fiberglas r-value loss in sub zero temps

wild01 · December 2009

I've come across several sites (and a book) that claim us gov't studies show that fiberglass insulation loses half it's R-value in sub zero (f) temps. anyone know anything about this?? most specified blown in fiberglass, but some didn't specify, just added an addition using fiberglass batts, wondering if I f-ed up.

supposedly foam and cellulose don't lose R-value in neg temps.

would explain why log cabins seem so much easier to heat than stick and frame...

stats for where I live

winter high temperatures: 34.3 degrees F.
winter low temperature: -2.1 degrees F.

this is seasonal average of course. highs in the 50s are not uncommon and lows in the -20s

oh and winter here starts in oct.

Cariboocoot · December 2009

Re: fiberglas r-value loss in sub zero temps

Fibreglass loses R value in sub zero temps? I've never heard that before, and can't see any logical reason why it should. The insulative factor is the trapped air between the strands, not the glass itself. The same for foam insulation; it's the bubbles, not the plastic. I've lived with sub-zero Winter temps most of my life. The only problems are with a poor insulation job, not the type of insulation. Air leakage, lack of vapour barrier, insufficient R value are the culprits.

As for heating a log cabin being easier .... Nope. I've got one with 12" walls; R value of about 12 (wood is approximately R1 per inch). Not much different from a conventional 2 x 4 stud wall, except for the gaps where the logs don't fit together properly (I'm working on it - it's 50+ years old). The only thing that makes it easy to heat is having acres of dead pine trees to cut down, split up, and shovel into the stove fast enough to keep the interior temp up ahead of the heat loss.

Now if you were in Baffin Bay, for example, with Winter temps around -70C ... Could be a whole 'nother issue. But I don't see how just dropping below 0 would suddenly cause a 50% drop in insulative value. If anything, it would be a gradual decline.

Solar Guppy · December 2009

Re: fiberglas r-value loss in sub zero temps

Never heard this either, links please

Earthandsol · December 2009

Re: fiberglas r-value loss in sub zero temps

This is discussed in Home Performance Contracting and Home Energy Rater circles.

Also Home Energy magazine ( the first link ) is a great resource ( I liken it to Home Power )

http://www.homeenergy.org/archive/hem.dis.anl.gov/eehem/92/920510.html

http://www.ornl.gov/sci/roofs+walls/insulation/ins_02.html

Cheers,
Chris

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

The reason a log structure "seems" easier to heat, is that the logs are a ton on thermal mass compared to GWB or Plaster wall. The building warms up slowly, heating the mass of the logs, and then cools down slowly. A big issue with "old fashioned" log buildings is they were very hard to get air tight, especially in the days before spray foam.

I have never heard of fiberglass losing r-value

Tony

On the other hand a quick search reveals:http://www.askthebuilder.com/B54_Insulation_-_Fiberglass_and_Cellulose.shtml

http://www.houleinsulation.com/cellulose_vs_fiberglass.html

http://www.insulationusa.com/cellulose_vs_fiberglass/

None of these after an admittedly quick scan explain why fiberglass loses R-value. It smells a bit like urban legend.

PS On yet another hand, from the previous posts link :"The investigations confirmed that natural convection occurred within the insulation itself. The reason is that air densities changed with the temperature differences. As warmer air from the heated space below reached the top of the insulation, the air cooled, became more dense and fell back into the insulation. As part of these observations, the researchers made infrared scans of the upper surface of the insulation. The heat patterns resembled the traditional hexagonal Benard cell pattern of natural convection occurring in fluids heated from below (see Figure 4)."

"In all cases with the convective cover experiments, thermal performance of the loose-fill insulation improved. Open questions remain about vapor entrapment, the additional costs of the covers, or the possibility of other remedies to unwanted convective flow. When researchers are able to answer these and similar questions, manufacturers will probably develop more "convection retarder" products. These will be useful for retrofitting fiber-glass insulation; but for new installations, cellulose alone, while less expensive, will out perform fiber-glass loose fill, especially in cold climates. Of course, only two types of insulation have been tested before and it is probably too soon to jump to conclusions."

My best guess is that it is a moisture related issue,

T

BB. · December 2009

Re: fiberglas r-value loss in sub zero temps

I think they are all adding leaky homes (no vapor barrier) and the potential for water logging.

This story is on a few websites and is used to justify spray-in urethane type foams.

