There is some similarity between wet and dry systems, in that both are capable of being left to Nature to operate, or may be given a mechanical boost. The one fact to be remembered about Nature, apart from the fact that it costs nothing, is that while it is inexorable it is slow and comparatively weak. For instance nothing will stop warm air at ground floor level from rising to the upper floor. But shutting a bedroom door will prove a serious barrier, and it could take a week for the warm air to get through. That is hardly a practical proposition, but it serves to show incidentally that free flowing warm air is most likely to succeed with open planning.
In a dwelling so constructed that free flowing warm air can circulate, there could be a lot of sense, and of economy in first and running costs, in having such an arrangement – it can scarcely be called a system. For best operation it calls for a good standard of loft insulation, and for double glazing in bedrooms. Anyone who does not fancy a well warmed bedroom has only to keep the bedroom door shut to stay out of the circulating zone. The human body is not seeking a specific temperature, but a comfortable zone of temperature above the minimum level of acceptance.
The principle upon which all air heaters are based is that there is a heat exchanger. Fuel is burned on one side of it, and air passes over the other side, becoming warmed The manner in which the principle is applied illustrates mainly the ingenuity of manufacturers. The commonest of the free flow units is still the so-called ‘brick central’, because that is what it is, both brick and central. It can house quite a large unit, which calls to mind that in the USA the air heater is called the furnace. No doubt the use of a brick chamber was first employed to make safety obvious, giving some thermal storage as a bonus. In later times units were built in which the chamber was more lightly constructed, for example of asbestos cement sheets. The cladding, brick or otherwise, is not a part of the heater. There is a flue, and grilles at high level to allow warmed air to escape. At low level, apart from the combustion air inlet, there is at least one inlet for cool air to be warmed, and a measure of control is exercised over the output by the extent of opening of lower air dampers.
In cases where the brick central was not situated in a way to give warm air an easy passage upstairs it became customary to add a short duct from the top of the unit to a convenient discharge point. The decline in popularity of the brick central is no doubt due to its massiveness, both real and apparent. But that apart it is capable of doing a good job.
The same function is being performed, for much less space taken, by an appliance which looks very like a large gas fire. Using fuel oil, which may be piped in or contained in an attached storage vessel, this unit must be connected to a chimney, so that choice of location is limited. Some models have a radiant section, even a back boiler but the bulk of their output is in warm air, discharged into the room where the chimney is. In some cases this warmed air is able to travel freely. Where it is not, its passage is aided by a simple extractor fan, let into the wall of the room at high level, and discharging for preference into a hall or passage way by which the warm air may go on its travels to the rest of the dwelling. Reports from users seem to be predominantly favourable.
For warm air heating which qualifies as a system we must turn to ducted air. It obeys all the rules laid down for heating description, in that it may be a full heating system, or partial, or selective or background. The full range is available only to full duct systems. A stub duct or abbreviated duct system cannot fulfil the requirements of full central heating in terms of individual room control.
The heart of a ducted system is the fan, almost always incorporated in the unit. The fan may be compared directly with the pump in small bore systems. It provides positive flow, speed, assurance, and it frees the system from the annoying limitations which Nature imposes upon a system without artificial power. Notably of course, in natural systems warm air must always be travelling in an upward direction. With fan power, it will travel for long distances horizontally, and even go downward, as into an underfloor duct system.
Ducting is large, not often an item which can be added to an existing house without becoming a nuisance. Consequently a full duct system is usually built into a new house, and the architect or builder should be given early notice so that he may plan it in. Ducting may be run under floors or in lofts, in walls. It may house the required cross sectional area in various shapes. A 6 x 8, for instance, giving an area of 48, may be made as 12 x 4 if it has to disappear in a four inch wall. Although the object is to make duct runs direct, short and economical, all flow ducts are insulated to guard against much unplanned heat loss.
