Is there an easy way to determine the U-value (or R-value) of an existing fitted double glazing system?

Two options spring to mind, looking up the value in some kind of database (sadly I have no idea who the manufacturer of my windows were as my flat had them when I moved in) or trying to measure it experimentally (maybe with a heater on one side and temperature sensors attached to the glass on both sides). As such I was wondering whether such databases even exist, whether measurement is practical or if there were options I hadn't considered.

  • I've made a few custom insulated doors. I just measured one of them with small infrared tool. Ambient outside temp 10*. Inside door 15.2-15.5, outside 9.6-10 (North facing). Compared to old multipane single glazed Sheltered South East facing door inside 11.8-12.2 outside 11.6 same door timber stiles read 13 inside 11.4 outside. 8 year old Pilkington Insulight Gas filled Double glazed unit South facing 15.2 inside 9.6-10 outside. (all +/- a few tenths) If nothing else that very crude / basic test shows me that my North facing custom Oak door with celotex fitted internally, is as good as the Sou
    – Flowboy
    Mar 6, 2019 at 11:45

5 Answers 5


First order approximation: R-1 (British units BTU/hr/ft2/°F) per sheet of glass.

So double pane is R2, triple pane R3.

Second order approximation depends on the frame material, and to some degree on coatings, and gas fills, and years since the gas fill; also depends on your heating system, exterior wind exposure. It gets complicated enough that I have never bothered beyond the first order approximation.

Note that financially, replacing existing non broken windows does not pay off in terms of saving on your heating bill.

Consider: Let's talk about a 1 square meter window -- 10 square feet. Suppose that the replacement window costs $100 and you install it yourself. Suppose that you found a wonder window that was rated as R-5, and that your old windows are R-2.

If you are in my climate you have a 10,000 degree day heating season. That's 240,000 degree hours. The old window: 10 square feet * 240000 degree hours/2 = 1.2 million BTU per year. That's about 1/8 gigajoule or about 1.25 therms. Last year at peak winter rates, my gas bill was running $6/Gigajoule. So the old window cost me 75c per year.

If the new window was PERFECT, it would save me 75 cents per year. So the minimum payback is 133 years at zero percent interest. If I have to borrow money at even 1% the new window will never pay for itself.

Flaws in this reasoning:

  1. If the old window is also a serious air leak, then it has a much smaller effective R value, and a correspondingly higher cost.

  2. Windows such as aluminum sliders have a lower R value, often being effectively R-1. This would double the annual cost. Still not enough to warrant replacement.

  3. While replacing the windows may not be cost effective they will make the house more comfortable. They will also make it easier to sell.

Window replacement is one of the LAST steps in making a house more energy efficient.

A much better strategy is to put shrink film plastic on the inside of your windows. This is only a few bucks per window, adds an R to the window, and also stops air leaks around the edges of the any windows that open. For a bit more you can get systems where you make frames of thin wood with magnets that stick to washers screwed to the existing frames. These are easy to remove for summer use. With more clever you make the add on frames 'double pane' with plastic on both faces, turning your R-2 window into an R4 window.

In winter when it is very cold out, the radiation through the window is significant. Having double layer window coverings can make a significant difference. In addition having a surface window treatment that reduces air circulation near the window allows the creation of a thicker boundary layer near the window, which will cool off, thus reducing the effective temperature differential between inside and outside.

Currently in the UK the electricity runs at 0.20£/kW-hr. My flat has 5 m2 of 30 years old double glazed windows, assuming a u-value of u = 3 W/m2/K, a winter outside T out of 4 °C and an inside T of 20 °C it gives 3(u) × 5(m2) × (20°C-4°C) × 24h × 0.20(£/kW-hr)/1000(W to kW conversion) = 1.15£/day. Make that for 180 days extended winter of the UK and you have 200£ spent each year. If I were to install new windows that would cost about 2500£ labour included and it would result in a saving of roughly a 100£ per year, as well as improve condensation, etc.


I suggest you use a U-value kit from greenTEG. It is easy to use, very accurate and you get the U-value of any position of your wall or glass within a few days of measurement. I tested it on my new house and it resulted in the same value as predicted by the energy engineer.


The U-value and R-value of glazing systems are not a single measure. Instead they are a composite number from R & U measures at several different places on the glazing system (center of glass, edge of glass, sash, frame, etc.) In Europe windows sold across member nation borders have to publish a U-value according to a predetermined formula. Duplicating it outside of a lab environment is impractical.

You can check for published numbers, but they are not likely to be directly comparable to those published by other manufacturers unless your comparing recent numbers by European manufacturers. The most likely place you are to find a make/model number is on a sticker placed on the outer sides of a moving sash. The stickers usually fall off vinyl windows, however, so don't get your hopes up if you've had them long.

