A busy winters fettling

Our erstwhile chairman has asked me to jot down a few notes about the recent fitment of a naturally aspirated Cosworth engine to my Sierra based RH7.

Having built the car originally with a corking 165BHP Pinto engine. I was quite pleased with the performance of the car, however following a day trip to the Sandown '95 show said engine shed a conrod bolt and threw a rod destroying the entire bottom end and bending two valves. The mighty bang scared the sh*t out of my passenger as oil got onto the exhaust manifold and she thought the car was on fire. Four litres of expensive synthetic oil and some twisted bits of metal were dumped in the middle of Slough high street.

The resulting 'phone call to the AA was quite amusing.

"I've broken down in the middle of Slough and will need the vehicle recovered"
"Are you sure sir, in most cases we can get your car running at the roadside"
"No - I think it will need recovering"
"In over 90% of cases sir we can fix the problem, what makes you so sure?"
"I'm actually holding one of the conrods in my hand at this moment"
"We'll send a recovery vehicle sir"

After a depressing ride home I vowed to make the engine unburstable by the addition of steel crank and rods (I had a Cosworth crank and rods in the garage from a friend who had melted a piston in his Rally Sierra Cosworth).

After pondering I decided to rebuild the engine as quickly as possible to get the car running and build a better engine over winter. The engine was duly rebuilt with 205 bottom end, titanium big-end bolts, wide beam rods and tuftrided crank, the valves and cam were replaced and all was well. This engine has now been sold to my business partner who is currently building an RHE 7.

Come the winter I started evaluating the options, staying within the 2LTR maximum: -

All steel Pinto                                    Maximum output <= 190 BHP
Warrior 16 valve steel Pinto              Maximum output <= 250 BHP
Naturally Aspirated Cosworth           Maximum output <= 240 BHP
Vauxhall 16 valve                              Maximum output <= 250 BHP
Hart                                                  Maximum output  <= 250 BHP
BDG                                                 Maximum output <= 250 BHP

I had discounted the turbo option for the Cosworth for two reasons, I did not like the all or nothing nature of the torque delivery, and the installation of the turbo would give space/heat problems in the engine bay. The naturally aspirated Cosworth option would entail scrapping the turbo/injection/engine management system and using carbs and conventional ignition. After all the engine was of sound design, a good solid steel bottom end topped with a DOHC 16 valve head, no reason why it should not make good power on carbs.

The Pinto was the cheapest option, but was 50+ BHP down on the other two, the Warrior was expensive and a rarity as were the Hart and BDG options. Although the Vauxhall is a charismatic engine and a more modern design than the Cosworth, there were disadvantages, carbs on the wrong side, bellhousing problems, flywheel problems, clutch problems, engine mounting problems, gearbox problems. In addition I felt a Ford engine was somehow better suited than a Vauxhall for this type of car and that the installation would be easier.

Given the choice again I would go for the Vauxhall. Although physically shoehorning the Cosworth engine into the engine bay was easy, as it is basically a beefed up Pinto with a 16-valve head, incompatibility of basic components made the engine build a nightmare.

A second-hand Cosworth engine with the bearings run was acquired and a stripdown revealed a scrap block, but serviceable components otherwise, a good 205 block was sourced (the non 4wd cossies used selected Pinto 205 blocks) and the planning started.

Problem one: Pistons

Turbo engines run very low compression to avoid detonation when the turbo cuts in, so the standard pistons that give only 7 to one compression had to go. The capacity of the combustion chamber in a Cosworth head is large, so domed pistons are required to bring the compression to a reasonable level or the engine would run like a herd of turtles. These are bloody expensive as they are forged and designed for racing, they also clatter like buggery when cold (like a diesel taxi).

Problem two: Flywheel

The Cosworth crank has a nine-bolt flywheel fixing, the Pinto flywheel is six bolt. There is no standard flywheel to mate a Cosworth to a normal gearbox/bell housing. Either use all Cosworth, flywheel, bellhousing, clutch, starter, gearbox etc or have a flywheel made. I chose the latter and a nice lightweight flywheel was made for 80 pounds.

Problem three: Cams

A turbo engine does not have much duration or overlap on the cams, as this would result in 'blow-through' where inlet mixture would blow through to the turbo causing detonation in the turbo. In order to extract reasonable power the cams would require much more duration, overlap and lift. The standard cams have about 260 degrees of duration; I would need about 300 degrees on inlet and 290 on exhaust for the right power curve. Kent cams to the rescue.