Another feature of long flow ducts is that the cross section undergoes gradual reduction. Each length of duct is sized according to the amount of air it has to carry, in order to maintain a steady air velocity. This is necessary in order to achieve a certain velocity of air delivery at the outlet points. It will be seen that as draw-offs are made from the main air duct, the quantity diminishes and the area of the duct must be reduced to suit the new reduced volume. Such calculations are made on the basis of all discharge points at work. In any other event if the discharge rate seems too strong it may be adjusted at the damper which controls the discharge.
It has already been suggested that an arrangement by which warm air goes where it pleases does not constitute a system. A system is a logical development, with beginning and end, and the logical development of a ducted system is to return the air, or at least a good deal of it, to the starting point. Thus, a ducted system includes a return air duct, and it is upon this recycling that the economy of the system mainly depends. By comparison with the flow ducting, a return air duct is a simple affair. Usually it collects from main points only, is not insulated, and not made leakproof to more than an elementary degree. The exception to the last point occurs within the heating chamber, or where the heater is situated. There the return air duct must be very effectively sealed in itself and into the heater. If the heater is a conventional flue type it relies upon access to a continuous supply of fresh air for combustion. It cannot tolerate strong competition such as would occur if there were an aperture in the return air duct, already under strong fan suction.
More than that, if incomplete combustion were to occur, with probably the formation of carbon monoxide, it would get into the duct system and be distributed throughout the house.
The fact that a duct system necessarily forms a communication channel from room to room has not escaped the attention of fire prevention officers, and it is advisable to check with the local authority in case their interpretation of the Building Regulations bears upon your proposals.
Sound also passes readily along this type of communication channel and this seems to be particularly true of the return air duct, no doubt because it is intentionally straight. Fan noise will travel in it unless checked, and the most effective check is a bend.
At this point we should mention that anyone seriously intending to install ducted warm air would need to go into the subject in greater depth than this post can, for ducted warm air has never been regarded as a very suitable matter for the amateur. Perhaps, with its new building context, it rarely offers the chance. However, all the leading manufacturers of warm air units, who together form the Warm Air Group of the Society of British Gas Industries, have collaborated in producing a design manual, which sets out the current British practice.
Anyone who objects that it is a gas document should be reminded that systems are more important than fuels. The majority of the manual concerns duct systems, which do not alter just because the fuel does.
We have seen how main ducts are stepped, I.e. progressively reduced in cross sectional area as the volume diminishes, in order to maintain a reasonably steady air velocity. If more than one main duct is run from the plenum then the design intention should be to give roughly equal duty to each main duct.
This intention can rarely be achieved in practice, and it is customary to add the final adjustment, known as balancing the system, by means of dampers. The comparison with wet systems, and balancing valves on radiators, will be apparent. Sometimes the need for an incorporated damper will be obvious. For instance if a radial system were to consist of one short duct and a number of longer ducts of about equal resistance, the short duct would have an advantage, and this would require a damper to neutralise it.
On a more general level, it is good practice to include a damper in every register. Formerly this called for a separate device, a stack damper, but nowadays it is much more common to include a balancing damper in the total structure of the register. This is independent of any modulating mechanism, including an ‘off position, which is provided for the user to control. Balancing dampers once set should never need alteration.
Where to put Registers and Grilles
Registers or diffusers are the terminal fittings through which warm air is discharged into the room. They should bring about as uniform a room air temperature as is possible, by low speed draught-free air movement. Their relationship to the room, and if more than one to each other, is therefore of first importance. Their ability to project air sideways and forward is limited by the air velocity, which must not create a draught, but their influence is greatly extended by secondary means, by disturbance and by entrainment of air surrounding the ejected warm air stream.
As a convenient rough rule we may assume that one register will deal effectively with a room, or part room, which has a small to medium square plan area. This would include such rooms as 4 x 4m, 5 x 5m or 4 x 5m. Rooms which are distinctly elongated or irregular are best treated by being visualised as squares, and each square treated separately.
