Lee - I guess that is where we differ. When the window has no shield (or shade) there is only a very slow convectional current. When a shield that is not totally seal is used it actually speeds up the air flow past the window as the cooling air is lineally directed down the glass as it drops out the bottom. I will keep search for the reports that I had seen.

•••That is what the theory says, and that is what my measurements have shown.•••

This is what I would like you to show from your test results. All I read was temp sensor averages. No air movement sensor. No high and low. No way of knowing that it is not the increased warm air flow entering the space that is giving the higher values.

I believe I could reproduce your test results by using a sheet metal shield (no r value) and inserting a small hair drier at the top to model the effects of the warm air being drawn in. This would not mean that the sheet metal increases the performance of the window.

Lee - please post the actual raw temperatures as measure by the sensors in each of the scenarios.

Posted By jonr on 06 Dec 2013 11:03 PM

This air cooling was a net heat loss to the building that did not occur when there was no shield. And of course it dramatically increases condensation.

Condensation releases heat (8K btu per gallon) - so you don't want it occurring on a window.

jonr- Who are you quoting here?

FBBP-

Let us simply the problem so that we can all understand it. Do NOT worry about the convective air flow from the room to the window. Consider ONLY the heat transfer from the innermost window surface to the outside world. So we have heat conduction through the panes of glass and convective heat transfer between the panes, and convective heat transfer off the outside surface. There is some radiation heat transfer thrown in too, but it will show the same trends as the other heat transfer mechanisms. Note that this heat transfer process is fixed in terms of the materials involved and the dimensions of the problem. For a given outdoor temperature, the only variable in the process is the temperature of the innermost window surface.

Conductive heat transfer (through the glass) is described by:
q = k * A * deltaT / s

where:
q is the heat transfer rate (W)
k is the thermal conductivity of the material (W/m K)
A is the heat transfer area (m^2)
deltaT is the temperature difference between the innermost glass surface and the outside world (K)
s is the material thickness (m)

Note that everything is fixed except for deltaT which only varies with the temp of the innermost glass surface

Convective heat transfer for the heat transfer by air or gas fills:

q = h A deltaT

where the new variable is:

h = convective heat transfer coefficient (W/m^2 K)

Again, for heat transfer from the innermost glass surface to the outside world, all the glass areas are fixed, the composition is fixed, the outside air flow is fixed. The only variable is the temperature difference, and that is all based on the temperature of the innermost glass surface.

So we do NOT need to worry about the air flows in the room, ONLY the average glass temperature of the innermost glass surface. If we put up a thin layer of perfect insulation from the inside, then the innermost glass surface equilibrates to the outside temperature and the heat transfer is zero. If we put up a thin layer of nothing, then the innermost glass temperature has the maximum possible value, and the heat transfer is maximized. So we can evaluate the effectiveness of the insulation of the shade simply by measuring the innermost glass surface. Good insulators cause the glass temperature to drop, since more of the temperature drop is across the shade. Poor insulators cause the temperature to be higher.

Note: No discussion of the airflow looping around in various tortuous paths is required to evaluate the insulation effectiveness of the shade. So please keep it simply and avoid that discussion.

So we can evaluate the effectiveness of the insulation of the shade simply by measuring the innermost glass surface. Good insulators cause the glass temperature to drop, since more of the temperature drop is across the shade. Poor insulators cause the temperature to be higher.

so please post those values.

Here are some sample data: (These sample data will be hard to follow, but all these have been processed and results presented at http://www.residentialenergylaboratory.com/r_value_cellular_shades.html.)

Sensor Serial Number Simple Label Location
3900000245B17E28 39 Midline bottom window, right edge
6A0000023C4E7628 6A Midline bottom window, center
810000023C563228 81 Midline top window, right edge
440000023C1EB928 44 Midline top window, center

Determine baseline temps by window when shade is up, windows #1
Temps. (degF)
Sensor Label 10/29/2010 8:15 10/29/2010 9:12
39 56.6 57.3
6A 58.1 58.9
81 57.9 58.9
44 56.6 57.3
Avg. 57.3 58.1
Inside house temp. 62 62
Outside temp. 28 33.6
Depression of window temps 13.8% 13.7%

Window is Pella Designer triple-pane, low-emissivity, low SHGC window rated at U=0.29 btu/(hr ft2 degF)
Then R=1/U = 3.4

Assume that heat flow through window and shade are the same. Then deltaH = Uwindow * (Tout - Tinside window) = Ushade * (Tinside window - Troom)
or in terms of R-values, deltaH = (Tout - Tinside window)/Rwindow = (Tinside window - Troom)/Rshade
so,
Rshade = Rwindow *(Tinside window - Troom) / (Tout - Tinside window)
Test on window #1 in Office (west side) with Pella Designer triple pane with low solar gain, speced at U=0.29 (so R=1/U = 3.4)
Tests with light-blocking shade down and side tracks in place
Rwindow = 3.4
Sensor Label 10/29/2010 6:31
39 43.2
6A 43.5
81 46.5
44 46
Avg. 44.8
Inside house temp. 62
Outside temp. 25.3
R_shade 3.0
R_shade_correct_baseline 2.0
Ratio 0.6
R_s 2.4
Average R_shade 3.2
Std. Deviation 0.32

Thanks Lee
If you could post similair data set for window with no side tracks?
Also if you have a set with higher delta-T?

