Why Not
Page 49
Page 50
Page 53
If you've noticed an error in this article please click here to report it so we can fix it.
Pressure Cooling?
BY employing a cooling system for internal-combustion engines with a blow-off pressure of 10 lb. per sq. in. above atmospheric, it is possible to reduce the fan horsepower by approximately 50 per cent. of that required when a normal system is used; alternatively, the size of the radiator may be reduced to approximately four-fifths of the normal area.
This has been proved by the Ministry of Supply, and after exhaustive tests, raising the pressure to 10 lb. per sq. in. has been accepted as standard practice for military vehicles. .Nevertheless, many engine designers consider that the disadvantages of pressure cooling outweigh its advantages.
The percentage saving in terms of the power developed by the engine is dependent .upon the efficiency of the fan. This introduces a subject which is to . be Teviewed in' a subsequent article, but for the present purpose it is sufficient to emphasize that fan efficiency is normally low and that one-third reduction in fan power will save fuel and provide a noticeable increase in performance throughout the life of the vehicle.
Scale Formation
Vaporization losses are greatly reduced by raising the pressure, and replenishment is required less frequently. Scale deposits are slow to form and the durability of the radiator is increased.
Space considerations on a vehicle are not, as a rule, of comparable importance, and if advantage be taken of the smaller area of a pressurecooling radiator, there is a saving in first cost.
Some of the objections raised are, in part, obviously justified. For example, it is claimed that a high coolant temperature reduces volumetric efficiency. This is correct, but the loss of power should not exceed 0.3 per cent. for every 10 degrees F. rise in temperature, the actual reduction depending upon the air flow .under the bonnet and the position of the air-filter intake.
A pressure of 10 lb. per sq.. in. raises the boiling point to 239 degrees F., and it follows that. the power loss of 'a 190 -bhp.engine will be about h.p., assuming that the running . temperature is raised by approximately117, degrees F.
Cylinder-wall -temperatures will increase approximately in proportion to the rise in water temperature, and may produce ring sticking at extreme pressures. The use of a detergent oil and larger piston-ring clearances should obviate the troublein the_ rairge :of' temperatures considered.
The effect of high water tempera tures on the exhaust-valve seats and other hot parts of the cylinder head varies with the design of the water passages and with the rate of flow
locally. The formation of steam. bubbles in the vicinity of the hot spots is normal, even when the toptank temperature is below 180 degrees F.
Steam is condensed before it reaches the outlet, and make-up water takes its place. If the steam be free 'to ekape without being trapped in a pocket, and if the rate of flow towards the hot spot be sufficient to ensure a good supply of make-up water, the temperature of
the metal surface may drop as the top-tank temperature rises above 175 degrees F. This is because the latent heat of steam is a favourable factor in heat dissipation, provided that the steam is immediately replenished by water.
Valve-seat Temperature in tests of a good design with an unpressurized system, the temperature of the exhaust-valve seats progressively decreased as the top-tank temperature rose above 176 degrees
F., and at 212 degrees F., the tern.
perature conditions were similar to those existing at a tpp-tank.temperature of about 140 degrees .F
Other objections to pressure cooling relate to the greater difficulty of preventing leaks at the pump gland and-hose connections, but means are available for overcoming these problems. It has been suggested that a
driver who is accustomed to topping tip the radiator after a run may be scalded by water being brought to the boil when the cap is removed and the pressure is released. A suitable routine, based on increased intervals between replenishments, could be used to prevent any possibility of physical danger from this cause.
Many engine manufacturers consider that commercial applications of pressurizing should be limited to the use of pressure-cap settings of about 3 lb. per sq. in, to obviate loss of coolant by surging and to maintain a uniform pump performance.
Overcoming Ring Sticking The opinion is also general that piston-ring troubles could not be avoided at higher coolant temperatures and that satisfactory pumpgland sealing would be difficult to obtain. The pressure drop across a typical pump is over 10 lb. per sq. in. and the gland would have to seal at more than 20 lb. per sq. in. with a pressure-cap setting of lo lb. per sq. in.
The views of a manufacturer of military vehicles do not support these contentions. In the Scammell breakdown tractor, whichhas an engine delivering 175 net h.p. at 2,400 r.p.m. and a Maximum torque of 5,700 lb.in. at 1,400 r.p.m., the cooling system is raised in pressure to 10 lb. per sq. in. and its frontal area is 8.9 sq. ft. The core comprises four rows bf Clayton Still tubes in staggered formation, in front of which is installed the oil cooler.
The cooling system functions satisfactorily with an ambient temperature of 110 degrees F. when the engine is developing its maximum power. At maximum torque the system stabilizes in an ambient temperature of 100 degrees F. These results were obtained during dynamometer tests of a complete vehicle.
Lower Running Temperatures
Road tests of a vehicle towing a load, with power absorption regulated by the brakes, have given improved results. In the latter instance, temperatures were reduced by as much as 10 degrees F.
