough to freeze skin on contact, so gloves and eye protection are a good idea.

If you are adding another can, immediately tighten the connection after bleeding the pressure. Have the can ready to go before proceeding. Open the can valve fully to retract the puncture needle. Remove the old can and install the new one promptly to avoid getting any air in the line. Close the valve to puncture the new can. If delay is unavoidable, purge the line of air as before.

How do I know when the system is full?

See, here’s the problem: we don’t want the system full. If it was full, there wouldn’t be any room for the refrigerant to expand and then we’d never get the cold air we desire. The trick is to get it just full enough, like our sponge that was just big enough to cover the bottom of the bucket without being squeezed. The ideal way is to remove all the refrigerant with a recovery machine and recharge the system by weight. The underhood sticker tells us exactly how much refrigerant to install. Unfortunately, few shadetree mechanics own an expensive recovery machine or the required cylinder or the scale that is both sensitive enough to measure fractions of an ounce yet robust enough to take the weight of a 40 lb cylinder (note: refrigerant scales were once a very expensive item, but Chinese postal or kitchen scales and eBay have changed all that!) It is both illegal and irresponsible to deliberately vent refrigerant into the air so if you want to start from scratch you should go to an AC shop and have them recover the refrigerant before you start.

The alternative is to fill by performance. Think back a few paragraphs to when we talked about the operation of the evaporator. Ideally, the last of the liquid we are spraying into the evaporator is almost – but not quite – all vaporized at the outlet. Those last few droplets vapourize in the accumulator. All the heat absorbed was in the form of latent heat – the heat energy required to vaporize a liquid. If you think back to your high-school chemistry, you will remember that the latent heat of vapourization represents a lot of energy being absorbed in converting a liquid to a vapour with no increase in temperature. This means that the temperature of the refrigerant coming out of the evaporator is virtually the same as the temperature going in!

So with a reasonably accurate temperature measuring device, and the engine at 1,500 to 2,000 rpm, we can verify the charge in the system by simply comparing the inlet and outlet temperatures at the firewall. If the evaporator inlet pipe (the one coming from the condenser) is colder than the outlet pipe (going to the accumulator), then the system is undercharged. We should start to see some superheat (warming of the vapour) at the accumulator itself and if the outlet of the accumulator is not warmer than the evaporator inlet, then the system is overcharged.

What’s the high pressure gauge good for?

The high pressure gauge is a diagnostic tool. It can provide us hints about what is going on in the system, especially what is going on with the compressor, condenser and orifice tube. There is no hard and fast ideal high pressure; it will fluctuate pretty dramatically depending on ambient temperature, interior temperature, humidity, engine rpm, fan speed, etc. As a rough rule of thumb, the high pressure should be approximately 2.4 times the ambient air temperature. The FSM provides some charts with greater detail based on temperature and humidity, which I have appended as R-12 performance chart and R-134a performance chart at the end of this series.

If the high pressure is significantly lower than expected, this suggests the compressor is failing (or in one case I know of: the orifice tube is missing!). Conversely, if the pressure is significantly higher than expected it could be the orifice tube screen is partially plugged or it could mean the condenser is not cooling properly. Misting the condenser with a garden hose should make the high pressure plummet almost immediately if it is a condenser cooling problem, and one would then look to possible air flow or fan problems rather than tear apart the system to look at the orifice tube.

Another very useful diagnostic tool is the electronic leak detector. Good ones, tuned to detect refrigerants and avoid false alarms are very expensive – out of the typical shadetree mechanic’s snack bracket. However, there are alternatives: most broad band pelletized catalyst combustible gas detectors will pick up on R-134a quite nicely as will cheap corona-discharge refrigerant detectors. Just be aware that they may false-alarm as they respond to a wide range of gasses. But they will get into hard-to-see places and don’t rely on having a dye already installed in the system.

If you know there is UV dye in the system, a compact fluorescent “black light” bulb in a drop-light (aka: handheld work light) is a cheap but effective way to make the leaks glow, especially in dim lighting. Don’t waste your time on an incandescent bulb though – those things just don’t kick out much UV light. LEDs have dropped in price over the last few years, and now you can buy LED based UV flashlights. They don’t cover the entire UV spectrum, but they do work reasonably well with UV dye and fit into tighter spaces than a drop-light.

