Two triple glazed windows get fitted....

[Ivan]

I've now replaced the last single-glazed window in my house (also the largest). Actually, it was secondary glazed, but now it's triple glazed (with two sheets of soft-coat glass, warm-edge spacers and argon-filled)+secondary glazing. I've also installed a triple glazed window in my eldest daughter's bedroom, which has always been cold. The effect was instantly noticeable - no cold zone near the window, and the roof feels a few degrees warmer. My IR thermometer showed some interesting figures.

Outside air temperature 5C.

Inside surface temperatures:
Old (6mm air gap) double-glazed window: 11C
Insulated Cavity Wall: 18C
Newish 20mm double-glazed window (soft-coat, air-filled) 15C
New Triple Glazed window: 17C

[desperate]

Ivan,

It is interesting that there is not so much difference between the double and triple glazed units, but the 1 degree difference with the insulated wall is most impressive. What were the internal air temps when you measured?

Do you glaze the frames after fitting? or is it factory assembled?

Desperate

[dtaylor]

Hi,

I have recently replaced one of 4 panes (DGUs) in one window (breakage). The previous panes were 4mm gap, and the new one is with a 12 mm gap, argon filled & low-e inner pane.

I was thinking of trying to organise a thermal imaging session to see if it is making a difference.

The window is in the old part of the house, which is clay lump - so it may be that the wall is losing more heat than even the old windows. According to http://books.google.co.uk/books?id=hGVIXqZjrTAC&pg=PA120&lpg=PA120&dq=clay+lump+u+value&source=bl&ots=3CJHiJqhF0&sig=2DdPXcnTlmWaNYSdu67P5vI-yDU&hl=en&ei=kludS67kJcq6jAehqZS0DA&sa=X&oi=book_result&ct=result&resnum=7&ved=0CB8Q6AEwBg#v=onepage&q=clay%20lump%20u%20value&f=false, I would need to have walls 1.15m thick to get a U value of 0.7. The walls are only about 1 foot (a third of that) thick.

The rooms are cold.

I am interested in your method of temperature measurement. Is it just a thermometer taped to the wall/window for a bit with some kind of insulator ?

Are you able to make inferences about U-values, given some temperature measurements inside & outside - both in free air & against the wall/window?

I want to work out if it is worth replacing the other DGUs or trying to insulate & render the clay lump first.

thanks in advance (tia),
Regards,
David

[Ivan]

I think the difference between double and triple glazing is more than you think:

Typical available U-values:

Single-glazing 5.0
Double – glazing 3.0
Triple-glazing 2.2
Double-glazing with low-e coating 1.7
Double-glazing with low-e coating and Argon filled 1.3
Triple-glazing with multiple low-e coatings and Xenon filled 0.4
(from http://www.greenspec.co.uk/html/materials/glass.html )

So my triple glazed low-e units have U-value three times lower than the equivalent double glazing. And should indeed compare favourably with the walls:

Typical U values in W/m2K
Solid brickwork, no insulation 2.5
Cavity wall, no insulation 1.6
Cavity wall with insulation 0.5
(from http://www.rooftherm.co.uk/index.php?option=com_content&task=view&id=39&Itemid=9 )

I've simply used an infra-red thermometer, and pointed it at the wall / window. I do have a thermal imaging camera. I'll take a few images sometime during the next week or so, to see visually what the improvement looks like.

I'm sure there's a way of deducing U-value from the temperature reading, if taken under steady-state conditions (actually quite tricky), and if outside and inside temperatures are known, including adjustments for turbulence etc. Djh knows much more about this kind of thing than me, and he did explain that this wouldn't be very straightforward, although I've certainly seen very strong correlation between good insulation and higher temperature readings, and vice versa. djh did suggest a method for fairly accurate determination of U-values using a small piece of insulation to be temporarily attached on the inside. I've saved a bit for doing exactly this, but yet to find the time to try it.

Desp, The frames were produced by a local joiner friend. The glazed units were produced by a local glazier who was prepared to listen to what I wanted and to produce it to my specifications. The frame was fitted first, then we installed the glass after. We'd intended to fit using dry-fit tape, but the type of tape I wanted didn't seem to be available, so in the end, carefully-applied silicone was used (the frame has been designed to be drained/vented.).

[dhaslam]

The difference between double and triple glazed windows overall is not quite so  great, more like an improvement of 50%. 

