GT40 Car

GT40 Replica Construction Article.
Development and Testing.
by: R.C.Sundt

Return to Robin Sundt's Family Home Page
Return to Robin Sundt's GT40 Home Page

Building the car is only the first step. When you build a car yourself, even from a manufacturer's kit, things will never be quite like you expect. Production cars go through months of testing and fine tuning to get them right. With a kit car, you have to do this process yourself. With a "Kit" there may be fewer major problems but there will still be a lot of fine tuning to get the best out of the car and to make sure that it is as safe and reliable as it can be.

First a little background. I started with a 351 Clevland 4V, rebuilt that and then bought a second hand KVA chassis and spent six months rebuilding that. Every item that I could think of was installed on the chassis, from engine and suspension down to the windscreen wiper motor just to make sure that everything fitted and all the holes and brackets were in the right place. The problem, of course, was that they weren't.

Nothing fitted where it was supposed to be. The engine was in the wrong place, the rear suspension got in the way and was violently asymetrical and the ride height at the front was double that at the back to mention just a few of the major problems. In the end the only fittings that were not modified or moved were the eight main front suspension brackets. Every other bracket and fitting needed to be changed to get everything to fit together and work together.

One of the difficulties of building a GT40 replica is that there is no right answer to most problems. Take the rear suspension. OK, so it had to be moved to get the engine in the right place, but where do you put it? Experts told me that one key component was the roll centre. "It shouldn't move too much from the car's centre-line as the car rolls." Very useful information - but how much is too much? 1cm, 10cm, 100cm? No-one knew. They just said that it was a case of trial and error.

In the end, I built a simple string computer, converted it into an Excel program and started playing. The result was a roll centre that moves quite a lot less than it did in the original KVA layout - but would it work in practice?

When the chassis was finished everything was removed and the chassis sand-blasted and enamelled. This was in September 1993. Then the next stage got under way - the build proper.

The first step was to put all the major bulkheads in. Each was double skinned with a foam filling betwen the skins. Then the rest of the components were added and the car slowly took shape. In July 1994 the engine was started for the first time and the decision was made to try to get it ready for the Kit-Car action day at Castle Coombe in mid September. To do this, the car had to be road legal - a tall order when you consider that little things like the doors, body, dashboard, electrics, windscreen, brakes etc. all had still to be done.

GT40 Car When I started building the car, I had decided that I wanted a good, safe, fast car rather than an accurate replica and that the best way to do this was to worry first about the engineering and to get that right without worrying too much about the bodywork. Only when the engineering was right would I start to work in ernest on the body and making it look pretty.

With the help of a number of good friends, we got it on the road two days before Castle Coombe. It failed the MOT first time because the front wheels fouled the chassis on full lock, but after a nights work rebuilding the rack with new stoppers, it sailed through on the second attempt. So, with 59 miles on the clock and a body still in its unpainted state, we arrived at Castle Coombe.

The first major problem I ran into was heat. The engine was new and tight and was producing a prodigious amount of heat. It would keep cool if you were moving but not when standing still. So initially we just kept moving. However the heat caused another couple of problems. On the first outing we melted the rear anti-roll bar mountings which are just under the exhaust headers. Then the heat began to melt the throttle cable which meant that the throttle moved in a series of jerks rather than smoothly. This caused quite a stir driving down Keynsham High Street that night in heavy traffic when the engine had only two positions left - idle and 4,000 revs! The brakes were also not good, partly I discovered later, due to the fact that the heat had melted the non-return valve in the suction line to the servos and welded it shut! Despite that, it was a great day and very informative.

One problem that was answered in dramatic style was the efficiency of the suspension set-up. As I was getting a little more confident of the car and beginning to drive a bit faster, I followed three Westfields into Quarry. The problems with the brakes and throttle dictated a cautious entry into the corner but then power could be added to clip the apex and accelerate out of the corner. The car felt very stable and, even though limited to 4,000 revs while running in, responded with a frightening amount of power.

