The First Lotus GT

LOTUS MK 11 SPORT   XJH 875   CHASSIS NO. 203

by Gilbert “Mac” McIntosh

 

 

 

The top could be opened to allow the driver to step inside.

The car behind is a Bond 3-wheeler.

the DeHavilland Vampire fighter jet.

 

 

 

 

Mac McIntosh and his Eleven

With the "FIA" full width 'screen.

Mac's sketch of the latch mechanism.

 

A view of the latch, with Venom canopy handle.

 

Mac's sketch of the hard top.

 

 

 

 

 

 

 

 

Mac's typical 'Sports' engine bay.

 

 

The original drawing of the Elite

 

 

 

 

 

 

 

 

These are a few notes on the background of the car I built for myself and used for several years as a road car.

The design of all the cars from the Mk 8 to the Elite was based on aircraft practice rather than the very ‘hit and miss’ practice of the sports car world.  I was horrified by the complete lack of data on loading cores for the structure so a lot of the time on the 8 and 9 was spent working back from failures to what must have been the applied loads.  By the time of the 11 we could design off the drawing board and get a light structure which didn’t fail from fatigue or from an acceptable degree of bump.

It was in this environment that the Elite idea first started:  we wanted a bread and butter money raiser to back up the Grand Prix and sports racing cars.  The problem was how to make a G.T. car down to a price – a poor man’s Porsche.  A paneled space frame was the Italian approach and we could do that, but the price would be too great, as would the pressed steel chassis body unit – the returns would be too small to pay off expensive dies.  We looked at buying Ford 100E body shells, which we could buy cheap, then fit our own suspension and engines – the route Lotus went with the Lotus Cortina.

The breakthrough came when Colin saw the little Bond sports car with a fibreglass chassis body unit at the Motor Show.  He was full of it and we really kicked the idea round.  The idea of a fibreglass chassis body unit answered all the problems on paper but there was a hell of a lot to do before we could design a car.   I was OK on the structural side as far as stressed fibreglass mouldings went because I had used it on Vampire nightfighter radomes – mark you there were different types of mouldings to the car body mould – bathtub mouldings we called them.  It may sound rude but tubs and most car fibreglass was 50% bulk filler with polyester resin and chopped strand mat whereas on the radomes and Elite [doors] we used satin weave cloth and epoxy heat cured resin.  Most fibreglass structures would collapse into powder if you hit them with a hammer whilst on ours it just bounced back.

I had to get round the problem of an odd shaped box with a lot of cutouts and no stress calculation background from the car world, they did it all by eye and testing in those days.  I was used to aircraft stressing where every problem is different, so that could be solved with a bit of basic stressing, but I saw another problem where there was a lack of knowledge, rather as we were with the stressing cores on the 8.

That problem was water-proofing because nobody would tolerate getting wet as you did in an open sports racer, it used to come in everywhere!  It may seem simple “you just use sealant,” but I had lived with the problem of sealing the integral wing tanks on the [DeHavilland] Comet.  It took DH years of test work before we knew how to keep kerosene inside a bolted aluminium wing without leaks on a production basis, and I could see a similar problem on any new civilized car.  Colin could see the point but didn’t want the hassle at the works so the plan was hatched that I would make my own civilized 11 Sports so we could learn the drill.  Where water flies around under a car is not generally known so we had to do our own homework – it took 6 months hard work before we got there.

My car had the standard Sports layout with a live back axle, 3 speed standard gearbox and Ford 100E side valve engine.  Exhaust was normal multi-tube and twin SU carbs were fitted, but no other tuning was done as it was to be a road car – replacing my 1928 Austin 7 Chummy!  At first I used the standard full width F.I.A. screen but a hard top was on the way as shown in the works brochure. 

I can’t remember where the screen came from.  I certainly didn’t buy it or make it so it must have come from Lotus but the top was all my design and manufacture and worked very well.  The top had a ˝” square 20 SWG mild steel perimeter frame with a 20 SWG aluminium skin riveted to it.  I can’t remember exactly how the rear lower window support went other than that it fixed to the rear bodywork with a few bolts and incorporated the pivot brackets for the hood.  There were two spigots on the top of the windscreen to locate the front of the hood.  These spigots stuck into holes in brackets at the ends of the front tube and locking was by a handle in the center and two rods to the locks at the spigot housings.  Again I can’t remember the details other than that it was very simple and that the handle was a [DH] Venom canopy release lever from the night fighter test fuselage. 

