Gen 3 A/C

Yes, familiar with the ND11 oil (instead of PAG) required for electric AC compressors so need to avoid cross contamination.

Getting an AC gauge set is probably my next move. I could blindly charge up the low side but given the leak evidence, there is no reason to think it will hold. Furthermore, don't know how much oil was lost so probably need to put some oil in as well.

Been reading up on repairs info I could find. thx Mendel Leisk for posting the directions... couldn't figure where this stuff was in my few thousand page Gen3 Factory Service Manual PDF haha. The most mysterious part of this repair seems to be - how does one know the right amount of oil is in the system without completely taking out all system components and either replace or confirm oil quantify left in various component. Sounds like vacuum removal of r134a doesn't remove the ND11 oil and they are mainly in the compressor and condenser. But clearly I have some oil capacity that got blown out on the high pressure line so the oil must travel through (and therefore deposited) the entire system? Seems getting the perfect oil capacity is rather tricky without all new components.

 

You can buy the ac o'rings, a set of gauges and rent a vacuum pump. This is relatively cheap compared to paying 1k+ for someone else to do it. You will still want to have shop recover the refrigerant for legal reasons.

 

Yes, got a AC manifold on order. On repairs, planning on shop recovery followed by my own repairs (fairly skilled). Sounds like recovery will also take out the oil (anyone know if just "some" or completely?) so need to get POE oil (looks like cheaper ND-11 compatible options are available, I know car manufacturers avoid detail specs on their branded oil but we all know they don't make the oil, read a post recovered ND-11 has higher viscosity to compatible aftermarket synthetic, will do more research given how expensive Toyota ND-11 is), find oil and r134a capacity info for recharge.

Thanks for the link on the o-rings. Got a black set but I'll go green since its pretty cheap and maybe HNBR material makes some kind of difference. I have found with the black set the assortment often don't have a good fit to what you need. ring diameter + tube's diameter result in a lot of combinations.

Probably take it slow on this one (to understand all steps and gather tools/parts) so won't have data real soon. But will definitely update as I make progress.

 

I'm impressed! When mine got low, I didn't have the time to research, so I had the dealer check it out. He said he recovered 5.8 oz and put back in 18.24 oz. Not sure what the spec is, but that's what he did and it appears fine over the past 3 months. No record of how much oil he used, though. He also added dye in case it leaks down again.

Good luck with the research and the big job. Sounds like a fun learning experience.

 

Slowly gathering equipment and got my halogen (freon is mostly halogen) tester today. I can detect some leak on the high/low pressure port when I remove the cap. In fact, I hear a small vacuum or pressure sound when unscrewing the pressure port caps (Wondering if other people's car are like this) This was the case a few hours after shutting the car down without AC use. But no more leak at the ports after say 6 hours or so.

Didn't detect any leak at the 2 seals where high and low pressure pipes enters the firewall but didn't run the AC to be safe in case of lacking compressor oil in the system.

Still getting the manifold, waited for one ordered from ebay but seller disappeared haha.

Perhaps the most challenging is to understand how to get the right amount of compressor oil in the system. Toyota direction (post #59, file "... AC compressor removal & install.pdf" page 8 of 11 of this PDF) suggest the system oil capacity is about 4..4-4.9 ounces. Then it says if there are too much oil in the system (for example remaining in the pipes), then the system will be inefficient. Since I seem to have some oil already leaked out, How do I know how much is in my pipes? Evacuating the AC doesn't remove the oil or not all of it from what I read. So how much would you guys put in? Maybe 3oz just to be at least safe in case its empty and not too over filled if not?

Anyhow, given the oil leak pattern, a friend is suspecting the high pressure line o-rings at the firewall which make sense. Oil seeping out of that seal will run along the slightly downward slope away from the firewall. So the sequence of steps are probably 1) evac the system 2) replace the 2 o-rings at firewall 3) evac the system again to remove air+moisture 4) refill r134 with guestimated compressor oil capacity.

 

Pros put in some dye and then come back later with a black light after it's run for a while.

As for how much oil, you can't know what's in there until you get it out. The vacuum pump should get out virtually all the oil, air, and water. There may be a residue of oil that didn't come unstuck but it's so little that it isn't a factor. But, if you have a leak, it won't hold the vacuum and moist air will leak in the same way the coolant leaked out.

So the steps as I understand it would be to:
1. Add some die and top off the freon without oil and run it for a while to find the leak.
2. Evacuate the system and fix the leak.
3. Reattach the vacuum pump and pump it down as far as it will go. Shut off the pump and wait for an hour or more (not sure how long the pros wait) to make sure the vacuum holds.
4. If vacuum holds, charge the system using the right procedures to not introduce air into the system.