Seam-lett engineers see no reason why connections should give trouble at radiator pressures up to 10 lb. per sq. in. and that pump glands of modern design should be able to seal satisfactorily at the resultant gland pressure. Given suitable piston and ring clearances, and the choice of a satisfactory lubricant, no engine trouble should be experienced at the higher temperatures.
Other interesting features of the system include an approximately B16
tangential water entry which reduces turbulence caused by the formation of steam when working near boiling point. The conical pot, which has been a characteristic of some Scammei systems for a number of years, was originally introduced in conjunction with a shallow header tank to ensure coverage of all tubes when negotiating severe transverse slopes.
The systems of the war-time vehicles Were fitted with a float and rod extension to provide a visible level gauge.
The reduction of under-bonnet temperatures is most desirable, whatever the type of cooling system employed. This is achie‘td on the Seaman11 by water cooling the exhaust manifold. The value of a pressure cap in reducing surge has already been Mentioned. Surge may be further reduced by submerging the tiller neck and so trapping a certain amount of air, which acts as a cushioning agent and provides the necessary space for expansion. This also applies to a non-pressurized system.
A temperature of 240 degrees F'. is well below that at which the strength of the solders used in the construction of the normal type of radiator matrix is materially reduced, or at which diffusion between the solder and copper tube takes place. A radiator with a coreformed by tubes with an interlocking seam can be safely employed with pressures exceeding 10 lb. per sq. in.
The heat-dissipating capacity of a cooling system is a direct function of the difference between ambient and radiator temperatures. The maximum top-tank temperature is that at which the water boils, and the increase in performance made possible by pressurizing is proportional to the amount by which the boiling point is raised.
Overload Capacity The advantage has already been expressed in terms of the possible
saving in fan power or radiator size: there are other considerations of importance related to operation in high ambient temperatures and at high altitudes.
Overload capacity is a term used to describe the radiator's reserve of cooling potential to meet extreme conditions. This capacity is ailimportant in many overseas countries, where hot day temperatures are experienced at altitudes which reduce the boiling point of an atmospheric system by 10 degrees F. or more. Steep gradients and following winds often increase the difficulties of adequate cooling.
For successful performance under such conditions, it is usual to provide a special tropical radiator, which is more costly to produce and tends to overcool the engine when operating at lower temperatures. Means must then be provided, for . optimum cooling, of restricting the air flow to the matrix, a practice which increases the fan. horsepower required. In some mountainous countries, day and night temperatures may vary by 70_ to 100 degrees, and a high overload capacity is essential.
Effect of Altitude
The pressure-cooled system is particularly advantageous in such districts. The effect of altitude can be discounted, and the greater, overload capacity for a given rise in temperature may make modifications to a standard system unn'ecessary.
If the temperature at sea level rises from 70 to 110 degrees F., the potential maximum difference between the ambient temperature and the boiling point of an atmospheric system is reduced from 142 degrees F. to 102 degrees F. The proportional reduction in heat-dissipating performance is, therefore, about 30 per cent.
If the radiator be pressurized to 10 lb. per sq. in. and the boiling point raised to 239 degrees F., the potentional difference, under these special conditions, decreases from 168 to 128 degrees F., a reduction of approximately 24 per cent. At higher altitudes, the relative safety margin is widened by the maintenance of a constant boiling point. Even if full advantage be taken of the system to educe fan horsepower or radiator dimensions, the comparative overload. capacity ifinprOves with an increase of ambient temperature.
The influence of altitude on engine output is relevant to the subject of cooling. For each 1,000 ft. increase, the power drop is 4 per cent, of the rated horsepower, the loss becoming proportionately less at high altitudes. The heat to .cooling water is also reduced, but not by the same percentage. Compression pressures fall and consequently the efficiency, This is more marked in a petrol engine than an oil engine.
When the engine is developing less power, a steep gradient may call for the use of a lower gear, and efficiency is further reduced by heat losses in the gearbox. In certain instances, the reduction in air velocity will also raise the radiator temperature. The claim that power loss offsets the disadvantage of a lower boiling point is not, therefore, borne out in practice.
Over-boiling Tank
The overload capacity of a radiator may also be increased by employing an over-boiling tank. This is nor-, mally used in conjunction with a cap• setting of about 3 lb. per sq. in. and comprises a separate water container into which the steam is passed if the coolant boils. '
Steam is condensed by the water and a fall in radiator temperature causes make-up water to be drawn back into he header tank. The tank can be fitted at a comparatively low cost and its size may be increased without affecting the normal running temperature of the engine.
Whilst a glycol-base anti-freeze solution has a higher boiling point than water, most alcohol-base solutions boil at a much lower temperature. A change in temperature from below freezing to over 100 degrees F., as may be experienced in some mountainous sub tropical areas, will not lead to premature boiling of a glycol mixture, but the use of an alcohol solution may .have serious consequences. In a pressurized system, the pressure cap could be set to raise the boiling point. sufficiently to overcome this problem. ' The potential of interior heaters is increased by raising the coolant temperature, and a saving should be possible in the size of the diffuser units and of the pipes required.