Capt. Alan Southon describes his rebuilt 1923 2-litre version of this rare vintage car and adds some technical and historical data on 4-cylinder H.E.s in general.
Since returning home from overseas I have been slightly surprised at the interest displayed by chance acquaintances in my rather elderly 2-litre H.E., which I have just relicensed. It occurs to me, therefore, that a description of this car may be of interest.
Before writing of my own car it might be as well to record some historical notes on H.E. 4-cylinder cars in general, from what I have learned during the past 12 years, though I do not claim any special knowledge, and no doubt there are others who have more intimate experience of this make.
To the best of my knowledge, all the 4-cylinder H.E.s had the following likenesses: 4-cylinder side-valve monobloc engine, with detachable cast-iron or aluminium head. Aluminium-alloy crankcase with detachable sump. Dry-plate clutch mounted on an exposed flywheel. Separate 4-speed and reverse gearbox with r.h. change-speed lever. Propeller shaft enclosed in a torque tube. Overhead worm drive rear axle with hand and foot brakes operating on separate shoes. Later models had front brakes. Bolt-on artillery-type or wire wheels shod with 815 by 105 tyres. Later models were fitted with Rudge wire wheels; 3/4-elliptic rear and 1/2-elliptic front springing.
The H.E. first appeared on the market in 1920 as a 75 by 120-mm. 4-cylinder, rated at 14/20 hp., and was manufactured by the Herbert Engineering Co., of Caversham, near Reading. I recollect seeing one of these cars at a breaker’s yard near Salisbury some considerable time ago. This model differed, as far as I can remember, from any of the succeeding ones, in that there was a magneto mounted above the dynamo on the r.h. side of the engine, on a sort of extended chaincase. Artillery-type road wheels were fitted and they did not improve the general appearance of the car, that was not of the most handsome proportions.
The “14/20” appears to have lasted until 1923, about 100 being made per year (according to chassis numbers).
In 1922 the “14/40” hp. model made its appearance; it possessed the same bore and stroke, but had coil ignition. One, owned by a close friend of mine, had two plugs per cylinder (one in each valve cap) and two ignition distributors (each with its own coil), driven by skew gears from the end of the dynamo. A single Zenith triple-diffuser carburetter fed from the l.h. side of the engine, between the centre pair of cylinders, to an induction pipe cast integral with the block. The cast-iron cylinder head was detachable, and was fitted with large brass valve caps in addition.
About 1923 the-2-litre was introduced, having a reduced bore of 72.5 mm., but retaining the stroke of 120 mm.; this model was rated at 13/30 h.p. and was definitely a Sports car in the accepted sense of the word, whereas its predecessors were more in the nature of touring cars. The “13/30” became a “13/40” in 1942, costing £490 as a chassis, while there was also a “13/45” h.p. model at £600 (chassis price) offered at the same time. What structural differences there were to cause the increased rating I do not know, but I suspect that the 2-litre was fitted with an aluminium head, amongst other improvements, in its later years, which would account for the rise in power. Both the “13/40” and “13/45” appear to have terminated in 1925 with chassis numbers of approximately 3000 and 5000, respectively, though this does not necessarily indicate the actual numbers built.
The “14/40” was withdrawn about 1927, at which time it had become a “14/50” and cost about £600 for the chassis. In this case the increased rating is known to be the direct result of fitting an aluminium alloy head, combined inlet and exhaust manifold, well-swept ports, etc.
On July 7th, 1925, the Motor reported the 2-litre H.E. “as a much improved chassis, with a good road performance.” This car was presumably a “13/40” or “13/45” model, I suspect the latter, as from the photographs illustrating, the report it is apparent that an aluminium cylinder head, combined manifold, etc., were fitted, together with such refinements. as Rudge-Whitworth wire wheels. The following information is reproduced from the data panel of that report, together with a graph showing the performance of the car on third and top gears. This represents, I suppose, the peak of the 2-litre performance in the form in which it was available to the public, though there is evidence of this model having been raced at Brooklands track, presumably in very much modified form :—
Type: 2-litre 3-seater. R.A.U. rating, 13 h.p.