Problem four: Cam followers

Cosworth engines use service-free hydraulic cam followers, and as the cam profiles are not very radical, and the standard engine is designed to rev to 6500 or thereabouts these are satisfactory. However when more extreme cam profiles and higher RPM bands are used, these tend to pump-up resulting in valve/piston contact. Every Cosworth engine I have seen has had 'kiss-marks' on the pistons due to over-revving. Valve piston contact is a definite no-no especially at the sort of RPM (8500+) I envisaged using.

Kent cams did solid followers at 274 pounds a set (AAAAAARGH), however a cheaper alternative is to strip out the hydraulic gubbins in the standard followers and shim them to make them solid. This makes them quite heavy and is generally OK for the exhaust valves that are smaller and lighter than the inlets, and have a tamer cam profile. On the inlet however I used proper light steel followers to avoid valve float. I have since discovered that Fiesta Diesel cam followers fit and these are of the solid shimmable type.

Problem five: Cylinder head

O lordy lordy, where to start. I have modified quite a few cylinder heads in my time, minis, imps, crossflows, Pinto, BDAs, but the Cosworth is quite a challenge. The ports are way too small, great for a turbo engine; it keeps the gas velocity high and the turbo spinning. This is bad for a N/A engine though and the inlet ports are VERY long, the valve seats and throat area are very badly finished, a blind man on a galloping horse could do better. The exhaust ports have a nasty sharp turn in them and are tiny. The inlet ports although two to a cylinder are not unlike the imp head and once you have worked out how to open up the port 4 inches into the head it is straightforward. The valve throats needed much relieving as I was fitting over sized inlet valves. The seats needed recutting on the inlets to accommodate the larger valves, fortunately the steel inserts are large enough. A full radius into a thin (1mm) seat was made on the inlets and a 1.5mm seat on the exhaust for better heat dissipation. As the cams being used had a lot of lift it was also necessary to relieve the valve spring seats to avoid coil binding in the valve springs. I am very fortunate to have all the necessary equipment to do most of the headwork myself, having acquired it over the years.

Problem six: Oil pump

The Standard Cosworth oil pump has an integral spray bar to spray oil onto the underside of the pistons to rid them of excessive heat generated by the turbo. This is unnecessary for N/A engines, so I used a Pinto high capacity pump. The Pinto con-rods have an oil spray drilling for small end lubrication, the Cosworth rods do not as they have the oil spray bar (which I was scrapping). I was a little concerned about lack of lubrication at the small end but experience and knowledge greater than mine have assured me that there is sufficient oil splashing around to avoid galling of the little end bores.


Problem seven: Carburation

Simple one this, two bloody great carbs (48 DCO/SP) on a custom made manifold should do the trick, and as the engine bay top panel already had a box cut-out to clear the carbs on the Pinto, they should be no problem. However the ports line up differently on the Pinto, and they are offset to the rear of the head, and the manifold is swept back. On the Cosworth the ports are central with the bores and the manifold is dead straight. This required the box to be enlarged forward to accommodate the carbs in their new position. The linkages then did not fit as the carbs were now 20mm further apart. Two minutes extending the carb linkages with a MIG and a day enlarging the carb box cut out using stainless and a TIG did the trick.

Problem eight: Exhaust

There was no way to fit the existing exhaust to the Cosworth head, and even if there were the manifold is a grim design and the box is not much better. Over in Brackley near where I live is a place called GDS exhausts that make Stainless steel exhausts for lots of exotic/competition cars. The owner is a loony who drives a Metro 6R4 with a 475BHP Turbocharged V8 installed as his everyday car.

My kind of guy! He soon had an exhaust fettled for the car that works very well and can be removed without disturbing any other components. It has a re packable silencer (which blows its packing out every 1000 miles). It is well made, effective and quite easy on the eye - but expensive.

Problem nine: Gearbox

The standard Sierra gearbox I had fitted had ratios better suited to a transit van than a performance car, the gap between 1st and 2nd was ludicrous. Even with the Pinto engine which gave power from 3000RPM it was necessary to rev the engine to 7500RPM to get in to the power band in 2nd gear and as 1st is so low wheelspin prevented this from happening. The result is falling off the cam at all sorts of embarrassing moments. The expected power band of the new engine was 4000RPM to 8500RPM so the problem would be worsened. Fortunately I know a chap called Brian Hill who has been making specialised gearboxes since the seventies, and he will produce a gearbox using the Sierra casing and internals with a wide variety of ratios. I plumped for the old 'rocket' ratios from the Lotus cortina together with a lower 5th gear of .88 instead of .82. And what a corking set of ratios these are! Brian achieves this miraculous conversion by machining gears off the layshaft and splining on new gears from Quaife. He also wreaked his magic on the gearshift making it quick and light.

