Hi all. I have a 12000 btu Mitsubishi minisplit that wasn't working. I went to reclaim refrigerant and got nothing out. Recovery machine sprayed a little r410-a when I was purging the hoses, but didn't add any weight to the tank. The pressure of the system was slightly negative afterwards. I added nitrogen to 545.5 psi to check for a leak and didn't find any with bubbles. It's been holding for about an hour. I was thinking maybe it was the Schrader valve? I already removed it. Any helpful insight or advice would be appreciated. Thanks.
Minisplit leak
- Thread starter Kingtut
- Start date
Look for traces of oil, did you inspect and bubble test the flare fittings at both ends of the lineset?
Everyone says they can make a great flare everytime but in reality many of them cannot. I once redid every flare on a 3 head system and not a single one passed my go nogo gauge.
Everyone says they can make a great flare everytime but in reality many of them cannot. I once redid every flare on a 3 head system and not a single one passed my go nogo gauge.
Do you have a good micron gauge for monitoring it when you pull a vacuum? I use pressure and vacuum for tough leaks, in my limited experience sometimes a micro-leak will only show with one or the other. Also a sniffer to look for leaks when you refill it? No disrespect meant to GJ peeps or you but there are probably better places to get hvac advice, like a dedicated hvac forum.
Thanks for the responses. I bubble tested the flares outside but not inside. I haven't taken them apart to look at though. There was no oil that I could feel around the fittings but they were tarnished in a way that I wasn't sure about - maybe oil from months ago? (the unit hasn't been working for close to a year) I have a sniffer coming so I'll definitely check again with that later on. I have a cps micron gauge to check my vacuum with too. Thanks again.
Airforce_NavyDAD
Active member
Pull it down to at least 500 microns. Close your Vac valve and watch your microns for (10) minutes. If she doesn't rise above 1200 you are generally ok.
I hesitate to inspect the flares because I don't want to mess with them if they are not the problem. The pressure seemed to have held overnight. I have it written down at home but I think the pressure only dropped around 30 psi with a temperature decrease of about 20 degrees overnight. Hopefully when I get home later I'll see the pressure back up to what it was. Would it be rare for it to be a bad schrader valve?
pcmeiners
Well-known member
If in doubt of a flare leak....these will seal with <1/3 the recommended torque @500 psi or low micron level; I tested. Cheep insurance if you doubt your flare skills, hell of a lot cheaper than a few ounces of gas and the aggravation involved.
www.flareseal.com
As to possible Shrader valve leakage, present or future, use blue or 554 Loctite on the king valve brass valve caps.
FlareSeal - Leak-Free HVAC Flare Connections for Mini Splits & Refrigeration
Prevent refrigerant leaks with FlareSeal - the permanent, serviceable flare connection solution for HVAC mini splits, refrigeration systems, and more. Easy clip-on installation.
As to possible Shrader valve leakage, present or future, use blue or 554 Loctite on the king valve brass valve caps.
SVibs
Well-known member
This is a handy Nitrogen Pressure Calculator. Per the calculator, if you started with a nitrogen pressure of 519 PSI at 70 degrees, at 80 degrees the pressure should have increased to 529.08 PSI.
pcmeiners
Well-known member
"If I can't find any leaks anywhere, would you recommend undoing the flares and installing the flareseals anyway?"
If your not positive your flares were not done 100% correct. These seals are coated with a thick Loctite type sealant and will seal the crappiest connection from possible leaks now and the future, you call for a few bucks per seal.
As to Nylog, do you want a thicken oil or a sealant on your threads?
If your not positive your flares were not done 100% correct. These seals are coated with a thick Loctite type sealant and will seal the crappiest connection from possible leaks now and the future, you call for a few bucks per seal.
As to Nylog, do you want a thicken oil or a sealant on your threads?
jskicet
Well-known member
I just had my mini split repaired by my local A/C guy. He found several leaks from original installation just 5 years ago. He asked me if I wanted the new fittings installed that would replace the older style. I said yep! He had a battery powered crimping tool that he used to install the new fittings that I believe were double oringed to boot. It's been rocking and rolling ever since. I had leaking fittings 3 times. Hopefully no more. It something like this. https://www.grainger.com/product/STREAMLINE-Refrigeration-Press-Fitting-787WN9
Thanks. I'm definitely going to do the Flare seal if I open up the connections - I got them today.
I'm having trouble finding the leak though. I've been using big Blu micro leak detecting fluid but I can't seem to apply it without getting tons of suds to begin with. I'm using the version with the dauber.
I'm also noticing a lot of green on my copper lines. I just had a friend over who said that a tech told him that units with white lineset (like mine) often get pinholes in the copper tubing from the mix of oxygen, moisture and acid from the white lineset. Does anyone know how common this is? I noticed green corrosion around a crack in my insulation outside and also in the tubing coming from my indoor unit before it meets the Flare connection. Thanks again for the help.
