,

Toroidal Internal-Combustion Engines.

Updated: 24 June 2008
BSA engine added

CONTENTS OF THIS PAGE

The Bradshaw Omega Engine: 1955 #2
The BSA Toroidal Engine: 1955 New
The Tschudi Engine: 1967. #1

Several inventors have persuaded thenselves that having curved pistons oscillating or rotating inside a cylinder block is a good idea. It is not. Here is the story of the toroidal internal-combustion engines.
An engine expert speaks:
"A great many ideas for engines in which toroidal pistons rotate or reciprocate within toroidal cylinders have been advanced. The difficulties of connecting such pistons to the output shaft by a simple and reliable mechanism, together with the problem of sealing the surfaces involved, make such ideas little more than amusing adventures in ingenuity."
Quote from The Internal-Combustion Engine in Theory and Practice by Charles Fayette Taylor, 2nd edition, pub MIT press 1985. This book is one of the standard works on the subject. Taylor was Professor of Automotive Engineering at MIT; he has little time for unconventional engines of any kind, and his arguments are persuasive.

THE BRADSHAW OMEGA ENGINE: 1955

Left: The Bradshaw Omega engine: 1955.
Originator: Granville Bradshaw (British)
The Bradshaw engine had a single toroidal cylinder, containing four double-ended curved pistons. The pistons reciprocated in pairs, while the cylinder rotated around them, carrying around the spark plug and inlet/exhaust ports. The connection to that spark plug must have involved some complications- a slip-ring, perhaps, or maybe non-touching contacts like the inside of a distributor cap?
Wouldn't that pair of single-helical gears generate end thrust? Perhaps not enough to worry about.

From Popular Science, date unknown.

Left: The Bradshaw Omega engine principle.
The cylinder was rotated by the central shaft while the cranks that reciprocated the pistons were driven by the output shaft, running at half the cylinder speed.

From Popular Science, date unknown.

Left: Granville Bradshaw.

From Popular Science, date unknown.

Granville Bradshaw was no solitary eccentric; he was a respected engine designer who began his career in 1910 and designed the first Pratt & Whitney radial aircraft engine during World War 1. At the time of the Popular Science article he was 67, and described as a "millionaire inventor". He worked extensively on motorbike engines ; see www.realclassic.co.uk (external link) for an interesting motorcycle engine in which the cylinder is cooled by oil rather than air or water.
I have so far not discovered what advantages Bradshaw thought his engine had over more conventional designs. Its less-than-compact layout is one snag, and surely those section-of-a-toroid pistons must have been expensive to make?

According to one commentator from the world of motorcycles:
"The Omega was the final fling of a man whose ideas were always clever and innovative, but who sadly failed to understand the commercial needs of the business. His designs were novel, but invariably costly and seldom trouble-free, so his long involvement with the industry made news and kept everyone intrigued, rather than producing machines for riding."
It appears his aeroplane engines were also problematic: see Wikipedia.

I unearthed this quote from Patrick Head, one of the great engineers of Grand Prix racing:
"They had a wonderful project while I was down at Harry Westlake's. Somebody had sent them an engine called a Bradshaw, and it arrived in a cardboard box. It was basically a toroidal chamber with pistons which filled a section of the toroid, attached to two crosses, such that, as they went round, gearing superimposed an oscillation on the rotary motion. Intakes, exhaust and spark plugs were arranged around the periphery of the chamber. I was asked to find out how this engine worked, assemble it and install it on a dynamometer so that it could be evaluated.
"There was no inlet manifold or carburettor, so I got an Amal and made an inlet manifold, assembled the whole contraption and put it on the dyno. It happened that Dan Gurney turned up on the day we were due to start it up, and watched from outside the dyno. room, behind the bullet-proof glass. Anyway, this Bradshaw engine started up - the guy had said that it was perfectly balanced and would rev to umpteen thousand RPM - the only problem was that in the gear casing, at the back of the engine, he had made all the gears himself, with a file or something! The pitch of the teeth were all irregular and I had to do an incredible amount of lapping - the quality of build was awful.
"Anyway it did start and run, and I fiddled about with the carburettor and sorted out a few things. Dan Gurney was outside and encouraging us to give it the berries. While it was idling at a few hundred RPM a cloud of smoke gradually built up in the dyno. room. Eventually we gave it some more RPM, only about 2,500, when suddenly there was a mighty BANG! and the whole of the glass window disappeared in a mess of oil and metal. Slowly the murk cleared and all we could see through the smoke were the feet of the gearcase, and the toroidal chamber, with bits of cast iron and aluminium all over the place.
"It all got collected up, put back in the box and sent back to Mr Bradshaw. I don't think his concept was properly evaluated at all....."
It is not clear from this description of a rather casual approach to engine testing what failed- perhaps the home-made gearwheels were not up to the job. At any rate, it is clear that it did run.
If Mr Bradshaw was paying to have his engine tested, I should think he was pretty angry.

