DIY Home Improvement Forum banner
1 - 20 of 50 Posts

Bigplanz

· Registered
Joined
·
1,769 Posts
Discussion starter · #1 · (Edited)
Since it was 80 degrees here yesterday, I was reminded my wife's Windstar needs the AC fixed. Me, or somebody, is going get the AC working!

Here is a video. Compressor will come on if I jump the low pressure switch. It has a leak somewhere, I guess, but I will add some refrigerent to see if the AC briefly comes back to life.

 
If you are sure there is a leak best approach would have been to apply vacuum to 29.9 mercury and watch guages. It they loose vac within 15 minutes then brush on soapy water at connections. Why bother to add refrigerant at this point. Let me know what happens.

Bewaare of a little known source of leaks- the black low pressure line. It could seep gas and diffficult to verify.
 
Discussion starter · #3 ·
If you are sure there is a leak best approach would have been to apply vacuum to 29.9 mercury and watch guages. It they loose vac within 15 minutes then brush on soapy water at connections. Why bother to add refrigerant at this point. Let me know what happens.

Bewaare of a little known source of leaks- the black low pressure line. It could seep gas and diffficult to verify.
I would pull a vacuum on it, if i had a vacuum pump. :) Since it snowed yesterday, I hit "pause" on the AC troubleshooting.
 
2000 Windstar is an 18 year old vehicle. Check the Schrader valves for leaks. But 9 times outa 10 it's the evaporator. See if you can put the green die in it. You may watch the compressor turn green.:vs_cool:
 
Discussion starter · #6 ·
Understand, Bigplanz. Hey, got a compressor? You can buy a $10 vac venture at HF. That's what I use! No need for a $250 vac pump!

Let me know how you progress as I have been around auto ac for a while......
I thought about that venturi vac. My compressor is too small, though. 3 gallon pancake. Might get a vac pump from harbor freight with a 20% coupon. Should be about $80.
 
No sense in vacuuming the system until you find your leak. If you vac the system and then spray something on it to see if it leaking, you'll draw it into the system. Charge the system first then spray soapy water on the connections, compressor, condenser, lines, and shrader valves where you hook up to charge it. Looks like a tire valve. The only way to check the evaporator without removing it is to test it with a A/C leak detector. Fords are bad about leaky evaporators.:vs_cool:
 
Discussion starter · #8 ·
No sense in vacuuming the system until you find your leak. If you vac the system and then spray something on it to see if it leaking, you'll draw it into the system. Charge the system first then spray soapy water on the connections, compressor, condenser, lines, and shrader valves where you hook up to charge it. Looks like a tire valve. The only way to check the evaporator without removing it is to test it with a A/C leak detector. Fords are bad about leaky evaporators.:vs_cool:
I am planning on doing just that. I am going to get a can of r-134a with dye in it, put it (and enough r-134a to get the system pressure up enough for the compressor to come on) and run the AC for enough time to distribute the dye. Then check with UV light and the funny amber glasses.

If the evaporator is leaking, yikes! whole dash has to come out! If it's a line or seal, eh, I think I can fix that. Compressor, eh, maybe? It's the wife's car so I consider it, um, a test platform!

Still cold here, so this isn't pressing. Plenty of time to scour Amazon for what I need.

I read sniffers aren't very reliable for AC leaks. Opinions? I found them on amazon for about $30-$40.
 
Discussion starter · #9 ·
It was sunny and 80 yesterday so I went to AZ and got a can of uv dye shots, a kit with the yellow glasses and a uv penlight and another 12 oz can of r-134a. Got sidetracked so I didn't get a chance to charge the system. Now it 55 and raining.

I assume only one shot of dye is needed, correct?
 

Attachments

If there's no gas in the system, don't add refrigerant and run it with air/moisture.



If you are sure there is a leak best approach would have been to apply vacuum to 29.9 mercury and watch guages. It they loose vac within 15 minutes then brush on soapy water at connections. Why bother to add refrigerant at this point. Let me know what happens.

Beware of a little known source of leaks- the black low pressure line. It could seep gas and diffficult to verify.
If automotive a/c system is anything like other refrigeration systems, the dumbest thing you can to do a system with a leak is pull a vacuum. Vacuums pull moisture in.

For non-automotive they usually do leak testing with nitrogen if there's no gas in the system.

If there is gas but it's low, could use a leak detector. or if it's flat, put a tracer gas plus nitrogen and use a leak detector.

The vacuum is only pulled to remove air/moisture from a leak free system before charging it up.
 
Discussion starter · #11 ·
The results of today's testing! Opinions, please!

 
The results of today's testing! Opinions, please!

https://youtu.be/u-rLsxqeA44
Big,

In my opinion, suspect a bad compressor if high side psi is way low. You didnt say if you indeed had any even slightly cool air ion the cab, but I assume you have absolutely no cool air. be happy if indeed it ends up being a bad comp and not a bad expansion valve, requiring yanking the dash. Just did such a job. Sheesh.
 
If there's no gas in the system, don't add refrigerant and run it with air/moisture.