Engine: 4-cylinder side valve, detachable head, 72.5-mm. bore, 120-mm. stroke, 1,981-c.c.
Gearbox: Four forward speeds, righthand control. Ratios : 4.2, 6.36, 9.35 and 15.4 to 1.
Suspension: Semi-elliptic springs in front, 3/4 elliptics at the rear.
Brakes:Fully compensated, operating on all four wheels, and independent shoes on rear wheels controlled by hand.
Maximum Speeds on Gears: Second, 36 m.p.h.; third, 52 m.p.h.; and top, 60 m.p.h.
Fuel Consumption: Petrol, 25-28 m.p.g. Oil, 1 quart in 250 miles.
Dimensions: Track, 4 ft. 2 in.; wheelbase, 9. ft. 9.5 in.; overall length, 13 ft. 4 in.; overall width, 5 ft. 3 in.; ground clearance 9.5 in.
Turning Circle: Left, 40 ft. 2 in.; Right, 45 ft. 6 in.
Chassis Weight:18 cwt.
Manufacturers: Herbert Engineering Co. (1924). Ltd., Wolseley Road, Caversham, Reading.
This car would accelerate from 10 m.p.h. to 50 m.p.h. in third gear in 33 1/2 sec. and from 10 m.p.h. to 40 m.p.h. in 20 sec.
As regards the car I now have, this was purchased in 1933 for the sum of £4 and was in original condition, in that, no alterations had been made since it left the works; the Chassis No. 2088 identified it as a 1923 2-litre (“13/30”), with a bore and stroke of 72.5 by 120 mm. Needless to say, the car was not in show-room condition, nor was it mechanically sound, but though a little rusty, it was complete in detail, which was unusual for one so cheap, even in I 933.
The engine was the conventional s. v. unit, with the valves on the l.h. side of the block. A single Zenith 30.HA-type carburetter, fitted to the r.h. side of the block, fed between the centre pair of cylinders to an integrally-cast induction pipe. A separate four-port exhaust manifold took off from the l.h. side, and a detachable cast-iron cylinder head was fitted, incorporating the unusual large brass valve caps.
The cylinder block was mounted on a straightforward two-piece crankcase and sump, in which a very “bent wire” crankshaft ran in three white-metal bearings secured to the upper half of the crankcase. Big-ends and main bearings were splash fed, the former scooping from troughs fed by a gear pump in the base of the sump and driven by a vertical shaft and skew gears from the rear of the camshaft.
Separate silent chains from the nose of the crankshaft drove a dynamo on the right, and the camshaft on the left side of the engine.
Water cooling was by thermo-syphon, assisted by a fan driven from the front of the crankshaft by Whittle belt, the fan shaft being extended inwards and fitted with an impeller that stirred the water where it entered the cylinder block.
Ignition was by coil, the distributor being driven by skew gears from the end of the dynamo. One sparking plug screwed into each inlet valve cap. The 6-volt lighting, starting and ignition were all by C.A.V.
Fuel feed was by autovac from a 12-gallon rear tank (that was covered with wood slats in a manner fashionable in 1928) and possessed a decent-sized screw filler cap.