Converting from R-12 to R-134a

Once upon a time R-12 (Freon) was considered so harmless it was commonly used for everything from air conditioning refrigerant to the propellant in hair sprays and asthma inhalers. It was cheap, it was plentiful, apparently safe and tons of it was sprayed into the air every day. There was a brief scare that it might cause Legionnaire’s Disease back in the mid-70’s when a leaky hotel AC unit was associated with a breakout, but that was quickly disproven and a bacterium growing in the AC system condensate was eventually identified as the culprit.  But then along came the 80’s when flourocarbons were blamed for the vanishing ozone layer. The Montreal Protocol banned their production and use world-wide and the scramble was on to find a replacement. DuPont came up with R-134a, which has similar physical properties, but doesn’t react with ozone layers and it quickly took over virtually all the roles previously filled by R-12. For a while. Now that the ozone layer has been saved, legislators are worried about greenhouse gasses and R-134a is being targeted. You can still readily buy it for the purpose of firing paintballs at each other but in some places it is illegal to buy it for refilling your AC system (at least, in the small cans preferred by shadetree mechanics).

This must be considered when debating the conversion. In Canada, R-12 is no longer available, period. Even R-134a is supposed to only be available to licensed professionals.  But in the US, I understand the situation is a little different: only licensed persons can get their hands on R-12, but the license isn’t that hard to get. R-12 is considerably more expensive, making the conversion quite tempting.

R-134a is very similar to R-12, but not quite identical. For one thing, it has a greater tendency to leak. For another the pressure-temperature curve is slightly different – R-134a boils at a slightly lower pressure down near the freezing point of water but a higher pressure near the boiling point of water, which means that it won’t get as cold in the evaporator unless the low pressure switch is adjusted and it will exert higher pressures on the compressor and condenser. Converting some vehicles means replacing all the rubber hoses with special barrier hoses and replacing the condenser with a more efficient parallel-flow design. Fortunately, the 91 –93 B-bodies don’t need that level of retrofit. Those parts of the puzzle are already in place. All that needs physically happen on these cars is to install readily available R-134a adapters on the existing R-12 ports. You will also need to install an R-134a compatible oil, because the new refrigerant will not circulate the old mineral oil.

A controversial topic is whether it is first necessary to remove the old mineral oil. At one time, it was thought that leaving the old oil in the system was a recipe for disaster, based on the reaction of PAG oil and mineral oil. However, a series of studies showed that when used with ester oil (POE) the mineral oil remains inert, causing no problems. It’s presence may even be of some benefit, as it tends to seal o-rings, rubber hoses and seals. For the most part, it just finds a low point in the system and stays there.

All that said, the presence of mineral oil does reduce the internal volume of the system slightly. In theory, this reduces the capacity of the system slightly. The alternative is to open up the system and drain it out. As it is distributed around the entire system, this means removing a lot of parts and/or flushing (more on that below). If the system is not currently functioning and major repairs are necessary, this is a moot point. But let us assume the system is still somewhat functional and this exercise is simply because the charge is too low.

My personal preference is a compromise: change out the accumulator. My reasoning is that the accumulator is relatively cheap, easy to access and change, holds a lot of the old oil in the dessicant bag and that since the two oils won’t mix, the desiccant bag won’t have the opportunity to remove any moisture present in the system unless the bag is coated with ester oil. Plus you have to take apart the accumulator outlet connection and pour the new oil into the line to the compressor anyway, and a new accumulator comes with new o-rings. GM says you should replace the accumulator if it is more than 5 years old, so I think we all qualify.

So, to convert the system first it should be evacuated of all R-12 using the proper recovery equipment. Next, install the new accumulator or remove the outlet hose from the existing one -- whichever way you decided to go – and pour 6 oz of oil in that hose. I say “pour” but really, a large plastic syringe is a heck of a lot easier and less messy. Pour out some of the mineral oil from the old accumulator into a clean container and use that to lubricate the o-rings before connecting the lines back up. It would be better to use new mineral based refrigerant oil to lube the o-rings, but you only need a smear and that stuff is expensive! Transfer the low pressure switch from the old accumulator if required.

Next, the low pressure port adapter should be installed (unless your new accumulator came with an actual R-134a low pressure port already). There are two styles available. The first is equipped with a simple plunger to extend the existing R-12 Schrader valve core. The little “core” is loose and not spring loaded. This style is simply screwed right on the existing port. The other design includes it’s own Schrader valve. With this style, the existing R-12 valve core is removed before the adapter is screwed onto the fitting. The core on this style is obviously spring loaded.

