Automotive Air Conditioning Information Forum

Hi,
I have a 2006 Honda Civic GX (runs on natural gas). A month or so ago the condenser was hit by a rock and vented all the refrigerant. Luckily dye was in system so I could find the location of the leak. After replacing the condenser, I vacuumed the system for 20 minutes, it held a vacuum at 30 inHg (for over an hour), no leaks. I then continued to vacuum the system for four hours to boil and remove any moisture that may have been introduced. Then I added 15.4 ounces of refrigerant.

Because it continued blowing warm air for a week afterward, I added another ounce or two of refrigerant an attempt to get the vent temp lower. This didn't help, although pressures did change.

Here are my current gauge readings:
68 degrees outside
engine at 2000 rpm
compressor on
readings Low 20 psi, High 168 psi
vent 59

When idling at 800 rpm with compressor on, still 68 degrees outside, vent still blows 59, gauge readings Low 35 psi, High 142 psi.

The Low side lines form condensation on the outside and the High side lines are very warm.

My basic question is why is the system only blowing 59 degrees F when it is 68 degrees F outside? Shouldn't it blow closer to 40 degrees F? When the outside temp is above 80 degrees F outside the AC system continues to horribly under perform. Could another component in the system have failed? Thank you in advance for your responses.

EDIT: I forgot to mention this Civic has 215,000 miles, just in case that matters in some way. I also replaced the condenser fan that wasn't spinning.

EDIT2: I'm thinking my problem could be with an aged, or partially obstructed, expansion valve. Does anyone concur?

I remain hopeful someone will be able to provide feedback. Please let me know if my suspicion is accurate, that the TXV needs to be replaced, or if something else may be the culprit. If more info is needed, I should be able to provide. I'm asking as in my Honda Civic it is quite a task to remove all the interior components to access the TXV and evaporator. If I can avoid dismantling the interior, due to some other component failing, I'd be grateful for that awareness. With that said, if the whole system needs to be dismantled and cleaned I'm up for the task. Thanks, Rick

It's always best to test and evaluate system performance when it's fully charged and the ambient temperature is at or above 80°F. Adding more refrigerant over the manufacturer's specification (overcharging) won't help, as you found out, and might actually damage the system. And, too, an engine that's idling with the compressor engaged won't reveal much because the compressor is moving very little refrigerant at that point. You also tested the pressures at 2,000 RPM, which is good.

However, the center-vent temperature is still warm.

Several issues come to mind. The first is at 215,000 miles, the scroll compressor might be "tired." If the compressor has never been changed, I would suspect that first. Second, you might have some reheating going on, That is, heated air is mixing with the cooled air. Third is that your car probably has a temperature sensor on the compressor that might be faulty. Fourth is the TXV, although with "wonky" pressures and warm vent temperatures, I would still suspect the scroll compressor first, which historically doesn't last much beyond 100,000 miles, if that.

Can you disconnect the liquid line (the skinnier of the two) that connects to the TXV at the firewall? If you can, check inside the line and at the inlet of the TXV for the presence of desiccant beads (that look like coarse sand), which would indicate a failed R/D element.

John, thank you for your reply. Years ago, and about 50k or 60k miles ago, the compressor/clutch was replaced due to a failed compressor. Since that service the AC hasn't blown as cold as it was prior to that failure. The repair shop at the time only replaced the compressor and possibly (hopefully) the dryer desiccant. They didn't go further mostly because the system "worked," just not as well.

This time around I have more time (I'm retired) so I personally replaced the condenser that has an integrated receiver/dryer. When I inspected the old receiver/dryer desiccant element it was intact yet the bottom of it was dark compared to the top. I'm not sure if that darkness is normal or it reflects remnants of the original compressor failure. I replaced the condenser with receiver/dryer last, just before I closed up the system to minimize exposure to the air (moisture). I mention this because it may be a sign of small particles circling around the system through the various tubes and components. That is why I suspected the TXV may be partially blocked.

I'm not sure how I can check the air handler blender door movement. Is there a way to easily check to see if it is moving to the proper position for just cooling? I would like to think that since it is typically hot this time of year the doors are fully in the cooling only position. With that said, when I checked last week it was 68 degrees outside. The floor, face and defrost vent positions do work properly.

I can observe the compressor clutch engaging when the controls are set to max cool. It stays engaged yet the vents never get cooler than 59 degrees F. This makes me think the temperature sensor on the compressor is functioning properly. With that said, prior to replacing the condenser the clutch was short cycling, engaged then disengaged every 10 seconds or so.

Today I drove the car about 30 miles in 80 degree weather. The vent temps remained wimpy at around 60 degrees, never colder. Also, they didn't fluctuate up and down as I drove on the freeway or through street stop & go traffic. I didn't have my gauge set to check pressures so can't report anything in that area. The rest of this week it is supposed to be 80+ degrees F in the Los Angeles area so I do have more chances to investigate and report low and high side pressures if that would be helpful.