I guess it could be true--in the end check references (trust, but verify).

-Bill

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

It is interesting that in the one test, blow in cellulose didn't suffer from the drop in recorded R-value.

Tony

russ · December 2009

Re: fiberglas r-value loss in sub zero temps

The polyurethane site is an interesting mishmash of misunderstood truths, old wives tales and sales blather.

They seem to say standard testing should cover all contingencies and conditions which is almost never the case.

They equate all expanded polystyrene as equal which it is not.

The example of the potato warehouse is a misapplication of fiberglass - nothing more or less. İt is not suitable for a wet or high humidity application.

The example of a home not really needing insulation in the walls if well sealed is pure BS. Of course more heat is lost through the ceiling as that is where more heat is concentrated. Heat will radiate equally in all directions.

The R values D.B. South (author of the opinionated sales pitch presented as technical fact) seem to be a mishmash of numbers collected over the years.

The comparison chart he offers uses the correct numbers though he never mentions if he is actually using closed cell polyurethane spray which he should be. Open cell is not of much use once moisture is present whatever type of insulation. The more moisture the more problem - potato warehouses have very high humidity.

They are not really saying that fiberglass loses R value at low temperatures - they are saying that with a higher temperature differential between the two sides the R value drops off. That would be correct. Whether the percentage they give is correct or not İ don't know.

His comparisons of polystyrene and polyurethane are the product of a vivid imagination! You can find the specs for Dow closed cell XPS high density board on the net.

To wind up a long story, İ used Dow XPS high density polystyrene board on the outside of our new home - 50mm on the walls and 150mm in the attic. This summer the attic stayed comfortable even on our hottest days. Now we are in winter and though it does not get all that cold here İ am quite satisfied (so far).

niel · December 2009

Re: fiberglas r-value loss in sub zero temps

i'm not an expert on insulations, but maybe the convective losses where there in the beginning as the air spaces are larger in fiberglass insulations. leakage from 1 space to another becomes more prominent as temperature differencials become larger. think of a 1/2in hole in a wall during various temperatures. as it gets lower in temperature the effects of that hole are more pronounced as airflow (convection) increases.

westbranch · December 2009

Re: fiberglas r-value loss in sub zero temps

from Earthandsol's second link, a gov document.

"When attic temperatures drop below 0°F, some low-density, fiberglass, loose-fill insulation installations may allow air to circulate between the top of your ceiling and the attic, decreasing the effectiveness of the insulation. You can eliminate this air circulation by covering the low-density, loose-fill insulation with a blanket insulation product or with a higher density loose-fill insulation."

I have seen this in my attic, where air from inside the house, over 30 years of living here , found small holes through which it could flow , and left trace marks of black dust, we used to heat with wood, in the fiberglass. Some of the marks were quite large when compared to the size of the 'hole' in the vapour barrier...

to stop the 'bridging effect' I laid a second course of batt fiberglass at 90 * to the batts between the trusses. seemed to work quite well.

Had the house 'assessed' for improvements we could make, the best was to plug all the holes where the house wiring goes through the header plates down to the switches and stop air infiltration inside the envelope. you would be amazed when you have a negative air pressure test done, how much cold February air leaks in in through those switch boxes...

As to the 'circulation ' within the fiberglass, I can believe it, but I will leave it to a thermal engineer types to advise on it.

As to the sealing effect of spay on foam, this I am an ardent supporter of, sealing everything, having witnessed personally the air flow into our new guest house (16x18 cabin) around the edges of some unsealed plywood sheets. The ones I used sealant on before I placed them had no air leaks. Also when the window casings/ openings were sealed with low expansion foam, bingo, the ambient temp went up immediately just from body heat...

Eric

Cariboocoot · December 2009

Re: fiberglas r-value loss in sub zero temps

As I first mentioned, it is the air that does the insulating not the material - so the theory that circulation can occur between the air pockets in loose fill or even batt fibreglass allowing heat loss is valid. But are their examples realistic?

One of the improvements I made to the Langley house was to cover over the exposed exterior surface of the fibreglass insulation. This is a Tudor-style place and the upper level's walls are exposed to air circulation which would 'suck the heat out'. But this can only be called an improper installation; had at been built right or were it a typical two-story structure, the exterior surface would have been sheathed and protected against such air evacuation.

