The S.U. Carburetter

SU Carburetter Company

By P.G.G KNIGHT

INTRODUCTION

THE VAST MAJORITY of modern spark ignition internal combustion engines rely upon carburetters to provide them with the finely atomized fuel/air mixture which is necessary to enable them to perform satisfactorily. It is the function of the carburetter to provide good atomization and the correct mixture strength, under all operating conditions of the engine. The method used to do this in all carburetters is to speed up the velocity of the air by means of a venturi or choke, and to use the consequent reduction of pressure in the venturi to draw fuel from the float chamber through a suitable jet orifice into the air stream. Since the power that an engine can develop is dependent upon the quantity of air consumed, it is desirable that the mixture of fuel and air should be carried out with as little restriction as possible to the overall air flow.

On a fixed choke carburetter the venturi must be sufficiently small to ensure adequate mixing during conditions of low air consumption. On engines that require to operate over a wide speed and power range, it is inevitable that a choke sufficiently small for use with low air consumption will give an excessive depression at maximum air flow. An obvious solution to this problem is to arrange for the size of the choke to be increased at high engine output and decreased for bottom end performance.

GENERAL DESCRIPTION

It is now well over fifty years since the Skinner brothers first introduced their constant depression carburetter. The increasing number of cars using carburetters made to this same basic principle indicate that the idea was sound. The near constant depression, maintained in the carburetter under all operating conditions, is obtained by means of automatic regulation of the choke size. It is sufficiently high to ensure that good atomization is obtained, yet the depression is kept as small as possible in order to maintain engine filling at a maximum. At the same time as the main passage above the jet is varied so too is the area of the fuel jet by means of a tapered needle, the whole being regulated by the rise and fall of the piston under the influence of throttle opening and manifold depression. This depression is itself dependent on the power requirement of the engine under the conditions at which it is being operated. In this system one jet only is used and the avoidance of multiple jets means that any possibility of flat spots, which may occur during the changeover from one jet to another, is eliminated. Correct jet discharge areas are obtained by the accurate dimensioning of the needle, which must be carefully matched to the requirement of the particular installation.

TYPES OF UNIT

Types of unit in current use are: (1) the H type, (2) the HD type and (3) the HS type. In addition, a dual choke instrument designated the DU6 has been manufactured in very limited quantities. This latter type has, however, never been made in large production quantities, nor is it anticipated that it ever will be.

The H type carburetter

This is, perhaps, the most familiar type of S.U. carburetter, being the one which has been in general use for the longest period and therefore in larger quantity than any other type. It is shown in diagrammatic form in Fig. 22.

It consists of body, piston/suction chamber, jet, and float chamber assemblies, the fuel being fed from the float chamber through a passageway in the float chamber arm into the fuel passages of the body, and thence through cross drillings in the jet assembly.

It has been made in horizontal and semi-downdraught form with throttle disc sizes ranging from 1½ in. to 2 in. diameter. As with all current S.U. carburetters, its size is designated by the type letter followed by a number. This number indicates the diameter of the butterfly, and is the number of eighths of an inch that the diameter is larger than 1 in., e.g. an H1 instrument has a bore of 1½ in. and an H8 has a bore of 2 in.

The height to which the piston is lifted is controlled by the amount of air passing beneath it. When the piston is at the bottom of its stroke, with the engine idling, opening the butterfly allows the manifold depression to be communicated to the main volume of the body and then through a cross drilling in the lower part of the piston into the suction chamber above the piston. This depression immediately lifts the piston, allowing a mixture of air and fuel to pass the lower side of the piston and relieve the depression. The piston height is therefore stabilized at a depression controlled by the weight of the piston, the load of the piston spring, and the areas of the large and small diameters of the piston. It will be noted that the underside of the large diameter of the piston is open to atmosphere. The air/fuel ratio is controlled by the diameter of the needle in the fuel jet. The optimum dimensions of this needle are normally found by experiment on an engine dynamometer and by road testing of the vehicle. Once determined, a profile cam is made to give the desired needle dimensions, and production needles can then be very accurately reproduced to the desired shape.

From this it will be seen that once the correct needle has been specified, the jet has to be set to a datum dimension on the needle to ensure that the desired mixture is obtained throughout the range of the engine operating conditions. This position is set when the carburetter is tuned for idling.

