Sprung a leak when pulling a deep vacuum - where to look?

[aump945]

Hi all,

I'm a first-time AC DIYer trying to fix the AC on my 2011 Prius. Thanks for the forum, and I hope to add a positive contribution.

Short version: I detected a condenser leak. I replaced the condenser with an OEM Denso unit. I also injected 40ml of ND-11 oil. I also replaced the schrader valves and torqued them to spec. I pressure tested to 150 PSI with dry nitrogen and did not detect any leaks.

I pulled a vacuum overnight to try to hit <100 microns as recommended by some experts. I woke up to see the image below. As you can see, the vacuum initially went great, reaching 72 microns (measured at the high side with the vacuum on the low side). However, at that point, the system suddenly began to lose vacuum, going up to about 150 microns before I stopped the process to think about the next step. Subsequent vacuum attempts will not go lower than about 250 microns, even after leaving the vacuum pump on for over 24 hours.

IMG_4855.PNG (82.76 KiB) Viewed 416 times

My inexpert guess is that the deep vacuum opened up a weak link in the system and I now have a new leak. I tried replacing all the o-rings (with Nylog) in front of the firewall (but not the ones at the firewall), but I still can't pull below about 250 microns. Also, the vacuum plot now shows a constant ping-pong / up-down / sawtooth micron reading instead of a smooth line (image attached). I think this is also evidence of a tiny leak that constantly opens and closes in response to the vacuum pulling on it. My guess is that this is either a leaking o-ring that I didn't replace (i.e., at the firewall/evaporator), or a pinhole with refrigerant oil covering it that continuously creates a bubble that pops and reforms.

IMG_4889.PNG (73.82 KiB) Viewed 416 times

Although my nitrogen pressure tests do not indicate a leak, a 30-minute vacuum leak test strongly suggests a leak, as evidenced by the constant rise in the attached image. The attached image is with the micron gauge on the low side, near the vacuum, so that is why the vacuum appears to have gone so low. For some reason, I don't get the sawtooth pattern when pulling a vacuum this way.

Where are the most likely places I should be looking for evidence of a leak now?

IMG_4905.PNG (145.92 KiB) Viewed 416 times

[JohnHere]

No, I don't think your system has a leak at all. I believe what you're seeing is outgassing, which is no cause for concern. If you leave the system under vacuum for an hour or so, chances are the micron gauge reading will stabilize and level off. Realize, too, that one micron is an extremely small measure of vacuum, hence the term 'micron.'

For a first-timer, you seem to know something about micron gauges. But you should be aware that readings as low as 500 microns are excellent.

Does the needle on your Manifold Gauge Set, low-side gauge, move at all after leaving the system under vacuum for an hour, staying at about 29 InHg? If it doesn't move, then I'd say that the system isn't leaking vacuum.

Moreover, you tested the system under 150 PSI of nitrogen without any leaks showing up. So I'd say that you're ready to evacuate the system one final time and recharge it (by weight, not pressures).

The Receiver/Dryer element is integral with the condenser, so you had to have changed that as well.

[aump945]

Thanks for your reply, John

Your suggestion that I am observing outgassing is a good one, and that is what I first thought also. But after many nitrogen purges and several 24 hour+ vacuums, I am fairly confident (for a beginner) in saying that outgassing at this point is likely to be extremely minimal. I'm quite confident in saying this because my initial digital gauge readouts definitely showed outgassing and took a long time to make progress, but eventually I was able to pull a vacuum and have it flatline at around 500 microns. That was before replacing the leaky condenser.

I know that a reading of 500 microns is generally regarded as sufficient, but based on the amount of moisture still in the lines at 500 microns, I strongly agree with Jim Bergman of Measure Quick when he insists that 500 microns is too high, and that a target of 100 microns was the original industry standard based on laboratory benchmarks and should therefore continue to be 100 microns:

https://youtu.be/sczq-I1Nfjk?si=RLqwGufSGH5DWC5b&t=194

The last image (#3) in my first post above shows the result of the vacuum leak test. With the micron gauge on the low side port, it begins at about 100 microns but rises to nearly 500 microns in 30 minutes. The shape of the rise is linear, indicating a leak as visually explained by Jim Bergman here:

https://youtu.be/sczq-I1Nfjk?si=rrVNxsyUqX3sTUFK&t=684

If I had outgassing / moisture in the system, I would see a curvilinear rise, not linear. A leak-free system should have no appreciable rise, i.e., a flat line.

Supposing that my logic above is correct and that I do have a leak, what are the most likely places for a leak to occur? Since this is my first time doing this, I don't have the practical experience of knowing which leak points are most vs. least likely. I'm dreading the thought of a leaking evaporator because this is a 14-hour job (as per Toyota) to remove the entire dashboard.

Thanks again for your help!

[JohnHere]

Watching part of the videos, it appears that the speaker is discussing HVAC for homes and businesses, not Mobile Vehicle Air Conditioning (MVAC), which are two very different disciplines. I'm not qualified to comment on the former, but I can and frequently do offer suggestions through this Forum concerning MVAC issues.

To simplify the process, allow me to propose the following: Evacuate and recharge the system on your 2011 Prius and see whether it holds. If it does, great! If not, then it's time to break out the electronic detector ("sniffer") and start scanning the system for leaks, which could be anywhere.

That said, some areas to leak-check include the compressor, its seams, and shaft seal; metal lines routed down low, particularly those having any kind of insulation wrapped around them that can hide corrosion, damage, and pinholes; defective hoses and those that might be subject to chafing; air coming from the dash vents; the evaporator drain(s), looking for signs of UV dye in the condensate; all o-rings, and "peanut fittings" sealed with o-rings, verifying that the correct sizes and type of o-rings are used; service port Schrader valves and caps; and the area around the TXV, if accessible.

The specifications that I have for your car are as follows: 16.5 ounces net weight of R-134a, and 4.4-4.9 fluid ounces of Hybrid Oil. If you have an under-hood decal with specs that differ, the decal takes precedence.