Hi All,

Last year in July we had a lot of upgrades done to our house in North Carolina. We replaced the old furnace&a/c with a 3-ton Waterfurnace geothermal heat pump, blew cellulose insulation in the walls (there was no insulation there before), and sealed the attic with open cell foam. We have been very happy with the upgrades until about a month ago when I started noticing that the air in the attic felt humid (there were no humidity issues during the last fall/winter/spring.

The humidity in the attic varies depending on the time of the day - it is lower in the morning and high in the late afternoon, as the temperature increases. In the morning the attic is 75F and RH is 63-65%, in the later afternoon the attic is 80-85F and RH is 75-78%, never quite reaching 80%. The house is a steady 75-76F and 50% RH. the outside temperature is 70-75 in the morning and 80-90 in the afternoon with the RH varying between 50% and 80+% at any time of the day.

We have about 6" of open cell SPF under the roof deck and 3.5" on the gable walls. The attic did not have a ridge vent, it had 3 passive vents, one fan vent, and two passive gable vents. All roof vents were boarded up with plywood and the gable vents were boarded up with blue insulation board with the spray foam covering everything. There are no roof leaks and both bathrooms and the kitchen are vented outside. There are 5 sewage vents, which appear to be well sealed. The moisture meter registers 9-10% moisture in the wood rafters and joists throughout the attic.

A couple weeks ago I brought in a dehumidifier in the attic and ran it for a couple days to bring the RH to about 50%, same as the house. A day or two later the humidity in the attic went back up to 65% morning - 78% afternoon.

I've read "Moisture Safe Unvented Wood Roof Systems" (www.buildingscience.com/documents/reports/rr-1001-moisture-safe-unvented-wood-roof-systems) and browsed around, but I do not quite exactly understand why so much moisture is being drawn in the attic. The roof is dark asphalt shingles and is rather simple - no dormers, just two chimneys and they are well sealed. In reading other stuff, I learned that there is such thing as solar vapor drive (http://www.buildingscience.com/documents/reports/rr-0104-solar-driven-moisture-in-brick-veneer), but I do not understand how so much moisture is driven in the attic so quickly (it took only 1-2 days for the dry attic to get this moisture), especially since vapor permeation through the open cell foam is a very slow process and since wood decking is not a water reservoir material unlike brick or stucco.

Since the house is always at a lower RH, I do not see how it can contribute to higher moisture in the attic. Now I am considering two options, but I am not sure if they will address the problem. I was thinking about having 1" of closed cell foam sprayed over the open cell foam under the roof deck to create an air/vapor barrier (and add another 6.5 R value). Another option I am thinking about is to install an impermeable self-sealing membrane when we replace the roof, which will create an air/vapor barrier above the roof decking. I would hate to spend the money on the cc SPF if I can solve this humidity issue when replacing the roof (which will come soon - within the next year or so).

Do you think I am on the right track or are there other humidity sources that I did not look at yet? Has anyone had a similar experience with the sealed attic? Also, if 6" (about 21R value) of oc SPF is not enough, then can I add rigid foam boards above the roof decking when I will be replacing the roof? And, in that case, should the membrane go between the board and roof decking or above the rigid insulation board?

I know that there are a lot of questions and I very much appreciate everyone's input.

PS. I talked to the contractor who installed the foam (he also did the geothermal system) and he basically said "you chose open cell, so don't blame me now" and did not really suggest anything except running the dehumidifier in the attic.

Vent the attic to the interior and/or a dehumidifier are the easy solutions.

Thanks. I thought of including the attic in the air conditioned space, but there is no easy way to add air supplies and returns - all the equipment is in the basement and no ductwork extends into the attic. In any case, though, I'd much rather solve the problem rather than mitigate the outcomes.

First, stop using RH%, since relative humidity is only relative to the temperature, and the temperatures in the attic won't track those of the outdoors or the fully-conditioned space very well.  Convert your temp & RH data to dew point, which is the measure of the absolute humidity of the air.  If your monitoring equipment only uses %RH &  temp, use a pyschrometric chart or an online calculator to convert to dew point. 

Then compare hourly dew point data from a nearby weather station (weatherspark.com or wunderground.com)  to verify that the attic is approximately tracking the outdoor humidity, as I think it is.

