Leaky Minisplit Heads
Hey all,
Has anyone had any quality issues with Fujitsu units? Trying to determine if this is a known issue, just really bad luck, or something else going on specific to my situation.
I have three Halcyon units that are only 4 or 5 years old. For a couple of years, my upstairs unit (a slim duct) has developed a very slow refrigerant leak. A friend of mine moonlights with HVAC and he would come out, check for leaks, would not be able to find anything, pump me up with refrigerant again and things would be fine for 6 months or so, when it would be out of refrigerant again. This past winter, one of my downstairs ductless units started doing the same thing, but this leak was quicker to drain the refrigerant. Finally pumped the lines up with nitrogena and found that on both units there were leaks in the indoor coils. For the ductless unit, there was a big hole in one of the copper coils (where it loops out and then back in). For the slim duct, all up and down both sides, where the copper loops are, there were leaks everywhere on both sides.
What could be going on here?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Open cell spray foam in the building?
I have some closed cell foam spray (Great Stuff)...but I can get some open cell if that is all it takes to fix it!
Walter, are you suggesting that off gasses form open cell can corrode hvac coils?
Oh, I'm glad you said something. I assumed he was suggesting I spray some on the leaks to seal it up.
If he was asking about my insulation type, I have blown cellulose in the walls and the attic. I have a storm shelter in the basement, under the back porch. The ceiling of that room as spray foam of some type, not sure which.
Wow, are they leaking at brazed joints?
I'm in the midst of a indoor ducted Fujitsu install.
My Fujitsu wall units have bee ok so far. The Mitsubishi has a tiny leak somewhere but I think it's at the flare that's slightly cocked only went real cold out.
We assumed the leak was coming from brazed joints. We checked and rechecked those flared fittings and even got the gaskets they sell to improve the seal. Then we found the actual leaks after doing a pressure test with nitrogen. For the ducted unit, they were all along the coil as the copper tube comes out of the coil and then bends 180* to go into the next line. I am not sure if that is one continuous tube or if those turns are brazed on. For the ductless unit, the hole appeared in the middle of the tube, as if there was some kind of impurity in the place that corroded.
What a pain. Are they covering an under warranty? The small leaks can be really hard to find
I don't believe so. Too much time has passed.
What is worse is that the ductless unit has been discontinued, and there aren't anymore any stock anywhere, AND they have also discontinued all the indoor units that are compatible with the outdoor unit. So I am going to have to replace the entire system. For the ducted unit, they can't find the coil in stock but they can find the indoor unit.
Fujitsu is bad with this. Mitsubishi doesn't swap models as often and you can almost always get a replacement head. Unfortunately Mitsubishi ducted heads aren't very good options for most. With Fujitsu I'd always try to get the latest model at the time.
Could be formicary corrosion—you can google it and find good explanations. Certain organic acids can cause pinhole corrosion in copper tubing. There are reports of association with open cell spray foam, which is what I think Walter’s question alluded to. There was also a rash of failures some years ago with a certain kind of white lineset insulation.
Anyway hope that’s helpful. Two units makes me think it’s unlikely to be a run of the mill manufacturing defect.
“Walter, are you suggesting that off gasses form open cell can corrode hvac coils? "
The expert guest on a BS + Beer pod cast I listened to a few weeks ago, suggested that there was a very strong and wildly recognized in the industry correlation between leaking evap coils and homes with open cell spray foam.
True or not I do not have an opinion only questions.
Walta
This is concerning. I have 8 inches of open cell against my roof deck under 4 inches of polyiso.
Crossing my finders I don’t find myself in this situation.
Edit: do you have a link to that podcast? I can’t seem to find it.
#38 A look at Refrigerants
Walta
Hi Tom,
Here is the link to the BS* + Beer Show episode: A Look at Refrigerants. I would also recommend this article from one of the guests: How to Reduce Refrigerant Leaks from Heat Pumps.
Any idea if it's a permanent problem? Or with sufficient time and offgassing, would a home with OC spray foam be just as harmless to evaporator coils as a home without any spray foam?
I ask b/c I had a new HVAC installed in my attic in January 2019. I had OC spray foam installed in the same attic in July 2020. I had to replace the evaporator coil in that HVAC unit in September 2021 -- when the unit was less than 3 years old. And the tech identified the leak as a small pinhole in the coil -- not at a braze or a joint. So pretty consistent with what you've posted.
Anyways I am hoping that since the new evaporator coil was installed more than a year after the spray foam was installed, that the spray foam will no longer be offgassing in a way that might affect the new evaporator coil.
Unless you see obvious signs of corrosion on the outside, I think the issue is with the original install.
It is a bit too late to check since the refrigerant is out of your system, but my guess is you have moisture in your system which caused the POE oil to turn acidic. This slowly corrodes away anything copper causing pinholes.
