Install ductless mini-split

[kevm14]

I have convinced myself, possibly wrongly, that the great room mini-split has a low charge. Some guy on youtube put together these kits which make it about as easy as charging your car with a can and a simple low side gauge. My mini-split doesn't even have a high side test port so...

Anyway, I bought a kit (for $200) and I am anxious to see what the pressure is. He claims you really can look at the color bands on the gauge. He recommends shooting for the middle of the blue zone. Green is a low charge. We shall see.

As for symptoms, it just doesn't seem like it has the A/C capacity I would expect for 18k BTU. I also feel like the speed the condenser runs compared to the amount of cold air produced is the wrong ratio. It IS still functioning though. I did spray some soapy water on the condenser fittings and didn't see anything. I may want to pull back the foam insulation and really get in there. I also used Ed's sniffer in there and didn't find anything. I did get a few odd hits coming off the inside unit airflow so maybe the inside coil is leaking???

First step is determine if the charge is even low.

[kevm14]

I have convinced myself, possibly wrongly, that the great room mini-split has a low charge. Some guy on youtube put together these kits which make it about as easy as charging your car with a can and a simple low side gauge. My mini-split doesn't even have a high side test port so...

Anyway, I bought a kit (for $200) and I am anxious to see what the pressure is. He claims you really can look at the color bands on the gauge. He recommends shooting for the middle of the blue zone. Green is a low charge. We shall see.

As for symptoms, it just doesn't seem like it has the A/C capacity I would expect for 18k BTU. I also feel like the speed the condenser runs compared to the amount of cold air produced is the wrong ratio. It IS still functioning though. I did spray some soapy water on the condenser fittings and didn't see anything. I may want to pull back the foam insulation and really get in there. I also used Ed's sniffer in there and didn't find anything. I did get a few odd hits coming off the inside unit airflow so maybe the inside coil is leaking???

First step is determine if the charge is even low.

[kevm14]

Charge doesn't seem low. Will post some pics.

This is what I wrote to Ingrams just now:

Refer to previous request #290388 for background info. In short I have a Mr. Cool 18k DIY single zone (Gen 3) and received a replacement condenser (gen 4) under warranty last year. It has been working well. However, as we enter our hot season here, I have observed that the A/C is not as powerful as I would expect for the capacity. It does work and blows out cool air but it is unable to keep the room at the set point. I usually set it to 71 or 72 and there were days a few weeks ago where the room got up to nearly 78 degrees. At that time I observed that the system was running and blowing cool air but NOT at the volume I would expect. I also observed that the condenser air output seems cooler than I expected. I have also noticed that when the weather is more moderate and it is able to keep up, it seems like the condenser runs at a higher speed than what I would expect for the volume of air coming out of the inside unit. I had a hunch that refrigerant was low so I checked the low side pressure and it is seemingly within spec, however the pressure gauge needle bounces which seems a bit odd to me. There is also a new hissing sound at the condenser when running and I don't think that was there before. Can someone assist with further diagnostics to confirm if I do in fact have an issue here? I guess my initial thought is there is a restriction within the refrigerant loop somewhere.

I don't know if there is going to be a way to really prove my theory here. At this point, if there is in fact an issue, I think the only acceptable resolution is a complete replacement system. We'll see how this goes.

[kevm14]

I've had the unit set to 65 degrees for the past hour and the app says it still thinks the room is 70. It is only 75 degrees outside. I would expect better than this. So I await Mr. Cool contacting me (which is what Ingrams told me would happen).

[kevm14]

Shifting back to the work on the new system, here's what I accomplished yesterday.

Goal: create a suitable area to pour a concrete foundation pad for mini-split condenser.

Approximate steps taken so far:
- Using marking paint, sketch out the possible places the mini-split stand might want to end up (and this probably depends on the line lengths so I do need the pad to be bigger than the stand so I can move it around at installation)
- Also using marking paint, sketch out a total area that will be crushed stone and make a contour so that mowing along this is easier (i.e. no right angles)
- Using a flat shovel, dig up the sod and place in a wheelbarrow
- (Optional) relocate sod to bare areas of need elsewhere in my lawn
- Work on general grade a bit
- Make final measurement for concrete forms. I settled on 66x26 inches which means the forms need to have that as an inside dimension. Don't forget that two of the form sides will need to be about 3" LONGER so the boards overlap and can be screwed together. Fortunately I realized this before cutting anything. I used various scrap 2x4s to do this.
- Assemble forms
- Lay down in area and begin using stone to level things out. For final pitch I was aiming for a bit downhill and also away from the house. During installation of the stand, I will fine tune the legs for perfect level of the unit itself, but this will allow water to shed.
- Buy more stone and buy garden edging
- Lay down edging and stake appropriately. I like how this came out actually. I should have used it on other stone projects.
- Lay down stone (I reused some stone I had laying around, I think 3/4" white, plus bought 6 40lb bags from Home Depot). Rake around and test the level of the forms.
- Once that is close, cut some stakes from scrap wood. I ended up using 5 stakes.
- Add stone in strategic places and also use the final screw locations in the stakes to get the desired level.
- Compact stone.