Most S.U. carburetters incorporate a piston damper, the function of which is to restrict the speed of lift of the piston on snap throttle openings, and to allow the piston to fall at its normal speed on throttle closure. This one-way damping is obtained by means of a non-return valve situated at the base of the damper.

When the speed of piston lift is retarded an additional air depression is put on the fuel in the jet resulting in an increase in the quantity of fuel discharged. A richer mixture is thus obtained until the piston resumes its position of equilibrium. This enrichment is necessary to provide satisfactory pick-up. The piston damper also improves cold starting and drivability from cold.

It is of the utmost importance that the fuel jet is assembled concentric with the needle. In order to allow for liberal adjustment, the complete jet assembly is manufactured with side clearance between the jet bearing and the carburetter body. Once the assembly has been correctly centred (with the jet full up and the piston resting on the bridge) it is locked in position by means of a large jet locking screw. The components of the H type jet are shown in order of assembly in Fig. 23. On most H type carburetters the float chamber is firmly secured to the body by means of a banjo bolt. On some engines, however, where vibration has led to malfunctioning of the float chamber, it has been found necessary to use a flexible mounting between the float chamber arm and the body of the carburetter. This has been achieved by means of rubber grommets secured by a shoulder bolt, or pillar bolt on later models.

Since the carburetter is of the constant vacuum type, the depression at the jet never falls below its normal operating value, and it is for this reason that variations of fuel level in the float chamber are unimportant.

On many cars it is found desirable to use vacuum-operated ignition advance to obtain optimum part-throttle consumption figures. The take-off point for this vacuum is arranged slightly to the air intake side of the butterfly, and in such a position that opening the butterfly allows the throttle disc to pass over the vacuum take-off point so that it then communicates with the manifold depression. By this means the vacuum is small at the distributor during idling and full-throttle conditions, and is large at part throttle, being at a maximum when the throttle is open a few degrees.

Enrichment of the mixture for cold starting and running is obtained by lowering the jet. The lever used to carry out this function is also provided with an additional link which operates the cam and opens the butterfly by a predetermined amount. Some lost motion is built into this system, usually by means of additional clearance at one of the pivot points. This allows the throttle to be opened a few degrees before the jet is dropped, to provide for the semi-warm condition when no additional enrichment is required but the engine has not yet warmed sufficiently to prevent stalling at idling.

One other item now fitted to all S.U. carburetters is the piston lifting pin. This is used when the engine is not running, to lift the piston assembly for the purpose of checking that the needle is correctly centred and that the piston falls freely. It is also used to lift the piston a small amount when the engine is idling in order to check the mixture strength.

On earlier models, before the fitting of the lifting pin, a hole was made in the carburetter body below the large diameter of the piston. The piston could thus be raised by means of a wire inserted through this hole.

The HD Type

The HD type of carburetter (Fig. 24) is very similar to the H type, the main difference being that the fuel to the jet is sealed by means of a diaphragm in place of cork glands. A separate by-pass idling passage is incorporated and the enrichment mechanism is redesigned.

This type of unit is made in three sizes, HD4 (1½ in.), HD6 (1¾ in.) and HD8 (2 in.), and all can be supplied as either horizontal or 30° semi-downdraught units, but no flexible mounting for the float chamber is available.

Whilst the jet centring is, in general, carried out as on the H type, it is less accessible and the float chamber and diaphragm sandwich pieces both have to be removed before the jet bearing locking screw (21) can be slackened or tightened. Once properly centred, however, the jet is well protected against extraneous blows and is therefore unlikely ever to require further adjustment.

In order to minimize the likelihood of fuel entering the vacuum advance unit on the distributor, the vacuum ignition take-off point is mounted at the top of the butterfly, and for this reason the butterfly opens in the opposite direction to that of the H type.

For idling, the butterfly is held completely closed and the volume of charge required is regulated by a metering valve (18) situated at the side of the body. The setting of the jet is by means of a pivoted fork (7) controlled by an external screw (8), and the additional throttle opening required for fast idling is obtained through a sliding rod (13) operating in a bearing cast into the body.

The HS type

This type of instrument (Fig. 25) is also similar to the H type; the main difference is, again, in the method of supplying fuel from the float chamber to the jet. In this instance a nylon tube (4) is used, connecting the base of the float chamber to the bottom of the jet so that, as with the HD type, cork gland washers are not required. The butterfly opens in the same direction as on the HD type, with the vacuum advance connection mounted above the throttle spindle.