Same window but side tracks removed:

Test on window #1 in Office (west side) with Pella Designer triple pane with low solar gain, speced at U=0.29 (so R=1/U = 3.4)
Tests with light-darkening shade down and side tracks removed
Rwindow = 3.4
Sensor Label 11/3/2010 6:55
39 51
6A 50.8
81 53.9
44 56.3
Avg. 53
Inside house temp. 63.3
Outside temp. 25.3
R_shade 1.3
R_shade_correct_baseline 0.7

Different window, different R-value, greater deltaT
Test on window #9 in Kitchen (facing South) with Pella Designer triple pane with high solar gain, speced at U=0.31 (so R=1/U = 3.2)
Tests with room-darkening shade down and side tracks in place
Rwindow = 3.2
Sensor Label 11/14/2010 6:56
39 39.9
6A 40.1
81 44.7
44 44.5
Avg. 42.3
Inside house temp. 61.3
Outside temp. 11.3
R_shade 2.0
R_shade_correct_baseline 1.3

FBBP-

Now it is time for you to post your references for measurements that show an increase in heat transfer with shades in place.

Posted By Lee Dodge on 07 Dec 2013 01:56 PM

Test on window #1 in Office (west side) with Pella Designer triple pane with low solar gain, speced at U=0.29 (so R=1/U = 3.4)
Tests with light-blocking shade down and side tracks in place
Rwindow = 3.4

I am a little confused. Why would a triple pane window only have a U-Value of 0.29/0.31 or R-3? There are double pane windows out there that get that U-Value.

Most triple pane windows are <0.20 or R-7. Is the third pane of the Pella glazing part of the sealed spacer assembly or is the third pane unsealed?

Lbear-

The reason for the low U-value is discussed in great detail in the Appendix section of the reference already provided twice in this same thread. Again, the link is http://www.residentialenergylaboratory.com/r_value_cellular_shades.html. It includes schematic pictures, graphs, etc., and even in your state of being "a little confused," it should be clear.

Just goes to prove that old adage, "you can lead a horse to water, but you can't make him read."

<> my post was accidentally cleared :(

This forum needs some software updates...

Posted By Lbear on 08 Dec 2013 01:05 PM


So the third pane is not sealed within the spacer unit of the window. That explains why the Pella windows you have only get a U-Value of 0.31 since the 3rd pane is not a sealed pane as one is able to open it and blinds can be installed. Therefore that third pane is not a sealed glazing unit. Only the first two panes are sealed.

No. You are still not reading and understanding the appendix referenced above. The window seals have to do with infiltration, not the thermal conductivity of the window as measured under standard testing. Look at Figure A-3, and use that to compute the U-value for the window. Once you can do that and present those results, we can go forward with this discussion.

Let me modify my request to remove the request for you to compute the U-value and just ask you to provide an intelligent discussion about the effect of window pane spacing on U-value (hint: this discussion will not involve the term "seals"), comparing optimum spacing and those chosen by Pella for this particular window. Then add to your discussion the effect of inserting a mini-blind in the spacing that they provide, and closing that mini-blind so that it becomes effectively another window pane.

For extra credit, you might even discuss the effect of those blinds on radiative heat transfer if the blinds were assumed to be white and fairly reflective, but this is not required.

Are there any good (reliable, cost effective, etc) designs for external window insulation that are either automatic (preferred) or controlled from the interior?

Lee-

I just wanted to know why didn't you go with a sealed triple pane window that has U-Values of <0.20? Instead of a R-3 window, you could have had a R-7 window. There are true triple pane windows that sell for around $20-$35 per glazing square foot and get better energy numbers than the Pella Designer Series.


Martin Holladay, GBA Advisor:

"Pella only offers the triple pane in the Designer series, which is not a true sealed triple pane."  (emphasis mine)

Lbear-

Again, U-values for the center-of-glass are not related to the edge sealing details. If you want to understand the U-values in the Pella Designer Series windows as well as windows in general, I have already provided most of the answers in the appendix previously referenced. Just look at that and discuss for this forum how the glass pane spacings result in the specified U-value, with a provision for the difference in center-of-glass U-values and those for the overall window. This effort will provide to you a much better understanding of the relative importance of sealing versus spacing. Your discussion should include how the choice of low solar gain versus high solar gain affects the U-value.

