Increased Humidity With Geothermal Split System
Hello GBA!
I’ve read numerous posts about humidity on your forums, but I wanted to run this doozy by you.
We had a geothermal split-system with 3 ground loops installed in our home in March ’20. They replaced two old air-condition condensers/air handlers, a 3.5 ton and a 2 ton, with a 4 ton and a 5-ton heat pump and matching air handlers made by Bosch.
We knew going into the project that the old insulation needed an upgrade and that the house had a good bit of air leakage (at least with heat).
The original heat was from an oil boiler that ran hydro lines to the air handlers, which I believe the new heat pump system does the same thing.
The heating was fine, and it brought up our low humidity up to 35-40% from 17-20% originally.
When we switched it from heat to cool/AC, then the issues began.
The humidity went all the way up around 70%. There’s some condensation on the windows, and it actually got in-between the panes, leaving a stain inside pretty much all of them.
The style of the house is mid-century modern, so it’s basically all windows and some pretty high cathedral ceilings. It broke our hearts when we realized the stains were permanent inside the panes and is probably going to break the wallet to replace all of them.
The geothermal company has been out a few times to look at it and slowed the fan speeds as much as they could.
Now we’re around 50-55% on one side, and 55-60% on the other, both Nests set at 72º. We have allergy issues, and the dust mites are clearly having their day. We live 30 miles north of New York City, and it’s fairly humid in the summer.
We wanted to replace the old batt insulation as a next step and were thinking about going with spray foam to achieve a “thermal envelope”.
However, with this humidity issue, we’re afraid to do anything until we’re sure it’ll help and not make the problem worse. Do you think that would make the situation worse, or better?
My hunch, based on my own research on people with similar issues, is that they oversized the system by a ton or two on each side.
Thanks in advance for any thoughts or help here.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
9 tons? Yes, unless your home is Biltmore Estate, it's a good bet the system is oversized.
I'm ready to admit the system is oversized, but we want to rule out that it isn't the old failing batt insulation, or that they sized for the heat correctly and not the A/C, as someone mentioned below.
50-60% is typical. But if you want lower, add dehumidifiers. You could look into limiting cycling with a buffer tank and a smaller air handler. But contrary to popular opinion, over-sizing has little effect on dehumidification.
Air sealing will help. Insulation will mean less AC run-time, leaving more work for the dehumidifiers.
55% starts to get uncomfortable, as the temperature is 72º but you feel hot. Also, the whole family is sneezing constantly. I'd be fine with 4-50% rh. I'm thinking whole-house dehumidification and air sealing or new insulation will help, but I'm afraid the system running so on/off constantly will wear out the machinery relatively quickly.
>There’s some condensation on the windows, and it actually got in-between the panes, leaving a stain inside pretty much all of them.
This is a curious detail. Condensation on the windows is usually associated with excess humidity in the winter, not in the summer. Are these double-glazed windows and the condensation is appearing between the glass layers? That would indicate a seal failure on the window. Is the condensation on the inside or the outside?
Mid-century modern double pane might mean they are site-built and unsealed. Condensation in that case would suggest that the inner pane is becoming chilled below the dewpoint of outdoor air, and that outdoor air is getting between the panes. We have had dewpoints in the 70's all summer in the greater NYC area this summer, and with 72 F indoor setpoints, you're getting close to the dewpoint. One thing to check is whether the A/C vents are directed towards the windows or out into the room. Towards the windows can be common in systems designed originally for heating. With 55F air coming out of the vents, chilling the windows below dewpoint is easy.
It appears the windows were replaced in the early '90s with standard double pane. The rest of the windows are mostly aluminum and pretty old, but double-pane as well.
There are a lot of custom windows and some aluminum sliding glass doors with double panes.
I guess the point I was trying to get at was that unless there's condensation on the inside of the glass it's not indicative of an interior humidity problem.
The OP wrote: "The geothermal company has been out a few times to look at it and slowed the fan speeds as much as they could."
That would mean very cold air coming out of the ducts. If the ducts are pointed at windows, that could lead to condensation on the window. I think that's a separate problem from your interior humidity issue.
I'd like to know more about the windows. Maybe they aren't true sealed IGUs, but are just unsealed air filled windows. In that case, it may be possible to disassemble them and clean them. It may also be possible to replace the glass with IGUs at lower cost than full replacement windows.