The float chamber assembly differs from the H and HD types in both its mounting method and the type of float used. The float chamber is bolted horizontally through the body and can be mounted either flexibly (on rubber) or rigidly. Mounting may be at 30°, 20°, or horizontally.

When a brass float is used, it is guided by flutes on the wall of the float chamber rather than by a stud through the float. More recent carburetters have a nylon float anchored to the lever arm. The float chamber lid is attached with three small screws, and fuel connections to the lid are by push-on fittings.

The DU6 carburetter

These dual choke carburetters with a 1½ in. diameter throttle bore were used almost exclusively on the Coventry Climax 1½ litre twin cam engines. New units are no longer available.

Although they operate on the basic S.U. principle, they include an additional full-throttle, diaphragm-operated weakening device to overcome excessive mixture spread between full and part throttle conditions. Without this device, correct part-throttle tuning would produce an over-rich mixture at full throttle, and conversely, correct full-throttle tuning would result in an over-weak mixture at part throttle.

Another feature of these instruments is the positioning of the float chamber between the two chokes of the carburetter. This requires a narrow, rectangular, doped cork float, which is satisfactory in petrol but unsuitable for methanol blends.

SETTING OF S.U. CARBURETTERS

Once the correct carburetter needle and spring have been established, tuning is normally confined to correct idling adjustment. Before any adjustment, however, it is advisable to ensure that the ignition system is in good order, the engine is mechanically sound (tappet clearances, compressions, etc.), and that the carburetter is mechanically sound with its jet correctly centred. The engine should also have been run to its normal operating temperature.

Setting of H type carburetters

After the engine has reached the correct operating temperature, close the throttle completely by unscrewing the throttle adjusting screw until it is just clear of its stop, then open it by screwing down this screw 1½ turns. Remove the piston and suction chamber, disconnect the mixture control wire, and screw the jet adjusting nut until the jet is flush with the carburetter bridge (or "full up" if that position is unattainable). Replace the piston and suction chamber assembly and check that the piston falls freely onto the bridge. Turn down the jet adjusting nut two complete turns (12 flats). Restart the engine and adjust the throttle adjusting screw to achieve the desired idling speed as indicated by the ignition warning light. Then, adjust the jet adjusting nut until the fastest idling speed is achieved with even firing. During this adjustment, ensure that the jet is pressed upward and in contact with its adjusting nut. As the mixture is adjusted, the engine may run faster; you may need to unscrew the throttle adjusting screw slightly to reduce the speed.

Check the mixture strength by lifting the carburetter piston (using the lift pin on the side of the carburetter body) by approximately 1/32 in. If:

• The engine speed increases and continues to run faster, the mixture is too rich.
• The engine speed immediately decreases, the mixture is too weak.
• The engine speed momentarily increases very slightly, the mixture is correct.

When the mixture is correct, the exhaust note should be regular and even. Irregular, splashy misfiring with a colorless exhaust indicates a weak mixture, while a rhythmic misfire with a blackish exhaust indicates a rich mixture. Reconnect the mixture control wire ensuring approximately 1/16 in. free movement before it pulls on the jet lever. Set the dash panel mixture control knob to its maximum movement without moving the carburetter jet (about 5/8 in.), and adjust the fast idling cam screw to give an engine speed between 800–1000 rpm (when hot). For engines with large valve overlap, a fast idle is required, and it may be difficult to obtain an extremely smooth tickover.

Setting of HD and HS carburetters

The technique for setting these carburetters follows the same general principle as for the H type, though you must familiarize yourself with the different adjustment methods specific to these units.

Setting multi-carburettor layouts

When setting two or more S.U. carburetters on an installation, the same general principles apply as for a single unit. However, note that most manifolds using multiple carburetters are manufactured with balance pipes that interconnect the manifold passageways of each carburetter. An adjustment to one carburetter may affect the mixture in all cylinders to some degree.

Recent multi-carburettor installations (HS type and some HD units) use a lost-motion linkage to operate the throttle spindles independently. This allows fast idle to be achieved without moving the accelerator controls, reducing the load on the choke control knob, and permits individual slow-running screws to be adjusted without removing throttle clamps.