Unfortunately, the high side adapter you get with a retrofit kit might not work with some gauge sets. GM didn’t use a Schrader valve on the high side – they used a stemless flat rubber check valve and the pintle of some R-134a hose valves just won’t reach far enough over an adapter. If this is the case with your set, then you need to unscrew the existing port (located on the high pressure line near the compressor) and replace it with AC Delco part number 15-30418 (which comes with a cap) or 15-5438 (port only, no cap). Fortunately they are only a few dollars. And before you tear your hair out in frustration – the new port has 8 flats, so a standard socket won’t fit. You have to use a wrench. When unscrewing the old port, make sure you use a wrench to hold the line fitting or you could bend, fold or mutilate your high pressure line.

Hook up your vacuum pump, evacuate the system and charge with R-134a. The total charge should be approximately 10% less by weight than the R12 charge listed on the underhood sticker, so a full charge in the converted system should be 44 ounces.

Once most of the charge is in and the compressor is cycling, take a note of the cut-out (ie: lowest) pressure on the LP gauge. Now, unplug the switch on the accumulator, and look in the centre between the two terminals. You should see a small screw. This is the adjustment screw. Clockwise raises the cut-out pressure, counter-clockwise lowers the cut-out pressure. Carefully adjust it, a tiny bit at a time (1/8 turn = 2psi), so the cut-out pressure is 22-23 psi. Of course, you have to plug the wires back in every time you make an adjustment because the compressor won’t run with that switch disconnected!

The final step is to make up a new sticker or change the old one for future reference.

At a minimum, the sticker should include:

NOTICE: RETROFITTED TO R-134A
RETROFIT PROCEDURE PERFORMED TO SAE J1661
USE ONLY R-134a REFRIGERANT AND SYNTHETIC OIL OR A/C SYSTEM WILL BE DAMAGED.
REFRIGERANT CHARGE / AMOUNT:______________
LUBRICANT AMOUNT: __________ PAG □ ESTER □

HC (Hydrocarbon) refrigerants (RedTek, Duracool, etc.)

Probably nothing is as controversial as the use of HC refrigerants in automotive air conditioning systems. In my opinion, the root of the controversy is money. DuPont, et. al., have some serious coin invested in proprietary refrigerants like R-134a. AC shops have serious coin invested in the expensive recovery and recycling machines mandated by law to handle those refrigerants. And the AC equipment manufacturers are making a nice living selling their machines to the shops. So there is a lot of money in Auto AC.

On the other side of the coin, HC refrigerants are nothing more than a blend of iso-butane and denatured propane. They might be a little more highly refined than the gasses that come in a disposable lighter or BBQ tank, but you can’t patent propane or butane and there’s very little R&D required to find a suitable mixture. So the HC refrigerant manufacturers can readily undercut the big boys and still make big profits. You can refill a 20 lb tank of propane for a dollar a pound, but put it in a can with a big “12a” on the front and the market might bear $2 an ounce! So let’s wander through the forest and examine some of the myths and issues.

Is it legal?
In general it is illegal to mix different refrigerants in a single AC system. Any R-12, R-134a, etc. must first be properly recovered in the approved manner with the required equipment. After that, it depends on which lobby group has been most effective in your particular jurisdiction. In Canada, it is not illegal to install an HC refrigerant, and in fact is the only refrigerant available to the shadetree mechanic in small cans. The captive market means Duracool (the major Canadian HC producer) charges Canadians a 50% premium over and above what they sell it to Americans for.

Under US federal law, it is not legal to install it in an R-12 system. And the EPA has caught on to pseudo-conversions, apparently going after some shops that simply screwed on new fittings and filled with the HC, never having installed R-134a in the system. (As if that intermediate step really makes some sort of difference!) However, that does not necessarily mean it is legal to install in an R-134a system either – certain individual states have banned it’s use in auto AC applications regardless of federal law.

Is it safe?
Not according to those with a vested interest in traditional refrigerants. Just as safe as gasoline, according to the HC manufacturers. Each side throws around terms like flammability and ignition temperature in their arguments. Let’s look at the two sides:

AC Industry: Propane and butane are flammable. At normal atmospheric temperatures and pressures, HC refrigerants have an auto-ignition temperature of less than 750°C. Therefore they are flammable and therefore they are dangerous. Under the same conditions, R-134a has an auto-ignition temperature above 750°, therefore it is non-flammable and therefore it is safe.