I know it may seem odd that I forget the details but you have to remember that it was a long time ago and I was up to my ears in the design and stressing of the new car, not to mention that this was only a hobby.  My real interest was aircraft design and I had a fairly large chunk of the new 600 mph airliner on my drawing board.  We were right at the sharp end of the new jet airliner development and life was very interesting technically.  Colin and I couldn’t understand people’s interest in past designs other than as a source of data to make the next step on the new cars – sad for the historians but fact!

Back to the hood after that digression! The new window was Vybak, a flexible, transparent, plastic sheet about 0.8 mm thick and it fastened to the rear edge of the hood and the frame bolted to the body.  The window collapsed when you opened the hood and it was a light, cheap, simple solution to the problem, but was not perfect as it tended to form bubbles which could crack in very cold weather – I think – as mine didn’t all the time I had it.  Maybe it was an acceptable solution if you accepted a 2 year replacement life, after all we do that with tyres and exhausts.

The other main modification from the standard ‘Sport’ spec was to the cooling and there I used the standard inlet duct on the front bodywork but added the dotted rear ducted to improve the efficiency and added an adjustable outlet flap.  This made the radiator ducting exactly as on the [DH] Mosquito and Hornet where the drag was minimal due to heating the air and getting through from the outlet jet – a bit like a ram-jet!  I was not after that effect but wanted the quick warm up when the flap was closed.  The position was fixed by spring opening and a Bowden cable plus ball chain location on the dash using wash and basin plug chain in a keyhole socket [see illustration].  The system worked a treat as the radiator was big enough for higher output engines so was OK for hot traffic conditions but could be throttled down for frosty mornings.

I painted the headlight tunnels matt black to stop upward glare in the Hertfordshire winter fogs but the car was so low I found most fogs were non-existent below 3 feet so you could see quite clearly in what was a pea-souper in a normal car.  I am sure I had a lot of comments on the line of “Mad sod, he’s driving far too fast!” when I was probably much more in my safe visual range than they were.  You did have to watch ups and downs in the road as you were driving in a wide shallow tunnel.  Before the paint I had tried a layer of grass and dirt in front of the the lights, when the covers were off, and once while parked, a policeman came up and lifted the grass out of the tunnel.  He said, "ah, clever!" and placed it back. 

The car was OK as a road car if you used your head and avoided things like high kerbs into drives, Green Line bus exhausts, and idiots who used knuckles to see how thin the Alumimium was on the wings.  If you have not worked it out the Green Line exhaust pipe was level with a passenger’s ear and was very hot so in traffic you always had to keep fore-and-aft space so you could move out of the way.

It was at the time of the Suez and petrol rationing so its low fuel consumption was great.  I got well over 40 mpg and could get much more if I slipped out of gear and coasted – it would run for miles with the low drag.  If you ask where was the odometer the answer was my log book and a record of all trips plus fuel and my other events.  I learned that one from Mick Costin who had his diary on a shelf with all the events on the car.  When a problem cropped up Mike would say “just a minute and I’ll get the details,” retire to his office and go through the diary.  He would re-appear with a sheet of paper with all the relevant events listed in date order – it was annoying how often the reasons were immediately obvious.  That discipline is why he was one of the best development engineers in the game and is so much the reason for Cosworth’s success as Keith’s design.

Can’t think of any more about XJH 875 so I will go and have my dinner off the dining room furniture that was part of the spend when we sold XJH to the Chequered Flag before I got married.  From what I hear XJH is still doing good service,  like the sideboard, table and chairs.  

 

postscript – After concluding his volunteer design work on the Elite, Mac left Lotus to devote himself to aircraft design.  His Eleven had been a gift from Lotus for his many hours of effort.  Mac was therefore surprised when, several years later, he received a bill for the car and Chapman was  ‘unavailable’ to intervene.  He met Chapman again in 1969 and was offered the job of Chief Designer for Team Lotus.  Mac declined the offer. 

(This information from Peter Ross’ book, LOTUS – The Early Years)

Mac passed away on November 20, 2006.  An honest man and a brilliant engineer, he will be fondly remembered.

BACK to www.lotuseleven.org