The condensation appears on the outside of some of the upper windows, but mostly the moisture got inside the panes of the double-pane windows when the rh was at 70%.
The oversizing sounds really unreasonable. Presumably their thinking was that they were sizing for winter and that it was OK to have them oversized for summer, but they are probably way oversized for winter too. The good news is that your ground loops are probably oversized too, and in a ground-coupled system, that only improves your performance.
Options include:
1. Get them to replace the heat pumps with lower capacity units, sized based on a "manual J" calculation for heating need. Ideally with variable speed compressors, such as the Water Furnace 7-series, or at least "two stage" compressors (two speed).
2. Reconfigure at least the thermostats, maybe also the ducts, to mostly run just one of the two units in the summer, so it will have longer run times.
3. Reconfigure to use the Hydron Module Cruise WV variable speed water-water heat pump and run water lines to the individual air handlers rather than refrigerant lines. You'd need a competent installer to set that up right, and it doesn't seem like you have one, but that could work very nicely and would use less refrigerant and has less risk of leaking refrigerant with the associated climate impact.
4. Add separate dehumidification, either free-standing units or ducted.
5. Seal air leaks in the envelope, and if you take that far enough to need to add mechanical ventilation, do so with an ERV system that helps reduce the moisture that gets in.
Confirmed, they did install two-stage compressors, albeit most likely oversized.
That geo looks to be way over sized, hopefully there is a buffer tank in between.
If there is a buffer tank, you can keep turning the airflow down until the system just barely handles the cooling load. This should get you much better dehumidification.
If there is no buffer tank, make sure the air handler is configured to keep the fan off between calls and that there is a delay between running the cooling coil and turning the fan on. This will prevent condensate from re-evaporating from the coil on startup and shutdown. With an over sized systems, this can help a lot with humidity control.
Here are some pics of what they installed, minus the 5-ton air handler. The house is a two-zone, split level house. Basically each zone is a rectangle but slightly offset from each other. All ceilings are sloped, mostly cathedral ceilings.
You units are two stage water to air, so you have no buffer tank.
I would make sure at least the units are set up properly to run on two stages, both geo and air handler.
Many times I've have seen where these were not wire up at all or not configured. You want to make sure that your air handler also switches to low stage when the geo is running on low, sometimes this is not configured properly and you'll get terrible humidity removal.
The difference between the two stages is not all that much (4 ton unit is 3 ton on low stage), so at least if they are running on low properly they would be slightly less oversized.
Provided you don't need the extra capacity for heating, getting these replaced with correct sized units would go a long way for fixing the humidity problem. Even if you have to undersize them a bit for heating, it is lot of times better to add in some resistance backup heat to help during those polar vortex days instead of over-sizing the unit. The runtime on the heating element will be low enough that it will barely budge your energy use.
Getting a mechanical engineer to do a proper heat load calc would be the best place to start.
It may be that the water isn't cold enough. You will never be able to reduce the dew point of the air below the temperature of the cooling coil. If your coil is at, say, 50F, the air exiting will have a dew point no lower than 50F. If the ambient is a dew point of 70+, the mixed air could be 60+, which is not going to be comfortable. This would particularly be a problem in a leaky building where there is a constant influx of humid air.
Air conditioning does two things -- cool the air, and dehumidify it. The ratio is determined by the coil temperature, the colder the coil the more dehumidification happens. In conventional systems you can lower the coil temperature by reducing the airflow. In a hydronic system you also have to look at the water temperature.
In conventional systems, you have to be careful not to reduce the airflow too much or the coil freezes up. In a hydronic system that's not an issue. You could reduce the airflow further by partially blocking the intake without ill effect.
This system doesn't use chilled water.
It would be worth posting your exact model numbers. Some Bosch geothermal heat pumps have an optional "humidity control thermostat" that modulates the fan.
Here's the model numbers for the 5-ton heat pump, 4-ton heat pump, and the 4-ton air handler.
Can you confirm that the fans go off completely when there is no cooling (ie, compressors aren't running)? Unlike sizing, this makes a big difference in dehumidification.
But with a leaky house in a humid climate, you need separate dehumidifiers.