HC industry: But refrigerants are not subjected them to normal atmospheric temperatures and pressures in normal use.  And AC systems don’t circulate pure R-134a through the system either – oil is carried in suspension. When the mixture of oil and R-134a is sprayed out of a pressurized system, the mixture has an autoignition temperature below 750°C. Besides, we got a laboratory in Australia to certify that our highly refined product really has an autoignition temperature of 890°F, so really, our product should be considered less flammable than R-134a.

My comment: Ummm, you forgot to have the Aussies mix oil in your product too. Besides, once it leaves the system, it’s no longer under higher pressure.

HC industry: our refrigerants are just as safe as the fuels already used to power a vehicle. And R-134a is a health hazard that can cause all kinds of problems if inhaled

AC industry: but fuel systems are designed to keep all the fuel out of the passenger compartment. The tank is placed at the rear of the vehicle centered between structural supports and the engine itself is separated from the passengers by a firewall. The AC condenser is right up in front of the radiator and one of the first things to be hit in a head-on collision and the lines penetrate the firewall into the passenger compartment. In 2008, one firefighter was killed and 7 more seriously injured when a leaking New Zealand cold-storage facility HC refrigeration system blew up.

My comment: Can you say “Pinto”, or “Volkswagen Beetle”? And just how much HC do you think you can squeeze into one little auto AC system anyway? Actually, there’s a hilarious video of two fools trying to show that the total volume of an HC refrigerant charge, exhausted into the total volume of air in a passenger compartment results in a concentration that is too lean to burn. These morons dumped 4 cans of HC refrigerant inside the vehicle one of them was sitting in. The windows were closed to prevent the gas from escaping. He then lit a match. Guess they forgot that it takes time for the gas to distribute itself evenly because the resulting explosion was impressive. However, the professor DID walk away with little more than singed eyebrows, burns to his hands, a cut on the forehead and a temporary loss of hearing, proving that while it’s not as safe as they thought, it isn’t necessarily life threatening either.  Here it is on you tube:

The reality is that there has been little or no serious scientific investigation about the safety of HC refrigerants in automobiles, one way or the other. Anecdotally, there have been very few vehicle fires blamed on HC refrigerants, despite their widespread use in Australia and Canada, in particular in older cars that probably leaked all their existing refrigerant away to begin with. One well documented fire resulted from a home-made hose being rubbed through during a dyno test and igniting. It was quickly doused because someone had a fire extinguisher handy. So they are almost certainly not as dangerous as the AC industry would have legislators believe, but then, there’s no evidence that they are perfectly safe as the HC industry claims either.

As for the health hazards of R-134a: HELLO! This stuff is used as the propellant in asthma inhalers for crying out loud. How gullible do you really think we are?

Is it effective?
HC Industry: HC refrigerants are now being used in more and more fixed building installations because of their great efficiency and lower cost. They are a drop-in replacement for R-12 and R-134a, fully compatible with all mineral and synthetic refrigerant oils currently on the market. Lower head pressures result in lower energy consumption for the same amount of cooling. Systems with HC refrigerants installed actually blow colder air than R-134a systems.

AC industry: yes, but they’re flammable. Besides, we are going to develop a new proprietary refrigerant called HFO-1234yf that will address all the environmental concerns surrounding R-134a. It will be perfect and there will never be a need to develop anything else ever again.

My comment: Really. Let’s look at the track record: when R-12 was found to be bad for the environment, DuPont developed a wonderful, safe, effective harmless refrigerant instead: R-134a. What an odd coincidence that it is discovered to be a problem and legislated out of existence just as the patent protection is about to expire…

As for getting colder air out of the HC charged system, who do you think you’re kidding now? The system already has to have provisions to keep the evaporator from freezing over using R-134a. Does HC have some magical property that prevents condensation from freezing on the evaporator? It may be that HC refrigerant will provide better cooling overall as a retrofit gas in an R-12 system with a linear condenser than R-134a would in that same system, particularly if the low pressure switch was not properly readjusted to suit the R-134a pressure/temperature curve. However, I can find no qualitative data for the claim.

One odd thing about HC refrigerant blends is that dew point and the vapour point are different. In other words, the condensing temperature is higher than the evaporating temperature – presumably because one of the gasses has a different vapour point than the other. They not be charged as a vapour, as they will separate and more of the gas with the lower vapour point will be left behind in the can, changing the ratio and drastically altering the overall properties. In my own subjective, non-scientific experience, the difference leads to a delayed cycling of the compressor in an R-134a system that was retrofitted. The overall cooling performance seemed slightly reduced, and I can’t say that I noticed any improvement in gas mileage either. But that was in one particular car, I didn’t take proper before-and-after measurements under controlled circumstances and I’ve never repeated the experiment on anything else. The AC unquestionably blows cold, so maybe it is just my perception.

