For some of us, the aquarium hobby means keeping organisms that can only thrive in temperatures significantly lower than normal indoor temperature. Not only is the maximum temperature they can tolerate lower, but they are also adapted to seasonal variations with very low temperatures during winter. This means that temperature control is one of the coldwater aquarist's main challenges. A good chiller is important for those who take this seriously. But it is not the only thing that is important. Insulating the aquarium and external components like sump/filters/hoses is a very good idea too. It prevents heat influx to the water and thus, saves energy. And most of all it prevents condensation of indoor moisture on glass and other components. Condensation is at its worst when it forms on the viewing panels. But it can be bad enough in other places too, causing corrosion and mold.

In this article I am using my modification of a 210 liters (56 US gallon) all glass aquarium as an example. It is easier to put double glazing on an all glass aquarium than an aluminium framed one. This is because the framing gets in the way of the process and limits how much the outer pane can cover, resulting in an awkward looking aquarium. I wanted my aquarium to be viewable from 3 sides.

One way to get better insualtion is to use an acrylic aquarium instead of a glass one. Acrylic insulates 5 times better than glass of the same thickness. So a perfectly normal acrylic aquarium should help greatly. In some places you may be able to get aquariums made with particularly thick acrylic. If the thickness is doubled, the insulation ability is doubled too. However, acrylic aquariums are more expensive, and can't even be found easily in all countries. So they are not an option for everyone. Here I describe a method for putting double glazing on ordinary glass aquariums. This will give results that insulate 3.5 times better than 19mm (3/4 inches) thick acrylic.

The principle behind double glazing, or insulating glass, is simple: Instead of a single glass pane, two panes with a space of about 10 millimeters between them, are used. The space is filled with gas and sealed around the edges so that air can't enter and the gas can't leave. The gas is usually argon, but air works well too. The gas, standing relatively still between the two panes, provides the insulation. One may think that it is very easy to make this kind og glass: Just glue something that is 10 millimeter thick on the surface of one pane, next to the edge. Then glue the other pane on top and seal around with silicone. There is a problem though: Water vapor condensates on the inside. This dries the air in the space and more water vapor seeps in through the silicone seal. Water vapor is a gas. Stopping a gas from slowly going through a seal is much harder than stopping liquid water from quickly going through it. I have tested pretty much every sealant I could get hold on that comes in a caulking gun cartridge. Water vapor goes through them all. It seems like these sealants all need some type of solvent that can dry out of them. I guess that if the solvent can come out, water can come in. So a sealant without any solvent, and a high resistance to vapor penetration is necessary. I have found that the kind of butyl rubber that is used for car headlamps works well. In addition, something that absorbs moisture must be be placed inside the space. I use silica gel placed inside the hollow spacer tube.

The type of glass you need for this is basically the cheapest type there is. It is the most common type of glass, regular clear float glass. It can be thin because it doesn't have any structural function. It is relatively easy to get for free used, because it is used in windows. But I recommend new to avoid wear and scratches. I bought 3 pieces of 4mm thick glass from a local glass service company. This cost a bit under 700 NOK (80 USD). The back wall and bottom would be insulated with styrofoam and did not need double glazing. The size of the glass sheets, except their thickness, should be exactly the same as the existing sheets on the aquarium. In other words, they should not be large enough to cover any silicone seams or the ends of perpendicular glass walls. I did a mistake here and made the sheets on the sides too large. This meant that the existing silicone seams on the aquarium came on the inside of the vapor seal. To fix this problem I had to put a vapor barrier over the seams on the inside of the aquarium. But even that may not be enough to fix it perfectly. So be aware of this issue. You must also make sure that the sheets don't go further down than the existing walls on the aquarium. If they do, the weight of the water could end up resting on the thin, outer glass glass sheets and the weaker silicon joints between them and the spacer. So it is best to err a millimeter or two on the short side when deciding the size of the sheets.

After glass has been cut the edges are extremely sharp. The sheets are very dangerous to handle at that point, so the edges must be treated before handling. If you use a glass service company you can get them to treat the edges. However, that may double the price of the glass. So I prefer to do this myself. It is very easy to do with 120 grit sand paper. I wet sand, but water is probably not necessary. Just do a few strokes with the paper across the razor sharp edges and you are good. Do it systematically. No point on the edges must be left untouched. You will "find" it later, trust me. Be extremely careful not to touch the surface of the glass with sand paper. Don't touch the surface with you fingers either while sanding because there will be sand particles on them. After the sanding, wash you hands and the glass very gently with water and dry it with clean, dry cotton fabric.

The spacers in a double glass window are the parts that hold the glass sheets apart and firmly in place. They also contain desiccant (moisture absorber). I used aluminium square tubing with a 10 by 10 millimeter cross section as spacers. Aluminium is moisture proof, which is good. It is also very heat conductive, which is a bad thing. I decided to go with this kind of tubing because it was easy to get on ebay and good to work with. I couldn't find anywhere online that sold genuine window spacer that was practical for me to order.