The AM 75F/64% air has a dew point of 62F, which isn't terrible, but it's on the sticky side.

The PM ~83F/76% RH has a dew point of 75F, which is a mold hazard if it stays there for weeks/months on end.

Simply heating up the attic air from 75F to 83F would not change the dew point of the air (but it's RELATIVE humidity would fall, not rise.) 

That moisture measured by the increased dew point has to come from somewhere.  Unless you have a humidity source in the attic, bathroom/kitchen vents or some other vent terminating illegitimately in the attic, the only way to see a dramatic a rise in dew point over the course of the day is air-leakage to a humid outdoors.   Outdoor dew points very often see shifts that large over the course of a hot sticky day, which further suggests that outdoor air leakage is the root of the problem. If that's the case, fixing the air leak would fix the problem. 

Unless you have a vapor barrier between attic & conditioned space, the attic dew point will track the average dew point of the conditioned space, once the attic is truly air-tight to the exterior.

The asphalt shingle roofing is quite vapor retardent, as is the 30# felt underlayment, but when a dew-wetted roof gets the first morning sun the vapor drives are briefly extremely high.  In some instances may (temporarily) raise the moisture content of the roof decking, but not the attic, since wood is also somewhat vapor-retardent (solid wood is about as vapor retardent as closed cell foam). 

With open cell foam on the inside the roof decking can dry slowly to the interior, which is fine.  Air sealing with closed cell foam is a bit easier than with open cell, but you should be able to get there with open cell.  A geo installer who posts here by the handle "engineer" uses open cell to air-seal & insulate FL attics all the time, but doesn't let the foam truck leave until they've tested it by pressurizing the attic with blower door and cranking up theatrical smoke machine to make the leakage points obvious.  Even when they THINK they've done a pretty good job air sealing with the foam, leaks will still abound in most homes, but finding the problem areas requires using the right tools for the job.

Stabbing in the dark with membrane roofs or closed cell foam oversprays are pointless. Your problem is an air-leakage problem, not a vapor-diffusion problem.  Blower-door testing & remediation sealing up the leaks with a 2-part kit-foam  (TigerFoam, FrothPak, etc.) would get you there more reliably and cheaply than "shooting in the dark w/prayer" methodology.

Dana,

Thank you so much for the detailed response - I have not thought of the blower-door testing of the attic, especially with a smoke machine. We'll do this before doing anything else, and we won't certainly be shooting in the dark, until we have a reasonable understanding of the humidity source.

We do not have a vapor barrier between the attic and the living space. In fact, the leaks from the living space to the attic are abundant: there is an actual stair going up with a door that has a gap on the bottom, a non-functioning dumbwaiter without a ceiling in the shaft (there are gaps in the waiter doors in the kitchen and basement so some air escapes in there and can travel up), and the regular light fixture holes, plumbing gaps, etc. So, from my understanding, the attic should be more in tune with the house rather than with the outside air. However, since water vapor travels up (it is lighter than the air molecules), is it possible that it just accumulates in the attic from the living area downstairs? Furthermore, I assume that before doing a blower door test in the attic we will have to somehow seal the top of the dumbwaiter, otherwise all the smoke will be exiting in the kitchen and the basement.

In the living area the temperature is set to 75-78F, depending on the time of the day/night. With RH 40-50%, dew point goes between 49F and 58F. In the attic, the temperature goes from 75 to 90F and with RH 63-77%, dew point goes between 62-82F. (That being said, the attic temperature never falls below or very close to the dew point - there is no condensation there).

There is no specific moisture source in the attic - the stove and the bathroom are vented outside. There is no ductwork or any equipment in the attic - it's literally empty.

So, here is what I will do: I will monitor the attic, outside, and inside temperature, RH, and dew points three times a day (9am, 1 pm, and 6 pm) for a week and then see if there is a pattern. Once I am done, I will post the numbers. I will the run the dehumidifier in the attic and do the measurements for another week to see how quickly the attic gains moisture levels comparing to the outside/inside moisture.

I have two more geeky questions.
1) If RH measures the humidity level relative to the temperature, will the mold grow when the humidity ratio exceeds a certain level or will it still depend on the relative humidity and the dew point?