In conclusion
If R-134a is not available to you, HC refrigerants are a viable option. They do have a much shorter lifespan if released to the atmosphere, so in that they must be considered a “greener” alternative. The reduced head pressures suggest reduced load on the compressor and while the net effect on gas mileage is still a question mark, it seems reasonable that reduced head pressures will reduce wear and leakage. It is extremely unlikely that they will turn your car into an instant fireball even in a head-on collision but it is probable they would accelerate an existing engine compartment fire that burns through a line. It is unknown whether that acceleration would be more intense than compared to an R-134a/oil combination. Also unanswered is the question of whether the distraction of driving a broiling car is a greater overall threat to life than the limited increased fire risk of cooling that interior with an HC refrigerant.

If R-134a is available to you, then you have to decide for yourself whether to believe the HC industry’s unsubstantiated claims of better efficiency or the AC industry’s unsubstantiated claims of increased risk.

If you retrofit HC into your system, follow all the usual requirements for any retrofit – evacuate and recover all existing refrigerant, change port fittings, clearly label the new contents, adjust the charge weight to suit as stipulated by the manufacturer, etc. Two other caveats:

1. do not mix oils. Yes, HC refrigerants are compatible with all refrigerant oils, including mineral based. But that doesn’t mean the oils are compatible with each other. If you are retrofitting an R-12 system, stick with mineral oil. If you are retrofitting an R-134 system, stick with PAG oil (or whatever was in there if this isn’t your Roadmaster). It is safe to put a charge of ester oil on top of the existing mineral oil in an R-12 to R134a conversion because R-134a will let the mineral oil settle to the low spots and leave it there. No one has fully investigated the long-term effects of HC refrigerant carrying both mineral and synthetic oil around the system simultaneously.

2. Never mind taking the car to a professional AC shop ever again. Most are now using refrigerant detectors to verify what is in the system, since it is illegal to mix refrigerants. Few will touch your car if they find evidence of an HC refrigerant inside. Once you go HC, consider yourself on your own.

3. (I know, I said 2, but I can’t count). Charge with liquid only: IE valve down. In fact, give the can a shake first. Otherwise the gasses will separate and who knows what the final ratio will be.

To Flush or Not to Flush, that is the question.

Another controversial subject. GM recommends against flushing, many AC shops swear by it.

Flushing is the act of flowing a solvent through the system to dissolve and remove any debris or contaminants. The entire system cannot be flushed as a whole, and some components cannot be flushed, period. You cannot flush a compressor. You cannot flush an orifice tube. You cannot flush a parallel flow condenser.

You can flush lines, you can flush the evaporator, you can flush a linear flow condenser.

Let’s start by looking at the arguments in favour of flushing: any debris remaining in the system after a failure, such as bits of metal shed by the compressor, will eventually make their way through the system and damage the new compressor and/or plug up the orifice tube. Getting them out will ensure longevity for the repaired system. All true.

Arguing against flushing, if the flushing wasn’t perfect, all you did was concentrate the debris and make it more likely it will cause problems as it travels in one large mass. If the flushing chemicals were not completely compatible with the refrigerant, there could be a chemical reaction that will cause even worse problems (eg: R-11 is often used as a flushing agent, it’s OK with R-12 but will react with R-134a). So it is vitally important that all the flush be completely removed from the system and that is a difficult thing to assure. Also all true.

GM’s position is that an inline filter is the best option. An inline filter is considered a warranty approved alternative to flushing when replacing a failed compressor*. In fact, there is one installed as part of the orifice tube, but it’s possible to add additional ones. There are filters available that can be cut into the line (eg: GM part number 89016656 / AC Delco part #15-10413 or GM part #88960977 / AC Delco part # 15-1747). Early factory service manuals also reference an aftermarket liquid-line filter. Note that if you install a liquid line filter, it may include an integral orifice tube. If so, you must remove and discard the original orifice tube for proper cooling. Also, a liquid line filter comes after the compressor, so it does NOT protect the compressor from debris.

