i am not sure either. but one thing i can tell you is that if that door is not 100% airtight against the opening when closed, you just wasted a bunch of money.

Perhaps the following will help. I did not write it so don't ask me to explain it.

R-value is a term predominantly used in the building industry to rate the insulative properties of construction materials and building assemblies.

It is derived from the U-factor (see below). The higher the R-value, the greater insulation value.

The relationship between U-factor or R-value and thickness is not always exactly linear and therefore its value cannot be precisely extrapolated for a material of different thickness, but assuming a linear relationship is often adequate. In any case, R-values of adjacent materials can be added to determine a final R-value of the entire construction assembly; e.g., R-value(brick) + R-value(fibreglass batt) + R-value(plasterboard) = R value(total)

The SI unit for R-value is kelvin square meters per watt (Km²/W).

The imperial unit for R-value is ft²·°F·h/Btu. The conversion factor is 1 ft²·°F·h/Btu â 0.1761 K·m²/W, or 1 K·m²/W â 5.67446 ft²·°F·h/Btu.

Sometimes the nomenclature 'RSI' is used to denote the SI form of the value. In contrast, the imperial unit is often written as RÂ31.4. To complicate matters, some countries that employ the SI system (e.g. New Zealand) retain the R but incorporate a dash e.g. RÂ5.53. One tenth of an RSI is called a tog.

Some sellers of radiant barrier products spread the rumor that R-value measures only heat transfer from conduction, ignoring radiation and convection. This oft-repeated statement is false. As David Yarbrough, an engineer and insulation expert at R & D Services in Cookeville, Tennessee, has written,"The R-value used to describe thermal insulating products includes heat being transferred by all three mechanisms -- conduction, radiation, and convection. The term used in the thermal insulation community is Âapparent thermal conductivity. A formal definition for apparent thermal conductivity is contained in document C168 published by the American Society for Testing and Materials. The term is applied to situations involving the simultaneous flow of heat by all three transport mechanisms. The statement ÂR-values are measures of conductive thermal resistance is incorrect, if it implies a limitation, since ÂR-value includes radiation and convection when they are present."

R-value should also not be confused with the intrinsic property of thermal resistivity and its inverse, thermal conductivity. The SI unit of thermal resistivity is K·m/W. Thermal conductivity assumes that the heat transfer of the material is linearly related to its thickness.

U-factor (also known as U-Value)

U-factor describes how well a building material conducts heat. Methodologically, it measures the heat transfer of a material of known thickness over a given area under standard conditions. The usual standard is at a temperature gradient of 24oC at 50% humidity in no wind conditions.[1]

U is the inverse of R i.e. U = 1/R and the SI unit for U is W/(K·m²).

For example, if the interior of your home is at 20 °C, and the roof cavity is at 10 °C, the temperature difference is 10 K. Assuming a ceiling insulated to RÂ2, energy will be lost at a rate of 10 K / 2 K·m²/W = 5 watts for every square metre of ceiling.

Holy cannoli!

I'm surprised someone makes a garage door with an R value that high. That's higher than 3-5/8" batt insulation (R-11) and close to 6" batt insulation (R-17). How thick is the door? As David said, your challenge now shifts to the door perimeter. That's a great insulating value, but unless it seals very tightly to the wall in the closed position you won't really make use of the insulating properties.

It really is just 4x times better.
This is tempered by infiltration losses that will reduce the possible improvement.
Higher R value is easily obtained with various foam type insulations, they just cost a lot more.
Polyiso foam runs R-7 per inch.

While it is 4 times better, what does that really mean? When you are talking about heat loss and insulation, you're talking about slowing down the speed the heat travels through something (a door, a wall, a window, etc.). A single pane of glass has about an R-1 value, so, in comparison, your R-4.5 door would allow 1/4.5, or about 22%, of the heat to pass through compared to a window of the same square footage. When you increase the door to an R-16.5 you allow 1/16.5, or about 6% of the energy to pass through compared to the window. You are saving more energy by going to the higher R-value door, but it's not a huge amount becasue the R-4.5 door was already slowing down the heat loss a fair amount.

What this illustrates is that increasing insulation has diminishing returns because, once you slow down heat loss, the amount left to save gets smaller. Even if you had a 15" thick door with an R-value of 100, it wouldn't be that much better than your R-16.5 door. It's nice that you have a highly insulated door, but, unfortunately, you won't see any noticeable difference in your energy bill because: 1) your original door had a decent R-value to start with, and 2) the square footage of your door amounts to only a very small portion of your overall heat loss surface (roof, walls, windows, and floors).

The most important thing is to be sure the door is adequately weather-stripped, because you can lose much more energy through the cracks around a door than you will gain through having a higher R-value door.

Thanks for all of this information and insights. Just checking back into this thread now.

I misspoke on our original door. Apparently, they were R-2 rated. The new doors (Amarr Olympus) are 15.67 or something in that range, for whatever that rating/methodology is worth. From my eyes, there are no gaps or other issues with the weatherstripping (looking outside when garage is dark). We have some empirical evidence that the doors make a 2-3 degree F difference on the floor above the garage (we did pre- and post measurements with the same thermometer). While the surface area of the door is certainly smaller than the rest of the surface area of the room, the floor is 24x24 with an unheated garage underneath it, so we figured upgrading the doors (which needed upgrading anyway) was worth it.

Of course, I'd be lying if I could say that we can "feel" a 2-3 degree difference in the room on the floor temperature, but the empirical pre- to post testing (relatively scientific, but not perfect) makes us feel better for the expense.

One question I missed above - it's a 2" door.