Problem ten: Clutch

The standard clutch fitted to the Pinto engine would slip when doing full bore gear changes, so I did not think it would stand up to the Cosworth torque output. I looked at paddle clutches, but the original in the Pinto had caused the pedal box to bend and was way too fierce. I investigated the Centreforce clutch which is a beauty, paddles on one side of the plate and normal on the other, it is light and strong and does not jerk or snatch. So far it has not slipped either. However it did take 4 months to arrive. Apparently they ship the clutches to the 'States to be re-lined.

Problem eleven: Ignition


The standard distributor from the Cosworth is useless for a NA engine, it is simply a trigger, and there is no advance mechanism as this is handled by the Engine Management System via a crank trigger. Therefore a Pinto mechanical distributor has to be used. The Cosworth engines burn characteristics are vastly different to the Pinto, it has a central spark plug (faster burn) and cleaner flame path travel, in addition the CR I was using was 11.75 to 1, giving an even faster burn. A competition Pinto requires maximum advance of around 38 degrees BTDC at 3800RPM; a Cosworth requires 28-32 degrees and a way different curve. This required some distributor magic from Graham Hickman (ex Aldon) who got it right first time, a bigger problem was the distributor cap. The STD Pinto Bosch distributor is 35mm taller than the Cosworth one as it has a mechanical advance in the base. This leaves the distributor perilously close to the inlet manifold. A side entry cap just fits, but the distributor cap has to point out toward the front wing, then the leads wont reach (longer leads can be made 30 pounds each!). With some judicious fiddling things just reached, but the cap was the last one in Britain with no more supplies expected unless specially ordered - I am VERY careful when removing the cap. The standard points were junked in favour of a very neat IGNITOR unit from Aldon.

In addition there were many other nasties waiting to bite my arse along the way, crankcase breathing, front pulley, cam timing, servo take-off, oil cooling, underbonnet clearances, pistons hitting head etc. etc. Still no one said it would be easy. Whenever the word Cosworth is mentioned you can hear the sharp intake of breath as the price doubles. Ignition leads are 100 YES 100 pounds a set. I have since discovered that Rover 216 8 valve and Honda 8 Valve leads with some minor modification will fit, these are 15 pounds a set from Halfords.

Was it worth it? 


The result has transformed the car, the engine is very tractable but the real power is made above 4000RPM. The engine howls like a banshee and sounds almost as good as a BDA on full chat. It makes around 200BHP at the wheels at 7800RPM which is around 230/240 at the flywheel and it goes like stink. The engine has around 180lb/ft of torque at 6300RPM and over 140 all the way from 4000 to 8500.

In addition the gearbox keeps the engine singing and avoids the dreaded 'off the cam' experience. Pulling away cleanly is quite difficult as the high first gear (2.38) and poor power below 4000RPM mean you either bog down or spin the wheels like a loony. But a good start can bring 60 up (in first!) in around 3.5 seconds with 100mph appearing around 8/9 seconds. After this things get fraught as the car is built like a brick and the CD factor is about the same. It is easily capable of 135+ MPH (a mere 6400RPM) but the front end gets a little out of shape at this speed. Fifth gear is now a driving gear rather than an overdrive as it is now lower and the engine has so much torque.

It has most certainly surprised a lot of the local boy racers, a few bikers and a lot of exotic local machinery, including my mechanic friend in his Porsche 911 who could not believe he was being passed especially by a kit car. He is now a kit car convert as after his humbling experience he followed me home to ask 'what the f*ck have you got under the bonnet?".

Overall the conversion has cost around 3600 pounds including the gearbox clutch etc. which is a lot of money. But the engine is great and is worth money if I don’t blow it up. The car has now cost around 9000 to build, which is a lot cheaper than a Caterham Vauxhall and is quicker if not so nimble. There is no way I would recover this cost if selling the car but then I wouldn't want to sell it, it is exactly as I want it.

If I was doing the job again I would probably use the Vauxhall and it would be 1000 pounds cheaper and the engine is lighter. Most of the points which counted against the Vauxhall, flywheel, clutch, exhaust, carbs etc. also eventually counted against the Cosworth, the advantage with the Vauxhall is that it is made to be naturally aspirated. The disadvantage is that it not very strong in standard form, conrod bolts are made of cheese, as are the pistons, but these things can be remedied. When properly built it is as powerful, but sweeter than the Cosworth. There is no doubt however that the sight of the Cosworth is evocative it looks just right under the bonnet, somehow more butch than the Vauxhall.

A prolonged period in hospital and in recuperation following major back surgery has kept me out of the car for 3 months, but I am now able to drive it with a little care. I am now enjoying to the full the fruits of my labours.


Dave Andrews