Latest Discussions : Windows & Doors


08:17AM | 07/22/05
Member Since: 07/21/05
4 lifetime posts
After Hurricane IVAN, we had our windows appraised because some "fogged." Later, I was told that one could perform an "ice test" to ascertain the integrity of the seal of double pane windows which has been compromised and they are destined to fail. The test is as follows: Rub an ice cube on the inside pane of the double pane window until one is sure that the temperature has fallen -- usually indicated by frost forming on the glass or definite condensation between swipes with a cloth to wipe off the ice melt. After several swipes to keep the window dry, if condensation forms on the iced area BETWEEN the panes, and cannot be wiped off, it indicates that water has entered the space between the panes since the seal has been compromised. It follows that the window needs to be replaced because fogging is inevitable, but it might take a while.

I have done this with our windows. Many more test “positive: for compromised seals than is evident through frank fogging. I have mixed opinions from window companies. Some have said it’s a very valid test, but they don’t do it. Others have never heard of it.

What is your take?


08:55AM | 07/23/05
Member Since: 01/28/03
693 lifetime posts
The test is not entirely legitimate. Here's why.

Condensation occurs whenever a surface that falls below the dew point for a particular volume of air has been reached by a falling temperature.

But the dew point is a function of relative humidity and temperature.

For example, if a volume of air is at 95F and the relative humidity is at 90%, then the dew point is at a balmy 92 degrees F.

This means you should get condensation between the panes when the glass temperature falls to below 92F if the seals have failed.

Or, If the air temperature is 40F and the relative humidity is 80%, then the dew point is 32 F and when the temp for the glass drops to below 32F you get condensation between the glass.

The problem with the ice test is that it can produce condensation even if the seals are intact as long as you drop the temperature of the glass below the dew point for the air that is trapped between it.

Let's say the air in the factory the day the glass was sealed was at 65F with a relative humidity of 40%, this means that although the glass is still perfectly sealed, you will produce condensation using the ice test because the dew point is 42F and you will have cooled the glass to about 32F.

The result, then, is a 'false positive' for broken seals.

You will have caused condensation between the panes, yet the panes remain perfectly sealed.

Unless you can know for sure what the relative humidity and temperature of the air trapped between the panes, you cannot use the ice test effectively to prove anything but that there is gaseous water present between the panes.


09:35AM | 07/23/05
Member Since: 07/21/05
4 lifetime posts

Thx so much for your prompt reply. It certainly makes sense. However, I thought double paned windows (except early on when they were "removable" -- and we have some of those, too) were sealed with krypton or argon which, to my knowledge, doesn't hold water, so there should be no condensation. Do you know when they started filling them with these gasses? Our windows are about 12-13 years old and perhaps the gas replacement for air is a more recent inovation? Or am I wrong about the inertness of argon and krypton as it pertains to holding water?

Again, thx for your advice. Gonna be a buncha' guys disappointed in "hurricane country."

Looking forward to your reply.


09:47AM | 07/23/05
Member Since: 04/25/05
1915 lifetime posts
Dry air is used and they put in a descant (sP? - drying agent). Don't forget that the windowns need to work in "nomral" conditions with cold temp on the outside down to the well below zero depending on the climate.

But it is not very objective test unless you know the dewpoint of the current ambiant air and how long it takes for the air inside the glass to reach equaliberiam with the outside air.


03:00PM | 07/23/05
Member Since: 07/21/05
4 lifetime posts
Hi Billhart --

Thx for educating me a little more. It looks like there is more room for error than is immediately apparent. July, in the "Deep Soooth," where we live, has never allowed the outside temp to be cooler than the inside, since the A/Cs run virtually 24X7 -- at least not since the last Ice Age. And I have tested only from the inside and haven't paid attention to the dewpoint at all. And I concur wholeheartedly that it is a very subjective test since I have found windows that are questionable.

I guess the thing that bugs me is that I can check two adjacent windows, both installed at the same time, same manufacturer, exposed to identical conditions re: sun/shade, etc., and one will test strongly positive while its "brother" is weakly positive or may even test negative.

Oh well, as we used to say in the Navy: "If you can't take a joke, you shouldn't have signed up."

Seriously, though, if you have any idea to explain the "brother" phenomenon, I'd appreciate hearing it.

Thx again.


03:52AM | 07/26/05
Member Since: 01/28/03
693 lifetime posts
Argon or other gases are not used in double pane windows unless specifically asked and paid for.

Otherwise it is simply 'dead air' space that assists with the thermal quality of the window.

And since factories do not try to control the humidity when manufacturing the windows, you could have air trapped in one window with 50% humidity and the one next to it trapped with 95% humidity. Both could have perfectly intact seals but the one would show condensation with the 'ice test' and the one next to it so nothing because they have different dew points for the air trapped between them.

