Why the number on the bag is not the number in the wall
Insulation is rated in isolation. A batt labelled R-21 really does resist heat at R-21 — in the middle of the cavity, where nothing but insulation stands between inside and outside. But a framed wall is not all cavity. Studs, plates, headers, corners, sills and partition intersections are wood, and wood conducts heat roughly three times faster than the fibreglass beside it.
The result is a thermal short circuit. Heat that meets a stud takes the easy path, and because U-factors add in parallel while R-values add in series, a relatively small area of wood does a disproportionate amount of the damage. A 2 × 6 wall with R-21 batts, a normal 23% framing factor and ordinary sheathing and finishes performs at about R-17.5 — a sixth of the labelled value lost before you account for a single air leak.
That is what this calculator returns alongside the material count. The bag count tells you what to buy; the assembly R tells you what you bought. Energy codes increasingly specify a maximum U-factor rather than a minimum R-value precisely because the U-factor is the number that describes the wall you actually built.
The parallel-path method, term by term
Take the wall apart into two paths and treat them as resistors in parallel.
The cavity path runs through the insulation. Its resistance is the batt's R-value plus everything outside the cavity: the interior air film, the gypsum board, the sheathing, the cladding, the exterior air film. Those layers are worth roughly R-2.5 on a typical sided wall and around R-1.1 on an attic floor, where there is no exterior cladding and the attic itself is vented.
The framing path runs through the wood. Softwood framing is about R-1.25 per inch, so a 5½ in stud is R-6.9. Add the same other layers and the framing path is about R-9.4 where the cavity path is R-23.5 — two and a half times worse.
The weighting. Convert each path to a U-factor (1 ÷ R), multiply by its share of the area, and add. Then invert the total to get back to an assembly R. Written out: U = f/Rframe + (1 − f)/Rcavity.
The framing factor, f. This is the input people get wrong. The studs alone at 16 in on centre are only 1.5 ÷ 16 = 9.4% of the wall. But a real wall also has top and bottom plates, a second top plate, headers over every opening, jack and king studs, corner assemblies and partition intersections. Add those and typical residential framing lands near 23% at 16 in on centre and near 20% at 24 in on centre. An attic floor is far leaner — only the joists interrupt the blanket — so 10% is a reasonable figure there, and less once the blanket buries the joists entirely.
What the method ignores. Parallel-path assumes heat flows straight through without spreading sideways. Real heat does spread, which makes true performance slightly better than this calculation for wood framing and considerably better than the alternative isothermal-planes method for steel. For wood-framed residential work, parallel-path is the standard approach and the one the ASHRAE Handbook — Fundamentals presents first.
Worked example: 1,000 ft² of 2 × 6 wall with R-21 batts
A 1,000 ft² wall area, 2 × 6 studs at 16 in on centre, R-21 fibreglass batts, bags covering 67.8 ft², a 23% framing factor, and R-2.5 of gypsum, sheathing, siding and air films.
- Cavity area. 1,000 × (1 − 0.23) = 770 ft². That is what the batts have to fill.
- Bags. 770 ÷ 67.8 = 11.36 → 12 bags.
- Cavity path R. 21 + 2.5 = 23.5.
- Framing path R. 5.5 in × 1.25 = 6.875, plus 2.5 = 9.375.
- Framing path U. 0.23 ÷ 9.375 = 0.024533.
- Cavity path U. 0.77 ÷ 23.5 = 0.032766.
- Assembly U. 0.024533 + 0.032766 = 0.057299 Btu/h·ft²·°F.
- Assembly R. 1 ÷ 0.057299 = R-17.45.
Read step 5 against step 6. The wood is 23% of the area but carries 43% of the heat flow (0.024533 ÷ 0.057299). That is the whole argument for continuous exterior insulation: a layer of rigid foam outside the sheathing sits in both paths, so it raises the framing path and the cavity path together and there is no short circuit left to exploit.
Now try the substitution most people reach for. Swap the R-21 batt for an R-23 mineral wool batt of the same thickness. The cavity path becomes 25.5, the framing path is unchanged at 9.375, and the assembly comes out at 1 ÷ (0.024533 + 0.77/25.5) = R-18.10. Two points of label bought you 0.65 points of wall — because the framing path never improved.
How to read the result
Start with the ratio of assembly R to nominal R. In the worked example it is 17.45 ÷ 21 = 0.83, so the wall delivers 83% of the label. A ratio in the low eighties is normal for a wood-framed wall with a cavity-only strategy. If yours is much lower, either your framing factor is high — a wall full of windows and corners — or your batt is thick relative to the framing, which is the regime where the wood path dominates.
Then look at the table this calculator builds. In batt mode it prices every common batt R-value in your assembly and shows the fraction of the label each one delivers. The fraction falls as you go up the list, and that decline is the honest answer to "should I upgrade the batt?" — past a point, the money belongs in exterior foam or in air sealing instead.
In blown mode, read the depth as well as the R. The bag chart on a loose-fill product gives coverage at a settled depth, and settling is real: fibreglass and cellulose both compact over the first months. Use the settled figures, install to the depth markers, and take the reading in the open field, not over a joist.
Finally, treat the U-factor as the number to quote. Energy codes give both a prescriptive R-value path and a U-factor alternative, and the U-factor path is where the assembly figure — not the label — is what counts.
