Double-glazed windows are usually made from pre-fabricated glass panels. Although this is a good design to reduce the cost and complexity of manufacture, there are drawbacks. For example, if a double-glazed window is broken, replacing it may be very difficult and expensive due to custom sizes, discontinued models, and companies going out of business. Most double-glazed windows are only expected to last 10 years in moderate climates, and less where there are temperature extremes due to the seasons, which is reason enough to be able to maintain a window.
The design shown below is a true DIY double-glazed window. It can be built and maintained by a person of modest skill; and at minimal cost. If someone breaks the glass, it can be replaced with readily available glass sheets from a home-supply store. If it begins to collect moisture, the seals can be replaced and new desiccant can be added.
One often-mentioned disadvantage of an air-gap (instead of noble-gas filled) double-glazed window is the convective loss due to air circulating around in between the sheets of glass. Other gases may lower the convective losses due to higher mass, which leads to slower diffusion. However, I think that most older windows have long since lost whatever special gas filling they may have once had. Air-filled windows have the advantage that a normal person can replace glass and replenish desiccant without any special equipment.
Please remember to rate this instructable by clicking on the little stars on the right!
Step 1: Making the frame
The sash of the window is the trick to this design. It acts as the spacer between the sheets of glass; it provides a place for the O-rings that seal the glass to the frame; it provides a place to add trim, which holds the glass in place; and it provides a structure that supports the glass.
The air gap is created by using the wood of the sash. I used a 1/2-inch gap. Note that the thermal performance of this window sash is improved by not using a metal spacer between the layers of glass.
I designed my frame to accommodate 1/8" glass and 3/8" trim. The O-ring groove is 1/8", which is a poor choice because there were no available O-rings with the appropriate diameter. You should find the O-ring and let that determine the width of the groove. The O-ring should be chosen based on it's cross-sectional area. That area should match the cross-sectional area of the groove. See the selection at http://www.mcmaster-carr.com. I was surprised at how much some of these materials cost. Find something weather-resistant so the UV-radiation doesn't destroy it immediately. For my prototype, I used silicon calk, which was cheap but will make the glass very difficult to remove. DO NOT USE CALK!!
The wood that I used for this demonstration is pine from framing-grade 2"x4" lumber. When I do this for a larger project, I will use something much nicer. I've read a lot about windows and there doesn't seem to be a consensus on what wood is best for windows. I would use white oak, but most any hard or soft wood can be found in commercial use, so I guess it doesn't much matter.
When you're cutting this profile, make sure that you produce enough for your project. Measure the outside perimeter of the windows and add 10% to the total for all of the windows. If you are making a frame for a very large window, or if your stock is short pieces, you may have to be more careful about the exact amount of stock. Also, for a window larger than 3'x3', you'll want to make something more sturdy than the design presented here. Just a thought, Mr. Fox.
One of you will want a drawing of the cross section---I just know it. I've decided not to provide a technical drawing because the point of this article is to demonstrate a concept; it's not a engineering consultation. You'll know the difference because concepts are free. The other comes with a fee---and usually 10% up front.