Later this week I'll get the system recovered (my local PepBoys does it for free, "payment" is they keep the R-134a they recover). Then I'll be able to check the liquid line for any weird stuff at the TXV. The TXV in my Civic is one of those "H" types, so disconnecting it also disconnects the vapor side. I suspect I won't see any beads yet may see remnants of the old compressor failure. Beads actually make more sense as I suspect that valve is partially blocked. Metal fragments might block it entirely. Do TXV's wear down over time, mine has never been replaced? My Civic is almost 20 years old and has high mileage.

While I'm in there, with access to everything and a willingness to spend extra time, what is a good procedure, to back flush the connecting tubes and possibly the evaporator & condenser? Any help on a cleaning fluid would be helpful too as I imagine I'll need more than compressed air. Once cleaned, any advice on replacing Pag-46 oil in the various components would be helpful, unless just counting on oil in the compressor to initiate cycling it around.

Thanks again for your insight. Gaining access to the TXV valve & Evaporator is no simple task, I'm sure experienced mechanics cringe when a Civic is brought in for major AC repairs. Getting to everything behind the glovebox is going to be a challenge for me as I'm six feet tall.

--Rick

rick wrote: Mon Aug 18, 2025 7:45 pm Years ago, and about 50k or 60k miles ago, the compressor/clutch was replaced due to a failed compressor. Since that service the AC hasn't blown as cold as it was prior to that failure. The repair shop at the time only replaced the compressor and possibly (hopefully) the dryer desiccant. They didn't go further mostly because the system "worked," just not as well.

Scroll compressors, like the one used in your car, are known to disintegrate internally, sending metallic and nonmetallic debris downstream. Because the condenser is considered a filter of sorts, the shop, at that time, should have replaced the condenser assembly (including the R/D) and flushed the remaining components to ensure all the debris and old oil were removed.

rick wrote: Mon Aug 18, 2025 7:45 pm This time around I have more time (I'm retired) so I personally replaced the condenser that has an integrated receiver/dryer. When I inspected the old receiver/dryer desiccant element it was intact yet the bottom of it was dark compared to the top. I'm not sure if that darkness is normal or it reflects remnants of the original compressor failure. I replaced the condenser with receiver/dryer last, just before I closed up the system to minimize exposure to the air (moisture). I mention this because it may be a sign of small particles circling around the system through the various tubes and components. That is why I suspected the TXV may be partially blocked.

When you replaced the condenser assembly, did you happen to take note of any oil contained therein? If so, was it black or dark amber, and did it have any "glitter" in it? If it did, I would suspect that the compressor is going out again.

The bottom of the desiccant element was probably dark from the oil, either good or bad.

Replacing the condenser assembly last is a smart move because of the reason you mentioned.

If any suspected compressor debris reached the TXV, you're right—the TXV could have been compromised.

rick wrote: Mon Aug 18, 2025 7:45 pm I'm not sure how I can check the air handler blender door movement. Is there a way to easily check to see if it is moving to the proper position for just cooling? I would like to think that since it is typically hot this time of year the doors are fully in the cooling only position. With that said, when I checked last week it was 68 degrees outside. The floor, face and defrost vent positions do work properly.

I believe that your car has cable controls for the blend doors. If you checked them as you mentioned, chances are they're working as they should. One final verification is to pinch-off the hose supplying the heater with hot water from the engine. You can do this with a pair of locking pliers and two thin pieces of wood to cushion the rubber hose. Then go for a drive with the A/C engaged and see whether it makes any difference in the vent temperatures. If it does, then you have a reheating condition. If not, then the problem is with the A/C system itself.

rick wrote: Mon Aug 18, 2025 7:45 pm I can observe the compressor clutch engaging when the controls are set to max cool. It stays engaged yet the vents never get cooler than 59 degrees F. This makes me think the temperature sensor on the compressor is functioning properly. With that said, prior to replacing the condenser the clutch was short cycling, engaged then disengaged every 10 seconds or so.

The sensor on the compressor senses whether the compressor is overheating. I believe your car also has a thermistor on the evaporator that prevents it from getting too cold, forming ice, and blocking airflow. If the evaporator gets too cold, the thermistor will cause the compressor to cycle, thus preventing ice-up.

rick wrote: Mon Aug 18, 2025 7:45 pm Today I drove the car about 30 miles in 80 degree weather. The vent temps remained wimpy at around 60 degrees, never colder. Also, they didn't fluctuate up and down as I drove on the freeway or through street stop & go traffic. I didn't have my gauge set to check pressures so can't report anything in that area. The rest of this week it is supposed to be 80+ degrees F in the Los Angeles area so I do have more chances to investigate and report low and high side pressures if that would be helpful.

Posting the pressures would help. Be sure to do so with the system at maximum heat load—that is, with the compressor engaged, engine at ~1,800 RPM, A/C set to "max cool" or "recirc," blower on the highest speed, and doors/windows open.

rick wrote: Mon Aug 18, 2025 7:45 pm Later this week I'll get the system recovered (my local PepBoys does it for free, "payment" is they keep the R-134a they recover). Then I'll be able to check the liquid line for any weird stuff at the TXV. The TXV in my Civic is one of those "H" types, so disconnecting it also disconnects the vapor side. I suspect I won't see any beads yet may see remnants of the old compressor failure. Beads actually make more sense as I suspect that valve is partially blocked. Metal fragments might block it entirely. Do TXV's wear down over time, mine has never been replaced? My Civic is almost 20 years old and has high mileage.