The simplest and least likely to leak is a filter screen that fits into the suction line at the compressor (and therefore also protects the compressor). Simply disconnect the block from the compressor and insert the screen. These were not developed until 1998 -- after the last B-body was built -- so the various FSMs make no mention of it, but they are completely compatible with our cars. GM of course offers a special tool (J-44551-5) and mandrel (J-44551-3) for the task but it’s a clumsy excuse for a tool and with some ingenuity you can press the screen into the block without it.

A wide variety of screen sizes are available to fit various vehicles. I have yet to find a definitive list of what vehicles require which screen size. I do know that a 1995 Roadmaster takes size “A” – 0.510” diameter, to fit lines with an ID from .499” to .507”. GM Part # J44551-10, AC Delco part # 15-21186 (not all sizes are available from Delco). I strongly suspect all B-bodies from 94 to 96 use the same, as they all use the same compressor. For the sake of possible others, here are all the available sizes:

J44551-	Size"	Range”	AC Delco Colour, #	Notes
-40	.395"	.384”-.392”
-60	.471"	.460”-468”	Blue, 15-21185
-70	.492"	.480”-489”
-80	.500"	.489”-.497”
-10	.510"	.499”-.507”	Natural, 15-21186	Originally size “A”, J44551-15
-90	.521"	.510”-.518”	Yellow, 15-21187
-110	.531"	.520”-.528”	Green, 15-21188
-120	.538"	.527”-.535”	Purple, 15-21189
-20	.552"	.541”-.549”	Black, 15-21190	Originally size “B”, J44551-16
-130	.561"	.550”-.558”
-140	.579"	.568”-.576”
-30	.595"	.584”-.592”	Natural, 15-21191	Originally size “C”, J44551-17
-150	.600"	.580”-.597”
-160	.618"	.607”-.615”	Green, 15-21192
-170	.649"	.638”-.646”

AC Delco sells a set, including the tool, under part # 1521184. The original GM kit, J-44551-A, includes the tool, mandrels, etc. and sizes A, B, and C. The rest of the sizes come in the update kit, J-44551-50. There is also a J-44551-800 kit, BUT it is Saturn specific and comes with the –90, -120 and –20 sizes only so it probably won’t work with a B-body car.

So much for the inlet filter, let’s look at the failure mode:

The most likely failure is the compressor. When it fails, debris flows out into the condenser and may make it as far as the orifice tube. All the lines between the compressor and the orifice tube must be considered compromised. The line to the evaporator, the evaporator itself, the accumulator and the lines back to the compressor should be protected by the orifice tube filter.

Since the compressor has to be replaced and can’t be flushed anyway, we don’t need to worry about that. Same for the parallel flow condenser fitted to the B-bodies – compressor failure means the condenser must be replaced. In theory, it might be possible to clean the orifice tube and screen, but for 93¢, is it really worth the effort? We will be replacing the accumulator as part of the repair – after all, it is designed to capture and hold any contamination or debris that develops as part of the normal operation of the system, and replacement is a requirement of any replacement compressor warranty.

This leaves the hoses. To replace them with new is under $100. To flush them requires a flushing machine and an hour per item (15 minutes of flushing followed by 45 minutes of trying to get all the flush chemical out). The cheapest warranty approved** flush system is around $400 from Hecat, and uses your shop air supply. Add another $25-$50 for the flushing solvent itself. Approved** solvents are the Hecat’s “Safe Flush”, Four Season’s “Dura-Flush” (R-141b), and Cliplight’s “Diamond Flush” but only when used with the respective manufacuter’s equipment.

Now, if there is reason to suspect contamination of the evaporator (for example, if a sealant chemical was added to the system), flushing becomes a more attractive option. The evaporator itself is less than $100, but replacement is intimidating. The B-bodies are not as bad as some cars where the entire dash must be removed down to the firewall, but it is a cramped and awkward job that will occupy several hours.

What to use as a flushing agent is the next debate. GM approves closed-loop flushing using R-12 only (for flushing R-12 systems) and R-134a only (for flushing R-134a systems) as a condition of warranty. On the other hand, Four Seasons says exactly the opposite: use of a closed-loop flushing machine with refrigerant as the flushing agent specifically voids their warranty!

Both agree that the following methods will void any warranty:

• Open loop flushing (hand held flush gun) with terpene based flush.
  
• Any flushing chemical that is oil, alcohol, mineral spirits, isopropanol, acetone, toluene, or heptane based.
  