If automotive a/c system is anything like other refrigeration systems, the dumbest thing you can to do a system with a leak is pull a vacuum. Vacuums pull moisture in.

For non-automotive they usually do leak testing with nitrogen if there's no gas in the system.

If there is gas but it's low, could use a leak detector. or if it's flat, put a tracer gas plus nitrogen and use a leak detector.

The vacuum is only pulled to remove air/moisture from a leak free system before charging it up.
Fearmongering. Ive done the soap bubbles test many times and my ac's have not exploded/quit/siezed/rusted/fell off the car.

remember- it's SOAP. The water content on a few soap bubbles is so insignificant- they are mostly air. And the drier will absorb up to 10 mls of water. Also, youre vacuuming the system so the moisture will boil off and be expelled by the vac pump/venturi.

Nitrogen? Yes, it is an "ideal" idea but it is so impractical to get a hold of some comprsessed N gas, which is under extreme psi. You need special fittings, etc. Not worth the hassle.

And a vac is not used ONLY to pull a vac from a leak free system. It is also used to VERIFY there is no leak by watching the guages' needles over a few hours.

Dont be afraid, Bigplanz.
 
Discussion starter · #14 ·
I ain't a skeert. :)

The air coming out wasn't cold. The compressor was on, the pressure on both sides was the same. Ergo, compressor wasn't compressing. I will replace the compressor, condenser, drier and orifice tube. I will buy a vacuum pump, pull a vacuum, recharge and see if it works. I ain't skeert.

If it don't work, i try something else!

This will be on video.
 
Fearmongering. Ive done the soap bubbles test many times and my ac's have not exploded/quit/siezed/rusted/fell off the car.

remember- it's SOAP. The water content on a few soap bubbles is so insignificant- they are mostly air. And the drier will absorb up to 10 mls of water. Also, youre vacuuming the system so the moisture will boil off and be expelled by the vac pump/venturi.

Nitrogen? Yes, it is an "ideal" idea but it is so impractical to get a hold of some comprsessed N gas, which is under extreme psi. You need special fittings, etc. Not worth the hassle.

And a vac is not used ONLY to pull a vac from a leak free system. It is also used to VERIFY there is no leak by watching the guages' needles over a few hours.

Dont be afraid, Bigplanz.
A vacuum is only 14.x psi below atmospheric so finding a leak with a vacuum in theory is less effective than with nitrogen where it can be like 150+ psig above atmospheric. In fairness, a micron guage would be way more accurate and pick up subtle changes that would never show on manifold gauges.

Yes, if there's a leak and a vacuum is pulled, air/moisture will be pulled in so it doesn't sound like an optimal tool to find leaks primarily.

I'm in no position to say that the practice is wrong, but it sounds like automotive work is held to a lower standard than ac or refrigeration.

Refrigerant and parts seem to be sold to anyone in the us and the average diy'er won't charge by weight and probably doesn't know the theory.
 
Fearmongering. Ive done the soap bubbles test many times and my ac's have not exploded/quit/siezed/rusted/fell off the car.

remember- it's SOAP. The water content on a few soap bubbles is so insignificant- they are mostly air. And the drier will absorb up to 10 mls of water. Also, youre vacuuming the system so the moisture will boil off and be expelled by the vac pump/venturi.

Nitrogen? Yes, it is an "ideal" idea but it is so impractical to get a hold of some comprsessed N gas, which is under extreme psi. You need special fittings, etc. Not worth the hassle.

And a vac is not used ONLY to pull a vac from a leak free system. It is also used to VERIFY there is no leak by watching the guages' needles over a few hours.

Dont be afraid, Bigplanz.
Please don't try to pull a vacuum and find the leak using soapy water. That soapy water won't blow any bubbles when the system is in a vacuum.

Please don't try to use a venturi device to pull enough vacuum to do anything useful to a refrigeration system.

You can find a leak by adding some dye, there are cans that include a few ounces of refrigerant and stop leak for around $8. The stop leak will only be effective for a shrunken o-ring, not an old condenser with pinholes.

You can find even very small leaks with a Halogen leak detector. The cheapest are over $30 on Amazon, over 70 at harbor freight.

If its been at zero or negative pressure, and if it was below the pressure cutoff on a warm day, it probably has been at negative pressure in very cold weather, you need to replace the filter dryer. The prices vary, but it would be difficult to find one for less than $12. The one in there will be saturated and then they can fail (when they can't remove all the water the screen that hold the desiccant in place can corrode) and release it. That could keep you busy cleaning out the expansion valve and flushing the rest the system.

You do need to use a vacuum pump to get rid of air and dry out the system. They run $100 at Harbor Freight now, but just over $50 on Amazon. The fill and vent method to purge trapped air has been unlawful for decades, not to mention expensive, and it was never as effective at removing air, it was all but useless for drying it out, and it's not much help finding leaks either. That said, some joint and o-ring leaks are pressure/vacuum sensitive and a vacuum test can fail to locate a pressure leak and a leak under vacuum be caused by o-rings that don't leak with positive pressure. Most (all) Schraeder valves will leak when you pull a vacuum, make sure any ports you have capped have good caps tightened down.