Mounted in the l.h. rear engine bearer was the self-starter. On the rear face of the flywheel was a clutch pot containing alternate steel and cast-iron plates, running dry. The spring compressing the clutch was mounted on an adjustable rod to the right of the drive, in such a way that continual pressure was exerted on the free end of a pivoted arm that extended across the chassis and straddled a large ball-race that was a sliding fit on an extension of the clutch housing. The centre of the ball-race, in turn, pressed on a series of toggles and levers, making contact with the clutch pressure plate through two small plungers. The whole assembly was remarkably similar to Bugatti, except that centrifugal force was not employed, and the remote clutch spring actually supplied all the pressure. Operation of the clutch pedal relieved the assembly of the spring pressure and the plates separated themselves naturally and, since they ran dry, they did not suffer from binding when engaging a gear while at rest. Although the clutch looks a bit of a monstrosity, it works very well and has an exceptionally smooth take-up under all ccinditions of operation; also, in spite of the heavy end-thrust continually applied to the ball-race, which is not of the ball-thrust variety, no trouble has been experienced in that department.
From the clutch the drive was taken through two pairs of spiders and fabric couplings to a 4-speed and reverse gearbox with straight-cut gears controlled by a r.h. change-speed lever working in a visible gate.
A spherical coupling at the rear of the gearbox secured the torque tube housing the propeller shaft, and also housed the universal joint, which latter was a very crude form of the well-known Hook’s Joint and possessed no form of regular lubrication.
Final drive was by means of an overhead-worm-drive assembly incorporating a conventional bevel-gear differential. A feature of this assembly was, that having withdrawn the fully floating axle shafts by the simple expedient of removing the hub caps (it was not necessary to remove the road wheels),. the whole worm and differential assembly could be withdrawn by unbolting the rear cover plate of the axle casing, this cover plate acting as an anchor for the differential housing bearings. This, incidentally, was a very timely feature, as these axles were, I believe, prone to star-wheel failure in the differential assembly, though personally I never experienced this trouble.
Large cast-iron brake drums were fitted at the ends of the axle and housed two pairs of shoes, one each for the foot and hand brakes, respectively. No front brakes were fitted. The brakes were rod-operated and fully-compensated, and not awfully effective with the thin-section tyres then provided.
The chassis was braced largely by the substantial engine bearers at the front, the very rigidly mounted gearbox in the centre, and a tubular cross member between the gearbox and the rear of the chassis, where a simple channel-section member closed its extremities. A U-shaped channel member was mounted at the front to support the radiator, in front of which was a single crossbar supporting the starting handle.
Half-elliptic springs were fitted at the front, shackled at their rear ends, and 3/4-elliptics at the rear. This gave a very comfortable ride and quite good stability. An early form of Luvax oil shock-absorber was fitted all round. Bolt-on wire wheels were fitted, carrying 815 by 105 tyres.
A. large, cast-aluminium steering wheel directed the motor-car through the medium of an enclosed column extending from a massive worm-and-nut steering box mounted on the r.h. rear engine bearer, with a limited adjustment for rake. A cast-aluminium throttle and ignition control quadrant was mounted in the centre of the wheel and was connected with the usual ball-jointed linkage at the base of the steering box by means of concentric tubes within the column.
The body was the familiar three-seat, boat-tailed type, painted bright yellow, with long sweeping wings in red, together with red wheels.
Instruments consisted of an oil gauge, speedometer and, I think, clock, all by Smith’s. A C.A.V. push-and-pull switchbox controlled head, side and tail lamps of current design. A separate switch controlled the ignition, and there was, of course, a large push-button for the electric starter. During the years of my ownership many changes have been made that have so far resulted in the following specification: –
Commencing with the engine, the 72.5mm. cylinder block was scrapped in favour of a 75-mm. one, with an aluminium alloy head and a combined and detachable inlet and exhaust manifold mounting a single 36-mm. horizontal carburetter. This alteration was unfortunate, in that it placed the car in the 2,000 to 3,000-c.c. class for competition work. There was, however, no alternative if I was to have the advantage of the alloy head, since I was unable to find a 72.5-mm. block possessing its desirable features. The compression ratio was raised to 6.5 to 1 from the original 5.0 to 1 by shaving off some of the cylinder head and new Covmo pistons fitted after one of the original ones had parted company with its crown.