• Any flush chemical that has a known health hazard (1.1.1. trichloroethylene, trichloroethene, tetrachloroethylene).
  
• Any other cleaning agent such as brake clean, degreasers, carburetor cleaners, etc. that can deteriorate rubber components and reduce lubrication.
  

So to recap:
The only universally accepted and warranty approved method of flushing is not to flush at all! Install a filter in the compressor suction line instead. Replace the orifice tube, the accumulator and the compressor hose. If there is a sealant in the system, you are better to flush the evaporator since the use of a sealant in the system will void any warranty by anyone anyway. If you are going to flush, don’t bother wasting your time and money on one of those cans of self-propelled flush – they may do more harm than good. And finally, make sure absolutely every drop of flush is cleansed from the system. It takes about 15 minutes to flush a component with the proper machine, and about 45 minutes to get the flush out of that same component.

*by Four Seasons, a major aftermarket AC components supplier and by AC Delco
**by Four Seasons only

Troubleshooting (or why do I need gauges if they won’t tell me the amount of charge?)

Refrigerant is considered a condensable, not a compressible. That is to say, the more refrigerant you put in a closed system, the more liquid you get. The pressure does not increase unless the temperature increases. Each refrigerant has a very specific relationship between pressure and temperature. Knowing this, the pressures can tell us a lot about what is going on inside the system.

Under normal conditions with the system not running and everything at an equalized temperature, the static pressures at both the HP port and the LP port are equal to expected pressure at ambient temperature as indicated on a temperature/pressure chart for that specific refrigerant.

Once the system is running, as described above, the system divides itself into two halves: the High Pressure side and the Low Pressure side, each with a port to monitor the pressure. Once running, typical pressures for R-134a are LP =30-31psi, HP 204-210psi. R-12 has a narrower range, typically LP = 32-33psi, HP=185-190psi.

The above is rule of thumb only. Actual normal pressures may vary by ambient temp and humidity. Reference should be made to the R-12 performance chart or R-134a performance chart for greater precision when deciding if a pressure should be considered low, normal or high under the exact circumstances. The same goes for the vent temperature – a system in perfect working order might blow near-freezing air out the centre vent when ambient temperatures are 75°F and RH is 40 or 50%, but might only achieve 70°F when the mercury climbs over 90° and humidity approaches 100%.

LP low, HP normal to slightly low.
Probable causes:
1. Incorrect adjustment of LP switch (note: some vehicles use a  TXV valve control to perform the function of the orifice tube used in the B-bodies. The symptoms associated with the orifice tube would apply, with modification, to the TXV valve in those vehicles)

2. Restriction in the low side of the system. EG: plugged orifice or screen.

3. Moisture in the system (may freeze at orifice inlet, causing very cold inlet).

To verify between 2 and 3, turn off AC and allow to stabilize for 15 minutes, then turn back on. If gauge reading immediately goes to abnormal condition, the screen is probably clogged. If the gauge readings are normal for a few minutes, then goes abnormal, it is probably excess moisture.

LP very low to low, HP low
Probable causes:
1. Low refrigerant charge
2. Clogged inlet screen
3. Defective or plugged orifice tube
4. Moisture in the system (as above)
5. High-side restriction in the high side before the orifice tube (eg: crushed condenser tube).
Loss of refrigerant is usually due to a leak, although all systems leak a tiny bit as a normal condition. R-134a systems are usually more sensitive to the correct charge than R-12 systems. R-12 systems usually have a sight glass, and low charge will show up as bubbles in the sight glass. The compressor will usually cycle rapidly, as there is insufficient liquid refrigerant available to supply the compressor intake. The evaporator outlet line may also be warm, as there is insufficient refrigerant to keep the evaporator full of liquid.

LP low, HP high to extremely high.
Probable cause:
1. Restriction in the high side of the system
2. excessive oil charge
In the case of a restriction, it could be anywhere from the compressor outlet to the fixed orifice tube. The closer to the compressor, the higher the high-side gauge pressure will be (because the refrigerant has less time and space to cool).  Probably a kinked or bent line or tube. A marked temperature change will often mark the location (as it acts like an orifice tube) with the cooler side downstream of the restriction.

Excessive oil charge may result in vibrating or pulsating hoses. The oil acts as both an insulator and takes up space. The compressor must work harder against higher pressure and the lines may vibrate as the refrigerant is pushed through a pool of oil in a line. The outside temperature of the high-side line might be cooler than the temperature/pressure chart would indicate as the oil is preventing heat transfer from the refrigerant to the line.