Before setting the carburetters, run the engine until it reaches its normal operating temperature, then slacken the clamping bolts on the throttle spindle connections and fully close all throttles by unscrewing the throttle adjusting screws. Remove the piston and suction chambers, disconnect the jet control linkage by removing the fork swivel pins, and screw the jet adjusting nuts until each jet is flush with its carburetter bridge. All jets must be in the same position relative to their bridges.

Replace the piston and suction chamber assemblies and check that the pistons fall freely. Top up the dampers with oil and replace the damper caps (ensuring the correct type is fitted). Then, turn down the jet adjusting nut two complete turns (12 flats), restart the engine, and adjust the throttle screws equally to achieve the desired idling speed. Use a tube to listen for a consistent hiss at each carburettor intake; adjust the throttle screws until the hiss is similar on all units to synchronize the throttles.

Finally, adjust the mixture by screwing the jet adjusting nuts uniformly until the fastest idling speed is achieved with even firing. If the engine runs too fast, unscrew the throttle screws slightly (equally for all units) to reduce speed. Check the mixture strength of each carburettor individually (using the previously described method), and re-check after all adjustments due to their mutual influence. Adjustments should be made primarily on the carburettor corresponding to the piston whose lift indicates a discrepancy.

Once the setting is correct, tighten the throttle spindle interconnection clamping bolts on the couplings, and reconnect the jet control linkage and mixture control knob as for a single carburettor.

Adjustment of Jet and Throttle Interconnection

With the cam-type (or the preceding rocker type) jet and throttle interconnection (as shown dotted in Fig. 22), the outer adjusting screw (1) should be about 1/64 in. (approximately the thickness of a visiting card) from the cam face or rocker face when the engine is warm and idling on a closed throttle. With the rocker type, this gap should not be exceeded, but with the cam type, a larger gap may be used if desired. If the jet adjusting nut is altered substantially in position, then the adjusting screw (1) may also require readjustment.

Piston Springs

On 1¼ in., 1½ in., and 1¾ in. diameter horizontal carburetters, the red (4 oz) spring is normally used for initial testing. In most installations, this spring load is effective if the carburettor size has been correctly chosen. When the correct spring is fitted, full piston lift is usually obtained at full throttle and approximately three-quarters of the maximum rpm. If maximum power is required, a slightly larger carburettor may be chosen, which will not achieve full piston lift until nearer maximum rpm.

Oil Dampers

After the carburetters have been correctly set, check that the oil damper reservoir in the piston rod contains sufficient oil. This check should be carried out approximately every three months. Use oil of grade SAE 20 (not thicker than SAE 30). Simply unscrew the damper unit and pour oil into the hollow piston rod until it is within 1/8 in. of the top.

Float Chamber Fuel Level

The fuel level on an S.U. carburettor is not critical and need not be measured with extreme accuracy. The normal level is 3/8 in. below the rectangular inner face known as the jet bridge. Because this is difficult to observe—even with the suction chamber and piston removed and the jet fully dropped—a simple mechanical check is used. This consists of sliding a check rod of a specific diameter between the lid face and the inside curve of the forked end of the hinged lever when the needle valve is in the "shut off" position. For a 1⅞ in. outside diameter smaller float chamber and a larger one of 2⅛ in. outside diameter, a 7/16 in. rod is used. For the HS type, a 5/16 in. rod is used with a brass float, and a 1/8 in. rod with a hinged nylon float. If the fork does not conform within 1/32 in. of these check figures, it must be carefully bent at the start of the fork section to correct the deviation, keeping both prongs level. Alter the fuel level only if flooding is a problem; note that a too-high level can cause slow flooding (especially on steep drives), though flooding may also be due to grit jamming the needle valve, friction in the float gear, excessive vibration, or a porous float.

Effect of Altitude and Climatic Extremes on Standard Tuning

The standard tuning employs a jet needle broadly suitable for temperate climates from sea level up to 6000 feet. Above that altitude, and in conditions of extreme heat or humidity, a weaker tuning may be necessary. Altitude, high temperatures, and humidity each tend to require a weaker setting, and in combination, they naturally emphasize this need. Because conditions vary widely, no hard and fast factory recommendation is possible; owners may need to experiment with alternative weaker needles until one yields satisfactory results. To assist with this, a recommended weak needle is usually listed.