Remaining work:
- I will need another bag of stone maybe just to make sure the form has a good base, before I pour
- Considering adding some steel reinforcement because the slab will be over 5 feet long.
- Purchase something on the order of 10-11 60lb bags
- Mix up and pour a bag at a time
- Do the trowel and screed process...with appropriate waiting times for each step. I will do the curved edges thing.
- Let cure, probably keep wet as needed.
- Remove forms
- Admire work
- Purchase additional bags of stone and basically fill in and level as much as I can up to and around the slab

Again on installation I will choose the final location for the stand, level it perfectly with the adjustable feet, and then concrete lag it into the slab.

[kevm14]

Charge doesn't seem low.

I had an e-mail exchange and then a phone call with the guy who sells the R-410a charging kits. He suggested that even though the needle was in the middle of the blue range, try adding some and target the high end of the blue range (i.e. 135psi). We discussed this for a while and eventually I figured what do I have to lose? Well, it's a funny thing because I added the ENTIRE 1.8lb cylinder (minus a small amount that won't flow based on a lack of differential pressure) and the needle pretty much didn't change. Weird, right?

Well, I checked it later in a hotter part of the day and I found the needle LOWER. It was probably reading around ~110 psi. So perhaps my unit was trying to compensate for the low refrigerant with the various ways it can do that, and making it seem like it was charged properly. But, after adding some, it gets more in the range of normal and actually reads low.

More importantly, I noticed the inside unit fan speed was higher and it was definitly outputting more cold air. So I ordered a second 1.8 lb cylinder which is kind of crazy (and also crazy expensive at $145). It has UV dye and a stop leak so if I'm really lucky this will get it back to 100% operation and also stop whatever the leak is. I'm not sure I'd bet on that but we'll see.

Beyond this experimentation (and hopefully repair) period, and especially if I need to vacuum and recharge the full capacity, I will buy a 5 lb cylinder and just do it that way.

[kevm14]

This unit makes no sense. When I checked the low side after receiving the second cylinder, it looked like this!

So I added about 4.5 oz. And the gauge didn't budge.

Later I checked again. Gauge looked like this:

So I added the remainder of the SECOND cylinder (or another 22.7 oz for a total of 54.3 oz between these two cylinders which is 2 oz short of the TOTAL CHARGE this unit is supposed to have). Gauge looked like this after:

This unit SHOULD be overcharged at this point. I don't think it leaked 27 oz in a week. And I don't think the issue is the gauge. I still think there is some kind of line or valve blockage and that is skewing the readings and ability of the unit to work efficiently. Today it is 81 with about 50% humidity. Good sun but nothing insane. It has been set on 68 since this morning. It says 73 in here and a digital thermometer I have in the room agrees. This isn't a bad temp but I just expect better performance from 18k BTU.

[kevm14]

Remaining work:
- I will need another bag of stone maybe just to make sure the form has a good base, before I pour
- Considering adding some steel reinforcement because the slab will be over 5 feet long.
- Purchase something on the order of 10-11 60lb bags
- Mix up and pour a bag at a time
- Do the trowel and screed process...with appropriate waiting times for each step. I will do the curved edges thing.
- Let cure, probably keep wet as needed.
- Remove forms
- Admire work
- Purchase additional bags of stone and basically fill in and level as much as I can up to and around the slab

Again on installation I will choose the final location for the stand, level it perfectly with the adjustable feet, and then concrete lag it into the slab.

Back to the new 2 zone system. It's done.

Received the freight shipment on a pallet. Initial inspection resulted in everything looking perfect. I couldn't find the reason this was a scratch and dent. I would later.

Next was the pour. I think I ended up using 9 60lb bags.

Notes for future self: use a float and also wedge a trowel down the edges of the form to help ensure no air pockets. Otherwise I think I did fine.