Force-feed lubrication was arranged for the main bearings, the necessary take-off from the pump being already provided, but blanked off. Splash-feed lubrication for the big-ends was retained and no trouble has been experienced since having the white-metal run in direct on to the steel instead of using the original die-cast liners. In any case, the crankshaft would not, in my opinion, stand drilling for lubrication owing to its sketchy proportions.
The oil pump maintains a steady 20 lb./sq. in. pressure at 2,500 r.p.m. when hot, using Castrol XXL. A tiresome feature of this conversion to pressure feed was an increase in the already annoying oil leak at the back of the crankcase where the crankshaft protrudes, the oil thrower ring and “catcher” being of such a half-hearted design that it would not even retain oil with the splash-feed arrangement. The result is that more oil is distributed around the inside of the bonnet and under-tray than is actually burnt, and I have had to devise suitable flywheel guards to keep it within limits.
Standard H.E. double valve springs were obtained and KE.965 steel exhaust valves were made up; the double valve springs effectively stopped the valve bounce that was evident at anything over 2,500 r.p.m. with the original single springs. Continual distortion of the valves and seats, particularly the exhaust, was experienced every 1,000 miles or so and was only cured by fitting new valve guides and valve seat inserts. The usual port polishing and lapping of the manifold has also been carried out.
The camshaft (as already stated, driven by silent chain from the front end of the crankshaft) with its high-lift cams with abrupt opening and long dwell periods, coupled with the incredibly strong double valve springs, no doubt accounted for some distortion troubles.
The valve timing desired, but not realised, is: —
Inlet opens 3° E. Exhaust opens 55° E.
Inlet closes 50° L. Exhaust closes 8° L.
This would give an overlap of 11° and a period of opening of 233° for the inlet and 243° for the exhaust valve. Examination of a worn camshaft indicates that of the two openings there is a period in each during which the valve is fully open for 80° of crankshaft rotation. The maximum valve lift for both inlet and exhaust is 9 mm. This is a most encouraging feature, and it is, therefore, unfortunate that, due to wear, etc., the valve timing varies a little on each cylinder and the overlap is reduced to a negligible amount. The tappet clearances are run at 0.004 in. and 0.008 in. (when hot) for the inlet and exhaust valves, respectively. This seems to be the optimum setting which allows adequate clearance at all times, since the tappets follow the cam contours by means of 25-mm. diam. rollers, and these are not as round as they might be, thereby altering the clearances slightly in a manner that it is quite impossible to predict. I have never been able to discover the actual clearance recommended by the manufacturers.
Carburation is looked after by a Zenith 36 HAKG carburetter bolted to a rather overheated two-branch induction pipe, feeding siamesed ports and cast in one with a bulbous four-port exhaust manifold. In connection with the exhaust porting, it is of interest to note that the actual ports are swept back at an angle to the centre line of the engine, thus deflecting the gases to the rear slightly before they actually enter the manifold. While this may have desirable features for the present arrangement, it does not facilitate the fitting of a sheet-metal manifold and the two carburetters that I wish to use.
The choke and jet settings for Cleveland Discol before the war were : –
Choke … 28
Main jet … 140
Compensating jet … 120
Idling jet … 70
Float needle seating … 3 x 4.4
For Pool petrol (1945 brand) the setting for economy is : —
Choke … 26
Main jet … 125
Compensating jet … 80 or 85
Idling jet … … 60 or 70
The latter setting permits reasonable performance and economy, whereas the former was not so economical.
Fuel feed is by air pressure (1.5 to 2 lb./sq. in.) from a 19-gallon rear tank, the latter salvaged from a Bristol “Fighter” aircraft some time before the H.E. was purchased. This tank fits very well on the tail of the chassis and, apart from leaks, which occur with depressing regularity, due to the light construction and whipping of the chassis, it is ideal for the job.
The disadvantage of not having a mechanical air pump is not as real as it would appear, since the system is very air tight and, once the initial pressure has been pumped up, a few strokes of the hand pump every six or eight miles are all that is required to maintain pressure.
