I was having a discussion elsewhere and asserted that blown cellulose is a better choice for ceiling insulation because its more dense and less prone to convection inside the blanket thereby reducing the R value...however I can't find proof.

Is there a technical reference that refers to that phenomenon or is it just anecdotal?

Hoping I was right :)

thanks in advance

http://www.foam-tech.com/theory/rvaluedrift.htm
http://www.osti.gov/bridge/servlets/purl/451183-MCW2Pf/webviewable/451183.pdf

Thanks

Greetings,

I don't think you want to use either.
FG is probably the most inefficient insulation available. With out ac on the interior temps can exceed the exterior temps on a 95+ deg day. Cellulose is somewhat better but can hold up to 3x the moisture FG does and that increases its energy flow by about 75% according to a NBS test report in 1977.

A reflective insulation sys (RI) is far more efficient and if the advanced installation method is used a 3 layer sys will only radiate about 2 btu/hr/sf to the drywall. See Mech. Eng. Handbook for calculations where as a FG installation can radiate about 37 btu/sf/hr on a 95 + deg day.

RI is not subject to the problems which cripples FG or cellulose insulation.

Show me the NBS report from 1977.

Cellulose doesn't lose much insulation value until/unless it's over 20% water by weight, and it can hold WELL over 3x the moisture of fiberglass (which holds very little.) It would take a roof leak or truly atrocious levels of air leakage at the ceiling to come anywhere near the level at which cellulose loses R value, but high levels of moisture cycling well below that can still induce settling in low density blown cellulose. At a sufficiently high density the dimensional creepage from moisture cycling is no longer an issue, which is one of several reason why dense-packing wall cavities is superior to low density installations when only cavity insulation is used. (With exterior insulation outside the structural sheathing sufficient to protect the sheathing from winter moisture absorption the seasonal moisture cycling within the cellulose is much reduced, and lower densities will hold up over time.)

It's fairly common to see settling of cellulose in attic installations over time, and the rated R is at the 10 year settled density (it'll have higher initial depth & performance , but topping it off after 20-30 years to bring it back up to spec is pretty cheap. "Stabilized" wet-sprayed cellulose blown with adhesives may reduce the rate of settling in low density attic apps, but won't eliminate it in high humidity climates. It does a much better job at halting settling in mid-density wall applications. (Cheaper than dense packing, with similar longevity, especially in wall stackups with exterior insulation as well.)

Thanks, I just had an energy evaluation done on my 35 year old house and its not good.

I have to add some serious insulation values to the attic, I've used fiberglass before but wasn't very impressed and am looking at cellulose instead. My current attic has R19 or so (wood chips and un-faced fiberglass bats, code is R50 and I would like a bit more than that.

I had a discussion with a guy at Home Depot and based on what I have read here (lurker for some time ) I insisted that cellulose would be my choice instead of fiberglass. The guy (salesman) was aghast when I suggested that that the looseness of the fiberglass layer made it susceptible to convection currents (in the insulation) at the coldest temperatures (we get -40c outside and +22c inside).

I appreciate the guidance and will be sticking with Home Depots blown in cellulose

Don't use a cellulose that has Ammonium sulfate in it as most of the cheaper celluloses do.

Put a combined 14-16" in the attic and you will see a world of difference. You should air seal prior to putting in more insulation.

What WindowsonWashington said- insist on "sulfate-free" or "borate-only" cellulose. If there's ever a roof leak the sulfates will cause severe corrosion on any metals- particularly copper (wiring or plumbing), but it'll attack nails too.

All stabilized formula versions are sulfate free (or at least supposed to be, although it's been alleged that with some manufacturers only the graphic on the bag changes), as they're designed to be wet-sprayed (but can still be dry blown.) I tend to use a local (to me) manufacturer's products since NONE of their goods use sulfates, and if the production facilites never buy the stuff, it can't accidentally or intentionally show up in the wrong bag. The price difference between sulfated and borate-only goods is miniscule, but the goods sold at my local Home Depot DO contain sulfates. YMMV. A handful of product in a plastic cup with a few Drano pellets and water will outgas some pretty pungent stuff (ammonia) if it contains sulfates, if you're not sure.

Greetings,

QUOTE>Show me the NBS report from 1977.

The report is: Retrofiring An Existing Wood Frame Residence for Energy Conservation --an Experimental Study NBSIR 77-1274 By: D.M. Burch & C.M. Hunt Date: 1977

QUOTE>Cellulose doesn't lose much insulation value until/unless it's over 20% water by weight,

SO, according to you the law of physics is suspended. No more conduction, no more storage, no more transfer. All of you that believe that, please raise your hand.

QUOTE>and it can hold WELL over 3x the moisture of fiberglass (which holds very little.)

According to the report: the inside moisture for FG was 3% the out side moisture 15.3%
The inside moisture for cellulose was, 3% i/s and 20% o/s

I have seen FG saturated, that is wet to touch, from condensation, under summer drought conditions.
Moisture in either product DRAINS and increases the moisture level at base of wall. With FG this causes the % of moist' to exceed the 15% level required for wood decomposition. This is why you see dryrot(?)at bottom level of FG walls and increased fire acceptability at upper levels of cellulose.
You can't fool Mother Nature.

Using RI eliminates all these problems and you save more energy. You still haven't responded to my 2btu/hr/sf challenge.

The report doesn't seem to be available online.

http://catalog.hathitrust.org/Record/009487465

http://www.worldcat.org/title/retrofitting-an-existing-wood-frame-residence-for-energy-conservation-an-experimental-study/oclc/003290701

It was referenced by one subsequent researcher at TAMU in 1985 and there may be others, but I haven't been able to find them.

At 20% moisture by weight cellulose is just starting to become saturated, and barely feels damp to the touch. Cellulose fibers are tubular at the micro-level, and will wick liquid moisture into the individual fibers with negligible effect on it's thermal conductivity. Only when the moisture can no longer wick does it begin to lose effectiveness. Below that point it's thermal properties are fairly constant.

"I have seen FG saturated, that is wet to touch, from condensation, under summer drought conditions."

I can believe that- happens all the time. Air leaks from the exterior into a f.g. insulated studwall in an air-conditioned building can eventually saturate when the outdoor dew points stay above the temperature of the interior wall, and the wall has an interior-side vapor retarder.

Similar air leaks in low-density cellulose wall-installations can also cause settling issues, but evidence of saturation & wood rot from anything but bulk-water leaks just isn't there- the cellulose is protective of the wood by it's hygrophyllic nature.

At 3lbs+ density the air-retardency of cellulose is high (higher than any fg batt, comparable to new-school micro-fg at ~2lbs density), and the buffering capacity high. With less than 1" of the air movement that would occur with low density batts it takes very long time for the volume of air-transported moisture that reaches material below the exterior dew point, and when it does it's absorbed by the fibers without consequence. When the outdoor dew points fall, the buffered moisture is released as water vapor. Interior-side vapor retarders usually increase rather than decrease the moisture content in wall assemblies though- it should be allowed to dry in both directions. Moisture transfer & buffering of cellulose (as well as other materials) is well-modeled with WUFI (a modeling tool for moisture transfer in building materials, based on data from years of academic investigation. Try it: http://www.ornl.gov/sci/btc/apps/moisture/ibpe_sof161.htm )