 

Thanks, I'm also generally aware of these steps. Logically, there are 2 potential problems to consider.

Step 1 probably require adding some oil to the system just in case its low. Want to avoid the compressor destroying itself without oil while searching for the leak.

Step 2 is a problem for DIYer if trying to get the oil level correct in a system that already lost some oil

- Watching several people use DIY equipment (AC manifold + vacuum pump) on youtube to evacuate the system, there is no recovery of any compressor oil so it appears to be left in the system. From reading various forum posts on AC service, professional AC vacuum equipment will extract the freon and compressor oil and separate them. Please correct me if my understanding is wrong here.

- So using DIY equipment, I will have to guesstimate how much oil is in the system and decide how much to add. If using pro equipment at a pro shop, once they've established vacuum, it seems sensible just letting them recharge + add compressor oil.

It appears guesstimating compressor oil level is the only possibility in my case for a DIYer (some oil already leaked out so what is in the system is unknown)

Any misunderstanding on my part before tackling this?

=====

An interesting update : Cleaned up the oil on the high pressure line, ran the AC and it blows cold... not super cold but cold. Low pressure line is cold. Sight glass shows liquid flowing... not sure what bubbles look like but looks like liquid flowing. No leak showing from my halogen detector at the firewall o-rings. If I expel some gas from the low pressure port, halogen detector goes off (beeps faster) so halogen tool is working.

Foam surrounding the high/low pressure line at the firewall shows a darker shade just below the high pressure line clearly indicating a fluid leak exiting the line at the o-ring in the past. Friend suggest once I increase the pressure in the system, likely will experience leak here again.

The difference btw today's usage and previous usage when I got no cold air is ambient temp. Currently in the 50s and probably in the 80s when I last tried it. PV=nRT suggest 30+F temp difference should yield quite a bit of pressure difference.

Probably the smart thing to do now is one of the following 2 steps

1. Given the strong evidence leak is at high pressure line o-ring at firewall (also make sense high pressure line would be the most likely candidate to develop a leak) Open the high pressure line at firewall and change the o-ring follow by evac air+moisture, recharge r134a, add a couple of oz of hybrid compressor oil and some dye

2. Just to recharge the system with some r134a with hybrid compressor oil + dye (which is step 1 anyways) and see how it behaves.

I'm inclined to go with #1 given the strong evidence of leak at high pressure o-ring. if doing by the book and evac freon at shop for containment, they charge about $200 for evac+recharge and okay with you doing the repair in between. Probably > $100 just for evac only. So it seems like no need to get own equipment if want to do this by the book.

Auto AC shop provided the following info

- get a little bit of oil during evac, not all.
- use a separate oil injector for hybrid to avoid contaminating hybrid AC systems. r-134a injection equipment is common for electric+mechanical pumps and uncontaminated by oil
- add couple of oz of oil during refill. They are also guesstimating since no one really knows how much oil is in a previously leaked system. Basically guesstimating not to much and not too little.

 

I'm no expert, so I can't say which approach is better. When my freon got low, I took it to the Toyota dealer a mile from my house because they've always done good work and I don't have the right tools anyway. If you diy, make sure that the gauges have never been used on systems with non-hybrid oil. I've read that even a tiny amount of contamination will damage it.

I am surprised that they say evacuation doesn't remove the oil since leaks do remove it. Hence the oil mess that often marks the leak. The vacuum pump should boil the oil like what happens to any liquid in a vacuum and most of it should come out. But that's just theory, not actual experience.

 

Finally got around to this job. Just replaced 2 o-rings on my AC system along with full moisture vacuum removal, oil injection and refill. Thought I share play by play of what happened.

Before I start, there are quite a few steps that unavoidably release a little bit of freon into atmosphere when DIY at home so just be forewarned since we Prius drivers seems a bit more of environmental warrior Not sure AC shops do much better. I'll outline where these are in my play by play.

After filling to factory spec (~1lb or 460g), I was surprised I don't even get to 100 psi on the high side (All the youtube videos I watched shows 200+ psi on high side with mechanical compressors. Prius repair manual also specs 199-228 psi) Not sure if my high side is defective or the AC gauge is defective. Low side is within Prius repair manual spec (21-36 psi) Vent blows cold, low pressure line is cold, sight glass has no bubbles when running and some bubbles after shutting off, AC compressor is quiet (was noisy when system was ~empty).



Tools + Supplies (all pictured at bottom)

- Bought an AC oil injector. advancedautoparts.com must have a pricing error. Serious machined piece of equipment with high quality valves for like $5 before 25% off online coupon haha.