Do air currents and/or heat convection occur inside properly installed fibreglass insulation? Physics says that to some extent they must do so. Is this likely to have a significant effect on the R value? Practical experience says no. Some of the testing examples cited clearly were not valid for reasons others have already mentioned. You have to exclude the possibility of moisture contamination for one thing: everybody knows the main purpose of vapour barriers is to keep the humidity from traveling into the insulation and condensing once it reaches dew point, thus saturating the fibreglass and turning it into a useless soggy mess.

Sounds to me like 10% real science and 90% sales hype.

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

I've seen what Eric describes as "the dust trail" many times. I think the issue of exfiltration (probably not a word) of air as well as heat through porous surfaces is to blame. If you figure that the dust get through (over time) think about how much entrapped moisture gets through. Good vapour barriers, (especially around ceiling penetrations like fans/lights etc are essential.

Remember, vented attics might produce a low pressure in the attic, which if the intake venting were not up to snuff, might force air (and heat) out from under the ceiling.

Tony

EDIT PS. The only problem with placing a batt over blow in, I think the weight of the batt would tend to compress the blow it. You would have the full R-value of the batt, but a somewhat reduced R-value of the blow in. Ideally you might have a batt on the bottom, with blow on top. Or add an ~r-11 batt on top of R-40 blow.

Cariboocoot · December 2009

Re: fiberglas r-value loss in sub zero temps

icarus wrote: »

EDIT PS. The only problem with placing a batt over blow in, I think the weight of the batt would tend to compress the blow it. You would have the full R-value of the batt, but a somewhat reduced R-value of the blow in. Ideally you might have a batt on the bottom, with blow on top. Or add an ~r-11 batt on top of R-40 blow.

I'm glad you mentioned that, Tony. Experiments with compressing fibreglas insulation have shown that slight compression actually improves its insulative value. I don't remember what the final number was they came up with, but I think it was from putting 3 1/2 batts into 2 1/2 studwalls (or that was the original question - I'm getting fuzzy again). Too much compression reduces the R value, as you said.

In a related note, the difference in R value between the studs in a wall and the insulation in the voids can also show up. In the frozen climes where 6" walls with R19 batts are now standard, you sometimes see lines on the walls where moisture condenses on the slightly colder stud locales and collects dirt. It's not enough to cause any real trouble, but is unsightly. If a foam insulation is used the difference is even more drastic, and may even be seen on a standard 4" wall.

Super-insulated houses sometimes go with two separate stud walls, and a space in between which is continuous insulation. Another popular work-around is to cover the exterior sheathing with rigid insulation, giving R10 on the outside in addition to the R11 standard. Since this first layer 'blocks most of the cold' (really it's heat loss it blocks) the interior 'stripes' don't show up as the wall is still warm enough at all points to prevent the mild condensation.

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

One of the best ways I've every used (IMHO of course) is 2x6 walls, with 1/2" foil face polyiso-board on the out side of the stud wall, in lieu of plywood sheeting. (Shear value of sheeting needs to be replicated by let in bracing, or structural panels or other engineering). This way you get a R-value of R-25 inclusive, R-21 hi-density batts, with a ~R-4 from the thermax board. The great advantage is that you can thermal break all the framing lumber using the thermax, including not only the studs, but the plates, the headers, as well as rim band joists. It may not seem like much, but the lumber itself, while being a fairly good insulator, is also conducts heat or cold through it.

An somewhat easier way to accomplish this is putting thermax on the inside instead (or as well!) but your run into trouble with jamb thicknesses, electrical box rises etc. (Same on the ceilings)

Tony

Merry Christmas, -5 with 30 cm of snow predicted tonight and another 30 or so in the next day or so!

wild01 · December 2009

Re: fiberglas r-value loss in sub zero temps

Building Green: A Complete How-To Guide to Alternative Building Methods (Paperback)
Clarke Snell (Author)
Tim Callahan (Author)
pp426 3) R-value. Weeding through the many different R-value claims listed for both materials, it seems that cellulose has a slightly higher per inch R-value rating than fiberglass batts. In addition, several studies, including one conducted by the united states department of energy, seem to document that fiberglass insulation looses R-value significantly as the temperature drops. In the DOE tests, for example, an exterior temperature only 25F lowerthan the interior temperature resulted in almost a 10% drop in R-value. When the temperature differential increased to 88f, the fiberglass R-value was reduced to less than half it's rated value. In the same test, the R-value of blown cellulose was unaffected as it got colder.

this is the original thing that got me looking.

from looking at the links provided it looks like it only takes place with blown in loose fill glass and not in batts, so I probably don't have to worry. I have a cut in roof and used batts between 16inch on center studs. I places 3 layers of r13 (got it way cheap store closing sale) in a 2x10 cavity.(which gives me a slight crush, about an inch and a half) and 1 layer radiant barrier followed by 1X6 t&g spruce on the inside. roof sheeted in 7/16" osb.