2) RH decreases when the temperature increases and vice versa if the absolute humidity stays the same. Why then the attic begins in the morning with 75F and 61%RH (.011lb/lb humidity ratio) and ends up on a very hot day 89-90F with 77%RH (.023lb/lb)? That seems to be possible only if more moisture is drawn into the attic, but can that much moisture (double the water vapor content in the air) come from the outside through the leaks? I'll keep monitoring, though and see what happens with the numbers.

Mold grows in constant high RH. You might also check that your measuring device is accurate.

The hygro-thermal geeky answers are:

1> An RH of over 65% increases the rate of mold growth, and over 70% that rate of growth increases exponentially. But biological activity also increases with temperature- the rate of growth at 40F/70%RH isn't anything like what it is at 80F/70% RH. Below 60% RH mold growth is pretty slow at any temp.

2>The outdoor air increases in humidity during the warmth of the day, and drops overnight as dew condenses moisture out of the air onto surfaces that are radiation-cooled to the dew point. When all else is equal, the lowest outdoor humidity occurs just before dawn, and the peak occurs in the late afternoon several hours after the peak temperture of the day The outdoor air humidity also varies with wind & weather patterns moving in different bodies of air, so it's not a closed system that will be identical every day. If you're attic is leaking outdoor air at 0.5-1 air exchanges per hour (ACH) it will have the lowest absolute humidity in the morning, and highest in the afternoon. roughly tracking the outdoor humidity. And yes the moisture content of outdoor air can easily be 2-3x that of the conditioned air.

If the attic reads consistently MORE humid than the outdoor air, you have significant moisture sources coming from internal sources. Bathroom fans, kitchen fans, and plumbing vent penetrations to the attic would be in the top 5 list. Sealing both the vent stacks themselves and where they penetrate the attic are important. In decades past it was fairly common (but bad practice) to terminate bath & kitchen exhaust directly into ventilated attics to avoid having to penetrate the roof, and when converting to a sealed unvented attic it could create this sort of an issue, boosting the humidity in the attic with every shower.

If outdoor air is leaking into the conditioned space from the first floor &/or basement and leaking out the ceiling/attic floor, the AC will dry it out, and the outdoor air leaks into the attic are diluted by the drier air from the conditioned space. The tighter the conditioned space is, the less dilution takes place, and the more closely the attic tracks the outdoor humidity.

Water vapor doesn't "travel up" or stratify due to it's buoyancy factor over the very small altitude differentials within a building, rather it moves from higher vapor pressure to lower. So when your attic is leaking humid outdoor air in, even if you have some amount of air leakage from the conditioned space, the vapor drive will be from the humid attic to the drier conditioned space. A vapor retarder/vapor barrier slows this already quite slow process within building assemblies, but even through latex paint on gypsum the amount of moisture that can move by vapor-diffusion alone is miniscule compared to even a square inch of air leak.

If you're going through the expense of a blower door test, first seal every air leak to the exterior that you know of (the most important being the foundation & foundation sill, and the attic floor/upper-floor ceiling penetrations, since those generate the largest stack-effect infiltration drives), and be prepared to seal all air leaks to the exterior, not just the attic. Any utility chases that run from basement to attic are also good to seal up, but not critical to your immediate goal of blocking outdoor air from entering the attic. My suspicion is that the soffits are leaking, since they're often hard to get to properly with the foam (the lower the roof pitch, the tougher it is), and that you have a wind-generated cross flow of outdoor air in to the attic combined with denser-cooler conditioned air leaking out on lower floors sucking humid air into the attic by those soffit leaks. ( cood b rong, offen am. ) Any air leakage at any level in the house is worth fixing, but the bigger leaks are obviously the highest priority. Don't get concerned about making the place "too tight", since in retrofits on older construction it's usually pretty tough to get down to even 3 ACH at 50 pascals (the IRC 2012 maximum leakage levels) in a single round of air sealing.

Thank you for the advice. We are replacing windows right now and a within a month or so it should be done. I will then make sure to seal all possible leaky spots and then we'll do the blower door test on the entire house and perhaps, add smoke in the attic to see if there are leaks. Meanwhile, I am collecting the data on the humidity - I'll post it sometime next week.