Ignition is by a Scintilla “Vertex” NV.4 magneto, mounted on the end of the dynamo in place of the coil ignition distributor previously fitted. The lower end of the magneto armature shaft is adapted to provide a take-off for a Jaeger rev.-counter drive.
The C.A.V. dynamo and starter are the originals, brought up to scratch by servicing and adjustment, and a small capacity 6-volt battery is mounted on the cockpit side of the cast-aluminium bulkhead, accessible by opening the hinged scuttle cover on the l.h. side.
The flywheel was lightened 5 lb. by shaving off the stiffening ring, though even now it weighs 30 lb. (less clutch). The clutch is the original, consisting of five driven steel plates, interspersed with six cast-iron driving plates, running dry. The clutch never gives any trouble, nor does it require any adjustment. The short connecting shaft and two fabric couplings between the clutch and gearbox cause a good deal of vibration and are due for a replacement in the form of more modern couplings when the necessary parts can be made. The gearbox is of generous dimensions and the inside is readily accessible through the lid, which can. be unbolted and lifted off without disturbing anything; the lay-shaft runs on the left of the main shaft and in the same plane. The gear ratios and speeds per 1,000 r.p.m. will be found at the end of this article. A study of these ratios and speeds will show that there is a rather undesirably wide space between the third and fourth gears at maximum speed; this is particularly noticeable in practice, and there is no advantage in “nipping into third” to overtake slower vehicles unless travelling very slowly yourself. A variation of the constant-mesh gears to produce a ratio of 30/22 would seem to be indicated and will be tried out one of these days. The gearbox is lubricated by Castrol R.
At the rear end of the gearbox is a spherical joint for the universal coupling, to which the front end of the H.E. torque tube is attached. Midway down the torque tube, however, a join is made to accommodate the rear of a “14/40” Vauxhall tube, which is attached to an axle from a car of that make. The prop.shaft and universal joint were made up from one taken from a Triumph of unknown size. The Vauxhall axle was chosen mainly because it was made of aluminium alloy and provided a bevel drive in place of the original H.E. overhead worm; further, owing to its excellent design, it is possible to run the axle on Castrol R without the rear brakes becoming saturated in oil. It also possesses the very desirable feature of Rudge 52 splined, centre-lock hubs.
The axle ratio (4.5 to 1) was changed for a specially-made, straight-cut pair of gears giving a ratio of 3.8 to 1, thus bringing the final ratio with 19-in. by 5.50-in, rear tyres to approximately the same as before, when the ratio was 4.2 to 1 and the tyres were 815 by 105.
Bugatti (2-litre) springs and chassis rear end were attached to the extremities of the H.E. chassis and, without much difficulty, the front of the reversed 1/4-elliptic springs were secured to the ball spring mountings on the Vauxhall axle; these spherical mountings had to be repositioned toward the centre of the axle to comply with the slightly narrower spring centres. Hartford shock-absorbers have been fitted all round.
A “14/40” Vauxhall front axle was also fitted and looks more handsome than it is useful, since the front brakes are of very doubtful temperament and, although the axle is tubular, it weighs the earth. The brake gear was coupled up, with due regard to leverages, etc., without great difficulty, but it cannot be said that the brakes, even now, are in any way remarkable, and in wet weather they have a disconcerting habit of fading out altogether until the sun shines again.
This type of front axle also suffers from a slow speed wheel wobble which I have been unable to cure, except by ensuring there is no back-lash in the steering; this is not easy, as the massively-constructed worm-and-nut assembly has a white-metal lined “nut” which wears rather too quickly to be pleasant and has no adjustment to compensate for wear. Also, the underslung axle, as against the overslung beam originally fitted, gives the chassis a pronounced slope to the front and accentuates the already rather naked hindquarters of the car.