I've turned off all of the scheduled fans. Each side has a Nest Learning Thermostat. There's a couple of features that I'm curious about: Cool to Dry, and the auto-fan Airwave feature. I didn't have much luck with Cool to Dry, but I don't have scheduled fans running anymore (though the company recommended it). Its possible fans are running based on the learning or auto feature, however.
You don't want the fan running when the coil isn't cold. The way moisture is removed from the air is that it condenses on the cold coil and drips down into a pan where it's collected and removed. If you have the fan going when the coil is no longer cooling, any water that is on the coil will tend to dry out rather than dripping into the pan. That water gets put right back into the air.
A certain amount of evaporation is inevitable, but this is why you want long run cycles for cooling. It takes a few minutes for enough condensation to build up on the coil to begin dripping, and at the end of the cycle some of the moisture on the coil will evaporate rather than dripping off. If the cycle is too short very little of the condensate ends up in the pan.
This is in contrast to heating. When you're heating it's OK to run the fan after the coil has stopped receiving heat, to squeeze every last drop of heat out of it.
Is your indoor set point below 68°?
What is the supply and return water temps to the air handler?
My wild guess is the cold water supplied is only a few degrees below your indoor dew point. I doubt that your system can produce water cold enough to remove much moisture from the air to make you happy.
Your system appears to have been designed without any the benefit of mathematic. The school of thought seems to be if I throw the biggest equipment imaginable in this house it should work.
The first thing I think you should try turning off the larger system so the smaller system can have longer run times giving it a chance to remove moisture.
If you intend to keep this equipment you are likely to need four 70 pint dehumidifiers to remove the moisture this will heat the air in the house but the oversized geo units will have no problem keeping the house cool as long as you pay the massive power bills.
If you are willing to replace this equipment pay an engineer to design a new system and without hydronic cooling.
Walta
We keep it at 72º. It gets absolutely freezing below that. It can reach set points 3-4º in very short amount of time (15-25 minutes).
We're definitely leaning towards having them install smaller units. Someone recommended having a mechanical engineer redesign it?
I believe they did load calculations, but they seem flawed for the cooling based on the issues we've been having.
Either way, we're considering getting new insulation first, so as to rule that out as the source.
There's no reason to believe the insulation is the source.
Let me expand on my earlier answer.
The reason it's called "air conditioning" instead of just cooling is that the cooling system does two things: it cools the air, but also removes humidity. Cooling the air without removing humidity results in a cool, clammy feeling that is not comfortable, as you've seen. Both cooling and dehumidification require the removal of heat, and a well-designed system removes them in a ratio that produces a comfortable environment However, this ratio does not happen automatically, it is the result of deliberate system design.
Your system is not producing dehumidification and cooling in an appropriate ratio, it's providing too much cooling and too little dehumidification. The point of my earlier post is that there's no reason to believe that better insulation will change that ration in a meaningful way, and in fact I can think of several ways in which it might make the ratio worse.
While for the most part the ratio is determined by system design, to a certain extent it can be adjusted in an installed system. The adjustments you want to make are to lower the coil temperature and increase run times. Usually this is done by reducing the airflow across the cooling coil.
At this point, the question before you is whether your system is misadjusted or even malfunctioning, and can be adjusted to give satisfactory performance, or whether the design is so hopeless that the only solution is to remove it and replace it with a properly designed system. If the system is indeed improperly designed this may well end with litigation against the installer.
They've been out several times to adjust. They've reduced the fan speeds in the air handlers as far down as possible on the 4-ton, and second from last setting on the 5-ton. I should have mentioned, but they left the oil boiler as an emergency backup, which is the reason they didn't turn the 5-ton all the way down.
Clearly the adjustments they've made aren't working. And from what you've described it sounds like they're taking a hit-or-miss approach.
You need the air coming out of the vents to be colder. To know if that's happening you need to be measuring it. So first step is to get a thermometer and measure the air temperature and report back here. Now, there are two ways of making the air colder -- reducing the airflow and making the water colder. The air can never be colder than the incoming water, so there is a limit to how much reducing the air flow can do for you. So you also want to measure the temperature of the water to see how close you are to the air temperature. If the water temperature is substantially below the air temperature try blocking off the air inlet with cardboard to see if that gets the air temperature down.
If the water is not very cold you need to see about making it colder.
>”We keep it at 72º. It gets absolutely freezing below that. It can reach set points 3-4º in very short amount of time (15-25 minutes).”