This is a quote from the  Greenspec  link above.

Three- or four-pane glazing units with multiple low-e coatings, gas fills and special frame construction reach U-values of 0,4-0,5 W/m2K for centre-of-glass values and 0,7-0,9 W/m2K for the total window U-value. Their overall energy performance provides net energy benefits for any orientation.

[Kombi]

I have just installed a couple of 3G Velux. It would be interesting to compare their performance with the 2G I already have installed but haven't got an infrared thermometer. Any other way possible?

[EccentricAnomaly]

I've simply used an infra-red thermometer, and pointed it at the wall / window.

From my experiments I think that might be a bit misleading because of the various emissivities and reflectivities of the surfaces.  I'd suggest putting a bit of masking tape or something on each one to get more consistent values.

[Ivan]

EA, Yes, it's a very good point -  Glass is very reflective to IR.  I guess the IR method can't be used reliably to compare glass with wall.....but it's still valid for comparing old glass with new glass, and in my case, it's showed a very big improvement, so I'm happy.

Good suggestion about using masking tape - it would give the same emissivity/reflectivity for all materials.

[wookey]

Kombi, those IR temp sensors are fairly cheap (£25 e.g. http://cpc.farnell.com/_/in05687/ir-thermometer-voltage-detector/dp/IN05687 ) and very handy so possibly worth just getting one. Otherwise you'd need to tape a sensor/thermometer to the windows. K-type thermocouple+multimeter or 1-wire sensor are most obvious options. Bit of thermal grease to help measure window rather than room air.

I've had trouble with emissivity going wrong. In our damp-corner downstairs loo our IR thermometer was reading -2C in Jan. on a plaster wall. It was cold, but not that cold! It was probably actually about 8C. Not sure what happened there - it is generally within a couple of C of the right answer, although it doesn't work well on some surfaces (such as copper pipes).

Don't forget that those U-values Ivan quoted are glass-only. You want whole-window U-values for the real world. Obviously on a huge window like Ivan's these will be quite similar, but on normal-size windows the frame is significant, and usually worse than the glass for modern low-e units. The best whole-window values are around 0.6-0.7 and that is for very expensive windows (3G+thermally broken frames). Unless you are trying to build a passive-house I reckon anything under 1.4 is worthwhile. Don't forget that for south-facing windows there is a complex tradeoff between insulation and solar gain (3G windows keep heat in better but also let less light in), and in fact 2G may be a sensible choice.

[guydewdney]

what design is a 'warm edge spacer' or 'thermally broken' window - compared to a 'normal' DG unit?

[rhys]

what design is a 'warm edge apcer' or 'thermally broken' window - compared to a 'normal' DG unit?

http://www.superspacer.co.uk/superspacer.asp
This explains the warm edge spacer.
http://www.greenbuildingstore.co.uk/page--ecopassiv-passivhaus-timber-windows-door.html
and here is a picture of a thermally broken frame

[Kombi]

Kombi, those IR temp sensors are fairly cheap (£25 e.g. http://cpc.farnell.com/_/in05687/ir-thermometer-voltage-detector/dp/IN05687 ) and very handy so possibly worth just getting one. Otherwise you'd need to tape a sensor/thermometer to the windows. K-type thermocouple+multimeter or 1-wire sensor are most obvious options. Bit of thermal grease to help measure window rather than room air.

Hi Wookey. Thanks for the link. I actually have a thermal sensor which plugs in my Fluke meter so I'll do some testing with that.

[Ivan]

My window is indeed large (the triple-glazed unit is around 70kg!), but nevertheless I looked into passiv-haus style wooden frames. The joiner was concerned about load-bearing capability, so in the end, we dropped the idea, but there were some groves routed into the profile, so there is a small degree of thermal break.

When we get a cold night and I remember to, I'll take some thermal images, to compare before/after.

Wookey, the light loss isn't as severe as you'd think. I deliberately went for the more expensive soft-coat (rather than K-glass), which isn't tinted - especially as I've got two layers of coated glass in the triple glaze unit. If I remember, I'll take a photo of the window with the opening window open, so the extent of light-loss will be apparent.

[Ivan]

OK, here's the picture. On the left hand side of the vertical rail, we're looking through an open window (ie no glass between the camera lens and the landscape). On the right hand side, we're looking through the triple glazing with two layers of soft-coat low-e glass. The picture is pretty accurate to what you see.