I'm having trouble finding the leak though. I've been using big Blu micro leak detecting fluid but I can't seem to apply it without getting tons of suds to begin with. I'm using the version with the dauber.
I'm also noticing a lot of green on my copper lines. I just had a friend over who said that a tech told him that units with white lineset (like mine) often get pinholes in the copper tubing from the mix of oxygen, moisture and acid from the white lineset. Does anyone know how common this is? I noticed green corrosion around a crack in my insulation outside and also in the tubing coming from my indoor unit before it meets the Flare connection. Thanks again for the help.
SVibs
Well-known member
It's called formicary corrosion and it does appear to be much more common with the white line set insulation used by most mini split manufacturers. Have a look at this video for more info.
pcmeiners
Well-known member
"that units with white lineset (like mine) often get pinholes in the copper tubing from the mix of oxygen, moisture and acid from the white lineset."
Unless copper tubing is poorly manufactured your not going to get "formicary corrosion". For oxygen, moisture and acid from an external source to damage properly manufactured copper tubing would takes years and eat at the copper until it was thin. Formicary corrosion is from forming impure/contaminated copper tubing; the cracks are deep in the impure copper when the tubing is created in the first place, so a weak acid and a little water, and a little oxygen is all that is needed to corrode the copper within the imperfections of the tubing; any copper when immersed in acid is eaten away evenly, acid does not form cracks or deep impressions in pure copper. The idea of "formicary corrosion" was created by suppliers of cheap, poorly manufactured tubing...the cheapest **** they can buy from the third world.
The green color on copper tubing is the formation of copper salts. It could be caused by number of things. As mention possibly from the lineset covering, or most likely the solution used for bubble testing, I have discoloration near the connections from the solution I used; again, unless the tube was poorly manufactured that will not be an issue. You can buy bubble solution which does not produce any copper salt discoloration, which is recommended. Hard to tell if you do have a sub grade of tubing unless you do get pin hole leaks with pressure testing. You can clean off the corrosion near the flare connections with Scotch-Brite, wipe off any dust created with a paper towel and alcohol.
As to leaks, bubble solution only works for very accessible areas, otherwise it is difficult to see leak bubbles. A good gas detector is needed otherwise.
Unless copper tubing is poorly manufactured your not going to get "formicary corrosion". For oxygen, moisture and acid from an external source to damage properly manufactured copper tubing would takes years and eat at the copper until it was thin. Formicary corrosion is from forming impure/contaminated copper tubing; the cracks are deep in the impure copper when the tubing is created in the first place, so a weak acid and a little water, and a little oxygen is all that is needed to corrode the copper within the imperfections of the tubing; any copper when immersed in acid is eaten away evenly, acid does not form cracks or deep impressions in pure copper. The idea of "formicary corrosion" was created by suppliers of cheap, poorly manufactured tubing...the cheapest **** they can buy from the third world.
The green color on copper tubing is the formation of copper salts. It could be caused by number of things. As mention possibly from the lineset covering, or most likely the solution used for bubble testing, I have discoloration near the connections from the solution I used; again, unless the tube was poorly manufactured that will not be an issue. You can buy bubble solution which does not produce any copper salt discoloration, which is recommended. Hard to tell if you do have a sub grade of tubing unless you do get pin hole leaks with pressure testing. You can clean off the corrosion near the flare connections with Scotch-Brite, wipe off any dust created with a paper towel and alcohol.
As to leaks, bubble solution only works for very accessible areas, otherwise it is difficult to see leak bubbles. A good gas detector is needed otherwise.
Last edited:
housewolf
Well-known member
Not to hijack the thread, but comments on UV dye?
I have a system I cannot find the leak in & im pretty experienced in finding leaks in piping systems. I’ve resigned myself to just topping it off as needed for the remainder of the summer but I’d really like to locate & repair this leak in the fall. I have a sniffer & have found the general area but can’t get anything to bubble.
I have a system I cannot find the leak in & im pretty experienced in finding leaks in piping systems. I’ve resigned myself to just topping it off as needed for the remainder of the summer but I’d really like to locate & repair this leak in the fall. I have a sniffer & have found the general area but can’t get anything to bubble.
After getting rid of the gas and go tech who did not want to look for a leak. I spent far too much time with a HVAC bubble leak detector, nothing. Looked the situation over and bought a Accutrac ultrasonic leak detector. Found a leak in the evaporator coil in a spot I never would have found with bubbles. 4 year old Daikin evaporator coil with microscopic holes in it. Shameful?
I have been lead to believe that die is a last resort for an old system, but what do I know.
I have been lead to believe that die is a last resort for an old system, but what do I know.