This effort will also illuminate the desirability of adding the mini-blinds into the window system.

Todd6286-

I have an experiment for you to try. When it is going to get cold enough to form condensation on the windows, raise the blinds up to leave an area open at the bottom that is approximately the same height as the highest point where the condensation forms on the window. In the case of my windows, this prevents the frost or condensation from forming. It will degrade the thermal performance of the shades, but it should still be much better than leaving the shades fully open.

Lee - you say •••Note: No discussion of the airflow looping around in various tortuous paths is required to evaluate the insulation effectiveness of the shade. So please keep it simply and avoid that discussion. •••
Here is a fictitious report based on your dataset that will show how much the outside air affects the program. The same will be true for convection set up by inside air.

Air infiltration and its sources examined for double hung windows.

Abstract
To determine if outside air infiltration affects the temperature of room air next to a double hung window and if possible determine the area of maximum infiltration.

Introduction
People spend a large share of their building budget on windows. Those looking at energy efficient homes tend to spend even more of the their budget on windows. Windows are often cited as the source of the single most heat loss in a building envelope. This report examines if there is any place for double hung windows in energy efficient home.

Background
Most windows today have stated R or U values. Most also state air infiltration rates.
Air infiltration rates for the subject windows from the window manufacture’s web site;

Fixed Frame = 0.05 (cfm / ft2 of frame @ 6.24 psf wind pressure)
Awning Window = 0.05 (cfm / ft2 of frame @ 1.57 psf wind pressure)
Double Hung Window = 0.3 (cfm / ft2 of frame @ 1.57 psf wind pressure)

Experimental Apparatus and Methodology
Four temperature sensors will be place on the glazing.
A shielding devise shall be placed so as to block some of the indoor heat from reaching the sensors.
Temperature will be recorded to determine how much the infiltrating air cools the space between the glazing and the shielding devise. A second recording will be taken with additional shield devises placed on the sides to further prevent the co-mingling of indoor air with the inter window space air.
Initial readings will be taken with no shielding devise in place.



Observations
Measurements taken with no shielding devise indicate a fairly tight grouping, probably too tight to determine where the maximum leakage is occurring. This data set also indicates that there is no convection current across the height of the window as the upper and lower middle sensor temps are the reverse of the window edge sensors temps.

When the main shielding devise is placed, we see a small cooling of the inter window space. We also see a strong gradient between the upper and lower sensors. This would indicate a convectional current setting up because of the cooling effect of the infiltrating air. It is likely that the temperature drop would have been larger were it not for the room air being drawn in by the convectional currents.

When the secondary shielding devises are placed we see a much greater cooling of the inter window space. It is clear that as we eliminate more of the room temperature air from the inter window space, we see the true effect of the cooling by infiltration. It is noted that the convectional airflow is still at play in the inter window space as verified by the top to bottom temperature gradient.
It is interesting that in one of the data sets, the top two measurements show a larger differential then in the rest of the sets. This would probably indicate a strong source of infiltration air at this point.

House temperatures are well below normal room temperature requirements and this may affect the usability of this report.

There is no way of know if indoor micro climates exist next to the windows as indoor temps are record from a single centrally located thermostat.

The air infiltration rate through the main shielding devise is not known.

Based on the manufactures spec’s we see that double hung windows have a much larger infiltration rate of outside air than awning windows do. And both have a much larger infiltration rate than fixed in place glazing.

It is unfortunate that a control dataset is not provided for a fixed glazing window for comparison purposes. However readers are cautioned that there may be a large discrepancy in the infiltration rate of different glazing set with glazing tape as the butt and pass method of installing these tapes is prone to a large degree of human error.



Conclusions
There is a high degree of air infiltration when using double hung sashes. This is due in part to the fact that the windows must slide along weather stripping rather then compress the weather strip. In many cases the mating of the upper and lower sashes are very difficult to seal.

Individuals interested in energy efficient home should be cautious about pick windows for looks alone.

Individuals interested in energy efficient home should be cautious about picking wood framed, wood sashed windows due to the possible infiltration between wood joints in the sashes and frames. The issue becomes worse as the windows age and go through multiple temperature and humidity changes.

If installing double hung windows, it is safe to ignore both the r value of the window and the r value of the shielding devise as it is well know that air infiltration will trump r value every day.

I wonder if anyone on the forum has access to the reports referenced under ThermalCurtains and Shades on page 284 of this report and or the complete text of this report.
http://books.google.ca/books?id=mcRK-xlKMYIC&pg=PA283&lpg=PA283&dq=movable+window+insulation&source=bl&ots=kKxKWWJyD8&sig=D0hayMbn15ndFLYnKr4DdytuWiw&hl=en&sa=X&ei=gXCiUp3YJceOrgHQt4CoBw&ved=0CH8Q6AEwCDge#v=onepage&q=movable%20window%20insulation&f=false