BASIC INITIAL TUNING OF S.U. CARBURETTERS

A correctly tuned carburettor on a perfect installation should supply the engine with the weakest mixture capable of providing maximum power at full throttle, and minimum fuel consumption at all part-throttle running conditions. For a given piston lift and average air flow through the carburettor, heavy air pulsations enrich the mixture due to the continued flow of fuel through the jet (by inertia) after the air flow decreases.

Under full open throttle conditions, the pulsations are large and depend on many factors such as engine speed, induction manifold length and cross-sectional area, valve timing, number of cylinders per inlet port, and exhaust system design. Any closure of the throttle dampens these pulsations and weakens the mixture. This allows a needle setting that provides the desired fuel flow throughout the operating range, achieving the correct mixture spread between full and part load.

Balance Pipe Size

Fortunately, optimum mixture spread is closely approached on most four-cylinder engines using a single S.U. carburettor. By choosing the right size balance pipe, one can arrange the air pulsation to achieve the correct mixture spread on multi-carburettor, multi-cylinder installations—and even on twin installations on four-cylinder engines.

In some cases, even the largest practical balance pipe may not prevent the mixture spread from being slightly greater than optimal. In such cases, it is preferable to achieve optimum part-throttle results while allowing full-throttle mixture to be slightly rich. Provided full-throttle performance does not suffer, this condition can be tolerated, and it may even benefit cold starting.

Weakening Device

Engines with six or more cylinders using a single carburettor often lack sufficient full-throttle pulsations to produce the required mixture if tuned for part throttle. In these cases, tuning is adjusted for full-throttle, with part-throttle economy achieved by a weakening device. This device creates a small depression on the fuel in the float chamber via a pipe taken from the edge of the butterfly (similar to that used for partial vacuum advance in ignition). Air is drawn through a venturi from the float chamber and replacement air enters through a fixed orifice of calibrated size, reducing fuel flow through the jet. The venturi limits the air flow so that maximum weakening is achieved with only a slight throttle closure from full throttle.

New Installations

When planning a new engine installation, the desirable tuning sequence is:

1. Tune the engine for all running conditions on a test brake.
2. Re-check the installed engine on a car using a roller dynamometer.
3. Carry out full road consumption and performance tests.
4. Conduct cold start and drive-away tests.
5. Operate the car over a prolonged period under varied terrain and climatic conditions.

For initial bench testing of a new engine type, 'hand-controlled' (or variable mixture) settings are used so that the effects of changes in camshaft, ignition, manifold, compression, exhaust system, etc., can be assessed. Once all engine particulars are finalized, serious carburettor tuning can commence.

The process of obtaining the correct needle dimensions for an S.U. carburettor installation is comparatively simple and is described later in this paper.

Roller Dynamometer Tests

Since vehicle installation conditions may differ from the test bed, it is useful to install the whole vehicle on a roller dynamometer for further tuning checks. These tests are typically conducted at full throttle and road load conditions. Because it is difficult to accurately determine losses through the transmission and tyres, the road load curve is usually obtained using a vacuum gauge fitted to the induction manifold. Record manifold depression at constant speeds and match these with the dynamometer loads. It is important that carburettor and ignition settings remain unchanged after the road test and before the dynamometer test.

Fuel for roller dynamometer testing is supplied through the car fuel pump system, which also verifies adequate flow. Additionally, a test should be conducted with the correct carburettor dampers installed to ensure they do not affect tuning.

Road Tests

Both engine and dynamometer tests are aimed at achieving optimum vehicle performance on the road, so full road tests are essential. Ideally, these tests should be carried out on a proving ground such as M.I.R.A. at Lindley.

The normal test procedure involves conducting acceleration and set-speed fuel consumption checks using the needle setting found to give optimum dynamometer results. These figures are then confirmed by repeating the tests with slightly richer and then slightly leaner needle settings, or by adjusting the jet accordingly.

Manifold Heat

If a richer needle improves acceleration at the expense of road load consumption, it may indicate insufficient hot spotting in the manifold, which might require manifold modifications and further tuning. In multi-carburettor installations, a smaller balance pipe may help overcome increased part-throttle consumption by increasing mixture spread and allowing the use of weaker needles.