- O'Reillys tool loaner for 1.8CFM vacuum pump and AC gauge manifold set (~$300 deposit, refunded with 48 hour return). Folks at this particular O'Reillys were really knowledgeable, They knew to order new rental tools for hybrids to avoid mechanical compressor oil contamination. Another nearby O'Reillys was clueless so YMMV.
- R134A can tap ($7 local O'Reillys auto parts store)
- 3 x 12oz cans of pure R134A ($6.50/can on amazon)
- 1 hybrid AC compressor oil (advancedautoparts.com $18.50 before 25% online coupon)
- 1 halogen detector from ebay ($20)
- 270 piece HNBR Auto AC o-ring kit ($5+ amazon)

Oil Injection

I injected 2oz of hybrid AC oil about a month ago because there was oil residue on the high pressure pipe coming out of the firewall. Wanted to make sure compressor didn't get damaged before repairs. When I called an AC shop to ask how they determine how much oil is in the system, they said they use an injector to inject a couple of oz during recharge. Its an guestimate but better than no oil in the system. Too much oil also makes the system inefficient. Anyhow, their explanation led me to purchase the injector as a necessary tool.

Change suspect leaky high+low pressure pipe o-ring at firewall

I had oil residue on the high pressure pipe at the firewall, I reasoned these o-rings were the most likely leak source. Just about all the R134A already leaked out and system had very low pressure so I just released the remaining system pressure (lasted maybe like 5s so not much freon remained)

Took out the 10mm bolt on where H+L pipe entered the fire wall to release the secure plate. Then just pulled both pipe out. Also removed the 10mm bolt on the low pressure pipe near front of the engine so I can pull the entire L pipe out to change the o-ring. Impossible to do it way back in the firewall with little access and deep recess for o-ring seat. Was able to do the small H pipe's smaller o-ring hear the firewall but took a while with one less dextrous left hand/fingers.

O-rings

As mentioned, the 2 original o-rings looked fine. Was pliable. I decided to put in new o-rings from an auto AC HNBR kit and sized o-ring walls slightly thicker to give it slightly more compression. It was slightly harder to put pipe back in but not too bad so probably ok. See diagram below on how these o-rings are compressed.

The low pressure pipe's connection in front of the engine had exact same o-ring as back in the fire wall. But I was not able to slip on a slightly thicker o-ring here. Clearly, the machined fittings are slightly different here than back in the firewall. Either by design or manufacturing defect. But clearly, some reasonable margins on o-ring thickness and amount of compression is okay. Given how easy the firewall side slipped in and out, I thought thicker o-ring would likely compress better.

Vacuum

Hooked up the AC gauge+manifold with vacuum pump (if brand new, need to fill with compressor oil) to vacuum the moisture out of the system. I first vacuumed just like 1/2 min and stopped. But the vacuum was eventually lost as moisture in the air evaporated and increased the pressure. This is why you have to vacuum for 1/2 to1 hour to remove all the moisture. Did that and vacuum held for entire night (was missing a R134A can tap so couldn’t fill until AM) Most people suggest vacuum 30min and confirm vacuum is held for 30min.

No significant amount of AC compressor oil came out during my vacuum process. The vacuum vent port felt a little oily but no droplets. Perhaps 1.8 CFM not enough to pull the oil out.

These AC gauge set are all made in China with error in directions. Told me to hook the H lines to low gauge vise versa so be careful if you encountered that.

Guy on youtube said avoid opening the AC line ports on the car all the way as they can damage the valve. I had no problem on this car. Just beware.

When done vacuuming, make sure AC gauge H+L valve is closed before recharging as supply line will be bled.

Recharge 1st can

Tapped a 12 ounce can and weight it with the tapping valve. Connected to the AC gauge supply line and open the can valve. Then open the supply line at the AC gauge manifold slightly to bleed out any air in the line so it doesn’t go into your system (a little freon will be release at this step and is allowable according the youtube folks) Opened both high and low valve until pressure gauge needles stopped moving. This basically allow can’s pressure to equalize with car’s AC system. Close the high side valve on the gauge (only want to refill via low port when compressor is running). Start the car and crank AC up to high (temp = LO, fan=max, AC light on). Car’s AC compressor when running suppose to draw in the freon from the L port.

High side pressure started out around 50psi and took forever to suck in the freon. I eventually lost patience and pickup the can and turn it upside down periodically (I read afterwards I'm not suppose to do this as it may put liquid into the compressor input and damage it. Hope thats not the case. ). This allows more liquid freon to enter and is visible in the AC gauge sight glass. Took awhile but eventually the can was empty (shook it and heard nothing inside and gauge pressures seemed fixed) High side is still between 50-75psi at this point.