The decision whether to fit the diffusers in the floor, at low or high level in the wall, or in the ceiling, is in part dictated by local circumstances, in part by local custom. British preference has always been for the low level wall diffuser, and if this is arranged to give an angular downward deflection to the issuing air it can achieve very good mixing, with even temperature throughout the vertical gradient. It requires the ability to conceal the ducting in the wall, but that is a common condition. Floor diffusers require that an insulated duct can be accommodated in the floor.
High level diffusers, and in particular the ceiling type, are not for general application, since their natural trend would be to promote a great temperature gradient, hot at ceiling level and cold at the feet. They would be well suited to a place where the floor is already well warmed, possibly by being over another warm room.
It must be pointed out that the air flow from a register must be unimpeded, furniture being the most likely impediment. But in nearly all cases it should be possible to accommodate the furniture to the heating, not the other way round.
The number of grilles used will be less than the number of registers, and the system does not even call for a grille in every room. But if a room is equipped with a register but no grille, then there must be a permanent opening at high level through which air can escape to the nearest grille. The typical site is one in which a central grille in a hall or passage collects from one or more adjacent rooms. The permanent opening is most easily provided by relieving the top of the door, though an opening may be made in the wall. No such arrangement must be made with the bathroom, kitchen and toilet, and if these abut a hall equipped with a return air grille steps should be taken to see that they cannot contribute accidentally to the return air supply. Their ventilation should be arranged separately, possibly via a window ventilator.
When grilles are fitted in the same room as registers it will be seen that the zone of negative pressure in their vicinity will influence the air flow pattern. This fact can be used with advantage if the grille is situated with this in mind to create air flow in the required direction.
Since the object is to involve as much of the room as possible in the warming current of air, it may be taken that broadly the air will enter at one end of the room and leave at the other.
The Compromise System
We have considered the free flow unit, which uses no ducting; and the full duct system, which is best suited to new building.
It is often possible to obtain many of the advantages of the second, in application to existing property and with a minimum of disturbance. The compromise is called a stub duct system, and as the name indicates it uses only short duct lengths. These should be long enough to send the warmed air in the main directions it is required to go. The feasibility is very largely dependent upon the layout of the house, and the ingenuity of the designer. If there is more than one position where the heater could go, the object is to choose the one which gives most chances to run effective stub ducts.
For houses of two storeys the object may be to transfer part of the air ouput to the upper floor by a single duct, then taking stub ducts from this riser to convenient nearby points.
The conditions imposed upon a return air duct are, as we have seen, not too onerous. A return air duct could for instance be run across a loft, which might well be unsuitable for the flow ducting. In such a case the return air duct could well be more extensive than the stub duct system, and so arranged that by putting grilles at negative pressure at the extreme ends to which it is hoped air will travel, it will induce air travel. Another very important matter concerned with ducting is fan power. Clearly it takes less fan power to operate a stub duct than a full duct system. Even nowadays, when the matter has been fairly well regularised, it is still desirable to ask, when buying an air heater, not only the rated heating power but also the fan power, whether for stub or full duct system.
Control of Warm Air Systems
In free flow units there is only the heater to control. This may be given automatic control in one or more respects, in addition to any fundamental controls, e.g. for flame safety, which may be incorporated in the burner assembly. A room thermostat, for example, may be situated in perhaps the living room, wired back to stop and start the burner. A clock will also decide when the unit is to be started or stopped.
Both these controls are applicable to ducted unit systems too, the room thermostat being more likely to give a sensible response in a system which is positively fed with predetermined amounts of warmth via ducts, than in conditions which can be affected by a draught from an open door.
It is possible to obtain thermostatically operated registers, though it is doubtful whether these contribute enough to the end result to justify their inclusion. The most important part of a duct system for the user, if only in the all-important matter of relating comfort to economy, is the register. It should become as much a routine to shut a register when leaving a room as it is to turn off the light.
Refreshing Warm Air
A warm air system was once regarded as one of diminishing vitality, in which the oxygen content grew less, the carbon dioxide, tobacco smoke and so on increased, presumably to a point at which it would not support life. To offset this, arrangements were made to introduce a controlled quantity of fresh air from outdoors into the circulation, and this would displace an equal volume of stale air.
