The evolution of the Ducati 996 engine
15 July 2001 00:00
If ever an engine proved worthy of evolutionary development instead of replacing it with a new design it has to be the Ducati 996 motor.
The eight-valve, liquid-cooled V-twin which has dominated World Superbike racing and won countless other races is based on a design called the 748IE, which made its debut in the 1986 Bol d’Or in the hands of Juan Garriga, Marco Lucchinelli and Virginio Ferrari. The bike was forced to withdraw while in seventh place after 15 hours when a conrod bolt failed, but in fact few people expected much of or even noticed this unassuming Ducati – the idea that it might be the ancestor of one of the most influential race and road bikes of the next 15 years would have been laughable.
The engine came about after Massimo Bordi, who started with Ducati in 1978 to work with the great engineer Fabio Taglioni, had developed his university thesis on the four-valve cylinder head with desmodromic valve control. Taglioni had never been in favour of the idea, but he went into semi-retirement in 1983, leaving Bordi to take charge of Ducati’s engineering. One of his first jobs was to get engineer Luigi Mengoli to turn his thesis into metal, and by the early summer of 1986 the motor was developing 93bhp.
It seems obvious now that four valves would be better, but it was only the use of water cooling which made it work. With air cooling the angle between inlet and exhaust valves had to be very wide to prevent overheating, but this meant the combustion chamber was too deep for four valves to have any advantage. So Bordi adopted liquid cooling and modelled the design on the Cosworth FVA Formula 2 car engine, which had a 40 degree included valve angle.
While the top end of the engine was new, the bottom end even then was related to older Ducatis , the crankcases being based on those of the contemporary Pantah 600. Ironically, the 748IE had the racing version of the Pantah to beat as well as rival factory machines – the air-cooled two valve engine was producing 95bhp at the time, and represented the powerful swansong of the old Taglioni school.
It seems amazing now, but there was a fair chance that Bordi’s engine would be dropped altogether, but he pressed on with revised crankcases took the bore and stroke out from 88m x 61.5mm to 92mm x 64mm, giving a capacity of 851cc and with it a serious power hike to 115bhp. This worked , and at the end of 1987 Ducati built a batch of 22 851s which it sold to racers in order to homologate the machine for the 1988 World Superbike championship, the first year of the competition.
Marco Lucchinelli won the first WSB race on the 851 at Donington, a portent of things to come, although that year he only managed fifth overall (to be fair, Ducati didn’t compete in all the races). Those first bikes incidentally were fitted with 17 inch wheels and slick tyres and were known as the ‘Kit’ machines – these days they’re much sought after, a lot more than the batch of 300 16 inch wheeled machines which followed.
The first road version of the 851, called the 851 Superbike, appeared in 1988, in Italian ‘Tricolore’ colours of red, white and green. Like the now familiar SP and SPS Ducatis, this was designed to homologate the race machines, but already an 888cc version had been made, the Lucchinelli Replica race bike with the bore taken out to 94mm.
A proper road version called the 851 Strada was also made with milder cam timing and a lightly lower compression ration resulting in a power output of 102bhp at 9000rpm compared to the 851 Superbike’s 120bhp a 10,000rpm.
In 1990 the 851 Strada (‘street’ version) was joined by the SP2, the homologation special which had the 94mm bore, 888cc engine, although fundamentally the engines were the same apart from the state of tune (which gave the SP2 20 per cent more power) and the conrods – the 851 used Pantah conrods, the SP2 racing Pankl conrods.
At this point Ducati began an expensive catch up game with its crankcases – failure was a relentless problem as power increased, and the SP3 version of 1991 features additional strengthening webs behind the rear cylinder to prevent the cracking the race bikes were experiencing. The production bikes followed a year behind always.
Yet other changes were only in detailing: clutches were redesigned and there were various stated of tune achieved through different cam profiles and valve sizes, but apart from an overbored race version of the 888 – its 96mm bore resulting in a 926cc capacity for Raymond Roche’s 1992 machine – it wasn’t until the 94mm x 66mm bore and stroke 916 of 1994 that the engine changed significantly, and even then much of it was still shared with the 888. The race bike of that year incidentally retained the 96mm bore to create a 955cc capacity.
Crankcase failures continued to be a problem, to the point where five years ago a set of cases would last only one race meeting, but this improved when a bearing at the bottom pulley from the timing belt was altered, reducing a major stress centre.
Finally, during 1995 the first 996cc versions of the race bikes were being tried, with their 98mm pistons. The first road version of the 996c engine appeared in the homologation special 916SPS of 1998, the full production version, the 996 Biposto, arriving in 1999.
What is remarkable is that even the current 996 still has the same old-fashioned 40 degree included valve angle as Bordi’s 1986 original 7481E, and shares more or less the same crankcases, conrod design, gearbox, crankshaft with all its predecessors back to that mid-1980s race bike.
In the new 996R Ducati has finally redesigned the cylinder heads using the latest thinking, but the bottom end of the motor still is more or less unchanged. Most people’s money is on that changing next year with the full 996 replacement. And if Ducati gets half the life out of that which it did from the current engine, it will be doing very well indeed.