The ice test is just plain unreliable unless you can know with precision what the relative humidity and dew point is between the panes of glass for all windows all the time.


06:10AM | 07/26/05
Member Since: 04/25/05
1915 lifetime posts
"And since factories do not try to control the humidity when manufacturing the windows, you could have air trapped in one window with 50% humidity and the one next to it trapped with 95% humidity."

They have to control the humidity (actaull the dewpoint) in the windows. Otherwise they could not sell them any place north of Altanta.

Having the glass as cold or colder than the ice cube test is common in the winter and you would see 10's of thousands of fogged windows.

Either a dry air purge is used or a desiccant to remove the last moisture.

" Another approach is to replace the metal with a design that uses materials that are better insulating. The most commonly used design incorporates spacer, sealer, and desiccant in a thermoplastic compound that contains a blend of desiccant materials and incorporates a thin, fluted metal shim of aluminum or stainless steel. Another approach uses an insulating silicone foam spacer that incorporates a desiccant and has a high-strength adhesive at its edges to bond to glass. The foam is backed with a secondary sealant. Both extruded vinyl and fiberglass spacers have also been used in place of metal designs."


Two sheets of glass bonded together in a unit to enclose a captive air space. Units are constructed with a metal spacer inside the outer perimeter. The outer edges are sealed. Spacers contain desiccant material for absorbing and holding any remaining moisture in the air space."


07:16PM | 07/26/05
Member Since: 01/28/03
693 lifetime posts
Billhart said:

"They have to control the humidity (actaull the dewpoint) in the windows. Otherwise they could not sell them any place north of Altanta."


You're absolutely right.

Otherwise you would get condensation any time the temperature dropped below the dew point for that air humidity.

Sorry for my silly statement to the contrary.

I was thinking that having toured window factories and never having seen any special practice used in the glass assembly areas to ensuring uniform humidity levels that "nothing" was done to control the humidty levels in the windows.

Never occurred to me about the use of desiccants, which makes perfect sense, and how they would chemically remove any gaseous water molecules from the air trapped between the panes.

Would you then think that the ice cube test is fully valid then?

I don't because if the seals have failed and if the test is conducted on differing days of differing humidities, you could actually in theory do the test on one day of low humidity and produce no condensation and do the test on another day of high humidity and get condensation.

The ice cube test 'might' work, but there doesn't appear to be a guarantee that it would.


08:09PM | 07/26/05
Member Since: 04/25/05
1915 lifetime posts
"The ice cube test 'might' work, but there doesn't appear to be a guarantee that it would."

As I said the test is not very objective.

My guess is if it shows condensation that the seals are blown.

But no condensation does not prove that they are not.


06:02AM | 07/27/05
Member Since: 07/21/05
4 lifetime posts
Thx Homebild and Bilhart . . .

You folks have been a wealth of info.

Living in Pensacola (with high ambient temps and high humidity -- read that "Thank God for air conditioning") we are all trying to ascertain which windows have failed because of hurricane(s) -- or are destined to fail but still appear "normal."

Insurance companies invariably try to deny claims and an insured can modify their claim for only up to one year after the hurricane. We note that many of our windows are starting to fail that have looked good until now. Some neighbors have used the ice test with variable responses from their insurance company. As you may be aware, this is not unusual. For example we had a $27,000 claim at a condo (damage didn't involve windows to any great extent). We received $3100. Our neighbor who had the same damage, the same coverage, and the same insurance company received their full amount. It appears it will be "lawyer time" to settle this as their explanation doesn't make any sense at all.

Our home, however, is loaded with windows and sliders -- different insurance company. Many still look good but several have become cloudy recently. IVAN hit us on 15 September 2004. My concern is whether or not there is any test available that might/could identify seals that have been compromised so that we could enter a claim prior to 15 September, the one year deadline -- and after which we will have to "eat" the expense of replacing a couple of dozen panes.

Any ideas? I have had several window companies tell me of the ice test and several others tell me to call them back when the windows fail. Thelatter say there is no test that can be run.

Thx so much for your help.



07:19AM | 07/27/05
Member Since: 04/25/05
1915 lifetime posts
I don't know if there is any clear positive test for windows other than waiting for them to cloud.

One of the problems with a cracked seal is the rate at which air moves into and out of them.

I just don't know.

But you might want to contact an Independent Adjuster. They are insurance claim adjusters, but they work for you instead of the insurance company.

I think that it is common that they work for a percentage of what is collected.

They are suppose to know how to put a lose so that it is covered.

You might see if they can do anything with the condo claim.

Post a reply as Anonymous

Photo must be in JPG, GIF or PNG format and less than 5MB.

Reply choose button


Post new button or Login button