Just as I clipped the apex with my foot on the floor, the second Westfield in the group spun on the exit to the corner. The third Westfield turned sharp right to avoid hitting him - straight across my bow. There was no room to go inside him so the only way to avoid T-boning the Westfield was to hit the brakes hard. At first I thought it was going to spin, the back broke away but as soon as I stopped braking and applied a little opposite lock it stabilised just like an oversized go-kart. The car drifted sideways but remained quite stable. As soon as the road ahead was no longer full of Westfield I put my foot back on the throttle and it pulled away without either pitch or roll in a nice straight line. I felt like James Bond's Martini, shaken but not stirred, the suspension set-up definitely worked!

At speed the car was a little light at the front. Someone suggested putting a baffle under the radiator to stop air from running under the radiator and lifting the car's front end. This would also force more air through the radiator and improve the cooling. This worked very well but I took it one step further and built a bulkhead around the radiator so that ALL the air that comes in through the front scoop has to go through the radiator. I also routed the heater inlet pipe into the radiator to increase the flow rate because the main pipes were only 1in. internal diameter to fit inside the central chassis tubes. This brought the temperature down to 70 - 80 deg. C on the motorway and about 85 deg. when idling with the fans on.

The next step was to build a duct so that all the air that comes through the radiator goes out through the two large nostrils on the top of the bonnet. This raised the temperature by about 10 deg. but still kept it within comfortable limits. The only remaining problem was that the fans would not switch on automatically. Despite having 85 deg. settings, they would not start until the engine was on the point of boiling over. A little temperature measurement revealed that the problem was quite simple.

The radiator is a hell of a long way from the engine and there is a considerable temperature drop as the coolant flows forward through its two relatively narrow pipes. So by the time that the radiator got hot enough to trigger the thermostat, the engine was about ready to blow up. The solution was quite simple. Move the thermostat for the fans onto the down-pipe that carries the coolant from the top of the engine down to the main cooling rails that run down the centre of the car. By moving the thermostat closer to the engine it is triggered earlier and the engine keeps within a nice temperature range automatically.

The other major effect of this ducting is that all the high pressure air that comes in through the nose is now ducted out from under the bonnet. The result is that the area under the front bonnet is now a relatively low pressure area as the only air that gets in is the air that comes up from under the car past the wheels.

This has improved the airflow under the car and made it much more stable at high speeds. It also puts less strain on the front latches and reduces the tendency of the front bonnet to blow open and it stops any hot air from getting into the cockpit. As a result the inside of the car is also much cooler. Before this, standing in traffic with the fans on turned the inside of the car into a sauna bath as I haven't fitted the bulkheads at the front of the doors yet!

A few days after Castle Coombe, a friend was driving the car slowly up the road when it hit a drain cover and one half shaft popped out. The Renault inners had been professionally mated with the Ford outers but the Renault UJ had separated. It transpired that while the Ford end has good stops to prevent it from over extending, the Renault end has little or nothing to stop the shaft from pulling out of the"cup" connected to the gearbox.

When the shafts were made the length was chosen to allow the main part of the shaft to float in the middle range of its movement. Unfortunately this meant that if the suspension was compressed to its maximum, which extended the half shaft to its maximum, and if the Ford end was fully compressed, then the Renault end could be pulled out of its socket. The solution was to machine up two 10mm shims that were fitted to the Hub to make the movement range of the shaft shorter. This solved the problem but provides a lesson that the length of the shaft should be as long as possible, allowing for the suspension movement, to prevent the Renault end from falling out.

Both the clutch and the two brake master cylinders were originally standard 0.7in. sizes. It turned out that the clutch was nearly impossible to adjust because there was not enough travel at the slave cylinder. Either it slipped under power or, if you adjusted it to grip properly, it made terrible grinding noises going into reverse. The solution was to fit a larger 0.85in. master to the clutch which gave a longer travel at the slave end and solved the problem. It did increase the pedal pressure slightly but it is still very light.

