Before decorating a radiator consider what you expect the radiator to do.
Do you want the room to warm up quickly or would you prefer more of the heat to remain within the system and be emitted slowly? Has the colour to match the general room décor or doesn’t it matter? Do you want a glossy, eggshell or matt surface?
Your choice will no doubt be a compromise between all these factors with, perhaps, an accent on the first. Bear in mind that the heat generated by a fuel is not variable, but the rate at which it is emitted is. If it does not come out it remains in, and the reason why most people install central heating is to warm a room, not the system.
Rather wild theories are current as to the best paint for this purpose. One used to be that matt black was the best; another that emulsion of any colour should be used. Yet another that the type of paint does not matter so long as it is not a metallic one, because the bronze used in gold paint and the aluminium in silver paint are supposed to decrease the rate of heat emission by some 15 per cent. And still another that a matt surface is more efficient than a glossy one.
The writer has carried out a simple and most interesting experiment which upsets some of these theories. You can do the experiment yourself before deciding on what kind, colour and texture of paint will be most economical in fuel after other considerations have been taken into account. Junior can do the job for you. It will keep him occupied for hours — but remember the danger of handling boiling water.
A number of empty containers of identical manufacture and size are needed. I used 198 g (7 oz) empty powdered milk cans of the same make and treated them as though they were miniature radiators by giving them a light, brisk rub with fine abrasive paper followed by a wipe over with white spirit.
After this preparatory work has been completed give the sides and bottom of each container two finishing coats of whatever paints you are testing, thus following the ideal process for an actual radiator. Leave for four days for paint to harden.
Place each can in turn in the same position on the same working surface and keep the ambient temperature of the room as uniform as possible. The writer found this a bit difficult; it varied by some 8°F.
Fahrenheit measurement was employed simply because that was the only type of thermometer available when the experiment was carried out. Use Centigrade by all means and, for convenience, alter your readings so that they show comparable intervals. Whatever readings you decide upon, the intervals must be kept consistent.
Insert the thermometer through a hole in a piece cut from a polystyrene tile and secure it in position with string or with several wraps of rubber band immediately above and below, so that the bulb will be kept at a constant depth — suspended about two thirds down the can. Care should be taken to insert the instrument plumb centre in the orifice of each container.
Pour boiling water into a household measure until the pint mark is reached and transfer to the container. Place the thermometer-and-tile ‘lid’ on top.
When the temperature drops to 190° make a note of the time, and then at every 10° until 120° is reached.
Here are specimen readings which apply to polyurethane olive green eggshell paint.
The third main column shows the first recorded time indicated in the second main column (7 hours, 8 min., 10 sec.),
reduced to zero and each succeeding time altered to correspond. That is, the time taken for the temperature to drop from 1900 to 160° is 18 minutes 46 seconds.
From these figures you can draw a graph similar to that shown. It shows that the time taken for a coarsely milled gold paint to drop 70° (190-120) is 84 minutes, whereas that of the polyurethane is 61 minutes — which indicates that the latter heats a room quicker and so gets the maximum usable heat out of every pound’s worth of fuel used. If you are out all day and require a slow heat to keep the house aired, the former might be preferable.
To ensure accuracy, keep your line of vision consistent by laying your head sideways on the working top and lifting the container on to small sheets of wood of thickness equal to the space between each 10° reading on the thermometer.
Although the division markings on a thermometer are fine the mercury could take as long as fifteen seconds to travel from the top to the bottom of one marking; so prearrange the exact position when you will take each reading. Immediately the mercury passes each division is about the most convenient time to do this.
You may be puzzled as to why your graphs curve instead of being in a straight line. Newton’s Law of Cooling states: the rate of loss of heat is proportional to the excess of temperature over the surroundings. That is: the less the difference in temperature as a body cools, the slower the rate of heat loss. Then again, heat emitted from the container warms up the immediately surrounding air and, if there is no draught to carry it away, it will naturally slow down the rate of temperature drop. You will have noticed how quickly a kettle comes off the boil, but it then remains just warm for a considerable time.
You can, if you like, make the graph straight by using logarithmic paper; but ordinary squared graph paper is more readily available from any good stationer or artists’ sundries shop.
Glossy or matt
Conflicting theories are often propounded about the comparative effectiveness of a glossy and matt surface. One reason given against the use of matt is that a rough surface holds on to a column of air closely and prevents it circulating. But that action would be so minute as not to matter and, in any case, the theory is debatable. A claim that can be made in favour of gloss is that it is easier to keep clean.
If you want a matt surface, matt oil paint is better than ordinary emulsion because the latter, being waterborne, could rust an iron or steel radiator unless the priming coat is completely unbroken.
The final coat is the one that counts
The graph does not divulge what exact part is played by the paints in causing differences in the emission rate of heat.
You can answer this query by giving each painted container two coats of a polyurethane varnish and, after the coating has hardened, repeating the experiment. This time the rate for the emission for all containers will be almost equal, varying only by some seven minutes.
Which goes to prove that it is the final coat that really counts.
Not that varnishing is particularly recommended because, through differences in co-efficients of expansion between metal and paint, the thicker the coating the greater danger there will be of its lifting or crazing.
Where varnishing would help is if you already have your radiators decorated in a coarsely milled gold paint, and wish to retain that colour and speed up the rate of heat emission: all that is necessary in that case is to varnish the surface. If you want to change the colour, over-paint with the colour of your choice. There is no need to remove the existing gold.
Here is a physicist’s explanation as to why the final coat is the important and operative one: Heat is emitted mainly by radiation (by rays, as with a gas fire) and convection (the warming of air which circulates by virtue of density difference).
If a surface is painted or varnished or both, conduction also is brought into play, since heat cannot reach the surface without travelling through the film. Where the coating is of normal thickness, the temperature drop is so small that the thermal conductivity is of little consequence. So we can forget about conduction.
Now heat is radiated in a wave motion similar in principle to X-rays and light and radio waves, though the wave length is different. Heat emitted from a hot surface varies according to wave length, some materials being good at one wave length and bad at others.
Sheets of metal such as bronze which, in its milled form, is used in gold paint, are bad emitters at the normal temperature of domestic radiators, whereas sheets of solidified paint medium and varnish are good.
Relatively large metallic particles tend to float to the surface of the liquid paint medium of a metallic paint to form a sheet when the medium has dried and hardened. Paints with particles of that size therefore take longer for heat to pass through than metallic paints with particles small enough to remain enveloped in the medium. That is to say, the latter presents a surface that is more medium than metal. Overcoat coarsely milled gold paint with varnish and you will achieve something similar to finely milled metallic paint.
How convection works
Convection is the transportation of heat by the movement of heated air. As air is heated it expands, so becoming lighter than the surrounding atmosphere. Because it is lighter it rises, and cool air takes its place to be heated up in turn. As the first lot of warm air rises and leaves the precincts of the radiator it cools. As it cools it shrinks in volume and so becomes heavier. The increase in weight causes it to sink to floor level and take the place of the second lot of air that is being heated up.
And so we get a circulation of relatively warm air mingling with the cool — and eventually the whole room will warm up. The quicker the rate of heat emission the sooner this will happen.