I got really creative with the walls after I realized 2 2x4 studs were cheaper than 1 2x6 stud. I built 2x4 exterior walls insulated them with r13 and then added radiant barrier and another 2x4 framed wall (with the studs offset from the first to prevent thermal bridging) cross braced the walls from in between the 2 to eliminate the need for sheeting, sided the outside with 1x6 shiplapped pine, added a second layer of r13 inside followed by t&g blue stain pine.

so all told
R26+radiant=walls +r3 for wood panelling and siding
R39-1.5"crush+radiant ceiling +r1.5 for paneling

so yeah after all that work I was a bit concerned with the above book chapter.
it is somewhat a relief to know they were only referring to loose fill.

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

Wild01

If I read you correctly you have three layers of r-13 in a 2X10 cavity, with a bit of compression? Is this right? If this is right, I would be very concerned with the lack of breathing air space on the unconditioned side of the batt. It has always been considered (in my experience) to have at least 1" free air space above the insulation, with intake vents at the soffit, and exhaust at the ridge.

If I am understanding your design properly, you have no vent space at all. I think this is a problem waiting to happen. Even with a "perfect" vapour barrier, some moisture will come through from the heated side. As the temperature in the batt nears the dew point, the water in the air will condense, causing moisture to build up under the roof deck, hastening rot, and encouraging mold. It will also make the roof material too retain too much heat as no air moves below the shingles, hastening shingle/asphalt failure. Metal roofing is not a problem in this scenario.

The only roof structure that I know off where it is acceptable to eliminate venting is when one uses rigid foam, because of it's basic vapour barrier nature.

Please let us know if I am understanding you design wrong,

Tony

wild01 · December 2009

Re: fiberglas r-value loss in sub zero temps

Not really misunderstanding my situation, just my environment, moisture here is only a problem in getting any. this is an extremely dry area, I realize that in 90% of the us this would be a problem but it is common practice here, and having re-sheeted very old roofs in the area I can say moisture retainment isn't a problem. (snow evaporates here, you can line dry your clothes in winter-when they start moving in the wind rather than being stiff-time to take them it)

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

Very good,, where's here?

wild01 · December 2009

Re: fiberglas r-value loss in sub zero temps

san luis valley colorado 7800 ft

Cariboocoot · December 2009

Re: fiberglas r-value loss in sub zero temps

Air circulation under roof sheathing is important not just for moisture considerations. Without it, the roof because very, very hot. Roofing material will fail prematurely and rafters/joists/trusses will dry rot. You can actually see sap boiled out of pine wood put in such an unventilated environment.

icarus · December 2009

Re: fiberglas r-value loss in sub zero temps

I would also add, that adding 3 layer of 3 1/2" R-13 into a 93/8" cavity probably yields no better net R-value than just 2 layer or R-26. Somewhere I had a table of the effect of compression on fiberglass, and it is surprising how fast you lose R-value as you compress. It is important to remember that an R-13 3 1/2" batt is already a compressed R-11 batt, known as a High density batt, so I suspect that doing three layers you might actually be lowering the R-value.

One might consider, using a R-19 5 1/2" batt, with a R-13 31/2" making a total of R-32 (with no decrease in R-value) filling the 9 3/8 cavity to 9". Not ideal, but I bet it would perform better.

It is also important to remember if you are going to stack batts, if these batts are vapor barrior faced (either foil or paper) that only the layer nearest the condition space should have a VB. If the second layer has a VB, slit it or remove it, or you run the risk of trapping moisture between the two VB layers.

On a related side note, all too often in floor assemblies, I see where people have install insulation under the floor, between the joists. More times than not, the have stapled the paper/foil to the under side of the joists, resulting in the vapor barrier on the wrong side. The proper installation is paper/foil up, holding the insulation in place with either insulation wires, or nylon string, or lath.

Tony

fiberglas r-value loss in sub zero temps

Comments

Categories