Another consequence of insufficient manifold heat is the need to set a rich slow running to avoid flat spots on throttle opening. This keeps the manifold walls damp during overrun and idling, improving throttle response but potentially leading to inconsistent idling and increased fuel consumption.

Manufacturers often hesitate to incorporate adequate hot spots due to concerns over volumetric efficiency, but minimal hot spotting that vaporizes the fuel can yield considerable benefits and even a power gain.

Cold Starting

In severe cold, effective starting depends primarily on the even distribution of liquid fuel to all cylinders rather than merely providing a sufficient fuel quantity. An excessive fuel quantity in one cylinder may wet the ignition plug and prevent proper firing. Good cold starting relies largely on the design of the induction manifold to distribute fuel evenly among all cylinder ports.

Although cold starting ability can be tested in a refrigerator, this method does not provide information on the vehicle’s drive-away performance. For this reason, the vehicle should ideally be tested in an environment with the required ambient conditions.

Once all obvious deficiencies are eliminated, as many vehicles as possible should be fitted with the proposed production layout and extensively road tested under various terrain and climatic conditions. The tests should be conducted by a range of drivers, not just trained test drivers, to reveal any latent defects.

NEEDLE DETERMINATION

To determine the dimensions of a needle that will produce the optimum results obtained by varying the jet drop during tests, first plot out the test needle. Needle diameters (in thousandths of an inch) are plotted on the horizontal axis and needle lengths (in 1/8 in. increments) on the vertical axis, following the standard needle chart method. Piston heights corresponding to various speeds are then superimposed on the vertical axis.

All needles with the same profile over the first 1/8 in. of their length will require the same jet drop for idling, allowing an immediate assessment of the effect on mixture of the remaining dimensions. If these idling dimensions are not maintained, accurate assessment of different needles cannot be made until the required jet drop adjustment to maintain consistent idling fuel discharge is determined.

In order to maintain a constant idling setting when using different needles, every effort is made to keep the shoulder and the first 1/8 in. dimensions at standard figures: for 0•090 in. jets, 0•089 in. and 0•085 in.; for 0•100 in. jets, 0•099 in. and 0•095 in.; and for 0•125 in. jets, 0•124 in. and 0•1205 in. The jet drop below the bridge and the piston lift above are determined under idling conditions. This jet drop (shown in Fig. 26) is then added to the piston lift to determine the actual distance from the needle shoulder to the metering diameter.

>Needle correction

To determine the required needle dimension at a given engine speed and load, first establish the piston height and jet drop necessary for optimum fuel flow. Then subtract the idling jet drop from the actual jet drop to obtain the vertical correction. If the result is positive (indicating a richer setting is required), the required needle diameter at that piston height is that of the original (or trial) needle at the point given by the piston height plus the vertical correction. If the result is negative (indicating a weaker setting is required), then the required diameter is at the point given by the piston height minus the vertical correction. Figs. 27a and 27b illustrate the correction for rich and weak conditions respectively. Fig. 28 shows a needle corrected for both full and part throttle requirements, and Table 1 outlines the method for establishing the necessary dimensional changes.

Once the new dimensions are established, either an existing book needle should be found or an experimental needle made. Repeat the test with the revised needle and the jet set for tickover. This will reveal any unexpected fuel discharge characteristics due to significant jet drop variations. If necessary, further tuning using the revised needle as a datum should be carried out.

Part-load performance is best evaluated by running a series of partial loops at road load (3/4, 1/2, and 1/4 load) at various engine speeds, adjusting the jet setting above or below the idling position (see Fig. 29). All tests should be conducted with hand-controlled ignition and complete air intake and exhaust systems.

Piston Lift Measurement

It has been assumed that either an engine test brake or a roller dynamometer is available for needle determination. In both cases, measuring piston lift and jet drop is relatively straightforward.

One effective method of checking piston lift is to insert a rod into a hole drilled vertically through the suction chamber cap, then cut the rod off flush with the top of the cap while the piston is resting on the carburettor bridge. Care must be taken to ensure that the rod does not introduce undue friction. For fully dustproofed carburetters, which are internally drilled to equalize air pressure between the damping chamber and the suction chamber interior, it is crucial that air does not escape past the indicator. A small amount of oil on the rod usually provides an adequate seal. The principle is shown in Fig. 30.