Jim greengo
Well-known member
Or just replace the Schrader valves while you're in there.If in doubt of a flare leak....these will seal with <1/3 the recommended torque @500 psi or low micron level; I tested. Cheep insurance if you doubt your flare skills, hell of a lot cheaper than a few ounces of gas and the aggravation involved.
FlareSeal - Leak-Free HVAC Flare Connections for Mini Splits & Refrigeration
Prevent refrigerant leaks with FlareSeal - the permanent, serviceable flare connection solution for HVAC mini splits, refrigeration systems, and more. Easy clip-on installation.
As to possible Shrader valve leakage, present or future, use blue or 554 Loctite on the king valve brass valve caps.
I wound up taking the interior unit off the wall to test everything seperately. Right now the small line is holding 465.0 psig for a half hour. Yesterday, the 3/8 line dropped only 4 psig over 24 hours, but the lines run through the attic, under the insulation and it was far less sunny today and at least 5 degrees cooler. I tested the heat exchanger inside and it dropped 5 psig in 80min. My cheapo sniffer also beeped (some of the time) when I put it over the drain pan. I still haven't gotten anything to bubble though. Is it realistic to find the leak if it is in the middle of the coil? At this point I'm thinking my only path forward is to replace the heat exchanger - I've already installed the flare seals on the outdoor flares and I'll do the same on the indoor ones when I replace the exchanger. Any advice appreciated. Thanks again.
Last edited:
American Locomotive
Well-known member
How are you testing only one line at a time?
American Locomotive
Well-known member
I would double check the coil then. Isolate both sides, and pressure test. Remember, the leak detector needs at least some refrigerant in the coil to work. It generally won't detect nitrogen (unless it's an ultrasonic one)
Yeah. I have the coil under pressure still and I recorded the last reading before taking off the gauge. I was losing 1 psig when I removed it from the core removal tool, so I know what the reading should be when I reattach. I have the gauge on the 3/8 line right now - I'm going to test overnight. Are you recommending I redo all the connections on the coil and repressurize to test again?
I realize that I need refrigerant for the sniffer to work, but I thought it might be able to detect residue from an earlier leak? It definitely was beeping when I checked the drain pan several different times.
I realize that I need refrigerant for the sniffer to work, but I thought it might be able to detect residue from an earlier leak? It definitely was beeping when I checked the drain pan several different times.
American Locomotive
Well-known member
I'm saying I would pressure test only the coil, if possible. You will need to get some gas into it somehow for your leak detector to work.
OK. Sorry. That's what I meant to say before. I pressure tested the coil by itself and the pressure dropped 5 psig in 80 min. It was inside at stable temp and the nitrogen was loaded into it the day before, so the temperature shouldn't be any factor. So far it's the only thing showing a possible leak - although it did hold a vacuum pretty well.
Thanks for all the help. I was able to get a new air exchanger and get it all hooked up and installed. Everything is holding pressure great and it held a vacuum down to 200 microns hardly raising at all in 15 minutes. I double vacuumed with nitrogen between and then weighed in refrigerant.
Here's the problem - I removed the valve core and have been trying to reinstall it. I've gotten it to screw in and I've tightened it as much as I could with my hands, I even added a little light tightening with a wrench one time, but I still feel pressure on the vcrt. Even after I close the valve, open the back of the vcrt to let out any pressure there, and reopen the valve, I feel the same pressure as I did before reinstalling the valve. Anyone have any idea why this might be? Thanks again.
Here's the problem - I removed the valve core and have been trying to reinstall it. I've gotten it to screw in and I've tightened it as much as I could with my hands, I even added a little light tightening with a wrench one time, but I still feel pressure on the vcrt. Even after I close the valve, open the back of the vcrt to let out any pressure there, and reopen the valve, I feel the same pressure as I did before reinstalling the valve. Anyone have any idea why this might be? Thanks again.
pcmeiners
Well-known member
"I even added a little light tightening with a wrench one time,"
Not something you want to do, even applying a lot of force just using your fingers is not a good idea. Always have spare Schrader valves on hand. It is a good idea to use blue loctite on the brass cap(s) threads.
Not something you want to do, even applying a lot of force just using your fingers is not a good idea. Always have spare Schrader valves on hand. It is a good idea to use blue loctite on the brass cap(s) threads.
Thanks. It felt like a little bit of a high risk manouver. I just couldn't figure out what was going wrong and got desperate. I had a little bag of valves and kept trying new ones, not realizing that they were probably all working fine. Anyway, the minisplit seems to be working well now. Thanks for all the help, I appreciate it!
scooterbum46
Well-known member
I don't do HVAC, although I was responsible for mainframe computer rooms that live or die by the cooling. I learned a lot in those days, but I gotta tell you, I learned a couple or three new things with this thread. Thanks for taking the time to follow up. 