Mitsu H2i 4 zone w/branch box - lineset sweating

[Alex138]

Hi,

I had a 36,000 btu outdoor unit, branch box, and 4 indoor units installed. The house is a 1200 sq/ft rowhouse. The linsets are the insulated with the heavy duty white insulation, and ran through an old A/C duct and through a soffit. The upstairs linsets collect water on the outside of the insulation, and cannot be enclosed in drywall in fear of creating a mold situation. The only humidity in the house is the humidity introduced by the system. Humidity with system on is consistently 60-70%.

The indoor units are:
6k
6k
12k
9k

Total of 120 feet of lines. 11 lbs of refrigerant was added to the 36000 btu H2i outdoor unit.

Could the refrigerant be low, causing exessive moisture?

What is the proper way to evacuate/charge a multi unit system with a branch box?

Could there be an issue with vertical rise when using a branch box? (There is 18 ft of vertical rise to the top floor)

What would you do?

 

[Dana]

"The only humidity in the house is the humidity introduced by the system. "

That is an absolute misconception!

First, in air conditioning mode the system removes humidity rather than adding humidity.

Second, you have both indoor sources of humidity (cooking, breathing, bathing) as well as humidity being introduced by outdoor air infiltration.

The relative humidity numbers without the air temperature it's relative to is a meaningless number. Outdoor air that is 50% RH at 90F has a dew point of 69F. Cooling that air to 69F would result in 100% RH without changing the moisture levels of the air at all. Cooling that air to only 75F becomes 81%.

Low refrigerant usually results in colder than-design temperatures at the coils, which removes more moisture, not less. Running the blowers at maximum speed results in higher coil temperatures, and generally less moisture removal. If you set the cooling mode to "DRY" the heads auto-adjust the blower speed to maximize the latent cooling (moisture removal.)

With indoor RH that high I can only assume you have copious air leakage in the house, or it is being mechanically over-ventilated.

If the linesets are in a vented attic, above the insulation exposed to the humid attic air that is also passing through the penetrations of the insulation & ceilings into conditioned space, that could be a major part of the problem. Wherever the linesets enter the house have to be reliably AIR sealed, not merely insulated. (Stuffing a bunch of wadded up insulation batt around it is not an air seal.) Any plumbing, duct, or flue penetrations of the attic also need to be air sealed, since that's a primary infiltration driver.

A decent primer on the merit-order of weatherization measures can be found in this video, but if you can't stand listening to anybody who talks that fast for 50 minutes the basics are:

1: Air sealing is more important than insulating

2: Attic air sealing the attic is the most important air leakage points in the house in a heating dominated or mixed climate (that would be you), followed by air-sealing the basement (particularly big leaks such as band joist/foundation sill seams, dryer vents, etc.), with all intermediate floor window/door/plumbing & wiring penetrations being the least important, but still more important than insulation.

 

[Sponsor]

 

[Dana]

The average outdoor dew point in Washington D.C. over the last half-month has been 70F. In July it averaged a somewhat more tolerable 66F.

Then the dew point is 70F any surface cooler than 70F exposed to that air would be sweating. Enclosing the linesets in air tight & painted drywall would not have that direct exposure- a thin film of condensation might form on the lineset while running, but there would not be continuous replenishment of new humid air coming in contact with the lineset. Retrofitting more insulation (say the cheap foamy stuff sold for plumbing insulation) around the linesets would result in a higher surface temperature, for less sweating too. But fixing the air leakage problem and getting the indoor humidity to 50% RH @ 75F (dew point = 55F) or lower would almost certainly stop the sweating too.

BTW: 3 tons of cooling capacity is almost ludicrously high for a 1200' town house. Most single family homes that size would have a 1% cooling load of about half that, and townhouses would usually com in lower. I suspect you went with the 3-tonner because you felt it needed 4 zones(?). If the heads are oversized for their zone loads (almost certainly are) the latent cooling you would get in normal cooling mode isn't very effective. Unlike individual mini-splits with a compressor unit for every head, the heads themselves won't modulate when married to a multi-split compressor. The compressor modulates up & down based on the number and size of the heads calling for refrigerant, but it's in steps, not continuous adjustment. The submittal sheet says the compressor can modulate down to 6000 BTU/hr in cooling mode (the output capacity of a single half-ton head), but the individual heads would all be cycling on/off when the sensible load is that low. If cycling, when the head turns off the coils warm back up, re-releasing a sizable fraction of the moisture that it had removed back into the room.