In cases where neither a test bed nor a dynamometer is available, satisfactory results can be obtained by road testing, during which the jet position can be observed via the movement of the jet control cable. However, measuring piston lift in the field is more challenging.

A very satisfactory solution is to use a manometer with a liquid of low specific gravity (such as paraffin), with one side connected to the piston damper cap via a small-bore tube (see Fig. 31). Movement of the piston rod displaces the air in the tube and then the liquid in the manometer. The bore of the glass tube can be selected to provide the desired magnification of piston movement. Since underbonnet temperature varies with driving, altering the air volume in the tube, the instrument must be zeroed periodically via an external valve in the air tube. Before taking a reading, release the accelerator (allowing the piston to fall onto the carburettor bridge) and briefly open the air valve to reset the manometer.

CHOICE OF CARBURETTER SIZE

The correct size of carburettor for any installation depends on many factors. As a rough guide, for single-carburettor installations on four-cylinder engines:

• 1¼ in. diameter for powers up to 45 bhp
• 1½ in. diameter for 45 to 64 bhp
• 1¾ in. diameter for 65 to 85 bhp
• 2 in. diameter for 85 to 110 bhp

In deciding the right size, the primary factor is the maximum air velocity through the unit during the induction cycle, not the mean velocity. Thus, for a given engine capacity or power, a single-cylinder engine requires a larger unit than a twin, and a twin requires a larger carburettor than a four-cylinder, etc. Furthermore, the greatest air volume that can be passed through a carburettor for any given maximum depression occurs when used with a centrifugal blower, where peak depression approximates mean depression.

When considering multiple carburettor installations, the size of the balance pipe connecting the various induction tracts must also be taken into account. A large-bore balance pipe allows each carburettor to contribute significantly to the air supply for the cylinders served by other units, while a small-bore balance limits that contribution. Therefore, the final carburettor size is often determined after preliminary bench tests, and financial factors may also influence the decision.

Fig. 32 illustrates the results obtained using single and twin carburettor installations of different sizes on a 1.6-litre four-cylinder engine.

INDUCTION MANIFOLD DESIGN

Even if a carburettor can provide the correct air/fuel ratio at optimum depression, good induction also depends on the design of the remainder of the induction and exhaust systems. An induction system with one port and carburettor choke per cylinder can provide excellent distribution and supports the use of ram pipes tuned for optimal power at specific engine speeds. Although overall b.m.e.p. may be increased with such a system, varying pipe lengths might cause a slight power loss at other speeds.

Similar results are possible with one choke serving two cylinders, but then the effective ramming must be achieved in the induction pipe between the carburettor and the ports.

Ideally, the two cylinders served by one carburettor should have equally spaced firing impulses. While this is possible on four-cylinder engines, it is extremely difficult with six-cylinder engines. Fortunately, excellent results are achievable even with uneven firing impulses.

Some advantages may be gained by using air trumpets on the air intake side of the carburettors, which help collect blowback and prevent fuel wastage or poor distribution caused by blowback being drawn into adjacent carburettors. This setup provides excellent distribution, though the necessary length of induction passage may be difficult to accommodate—especially on six-cylinder engines.

Induction systems employing two carburettors can offer good breathing and distribution without excessive heat. Effective ramming is possible by extending the individual induction passages between the main gallery of the induction pipe and the ports. Sufficient heat must be applied to the main gallery to ensure that fuel separation does not occur before the fuel/air mixture reaches the individual passageways.

Installing a single carburettor on a six-cylinder engine to achieve good distribution is challenging and often requires extensive experimental work. While satisfactory results at a specific speed may be achieved, obtaining consistent results over the full speed range is more difficult. Fig. 33 indicates the full-throttle performance of various carburettor installations on a six-cylinder engine.

Four-cylinder induction pipe design issues are similar, though the likelihood of poor distribution with a single carburettor is lower.

Regardless of the number of cylinders, high-performance induction systems are more readily adapted to engines with one port per cylinder. Siamese ports preclude individual port ramming, increasing the likelihood of one cylinder "robbing" its neighbor and causing uneven distribution. With three siamese ports on a six-cylinder engine, twin carburettor installation is impossible due to the inability to incorporate a balance pipe restriction.

ACKNOWLEDGEMENTS

The author would like to thank the S.U. Carburetter Co. Ltd for permission to publish this paper, and to acknowledge the assistance of his colleagues during its preparation.