TXV's can wear out, lose their bellows charge, and can become partially blocked by compressor debris or desiccant beads. A primary symptom of partial or complete blockage is very low pressure on the low side, or even pulling into a vacuum.

rick wrote: Mon Aug 18, 2025 7:45 pm While I'm in there, with access to everything and a willingness to spend extra time, what is a good procedure, to back flush the connecting tubes and possibly the evaporator & condenser? Any help on a cleaning fluid would be helpful too as I imagine I'll need more than compressed air. Once cleaned, any advice on replacing Pag-46 oil in the various components would be helpful, unless just counting on oil in the compressor to initiate cycling it around.

Back-flushing the components is great if you want to start with a clean and "dry" system. A few notes on that, though: The condenser can't be flushed because it's the parallel-flow type having many micro-passages that no amount of flushing will clear. The only option is to replace it.

The compressor probably should be replaced as well.

The evaporator CAN be flushed, but only if you remove the TXV first. Because you'll have to disassemble the dash to get the TXV out, you might as well replace the TXV and evaporator with new parts.

As for the flushing solvent, contact Tim at ACKits.com, this Forum's sponsor, for a suitable product.

There are two ways to handle the oil: Either fill the compressor with the required amount, if it will all fit (after draining-out whatever is in there from the factory), or "Oil Balance" the system, which involves adding small amounts into each component. The latter has been discussed a number of times on this Forum.

rick wrote: Mon Aug 18, 2025 7:45 pm Gaining access to the TXV valve & Evaporator is no simple task, I'm sure experienced mechanics cringe when a Civic is brought in for major AC repairs. Getting to everything behind the glovebox is going to be a challenge for me as I'm six feet tall.

Most vehicles, not just your Civic, will require roughly a day's work just to remove the evaporator. It's always wise to take pictures at each step in the process so that you'll have references to aid in reassembly.

Hi John,

It is nice to think they cleaned and properly flushed the lines, yet I wasn't there to witness their work. I don't think they replaced the condenser.

When I replaced the condenser, receiver/dryer assembly combo I did not notice any debris. I did use my air compressor to blow out the condenser in an attempt to see how much oil was in there yet nothing came out - no oil or debris.

I have a temperature gun on order, once I get it I'll be able to spot check line temperatures. Right now I can only go by touch. Coming from the firewall the low side temp is cool and the high side is luke warm (barely above ambient temp). Nothing is anywhere near freezing up. Even though I now suspect the TXV is okay as it's clear and apparently doing its job, I will replace it. I did watch a YouTube video where the tech was able to remove the TXV from the engine bay side. If true that may save me a fair amount of effort inside the car when I replace the TXV. I think the evaporator is okay as not too much debris could get past the TXV in the event of a compressor failure. Hoping I'll still be able to flush it if I don't end up tearing apart the inside.

Thanks for the tip on the compressor sensor. That too must be okay as I didn't see it shutting down, due to over heating, when I believed it should be running.

I currently don't have the cowl panel between the engine and windshield off, so don't have access to pinch off the heater hose. At the moment I'm going to think the blend door control is okay. After I get the system recovered I'll remove the cowl to access the TXV.

I hooked up my gauge set today. It was 86F outside. The vent only blew 70F.

Here are current readings:
86 degrees F outside
engine at 2000 rpm
compressor on
readings Low 23 psi, High 340 psi
vent 70

When the engine was idling at 800 rpm, low side was 43 and the high side pressure was 230. At 230, the needle on the high side was flickering, not sure what that would indicate. The flickering went away with increased rpm. Perhaps that flickering was due to me adding about an extra ounce or 2 of extra refrigerant. My system only needs 15 ounces of R-134a. I should know how much R-134a is currently in there when they recover it. After turning off the engine it took about 10 minutes for the low & high side pressures to balance out at roughly 125 psi. Before starting the car for this test the pressures were balanced at roughly 110 psi.

I'm currently thinking, based in your feedback, I need a new compressor, TXV, and another receiver/dryer desiccant. If I end up having to disassemble the dash I'll also replace the evaporator. I'll back flush the lines. Since the condenser is new I'll stick with that one. I think I'll be okay based on the lack of debris in the old one.

My Honda Civic needs 5.5 ounces of PAG-46 oil. I'm going to go with fully draining the oil from the old compressor, measuring the amount. Then fully drain the new before adding in the amount removed from the old. AutoZone sells PAG-46 oil that already has dye in it so I'll purchase a bottle of that to use.

Thank you again for sharing your knowledge and insight. I'm hopeful my comments and the updated gauge readings may trigger more helpful thoughts. It may be a couple more weekends before I have time to tackle this project.

--Rick