The spacers should be cut so that they can form a frame along the edge of the glass sheet. The glass should stick out 10mm outside the frame all around. This forms the channel to put the butyl sealer into. It is important to form the frame so that one end of each length og spacer tube is open. This is so that desiccant can be filled into it after it is glued in place.

It is necessary to make holes in one side of the spacers. More precisely, the side that is to face into the space between the glass sheets. This is so that the moisture can reach the desiccant. I used 2mm in diameter holes spaced 20mm apart.

After cleaning the glass I glued the spacers on to the aquarium glass wall with silicone. It would be better to use butyl rubber as glue as well as a sealer. I will probably do that in future projects. After the silicone had cured for one day, I thoroughly cleaned both glass pieces and glued the outer sheet in place. I now had to wait a week for the silicone to cure. The silicone can not cure after it has been sealed off in a very dry environment.

Cleaning the glass is one of the most important parts of the job. Once the glass sheets have been glued together there is no way to reach the two internal faces. Any dirt on them will remain there forever. It is very hard to se all dirt on glass. It may look clean from one direction and dirty from another. A way to really see all the dirt is to use a flashlight and hold it close to the glass. This is almost too effective, because you will find that you can never get the glass perfectly clean. Every time you clean away some dirt, some other dirt appears. So the key here is to get the glass acceptably clean without going insane. I use ordinary glass cleaning soap and rub with a wet cotton cloth. Then I rinse the cotton cloth in clean water and clean away the soap. I finally use a squeegee to dry the glass. I also dry a bit with a dry cotton cloth, if needed. But only minimally, because it leaves dust on the surface. I have tried cleaning with solvents like acetone and denatured alcohol before. But I find that these chemicals always leave residues on the surface and seem to just make it worse. So now I only use these if there is some specific reason for it.

After carefully cleaning the glass, it is time to glue the outer pane on. I used a small string of silicone as glue, just like on the other side of the spacers. Then the whole thing must sit and cure for a week.

There will always be a small amount of water vapor in air. That means that even though the space between the glass panes is sealed off, condensation can still occur inside. In addition, even a good seal can still leak a little. Beacuse of this, it is necessary to let the air in the space be in contact with a moisture absorbing material. I used silica gel. It is available on Aliexpress and eBay for very little. I bought a type that is orange in color, but becomes green when it is saturated with water. I filled this into the hollow spacers. A bit of care must be taken when doing this: It must be done after the silicone, or any other glues that are to be located inside the vapor seal, are fully cured. Immediately after opening the sealed container and filling the desiccant into the spacer, the vapor seal must be installed to prevent the desiccant to expend itself by drying outside air. I needed to modify a funnel by adding on the tip of a caulking cartridge to have a good tool for filling. It is a bit of a hassle to do because the beads will form a congestion and block the tube if you pour too fast. It is smart to use a funnel with a relatively narrow tip to make sure that congestion happens in the funnel and not in the spacer itself. Because there is no way to resolve it in there. A funny and interesting thing happened when I poured in the silica gel: Two small, broken off, pieces of silica gel ended up in the space between the glass panes so that they are visible. These two pieces can be used as indicators. As long as they stay orange I know that the air in there is dry.

The vapor seal was one of the things I struggled the most with when it came to making insulating glass. I believed that silicone was water proof, or at least that it was possible to find something that came in a caulking gun cartridge that was water proof. But, as I explained in the theory section above, the only thing I could find was butyl rubber that comes on rolls for headlight sealing. This is easy to get on eBay or Aliexpress, and while not cheap, it isn't extremely expensive either.

I left a 10 by 10 millimeter channel outside the spacer, all around the edge of the glass, for the vapor seal. This was exactly enough space for two rounds of 9 mm diameter butyl bead. The bead gets slightly stretched while being applied. So even though two beads of 9 mm diameter should be about 130 square millimeters in cross section, they fit into the 100 square millimeters actually available. The butyl is much easier to work with if it is warm. It gets softer then. I heated the whole room to 30C and let it stay like that over night, before working on the seal. It could also be smart to put a fan oven pointed at the glass while working with it.

The butyl rubber is a bit tricky to work with. It tends to stick extremely hard to everything it touches. If you press it on to something, it may stick harder to your fingers than to what you press it to. You don't want it to get stretched, because you can't press it together again without risking to introduce air pockets. These things make it hard to put it in place up side down and other places where access is difficult. It is probably best to go one time around the whole window, cut the bead and then squeeze it into the groove all around. I found that using a short piece of aluminium square tube plus some of the plastic foil that came with the butyl were perfect for squeezing it firmly into place, one small step at a time. After that I put the second bead outside the first in the same way. It is important to work it so that there are as few air pockets as possible, and no gaps, small channels to the outside. One such gap can ruin the whole thing.

The sealing butyl and the glue on the spacer do not look very attractive. In addition, the seal is very sticky and needs to be covered with something to prevent things from touching it. So some sort of moulding is needed. I used a minimalistic black PVC moulding that I found at a local building store. Silicone didn't stick properly to it. But I could use a polyurethane based caulking with primer to attach it.