Ended up covering it in plastic and that made the weeklong process of keeping it wet pretty easy. Once every 24 hours was enough. And at the end of the week it was even staying wet longer than that.

The actual install began on 7/29 and into 7/30.

I spent another few hours the third and fourth day tidying up stuff, but I got it running on the second day.

The last step was just more insulation on the lineset. I ordered this giant 2" inside diameter rubber pipe insulation (3/4" walls!).

I could take one more step and cut the electrical feed to the inside units down to the appropriate length. That would require me to get the proper tool to cut armor, as well as reterminate the ends with the right ends. I don't see myself doing this anytime soon, but it would improve the installation to not looking as much like a DIY install.

[kevm14]

This unit makes no sense.

So I added the remainder of the SECOND cylinder (or another 22.7 oz for a total of 54.3 oz between these two cylinders which is 2 oz short of the TOTAL CHARGE this unit is supposed to have).

This unit SHOULD be overcharged at this point. I don't think it leaked 27 oz in a week. And I don't think the issue is the gauge. I still think there is some kind of line or valve blockage and that is skewing the readings and ability of the unit to work efficiently. Today it is 81 with about 50% humidity. Good sun but nothing insane. It has been set on 68 since this morning. It says 73 in here and a digital thermometer I have in the room agrees. This isn't a bad temp but I just expect better performance from 18k BTU.

New development. I opened a warranty ticket with Mr Cool and during the process they had me put the inside unit into Forced Cooling mode. This is done by first turning the unit off, and then double pressing the manual control switch inside the main cover. This supposedly is max cooling.

Well, the fan speed was low. Also when I checked via the app to see what it showed there, it showed this which seems pretty bizarre:

76 and low fan?? I mean that almost matches what I would get in FC. Might be a controller issue?

As I investigated further, I realized that air wasn't even coming out of the middle part of the unit, just the sides. First I thought FC was freezing up my coil. But then I looked at the fan.

What is that schmoo?? Some kind of like....mildew goo? Yuck. At this point I was forced to do this maintenance that I've apparently been neglecting. To accomplish this I was GOING to remove the fan but that seemed impossible because I couldn't find the set screw and was tired of flexing the coil around. So ultimately I just carefully vacuumed the fins of the fan all the way across until the big buildup was removed.

Then I foamed up the coil, let it soak, and rinsed by spraying water all over it.

Some gunk came out of the condensate line so, I am going to say this is approximately annual maintenance. This air handler has been installed since Sept 2021 so coming up on 2 years. Probably an annual cleaning like this would be good.

The results were immediate. The air output was improved so dramatically. I didn't realize how choked off it was getting. Furthermore, the unit had no issues maintaining my setpoint which means this was a MAJOR contributor to my issue.

The big question of the day is, what about the refrigerant situation? Was it really low before? Well, maybe? I mean I put so much in that if it was vastly over charged wouldn't it not work correctly? Still makes me want to replace the valve core, vacuum it down, and charge up with the factory amount. I might just try running the rest of the cooling season. You have to charge in cool mode so if I charge I have to be able to run the A/C mode. I could also try to get through the entire winter and see what I want to do in the spring. Either way, I feel like my original complaint has been resolved. Interesting, eh?

[kevm14]

Did some poking around mostly comparing AHRI specs for my single zone 18k (for a point of reference) as well as a dual zone 18k plus a tri-zone 27k. The reason to include the 27k is because for one of my 2 zone setups, I plan to run a 9k upstairs and probably a 12k downstairs. That can be paired with either the 18k or 27k outside unit. I wanted to compare efficiencies and output.

Current great room single zone 18k: https://ashp.neep.org/#!/product/62013/7/25000///0
18k dual zone: https://ashp.neep.org/#!/product/62016/7/25000///0
27k tri-zone: https://ashp.neep.org/#!/product/62017/7/25000///0

There are some interesting differences among these three units but the overall conclusion I came to is that the 18k outside unit is probably fine for my 9k+12k setup. Also, if I decided to run a 27k outside unit for this 2 zone installation, I would think an 18k for the downstairs would make sense. That way I could at least tap into the capacity of the 27k. I don't know if there is any performance downside (like dehumidification). I think any limitation on throttling down is with the outside unit. Obviously the 18k is larger and more expensive. It's probably overkill as this would give me a total of 5-1/4 tons household A/C capacity. But oversizing A/C to get better heat output is a thing with these.