After you find and fix the leak, then its time to refill the system. If you can't find it, try patience. It can take a while for a slow leak to spit out enough dye to find it.

When you refill the system, don't use any of the crackpot junk that says its better than R134a. Yes, 134a is a blend only a bureaucrat could love, but tossing in a few percent propane (that's the secret ingredient, and the secret is its cheaper) doesn't really work, its an OK working fluid at 2x higher pressures than a 134a system uses, albeit flammable. Be sure to shake the can and turn it upside down and let it shoot it in as a liquid. If you let it boil in the can and go in as a gas the blend fractions itself in the can and you get the more volatile part until the halfway mark and you get lousy cooling. Try to run in as much of the charge weight as you can as liquid into the high pressure side with the car and A/C off, then wait a couple minutes before starting the car and running the A/C and finishing with more liquid (slowly) into the low pressure side. You won't be able to weigh the charge, but the label on the car will show a weight and you can figure it as so many full cans and then a fraction of a can. Try to get the fraction as close as you can manage. Look at the sweat/frost level, use a non contact thermometer to find the liquid level, shake it and guess, whatever works. You can practice on the cans you run in entirely.
 
When you refill the system, don't use any of the crackpot junk that says its better than R134a. Yes, 134a is a blend only a bureaucrat could love
134a is not a blend unlike most ozone friendly refrigerants like 407C and 410A.

So should be able to take it from the can as a gas without any issues.
 
134a is not a blend unlike most ozone friendly refrigerants like 407C and 410A.

So should be able to take it from the can as a gas without any issues.
Yes, 134a is not an azeotropic blend, I'm not sure why I cross wired that thought.

What I failed to say was that if it's lost the entire refrigerant charge it will have lost some compressor oil with it and it would be safer to use one of the automobile refills with a few oz of the correct oil for the system as the 1st can, hence the shaking and allowing time to let the oil in the 1st can move through the expansion valve. Its possible this will cause a problem with too much oil moving around, but its not likely and curable but the MTTF of an unlubricated compressor is roughly the time it takes to put all the tools away and get stuck in traffic.
 
Discussion starter · #19 ·
It still showed about 20 psi on the system before i filled it. Before it started raining, i got my yellow glasses and UV lights and saw green dye on the underside of the compressor, a tiny amount on the hi and low fittings and little green flecks on the condenser. I will take some pictures.
 
It still showed about 20 psi on the system before i filled it. Before it started raining, i got my yellow glasses and UV lights and saw green dye on the underside of the compressor, a tiny amount on the hi and low fittings and little green flecks on the condenser. I will take some pictures.
I’m impressed you’re attempting to photograph the black light fluorescence. I’d have assumed that needs special filters.

From the description I’d guess the fittings show dye from when the gauges are attached and removed.

The condenser isn’t certain, some bug splat and other debris fluoresces. If there’s also or exclusively fluorescence that’s not the exact same color as the dye, or if you don’t see anything on the fan blades and fairing, then it’s very unlikely to be leaking at all. If you had seen patches or spots rather than flecks, or there was dye on the fan or fairing, I’d suspect the condenser was losing refrigerant.

Assuming the compressor isn’t just showing material that was ejected from the condenser and landed on it (the fan etc would also show it), then the questions are;
1) Are the line fittings leaking at the compressor? The car’s old enough for the o-rings to have failed.
2) Is the leakage from the shaft? It’s possible that the shaft seals leak at zero pressure and it’s meaningless, but it could be a pressure leak.

The 20psi residual pressure you found is well below the vapor pressure of R134a refrigerant at any temperature most people think of their A/C. There may be a small residue of refrigerant dissolved in the compressor oil that’s a source for a dwindling leak. You may be measuring the vapor pressure of a minor constituent or contaminant - a tiny amount of solvent added to improve solubility of PAG oil to improve oil return, or an oil intended to swell o-rings in a leak-stop.

If the charge was lost at the compressor fitting or shaft definintely find out how to get the oil level in the compressor right. There’s always a way, even it it’s flush it out with compressed gas and add X ounces. As mentioned before, change the filter dryer.

If you don’t see anything elsewhere on the lines, and nothing where the condensation drips, then I’d replace the o-rings in the line set connections, and while it’s at zero pressure I’d change every other line set connection o-ring I could reach without taking the dashboard apart. Then I’d clean up the compressor, and every other connection and then add just enough gas to make it leak if it plans to, say 40psi. You can use R134a, Nitrogen, CO2, Argon, almost anything is fine that’s lawful to use as long as it’s dry and oxygen free and you remove all of it before you refill the system. I wouldn’t use propane though, whatever other considerations apply, the job is enough to to without smell of mercaptan.

If you don’t see any more leaking, then pull a vacuum and dry the system out. If it’s not leaking you may need to run the pump for an hour or 4 to dry it or trapped water and refrigerant dissolved in oil everywhere in the system will vaporize and look like a leak. If you can close it off and it’s still in a deep vacuum the next day, fill it , look for leaks, test it, and look for leaks one last time.
 
1 - 20 of 50 Posts