One way to improve the latent cooling is to turn the heads off on the lower levels and let the upper floor zones run much longer duty cycles, leaving the doors open so that convection inside the house allows the cool air to flow to the lower floors. Even then you may have to run the active heads in DRY mode during muggy/oppressive weather to keep the indoor RH at more comfortable & healthy levels. But whether turning off some zones or keeping them all on, running it in DRY mode should make a difference unless your house leaks air like a barn.

 

[Alex138]

"The only humidity in the house is the humidity introduced by the system. "

That is an absolute misconception!

Sorry, basically I meant that the humidity only increased with the addition of the system.

 

[Alex138]

The average outdoor dew point in Washington D.C. over the last half-month has been 70F. In July it averaged a somewhat more tolerable 66F.

Then the dew point is 70F any surface cooler than 70F exposed to that air would be sweating. Enclosing the linesets in air tight & painted drywall would not have that direct exposure- a thin film of condensation might form on the lineset while running, but there would not be continuous replenishment of new humid air coming in contact with the lineset. Retrofitting more insulation (say the cheap foamy stuff sold for plumbing insulation) around the linesets would result in a higher surface temperature, for less sweating too. But fixing the air leakage problem and getting the indoor humidity to 50% RH @ 75F (dew point = 55F) or lower would almost certainly stop the sweating too.

BTW: 3 tons of cooling capacity is almost ludicrously high for a 1200' town house. I suspect you went with the 3-tonner because you felt it needed 4 zones(?). If the heads are oversized for their zone loads (almost certainly are) the latent cooling you would get in normal cooling mode isn't very effective. Unlike individual mini-splits with a compressor unit for every head, the heads themselves won't modulate when married to a multi-split compressor. The compressor modulates up & down based on the number and size of the heads calling for refrigerant, but it's in steps, not continuous adjustment. The submittal sheet says the compressor can modulate down to 6000 BTU/hr in cooling mode (the output capacity of a single half-ton head), but the individual heads would all be cycling on/off when the sensible load is that low. If cycling, when the head turns off the coils warm back up, re-releasing a sizable fraction of the moisture that it had removed back into the room.

One way to improve the latent cooling is to turn the heads off on the lower levels and let the upper floor zones run much longer duty cycles, leaving the doors open so that convection inside the house allows the cool air to flow to the lower floors. Even then you may have to run the active heads in DRY mode during muggy/oppressive weather to keep the indoor RH at more comfortable & healthy levels. But whether turning off some zones or keeping them all on, running it in DRY mode should make a difference unless your house leaks air like a barn.

Thanks! When its blazing hot and humid, the house is good, 72 degrees and 50% humidity. When the outside temp drops below 80, it gets muggy inside. The lines are cold to the touch through the best "white" insulation, and some places get condensation. Often, it seems that the "cold" gets directed to one unit (like there isn't enough for the other units). If I run all the indoor units in DRY, only gets the house down to 55%.

After I had my system installed, I have checked every system I come across and none have line set sweat like mine (including line sets with the black insulation going through attics and closets). All were multi zone, non hyper heat, non branch box.

This leads me to believe there is air in the system, which is why I asked about the charging procedure for systems with the branch box. The installers added the correct amount of refrigerant for the line set and load, but is it possible the outdoor unit don't have enough in it? Or air wasn't removed from all of the lines (from the branch box on)? Do the indoor units have to "call" the branch box when it is evacuated and charged?

 

[Dana]

It's unlike that there is air in the refrigerant, even if the refrigerant levels are below spec. If there were enough air in the system to cause symptoms like this there would be enough WATER in that air to cause real damage.

All refrigerant in the branch box goes through the outdoor unit and indoor heads. ALL of it. If any one part is under-charged, the whole system is under-charged, and that is something the installer can check, and SHOULD check when commissioning the system. The installer is responsible for getting it right, and if you really think they screwed it up they deserve a shot a diagnosing and correcting their work.

But the real problem is still far more likely to be the air leaks in your house.