When the servo suction problem was solved by replacing and insulating the non-return valve, the brakes worked better but to stop the car quickly you had to brace yourself against the seat back, put both feet on the pedal and HEAVE! Now the two master cylinders have been reduced to 0.65in. each and they are much better. There is more pedal travel but there is better progressive braking action and the pedal force is much more reasonable.

However, the brakes still don't "feel" right and this is a complaint that I have heard from many owners. If you hit the brakes hard, then it is quite easy to lock up all four wheels, but it still doesn't feel right. Certainly part of the reason is that the car brakes very flat, there is no nose dive, and because of the angle at which you sit there is less feeling of being "thrown forward" that you get in most cars. But I would still be happier if there was that feeling of hitting a brick wall when you stamp on the brakes. So more work to be done in this area.

The problem is not helped by the fact that the brakes are very difficult to bleed properly. In particular the rear Scorpio calipers are designed to be mounted on the top of the disc. Mine are mounted at the rear of the disc and a very small part of the slave cylinder remains above the bleed nipple. The only way to bleed them fully is to take the hub carriers off and rotate them through 90 deg. so that the bleed nipple is above the slave cylinder - not a simple task. The brakes are still a bit spongy so I guess I am going to have to bleed them again. Ah well!

All things considered, the car has been incredibly reliable. It has now done 1,500 miles and, apart from a stuck float which made it belch flame from the rear, flooded the top of the engine with petrol and nearly gave me a heart attack, it has behaved impeccably.

It has taken a long time to get used to having so much power under your right foot, to increase your anticipation when travelling quickly and not to get carried away. But it has been a learning process and a development process which has resulted in a very drivable car with, I hope, very wide safety margins.

The next step was to make it look pretty. At the end of August, 1995 I took it off the road to start its make-over. The first problem was gaps between the main body panels and around the doors that made the Grand Canyon look narrow and parallel sided. The hinges on both the front and back decks were rebuilt, and in the case of the rear hinges moved 2 inches to stop the bodywork from fouling the rear tyres as the deck opened. The hinge height was adjusted to make the panels fit better. Then 3/16th in. balsa wood was stuck into the gaps and the remaining holes filled with resin, mat and gelcote. The rear of this "patch" was then reinforced with more resin and mat and the surface blended into the original surface.

The doors fitted quite well (all things considered) but they were not waterproof or draftproof. The answer was to fit external "keepers" over the door tops. These were fitted to many original cars but in my case, I extended them along the front edge of the door top which alsmost eliminated wind noise and leakage from the doors.

Once the panels fitted correctly, and the gaps were at least small enough to jump across, the surface preparation started.

This body had been around for some years, in fact in the legendary gales the rear deck got blown across the garden, so they were not without the odd blemish. Each "star" crack had to be cut out front and back, backed with new resin and mat and then refilled from the front. In addition, two cuts were made at the narrow part next to the fuel fillers so that the front deck could be widened by 3/8ths in. to match the doors and centre section.

Then rub everything down so it was perfectly smooth and etch prime it. With primer on, for the first time it was all the same colour - instead of the blotchy faded white of misused, weathered and ancient gelcoat - it really began to look like a real car! Problem was you could now see all the surface blemishes - and there were lots.

So fill all the bad ones and rub them down then prime it, rub it down, prime it, rub it down etc. etc. This gets VERY, VERY boring. For every blemish you find and correct another two seem to appear next to them. This was about the time of the Kit Car Action Day at Castle Coombe, and at one point I harboured hopes of making it - but the blemishes won and there was no way that the car was going to be ready and it isn't worth trying to rush it. In the end I ran out of - blemishes; - eyesight; - patience; - and excuses. It was time to take a crack at the top coat.

This in itself was a slight problem. I have never sprayed a car before and, although I was learning from the many primer coats, I was not confident about doing the top coat. The car was going to be sprayed in my garage which is fairly large and well lit but not ideal. I had a friend who was a professional, but he lived in Bristol 132 miles away and, although we tried to arrange it, he couldn't make it.