The remainder of the car is essentially standard, and what is left of the original body has never been lifted from the chassis. I have endeavoured, as far as possible, to retain as much of the original car as possible, as I do not agree with wholesale changes in the specification of vintage cars. In my case, the axle changes were unfortunate, but necessary if the car was to be practicable and fit for everyday use. All these changes and alterations may seem to be a little misguided on such an ordinary chassis, they were, however, somewhat naturally, not all carried out at once, but were spread over some six years before the war commenced. There were also many interim periods during which various things were tried out that I have not mentioned here. I now have a car that affords me an enormous amount of pleasure to drive and which is astonishingly easy to take to pieces when necessary. It has taught me considerably more than I should have ever learned from working on other people’s cars.
In its present form it has quite a reasonable performance and it is possible to cruise at 2,500 r.p.m. (about 58 m.p.h. in top) all day long, while it will run up to 3,000 r.p.m. (in top) under quite ordinary conditions. The absolute maximum r.p.m. realised so far has been 3,500 in third gear, but this is not often indulged in, since I wish the engine to keep running, rather than have meteoric performance for brief periods between “blow-ups.”
During 1939 I towed a trailer with a racing car on it quite frequently, and without much difficulty, although it was necessary to fit smaller diameter wheels at the rear as the car is slightly overgeared, or under-powered, according to how one looks at it.
Its competition value in speed trials and sprint events is, naturally enough, a dead loss, but it is better to drive a slow car than no car at all. A number of persons has asked me what there is to be had in this H.E. motoring, which is rather like asking one what there is to be found in vintage motoring-something that cannot be easily explained. There are snags, of course, to this car, one of which is the lack of a hood in wet weather, combined with rather spartan coachwork, due to lack of finance and time, before the war, to worry about refinements, with the result that the body has had no attention whatsoever, other than cutting pieces out of it; there was a very good windscreen and electric wiper at one time, but this was, unfortunately, crushed when I was so foolish as to turn the car over one night.
As regards cost, the total outlay, including purchase but excluding petrol, oil, tax, etc., for the six years prior to the war, has been just under £90 (accurately accounted for). This figure includes all alterations and modifications past and present, and also includes two major accidents that I committed all on my own and which were entirely due to my own carelessness. Mileage covered during this period was about 22,500 up to the outbreak of war. Not very much, perhaps, compared with the mileage of some, but the car was seldom licensed for more than two “quarters” in succession; it has been highly satisfactory from a personal point of view, intermingled with the exasperation and awful periods of gloom associated with vintage motoring.
Here is a summary of some of the more technical features. I have probably left out the most important details, but if anyone is sufficiently interested I can always supply further information if they take the trouble to write to me.
4-cylinder, 75 x 120-mm. bore and stroke, 2,121 c.c.
Compression ratio, 6.5 to 1.
Bore/stroke, ratio, 1.6.
Area of one piston, 44.15 sq. cm.
Area of inlet port (Pi x d x h), 11 sq. cm.
Ratio, piston area/inlet valve port area, 0.245.
Average piston speed (at 3,000 r.p.m.), 2,360 ft./min.
H.p. 45(?) or 50(?).
Note. — The speed per 1,000 r.p.m. was arrived at by Cecil Clutton’s formula :— (Wheel diam. in inches / Overall gear ratio)x 3 = m.p.h. per 1,000 r.p.m. and is fractionally greater than my reckoning.
Rudge 52 centre-lock with well-base rims.
Front – 18 in. x 5.50 in. (running pressure, 30 lb./sq. in.).
Rear – 19 in. x 5.50 in. (running pressure, 25 lb./sq. in.).
Wheelbase, 9 ft. 4 in.
Track, front, 4 ft. 10 in,
Track, rear, 4 ft. 8 in.
Overall length, 13 ft. in
width, 5 ft. 8 in.
height (to top of body), 3 ft. 8 in.
Ground clearance (at sump drain plug), 5 in.