That’s probably your problem. Oversized units running short cycles and dramatically cooling the air doesn’t achieve much dehumidification. What you want is very cold water in the coil along with a low airflow over the coil in the air handler. That is the setup for maximum dehumidification with a hydronic cooling system. This will probably require some configuration changes on the equipment.
I’m most familiar with larger chiller systems which run water temperatures around 40-45 degrees F, which can accomplish good dehumidification if the air handlers aren’t running excessive airflow rates. I don’t know if your heat pumps can get the water temperature down that far, but even if you can only get to 50F you’re provably ok. You want to slow the airflow down in the air handler so that you have the air passing over a very cold coil. Does the air handler have a variable speed blower, or just a two stage high/low setting?
Our resident HVAC guy on here (Dana) appears to have taken a leave of absence recently, but he would have told you that “thermal calculations need to make aggressive assumptions about the structure”. He has said that on here many times regarding manual J calcs. Chances are your installer either used rules of thumb or was to cautious with their assumptions and ended up with an oversized unit. You’ll get the best size estimate from an engineer or an energy rater, not an HVAC system installer.
Bill
This is not a chilled water system. One of my recommendations earlier was to replace it will a chilled water system, at which point it would be possible to use a buffer tank to mitigate the oversizing and configure the coils to run low temperature water at low airflow rate and get excellent dehumidification. But as it is, the compressor is way oversized and you are pretty much stuck with short cycling.
From the original post.
“The original heat was from an oil boiler that ran hydro lines to the air handlers, which I believe the new heat pump system does the same thing.”
Sure sounds like a hydro system!
This photo sure looks like a geo hydro system to my eye.
Walta
Look at post #17 and look up those model numbers. It's a split system not a water-water system.
There's a lot of confusion in the comments here because there are a few people assuming this is a chilled water system, which it's not. As I said earlier, replacing it with a chilled water system would enable a skilled engineer or installer to tune it up to do what you want beautifully, but you'd need to find the right person to do that and convince that folks you are working with to take back the oversized equipment they sold you, or you'd need to sell it on ebay and take a big loss.
Short of that, the easiest things to do would be:
1. Make sure they really have the two-stage compressor set to run at the lower power level when its cooling.
2. Consider going back to a more straightforward thermostat. Nest tends to think it's being smart but not really let you configure directly what you want it to do.
3. Set the fans to their lowest setting, but just for cooling mode. Automatically or manually switch to higher speed for heating.
4. Set the thermostat for one of the zones, the one that includes the upstairs, to a lower temperature and the other one to a higher temperature. That way, you'll get one system running for longer run times and doing more dehumidification.
5. Add dehumidifiers.
The next step would be to get a heat load calculation done and see how far oversized the units are, and use that to negotiate for swapping them out for smaller units.
As far as insulation, the one thing that improved insulation could offer is the ability to downsize the units further, but if you aren't going to be able to get them swapped out, insulation will only hurt, not help, as DCContrarian explained. But air sealing can help!
You might be able to find someone to do an independent heat load calculation and a blower-door test to find opportunities for air sealing.
Thanks, Charlie (and everyone).
I think this is an excellent summary of the next steps and questions we have for the company.
I really appreciate the thought and inputs here, it's definitely helping me as the homeowner with a limited and new understanding of this complex issue.
We'll see how the conversation progresses, as it really could go either way at this stage.
Good suggestions from Charlie.
Turn off the Nest airwave feature. That's only useful in low humidity climates like Arizona. It will just re-evaporate the moisture off the cold coil.
The control band on the nest was far too wide and not adjustable when I had one. Enough that you would start to feel hot and tempted to fudge the stat to get the cooling to come on, but once it was running it would run and run and start to feel cold. With my basic Honeywell stat (and later the Ecobee) this wasn't a problem at all.
Make sure major air leaks like fireplaces are blocked off.
Second the idea to only enable the AC on the upstairs unit. On the lower unit set it to run the fan constantly but keep the AC off. I would ask for verification that it is running on 1st stage on the compressor and fan. You should be able to verify this with a amp meter. The measure the compressor and fan amps separately and verify they are different on the different stages. Don't take their word on it, I would want to witness the demonstration.
Also, verify you supply air temperature off the units (you want to make sure it's getting cold enough to dehumidify).