Finally the weather forecast was great for the whole weekend, the temperature was high (thank God for the Indian Summer) and the wind low. It was now or never.

A final 1200 rub down on the primer, clean it off, degrease it, check the masking and it was time to take that fateful step.

Getting the right amount of paint on the car is actually quite easy. All you have to do is watch the reflection of the lights in the bit you are spraying. At the critical point it starts to reflect the light on the new flat top surface of the paint. Stop as soon as the reflection is clear and you will have enough paint - but not too much. The problem was getting into a position to see the reflection. You need to be a contortionist and to have about 20 more lights than I had.

All things considered, I was very happy with the result. Two small runs (on the front bonnet cover) and only a couple of slightly dry spots after three coats. Then I masked up and sprayed the dash board and then re-masked the car to do the stripes. There was a lesson here. Mask everything and stick all the edges of the newspaper down so that there is no way for the paint to get in - because it will - and then you spend a day with rubbing compound getting blue overspray off the white!

The next step was to finish off some internal bulkheads around the doors and bolt everything back on the car. While I was doing this, I fitted a new adjustable thermostat in the radiator hose as the original one had stopped working (actually it did work after being in boiling water for 10 minutes). Then new carpets were fitted and the car was again ready for the road.

The first trip was a disaster. The engine started getting very hot and the fans were going full blast. Luckily I had fitted a more powerful alternator, but even that was having trouble keeping up with the load. I stopped at a petrol station to let it cool down, but then it wouldn't turn over and had to be bump started. Eventually I got home, put my thinking cap on and took the radiator off. I found the thermostat bulb was restricting the coolant flow into the radiator as it was in the hose, so I put it into the radiator itself. I had previously routed the heater inlet hose into the radiator to increase the flow and now I put the heater outlet hose into the other side of the radiator.

The next trip, down to Bristol, was much better, but still very marginal in slow moving traffic. In Bristol some friends and I polished it up with an electric buffing machine and it had its second MOT. The MOT was OK after we replaced a split boot on the steering rack.

Incidentally, this car is very consistent. We broke down in exactly the same place as before (to the inch) with - you guessed it - a jammed float valve flooding the engine, this time a small bit of rubber from inside the fuel hose. It must have broken lose when I was putting the new carburettor on, I changed the 600 for a 750 which helped to keep the engine power flowing at higher revs.

On returning back to Surrey, I was beginning to despair about the marginal engine cooling and was even entertaining thoughts of getting a "one-hole" nostril to improve the airflow. I wondered if perhaps the flow had been improved too much? Was it not cooling enough in the radiator and going back too hot? Surely not, but why not check it anyway. I took it out and found some slow traffic (this is VERY easy in Surrey) to get it good and warm, then I brought it back and started to measure the radiator inlet and outlet temperatures. There was a blast of hot air coming out of the fan on the inlet side and cold air coming out of the outlet side. But it was a trickle - not a blast. I opened up the front and took a closer look. The fan on the outlet side was turning the wrong way. It was blowing the air back through the radiator while the other fan was sucking it out. Very incestuous and not very efficient!

While rebuilding the car, I had accidentally wired one of the fans up back to front. Hence the cooling problem. 10 minutes later the car was cool as a cucumber.

Since fitting the carpets and sealing the doors, the car is completely different to drive. It is much more civilised and you can even have a comfortable conversation with your passenger without getting a sore throat from yelling. It makes the performance even more shattering because it hums along (quite a loud hum admittedly) in traffic and then explodes when you put your foot down with a roar which still thrills the senses but no longer makes your eyes cross.

We tested the brakes again during the MOT and tried them out in emergency stops and confirmed that they actually work very well and are well balanced fore and aft - they are fine when you get used to them but I still think that they can be made better.

Now the car not only works well but it also looks nice. Thinks - perhaps I'll take it to Le Mans next year? See you there I hope.

Email to Robin Sundt:

Mail email to

Return to Robin Sundt's Family Home Page