Left: Animation of the Bradshaw Omega engine.
Note how the spark plug goes round with the toroid, passing the pistons as they come together at the end of the compression part of the cycle. The white comet-shaped thing going round is the exhaust port.

Another stunning animation by Bill Todd.

Left: Another pic of the Bradshaw Omega engine.
As in all toroidal engines, there have to be slots in the cylinder wall to couple with the piston, and these need to be sealed. How this was done in the Bradshaw engine is not known.

Source unknown, but from the style, probably Motorcycle magazine.

THE BSA TOROIDAL ENGINE: 1955
This toroidal two-stroke engine was designed by the British motorcycle company BSA, and was intended for powering mopeds or cyclemotors. Development began in late 1938, and since it was reported by The Motor Cycle in August 1955, appears to have proceeded very slowy, though no doubt the project was shelved during World War Two. The swept volume was only 34cc. It definitely ran, but by all accounts, a prototype produced little more than enough power to overcome its own friction and pumping losses.

Left: The BSA Toroidal engine: 1955.
There is an obvious resemblance to the Bradshaw Omega engine, but I have not so far found any reports that Bradshaw worked with BSA. The two halves of the rotating cylinder assembly were made in cast iron and was fitted with curved fins for cooling.
As in all toroidal engines, there have to be slots in the cylinder wall to couple with the piston, and these need to be sealed. How this was done in this engine is not known; possibly the slot was short enough so it stayed in the centre section between the two sets of rings on each piston.

From The Motor Cycle Aug 1955

Research has unearthed no other reference to this engine, and the assumption must be that development work was abandoned in 1955 or shortly after.

Left: The BSA Toroidal engine: the principle.
The mechanism is clearly different from the Bradshaw, but the exact operation is rather obscure.
According to The Motor Cycle: "Dividing the toroid are two partitions at 180 degrees; within each half is a long double-ended piston and two pistons are coupled by a diametrical arm."
If there are fixed partitions, then clearly the pistons must be going round with the cylinder block, as otherwise they would whack into them. This appears to be indicated by the arrow in the middle.
The functioning to the two pairs of gears to the left is also a bit mysterious, as one of the pairs seems to be redundant. If anyone knows anything more about this engine I would be glad to hear from them.

From The Motor Cycle Aug 1955

THE TSCHUDI ENGINE: 1967
Originator: Traugott Tschudi (Swiss). Work began on the engine in 1927. US patent number 3381669; May 1968.

Left: The Tschudi engine: 1967.
The Tschudi engine works on the four-stroke cycle. It has four curved-cylindrical pistons moving in a toroidal cylinder; the toroid contains two rotors that each carry a pair of pistons, and two rollers that bear on cams fixed to the output shaft. The pistons stop and start as the rollers either press against the sides of the cam or drop into a groove in it.

From Popular Science, Jan 1967

Left: The operation of the Tschudi engine: 1967.
There are two rollers and a cam in black, and another set in white. The output shaft is eccentric to the rotors and toroids; it turns through 1.2 revolutions for every revolution of the pistons.

From Popular Science, Jan 1967.

Left: The stop-start action of the Tschudi pistons.
Only one cam and one pair of rollers can be seen from this side.

Classy animation by by Bill Todd.

Left: The roller & cam action of the Tschudi engine.
Bill Todd says:
"As the cam approaches and passes TDC, and the rotor stops, the follower rollers bear on a 'circular' segment on the cam. Since the cam axis is offset from the rotor/follower axis, any rotor movement would also try to move the cam axis which is fixed. However, the cam is free to revolve without moving the rotor, at least for that short section of the cam profile.
Tschudi added the asymmetry or 'dwell faces' to alter the inlet/exhaust timings. (according to the May 1968 patent)"

Another classy animation by by Bill Todd.

Left: The roller & cam action of the Tschudi engine.
Showing how a circular cam profile locks the rollers, although the cam itself can still rotate.

Diagram by Bill Todd.

An obvious objection to this design is that the stresses on the rollers and cams are going to be very high.
Bill Todd says:
"The cam is at a considerable mechanical disadvantage, so both it and the follower rollers are under enormous stress. The patent drawing shows the follower/roller assembly spring mounted to the rotor to 'decease friction and ware', presumably because Tschudi couldn't get the cam shape quite right. I can't help thinking this would have just made the thing clank like a bag of tools."
Unlike the Kauertz engine, the Tschudi only gives two power impulses for each revolution of the output shaft, and so a practical design requires two toroid assemblies, greatly complicating things.
Research into the history of the Tschudi engine has so far yielded very little. It does not seem to have made any news since 1968. Given the painfully obvious mechanical problems and the absence of any advantages, it seems unlikely that Tschudi found any financial backing.