Experience quite quickly showed that what might be called a laboratory view of the system at work had little relevance to what happened in practice. This is, to put it crudely, that a typical house leaks like a basket at most of its joints, floor boards and so on. The real problem became not one of introducing fresh air but of excluding too much of it, for it will be seen that fresh air is cold air and the need to treat it lowers the efficiency and therefore the economy of the system at work.
Although the introduction of fresh air receives a mention in the latest official document on ducted air design, our recommendation is to ignore it, unless you are convinced that you have a remarkably well constructed and weatherproofed house entirely free from draught leakage at any point.
The Nature of Warm Air Heating
Wet systems, we have seen, work ultimately by warming air, but it is an indirect process and, particularly in the case of radiators, the total effort is divided to give a radiant factor. By their nature they are, too, rather slow, even in spite of pumped circulation. It can take half an hour or more from starting the system before an improvement is noticed in the room temperature. Even then the convected or warm air part of the improvement lags behind the radiant one.
A warm air system is quite different. It is direct, the warm air coming ‘from factory to user’ with no intermediate process. It is simple in action, being convected only, and it is rapid. The room near to the register will begin to feel warm minutes after the apparatus goes to work, and it is quite possible to have the room up to temperature in about 15 minutes. But that is conditional upon limiting heat loss.
So in return for good insulation you will save money and will get the greatest benefit from a warm air system, in speed of response. The good sense and economy of insulation apply of course with equal force to wet systems. But for anyone specially interested in taking advantage of the speedy action of warm air it might pay to think about the detail of insulation. We shall go into the subject more fully later on, but a brief mention here is not out of place.
Cavity wall insulation is a very important part of house treatment, since it achieves a spectacular heat saving over the greatest surface area the house offers to the outside world. It is also very convenient. The cavity is a ready made former for the insulation, giving it standard thickness but not allowing it to take up any useful room whatever.
Cavity wall insulation stands in the way of that heat which used to pass through the inner brick leaf, over the air gap, through the outer leaf and away on the wind. Now most of the warmth passes through the inner leaf, and is stopped. Consequently the temperature of the inner leaf gradually builds up until it is approaching room temperature. This is both good and open to question. It is good because it is saving heat, also because once the wall is warm it begins to act as a low temperature radiant panel, giving back to the room in acceptable form that heat which it has absorbed.
It is open to question because it has a built-in time lag. The inner wall or leaf weighs several tons and will soak up a lot of heat, which during the warming period is not available for the main job of warming the room.
If heating is continuous over several days at least then the incidence of that amount of heat in the total output diminishes. But if the heating is on for long enough to warm the wall, then shut off for quite a time, the warmth in the wall will gradually dissipate into the room and be lost in air changes and the like.
Perhaps it will be clear, then, that in spite of our gratitude to cavity wall insulation, there are times when we would like to exclude even the inner leaf from our heating system. Such times are particularly bound up with speed of response, and this brings us back to air heating. The way to bring insulation right into the room is by the use of a thermal inner lining. This in its most elementary form may be seen in expanded polystyrene tiles, about 3mm thick, which can be stuck to the walls. They do a very useful job, and one should always buy a fireproofed grade.
A full scale inner lining is usually a little more thermally resistant than that, also more mechanically strong. It need not make more than one inch difference to wall projection, and a good way to construct one is to fix 15mm or 25 mm battens to the wall and then clad the wall in hardboard, with mineral wool packing in the gap. The wall must not be subject to dampness, though any such wall may be lined if damp is first totally excluded by a heavy and continuous coating of a bitumen compound or other wholly waterproof substance. The hardboard may then be treated as the wall surface, and papered or painted.
Tests carried out by a firm now unfortunately defunct showed that a warm air system could be shut off overnight, the room temperature falling only a few degrees by morning, and within 15 minutes of starting the room temperature was back to its prescribed level. The overnight maintenance of temperature could be due in part to the fact that the furniture in the room had acted as the heat buffer or reservoir. It was a remarkably good demonstration of comfort with economy.