R-value is one of the most quoted numbers in building design and one of the most misleading. A wall specification says "R-19." A load calculation uses "R-19." An architect draws "R-19." A contractor installs R-19 batts between studs, and everybody moves on. But the actual thermal performance of that assembled wall, the number that matters for comfort, energy, and moisture, is often 60 or 70 percent of what the specification implied. This note is about why that gap exists, what it does to the building, and what to ask so the drawings and reality end up closer together.

R-value versus U-value, in plain terms

R-value is the resistance of a single material or product to heat flow. Higher R means better insulator. R-19 fiberglass batts and R-19 mineral wool are both, in a lab, R-19.

U-value is the overall thermal transmittance of a whole assembly (a wall or roof or window with all its layers, studs, fasteners, and edges). Lower U means less heat flow. This is the number that actually shows up in a load calculation.

The two are related but they are not the same. You can have an R-19 batt inside a wall assembly whose overall U-value corresponds to an effective R of maybe 13 or 14. That gap is the difference between the number in the specification and the number the building actually experiences.

A load calculation is only as honest as the wall U-value it uses. When engineers use nominal R-values instead of assembly U-values, the load is understated and the equipment is undersized. Then the humidity note becomes the humidity complaint.

Where the R-value goes

Thermal bridging through studs. A 2x6 wood-framed wall with R-19 in the cavities looks like R-19 on paper. But the studs themselves are only R-1 per inch, and they cover roughly 25 percent of the wall area once you count corners, headers, plates, and rough openings. Averaged across the whole wall, effective R is closer to R-13 to R-15. Steel studs are worse. Steel conducts heat about 400 times better than wood, so a steel-framed R-19 wall can perform like R-7 to R-9. This is why exterior continuous insulation is now required by code in most climate zones. It breaks the thermal bridge.

Compression and gaps. Fiberglass batts stuffed around plumbing lines, electrical boxes, and blocking lose R-value in the compressed and cut areas. A wall installed at 90 percent of designed insulation coverage performs at 90 percent, not 100.

Air leakage through the assembly. Insulation is a heat resistor, not an air seal. An R-19 wall with air leaking through it around the top plate, bottom plate, and penetrations performs as if it were R-8 or worse. This is why blower-door testing on new construction has become standard: without it, air leakage silently wipes out a lot of the design intent.

Fasteners and z-girts. Exterior continuous insulation held to the structure with steel screws or z-girts loses performance through those fasteners. A well-designed detail uses thermally broken clips or minimizes penetrations. A poorly designed one turns R-continuous into R-thermally-shorted.

Edges and corners. Windows, door frames, corners, and roof-to-wall transitions have higher heat loss per foot than the wall field. On typical commercial buildings, edges are 10 to 15 percent of the total heat loss even though they are a small percentage of the surface area.

A concrete example

Figure 01 · Nominal vs assembly R-value on common wall types
Wall typeNominal REffective assembly RLoss
2x4 wood stud, R-13 cavity onlyR-13R-10 to R-1115 to 23%
2x6 wood stud, R-19 cavity onlyR-19R-13 to R-1521 to 32%
Steel stud, R-19 cavity onlyR-19R-7 to R-953 to 63%
Steel stud + R-10 continuous exteriorR-29 nominalR-17 to R-1935 to 40%
Steel stud + R-15 continuous exteriorR-34 nominalR-22 to R-2525 to 35%
The gap is real, and it is large. Continuous exterior insulation helps but does not eliminate it. The load calc should use the assembly U-value, not the cavity R-value.

What actually fixes it

Use assembly U-values in the load calculation. ASHRAE 90.1 publishes assembly U-values that account for studs and typical construction. Load calculations should use those tables or (better) a series-parallel calculation for the specific wall being built. If your engineer says "R-19 in the wall" as the load input, that is the wrong number.

Specify continuous exterior insulation and detail it well. Continuous means continuous. A perfect continuous layer interrupted by 200 steel fasteners is not continuous anymore.

Air seal on purpose. The air barrier is a separate assembly from the insulation. On modern commercial construction, this is typically a sheet-good or fluid-applied membrane at the exterior sheathing. The details at penetrations, transitions, and rough openings determine whether it works.

Blower-door test at construction turnover. The test verifies whether the envelope was actually built to the drawings. It costs a few thousand dollars on a small commercial building and it is the only way to know before the space is occupied.

Design for real edges. Include perimeter heat loss in the calculation. Do not model an all-glass storefront as if it were a wall.

What to ask your engineer

  • What assembly U-values were used in the load calculation, and where did they come from?
  • Was thermal bridging through studs included, or was the cavity R-value used as-is?
  • Is continuous exterior insulation specified, and how are the fasteners detailed?
  • Is there an air barrier detail called out in the drawings, and is a blower-door test in the specifications?
  • How were window frames and rough opening losses modeled?

The one-line version

R-value is what the product is. U-value is what the assembly does. Design and calculate against U-value, verify with a blower-door test, and detail the air barrier as carefully as the insulation.