There is a test kit you can buy to check for this by burping the service port check valve through a vial with a test strip.
Acidic refrigerant usually happens from a sloppy installs where the system is not purged properly and moisture is not fully eliminated from the system.
It can also happen if the system has a leak that is not found in time and moisture makes its way into the refrigeration loop. If the system is refilled, it will run for a while but eventually something will corrode and cause an even bigger leak.
Unfortunately once you have this type of damage, the whole system needs to be replaced. Replacing one part will just turn into a game of leak whack-a-mole as lot of the components have been partially corroded.
Thanks Akos. I'm not sure what to look for in corrosion on the outside. I don't see anything obvious. I see the copper has tarnished, but I assume that is natural. I have attached a pic.
Moisture in the line would be a logical, and alarming explanation. I hope that is not the case. I watched the installation process and remember them putting the lines on a high vacuum in order to draw out the moisture. But, obviously, there has been a leak from very early on, and maybe the leak allowed moisture in later and now I have even more leaks.
Is the moisture only a concern while the refrigerant is present in the line? I have the ductless unit out of commission while we figure out what our next step is, but it still has the hole in the line, exposing the line to moisture. However, there is no refridgerant in the line.
Akos,
Would even the lineset need to be replaced in this scenario? So many houses around me have linesets running through walls or in otherwise unaccessible places.
If the problem is corrosive oil confirmed with a test, then everything copper in the system would be damaged. Not need to take chances with pinholes, get new lines.
Thanks for the response. That is what I feared, and it makes sense. I suppose corrosive oil won't selectively impact the evaporator coil and leave the lineset untouched.
I think Carl Seville (sp?) GBA blogger had similar issues. You may wan to do a quick search.
Thanks. I found this one? It might be the same thing, but he doesn't mention the cause.
https://www.greenbuildingadvisor.com/article/learning-from-our-mistakes
Ya that's it.
Anyone have any updates here. I had my Fujitsu system installed 3 years ago. 2 compressors 11 heads. I’ve had line leaks flare leaks wall mounted unit leaks. Ceiling units corroded completely. What should I do!!!
My HVAC guy was unable to make repairs to all the leaks. We had to completely replace the downstairs unit. They discontinued the indoor unit so I had to buy both the indoor and outdoor units. Still waiting on a fix for the upstairs unit. They have discontinued that unit as well, so I'll probably have to replace that whole system as well.
I’m not sure if I can do that. I have 11 indoor heads. The compressors thank goodness have been ok. But flares and indoor units constantly leaking. I’ve had a few hvac companies come out. All have different issues they find. Charge me thousands and six months later out of Freon again
This is why I will never buy a house that has more than 1 or 2 units. 11 heads sounds like a maintenance nightmare. Even with a halfway decent installer, seems likely that you'd have at least 1 leak per year.
Unfortunately I believe it is the poor quality materials on certain lots, and a process in the manufacturing of coils.
I personally have installed and replaced many coils. From new construction with spray foam , to old homes with older insulation dating back to the 50’s.
I have since switched brands. Unfortunately warranty issues, lack of materials, supply chain, etc. That said I have also replaced many conventional ac coils of different brands. This is why I believe it is poor raw materials or a process in the manufacturing which is causing leaky coils
I use to be involved in the HVAC industry in Maine and it was very known that Fujitsu had coil leak issues. This was common knowledge that was used to sell Mitsubishi and other makes over Fujitsu. Reach out to a supply house that sell Fujitsu and see if they can help.
If your friend really is “pumping you up with refrigerant” I’d almost guarantee the problem is moisture in the lines causing internal corrosion. Every mini split I’ve ever worked on calls for purging the system of refrigerant, doing a triple evac, and then weighing in a replacement refrigerant charge. You can’t just add more refrigerant like a typical split system.
Fujitsu is still using the older POE oil even on their latest units that is hydroscopic and will absorb water which then creates acid. Many manufacturers have switched to polyvinyl ester (PVE) that is not hydroscopic and will not react to water. Mitsubishi, Daikin, Midea/Carrier, Gree and others are using it.
Daikin has an awesome presentation here on how much better it is than old mineral poe oil.
https://hvacrassets.net/handouts2/PVE_vs_POE.pdf
“It has been found from experience that once the moisture level reaches about 800 ppm there is a likelihood of incipient ester hydrolysis”
Further:
“The carboxylic acid components manifest themselves by an increase in the acid value and may lead to copper plating reactions in the system”
My guess is there were a lot of bad installs that let moisture into the system and the acid that is created is eating away at the coils.
As the other poster noted just adding refrigerant is a really easy way to dump a ton of moisture into the system.