Another interesting thing is that the 2 zone 18k has significantly more dynamic range of output than my single zone 18k. It can cool and heat to a lower minimum amount than my single zone. It also has a higher max output. In fact, for cooling, even the 27k tri-zone has similar minimums as my 18k single zone! That explains why I thought the dehumidification was kind of weak. It tends to overcool if you want the system to run. It can't really throttle back and keep that coil cold with low airflow. For heating, the 27k does have higher minimums than my 18k.

To see if the cost difference helps make the decision, here are the different kit options (not including extras but those are mostly a wash aside from the extra cost to run 10ga wire for the 27k outside unit):
Lowest option
18k outside
9k+12k inside
$2,854

Middle option (probably a wasted option)
27k outside
9k+12k inside
$3,182

High option (between this and low I think)
27k outside
9k+18k inside
$3,332

So really the cost difference I am looking at is between the high and low option. $478 extra plus the extra cost to run the 10ga. In the long run I suppose this is irrelevant if I really thought this was the right option. But I kind of don't. I think the lower floor of the 18k outside unit will be better, because there could be situations where the downstairs isn't calling for A/C but the bedroom is and with a 27k outside, the min BTU output is still on the order of 7.2k to 8k. That is a real problem because it would mean that 9k inside unit would need to run nearly wide open and either make the room too cold or shut off. The 18k dual zone can throttle down to 5.4k to 5.8k which is still high but obviously a better range for many conditions. In a normal cooling scenario, those min BTUs wouid be split in some way between the bedroom and downstairs so no issues. Heating is a similar concern I guess. Now I will say that 9k+12k is 21k not 18k so I will miss 3k of theoretical output because the outside unit will be 18k nominal. This is OK because I think the 9k bedroom unit will often have that 3k to spare while the downstairs is using more of that 12k capacity. Furthermore, by spec the 18k dual zone outside unit actually can beat its rated capacity in certain scenarios. For cooling it actually shows a max capacity of 21.6 to 22k so it can totally handle those transient peak scenarios. Even for heating it will do 24k with a mild 47F outside temp!! I think that makes my decision easy right there, and the low option is cheaper anyway.

Also from what I've seen from our 8k window A/C (after insulating), that unit actually cycles even on the hottest days. So that 9k unit will almost always be throttled down except in extreme situations where the house happens to be super hot and we want to cool it down quickly. These systems work best when you have a more consistent temperature so I guess I just wouldn't let the house get like that, and if we aren't home I could always cycle the system on remotely to get temps more under control before we get home. Many options here.

All that said, I am considering doing the kids' rooms 9k+9k setup first (which would obviously be paired with an 18k outside unit) because the windows where their window A/Cs usually go will be replaced in the next month or two and I'd rather not do a window unit in brand new windows. Plus their rooms run older window A/Cs. They have less A/C load in the summer but probably more heating load in the winter, than some other rooms.

I think a prudent plan is to do the 9k+9k and basically go an entire season with it. I would have the great room on an 18k, plus an upstairs and downstairs bedroom on a 9k each. That's 3 tons of nominal cooling capacity right there. I would think that would allow me to estimate the needed remaining capacity for the master bed plus downstairs unit (even though I pretty much already made the decision above).

So now that I have the 9k + 9k I already have thoughts. I have a total of 3 tons of mini-split plus an 8k window unit in the master bed. That is a total of 44k BTU (3.67 tons). And basically my current observation is that this is adequate. It doesn't even run wide open to keep up and the entire house is comfortable.

So if I added the low option from above to fill in the last 2 zones, I'd remove the 8k window unit but add 18k back. So 9k in the master bed plus 12k (the outside unit could still only provide 18k despite that adding up to 21k). That would give me a total of 54k (4.5 tons) compared to 44k now, or an additional 10k. I can envision that 10k in the main downstairs and again it isn't even really needed right now so that's just 10k of extra capacity for more extreme conditions I guess.

The only other option I am considering is a 27k outside unit with a 9k in the master bed but an 18k downstairs. That would give me a total of 63k BTU (5.25 tons), or 19k more than I have now. So do I want 10k more than I have now or 19k?

I re-read my stuff above and I reminded myself that the 27k powering a single 9k zone is going to over cool. So I guess it is 18k with 9k + 12k heads. I probably will go all winter like it is now (44k, 36k of it can heat). For heating, I suspect I would want the 27k but for cooling, not really. Since I do have a furnace I'd rather design for cooling than for heating, especially if designing for heating negatively impacts the cooling situation. The 27k is also physically larger and it will be the most visible of all of my condensers so there's that, too. Seems like I am confirming my earlier thoughts.