A simple home test can be done with a heat lamp. Remove the sash and brace it into a standing position on a table. Place the lamp 1 inch from the exterior glass perpendicular to the window. Wait five minutes then turn off the lamp and quickly record the temperature of the interior glass with an IR thermometer. Wait an hour for the window to return to ambient temperature then repeat on the opposite side. You switch sides, because some low-e coatings are unidirectional and some are only applied to one plate. Now subtract the ambient temperature from your two recorded numbers so you are left with temperature gain. This is quite limited and your only testing the glass pack, but will allow you to test prospective replacement windows using the same methodology. Quantifying window performance outside a lab is rather difficult.


It's probably easiest to find a datasheet for the nearest comparable system to yours. The things to look for will be:

  • the material of the frame (probably either aluminium or uPVC)
  • if it's aluminium, whether the frames have a thermal break in them
  • the spacing between the panes of glass
  • the approximate age of the double glazing
  • whether there are trickle vents or not
  • the ratio of glazing surface area to frame surface area.

I don't know of a database per se, but there are various listings around of example U-values, so you should be able to make a reasonable estimate of it.

Or you could do a hand-wavy thing and say the U-value is around 3, and you won't be too far off, probably.

It is conceivable that you could make measurements to estimate it in situ, but it's pretty difficult. You'd have to build an enclosure either side of the window, keep one side at a specific high temperature, one side at a specific low temperature, and then use heat-flux meters to measure the rate of heat flow through glass and frame separately. If you really wanted to go down this route, it might be worth getting in touch with the closest university with a building-physics group, and asking if they've got student that would like to attempt this as a project.

A simpler and cruder way to do it, is to wait til it's pretty cold outside (which is right now, if you're in the UK), and about 21°C inside, and then measure the temperature of the glass inside the room. That will allow you to estimate the central-pane U-value (NB that's only for the glazing - the frames will have a different value)

  • Next to a single-glazed window, the internal surface temperature is around 1°C (U>5).
  • Next to a double-glazed window (2000 vintage), the surface temperature is around 11°C (U~3).
  • Next to a modern, energy-efficient double-glazed window, the surface temperature is 16°C (U~1.6).
  • Next to a well-made modern triple-glazed window, the temperature is 18°C (U~0.65).

(frustatingly, the source I've linked to doesn't say what the outside temperature is supposed to be, for these measurements. Minus a few Celsius, I guess)


My challenge was slightly different. I was purchasing new NorthStar windows to replace all of the old windows in my house but wanted to make sure I was getting what I paid for, as the windows would look fine, but was their R-value or U-factor and SHGC performing as advertised.

I used an IR scanner around the edges after install to look for possible insulation gaps - digital read compared to base value allowed me to detect to windows where the insulation was missed near a corner and they fixed the issue. The problem with this digital scanner is that it does not work well against a reflective or clear surface, such as the glass window pane. But what about how efficient the window was performing - if it had been jostled and a seal cracked, losing argon gas, how would I know. I did two things:

  1. I used an IR thermal camera to look at the house and compare visually if the windows were different colors. I had one window that had not yet been replaced and from outside (in winter) it was visibly different which told me the new windows appeared to have less heat loss. I then went inside and performed the same thermal imaging, and could tell that the old window was colder and therefore more cold was coming in through that window than the others. I then used a solar meter and compared the old window value with the new window value and could tell that the new windows had a lower value, so the coating was reducing UV and solar light.
  2. The final test I did was use a heat source against the outside of the window (can be a heat lamp or hair drier) and checked the IR heat and compared that to now window with the IR heat reader same distance, and also on the old window. The IR heat reading was considerably lower on the new windows. No one tool or test would give an easy ability to test the manufacturers claimed U-factor and SGHC on the new "energy efficient windows.

Even on the cheap, it cost me about $600.00 in purchasing the thermal- imager to attach to my iPhone, IR scanner, and heat scanner. IT could have cost several thousand if I had purchased better quality tools. However, to me, the verification of window integrity on its E-value U-Factor SGHC is something that the window manufacturer/ installer should do for you as part of the install, especially when you are paying high value to get the best E-value windows installed to reduce your heating cooling costs. Although I have a lifetime warranty on the windows. I would have had no way to know if they were not performing properly unless there was visible evidence of a crack, broken seal, or condensation/fogging on the inside of the double pane glass.

  • Hi Bob, welcome to the site. Can you edit that large blob of text to make it more readable? <br /> for line breaks, empty line for paragraph breaks.
    – user2451
    Nov 15, 2018 at 20:41

Not the answer you're looking for? Browse other questions tagged or ask your own question.