Close the H+L AC gauge valves (supply line will be opened to move to 2nd can). Close the can tap valve and disconnect the supply line from the can. Some compressed freon in the line will leak out. When removing the tap from the can, what little freon remain in the can will leak out.

At this point, about 12oz (340 grams) are in the system. Engine bay sticker calls for about 460 grams so another 4oz or 115grams from the 2nd bottle.

Recharge 2nd can

Connect the 2nd can and connect to supply line. Open the tap valve and release the supply line at the manifold slightly to bleed out the air again (a little freon leaked out here). Then place the can on a sensitive scale (used my shipping scale that measures oz and grams) Unfortunately the scale won’t be precise as the supply line is attached so just an approximation.

Started the car and put AC on high again and then open the low side valve on the gauge. Turn the can upside down a couple of times until I thought about 100+ grams went in. At this point, high side PSI was 75 and car’s high side AC sight glass showed no bubbles. Closed off the H+L valves on the manifold. Shut off the car/AC. Closed the can tap, and closed the AC gauge valves on car’s AC ports and removed them. Put the car’s AC port’s caps back on and all done.

I think I did everything correctly except for some liquid charging rather than vapor. Anyhow, vent blows cold but we’ll know better by summer time.

 

well done, thanks for the write up!

 

For others reading, that part is worth repeating in bold face. Close the high side valve on the gauge (only want to refill via low port when compressor is running). Then make sure you did it.

There's a reason ... if that valve is open and the compressor starts, high side pressure will flow back to your refrigerant can, which will blow up, possibly taking bits of you with it.

Chances are you lucked out, or you would know by now, but yes, that's luck. Don't do that. When an incompressible fluid gets sucked into a running compressor, something will happen ... and it won't be the fluid compressing.

-Chap

 

Yes, got impatient because not knowing what the normal behavior is. High side might have issues as well causing non normal behaviors.

- Initially charging an empty system. The compressor didn't even come on. I read compressor may be protected under low pressure condition and need to be jumped but didn't research how to do that. It eventually started running. Don't remember the exact sequence but perhaps initial liquid charge effort might have kick started it. I saw some liquid charge and its done with very short bursts (this is what I did). I guess this allows the system to turn freon into gas before hitting the compressor?

- High side pressure started really low and ended with < 100 psi after charging 1 lb of freon. Probably indicative of high side problems (failing compressor, blockage) I suppose with lower than 200psi high side pressure, the cooling efficiency will be low.

- When doing vapor charge, freon can sat on a scale but showed no change. I took this as nothing was getting sucked in. Maybe its due to low high side pressure. Maybe shaking the can side to side rather than upside down (liquid charging) would have helped. Can is connected to the supply line have interference for accurate weight readings. This might have also contributed to errors in charging observations.

Anyhow, not sure what is the proper behavior. How slow should we expect vapor charging the first 12 oz on an empty system? And what psi reading should we expect with after AC compressor starts up with only the initial freon that entered the system through can's own pressure? How fast should one expect the psi to rise?

Finally, I'm starting to think I have a problem with generating high side pressure which contributes to my observation.

 

More update for everyone. Here is the Techstream info on AC system and HVAC settings after running it for a few min.




Some interesting readings to note

- Evaporator Fin Thermister temperature reached the target
- Compressor near 4k RPM
- Regulator Pressure Sensor reads 87 psi (I've seen this just above 90). I believe this is the high side reading.

Vent does blow cold, Low pressure pipe is cold with condensation on it. But AC doesn't have to work too hard in the dead of winter here in the northwest winter.

I put in about 1 lb R134A (spec) but lost some that got trapped in the AC manifold gauge's high side line.

Most PSI charts for R134A on high and low side does show probably closer to 100psi at 45F ambient temp. So perhaps 90psi shows I'm just a little bit low on freon. R-134a System Pressure Chart - AC Pro<?php/* bloginfo('name'); */?>

What does everyone else get on techstream at max cooling? In particular, regulator pressure sensor.

 

Willing to bet @lech auto air conditionin could answer the posted questions in this thread.

 

Yes your readings on the high side sensor are correct for a cold ambient temperature you can have well below 100 psi on the high side while you’re driving down the road on a cold morning.

I believe they released a video on Prius chat also on my YouTube channel of me driving my Prius with remote Bluetooth temperature sensors in pressure gauge is showing extremely low pressures on the high side while you’re in real world driving down the street. That is normal.

Real world driving with 25 35 mph air blowing against the condenser is vastly different than a car sitting in your garage with the hood up sitting still.

Under low load conditions like the dead of winter you can nearly be half empty on refrigerant and still get condensate on your suction line and cold temperatures out your dash.