R-value per inch of common insulation materials
| Material | R per inch | Depth for R-49 | Where it is normally used |
|---|---|---|---|
| Fibreglass batt, standard density | 3.1–3.4 | 15.8 in at 3.1 | Framed walls, floors, rafter bays |
| Fibreglass batt, high density | 3.6–4.3 | 13.6 in at 3.6 | 2 × 4 walls needing R-15 |
| Mineral wool batt | 3.0–3.3 | 16.3 in at 3.0 | Fire and acoustic separations, exterior board |
| Loose-fill fibreglass, blown | 2.2–2.9 | 22.3 in at 2.2 | Attic floors |
| Loose-fill cellulose | 3.2–3.8 | 15.3 in at 3.2 | Attic floors, dense-packed cavities |
| Open-cell spray foam | 3.5–3.7 | 14.0 in at 3.5 | Unvented roof decks, rim joists |
| Closed-cell spray foam | 5.8–6.8 | 8.4 in at 5.8 | Rim joists, crawl spaces, tight cavities |
| Expanded polystyrene (EPS) board | 3.6–4.2 | 13.6 in at 3.6 | Continuous exterior insulation, under slabs |
| Extruded polystyrene (XPS) board | 5.0 | 9.8 in | Below grade, under slabs |
| Polyisocyanurate board | 5.6–6.5 | 8.8 in at 5.6 | Continuous exterior insulation, low-slope roofs |
The depth column is simply 49 ÷ the lower end of each range, which is the conservative choice. Polyisocyanurate is the exception to trusting a single figure: its R per inch falls at low temperatures, so a roof assembly is normally derated in cold climates.
A compressed batt loses more than you think
Batts are rated at a specific loft. Push an R-19 batt, designed for 6¼ in, into a 5½ in 2 × 6 cavity and you do not get R-19 — you get roughly R-18, because you have lost thickness faster than you have gained density. That is why R-21 exists: it is the batt engineered to hit its rating at 5½ in.
The opposite error is worse. A batt that does not fill its cavity leaves an air gap, and an air gap in a vertical cavity sets up a convection loop that carries heat from the warm face to the cold one. Wiring, plumbing and electrical boxes create the same voids on a smaller scale. Split the batt around obstructions rather than compressing it past them, and treat gaps and compression as an installation-quality issue — the difference between a Grade I and a Grade III installation is a substantial fraction of the label.
Mistakes that make the number wrong
- Averaging two R-values. R-values are not additive across parallel paths. You must convert to U, weight by area, add, and invert.
- Using the stud-only framing factor. Studs at 16 in on centre are 9.4% of the wall; a real wall with plates, headers and corners is closer to 23%. Using the low figure overstates the assembly by several points.
- Ignoring the other layers. Air films, gypsum and sheathing are worth around R-2.5 on a wall, and they help the framing path proportionally more than the cavity path.
- Sizing blown-in by bag count instead of by depth. Bags are sold by weight, and coverage depends entirely on the depth you install. Set depth markers and check them; an under-blown attic is invisible from below.
- Forgetting the attic floor is not the ceiling area. Knee walls, dropped soffits and mechanical chases all need insulating too, and none of them appear in a flat plan measurement.
- Blocking the soffit vents. Loose fill pushed into the eaves cuts the ventilation path and invites condensation. Baffles go in first, always.
- Treating R-value as the whole story. Air leakage moves more heat than conduction in many older houses. Seal the top plates, chases and can lights before you add depth.
What this calculator assumes
It assumes one framing depth, one insulation product, a single uniform framing factor, and heat flow perpendicular to the assembly. Framing is taken at R-1.25 per inch, which is the usual figure for softwood at typical moisture content. It applies the parallel-path method, not the isothermal-planes method — for steel framing, where lateral heat spreading is dominant, parallel-path substantially overstates performance and you should use the correction factors published for cold-formed steel assemblies instead.
It does not model thermal bridging at rim joists, band joists, cantilevers or slab edges, all of which are separate details. It does not model air leakage, moisture, or the temperature dependence of foam R-values. It does not check code compliance: your climate zone sets the required R-value or U-factor, and that requirement is a legal minimum, not a design target.
Bag counts assume the coverage figure you enter is accurate for your product and that installation is to the rated loft. In blown-in work the single biggest error source is depth, not arithmetic — measure it.
Where insulation sits in the build
Insulation goes in after rough-in and before board, which means the framing decisions that set your framing factor were made weeks earlier. Advanced framing — studs at 24 in on centre, two-stud corners, single top plates, insulated headers — exists precisely to cut that factor, and it is the cheapest R-value in the building because it removes wood rather than adding material. If you are still laying out walls, the wall stud count calculator shows what the spacing change does to the stud order.
Once the cavities are filled, the next trade is board. The drywall sheet calculator takes the same room dimensions, and the paint coverage calculator follows it. On the exterior, continuous insulation sits between the sheathing and the cladding, so it changes the fastener lengths and the trim details for the siding takeoff.
For a whole-house retrofit, the highest return is almost never a thicker batt. It is air sealing the attic plane, then bringing the attic to the depth your climate zone requires, then dealing with the walls — in that order, because attic work is cheap per square foot and the attic is where the stack effect drives the largest losses. Price the labour side with the construction labor hours calculator before you commit to a wall retrofit.