LP high, HP low
Probable causes (electrical):
1. compressor LP cycling switch
2. PCM pressure sensor
3. ambient air temperature switch
4. engine coolant temperature sensor
5. throttle position sensor

(mechanical)
6. defective/misadjusted clutch
7. defective compressor

The compressor may or may not be running – look closely at the centre hub. If it is not turning, or not turning continuously with the pulley then there is an electrical problem or a clutch problem. If it is turning, it is probably the compressor itself or a mechanical problem with the clutch.

Disconnect the LP cycling switch connector (on the accumulator). With a piece of jumper wire, briefly short the two wires and verify if the compressor starts. The switch is adjustable, using the small slot-headed screw inside the connector. The switch can also be replaced without evacuating the system as it has a Schrader valve beneath it.

Using a scan tool, verify the AC pressure as seen by the PCM and compare the reading to the actual measured pressure. If the indicated pressure is much higher than the measured pressure, the PCM may be cycling the compressor each time the sensor hits the high limit. Similarly, the PCM will shut down the compressor if it detects an overheated engine, full-throttle operation, or low ambient temperature.

If the clutch fails to engage, only engages when travelling uphill or only engages when the centre is tapped then the clutch gap may be excessive and should be reduced. The clutch coil may be burnt out or shorted. A shorted coil will draw excessive current and usually blow it’s fuse. An open coil can be determined with an ohm meter. A slipping clutch is rare, but can occur if a rivet or arm has broken and it will make a racket. You won’t need gauges to figure out what’s wrong if the clutch is slipping.

Once you have ruled everything else out, the only thing left is a worn out or damaged compressor. Unless someone else worked on the system and forgot to reinstall the orifice tube.

LP high, HP high to extremely high
Probable causes:
1. Air (or other contaminants) in the system
2. Overcharge of refrigerant
3. Excess oil in system
4. Condenser fins clogged or obstructed or debris trapped between condenser and radiator
5. Defective cooling fan(s)
6. Overheating engine
7. Incorrect refrigerant

An overcharge of refrigerant will often result in a cool to warm evaporator outlet pipe and while cooling will be poor at idle, it may be OK on the highway. Air in the system can be very similar, but can sometimes be identified by turning the system off while watching the gauges – the pressure will drop 20 or 30 psi very quickly, then taper off very gradually as the two sides equalize.

Poor cooling of the condenser (fans, obstructions to airflow) can usually be temporarily corrected by misting the condenser with a garden hose and sprayer. The HP will usually plummet to near-normal almost immediately when the water hits the condenser.

An overheating engine causes additional heat load, keeping the condenser warm by radiation.

Air contamination can result from improper charging practices of course. However, one Florida study suggested that as many as 25% of commercial AC shops have contaminated recovery equipment, probably by servicing vehicles after the owner tried and failed. Similarly, the owner may have tried to recharge the system with some other refrigerant. If the shop does not test the system contents with an expensive refrigerant identifier, the contaminants get spread to successive customers.

Exactly what I said!Bet you are a blast at a party.Lot of effort went into this and its easier for me to understand than the manual,thanks a bunch Dwayne!

Flasheroo wrote:

Exactly what I said!Bet you are a blast at a party.

Psst. See that guy all by himself in the corner, the one that came in the old station wagon? Don't make eye contact unless your AC is broken. He'll go on about it all night.  

Actually, most of the work seemed to be getting the tables to format properly. Maybe the moderators could investigate TinyMCE for the WYSIWYG text editor at the next upgrade? Please?

I don't know if this is forum legal or not but Here is a great reference from www.AutoAcForum.com on what type of oil, and how much refrigerant is required for vehicles. It appears to be all inclusive. I used it as a guide when re-doing the retrofit on Ms. Roadie this summer.

http://www.autoacforum.com/speclisting.pdf

Mods if we can't link this info please just delete my entry on this thread thanks.

They got that listing from Four Seasons -- the numbers in the 2 right hand columns are Four Season's part numbers for oil. So they don't hold any copyright on the material.

A very useful forum though.

Thank you! Making a Sticky!

Chris

Informative thread, thank you for all the work!

As to the editor updates, those aren't controlled by us since this software is provided free.

so if my sponge is absorbing and is not compressed, do i need to flush?

great thread, I only read a couple parts though as I've done at least 5 or 6 systems now. Interesting reading on the HC stuff...