During the summer of 1980, a car that would push Formula 1 into a new era was starting to take shape. The McLaren MP4 would famously laminate a new technology into the fabric of F1 with the first carbon-fibre monocoque.

While that story is well-chronicled, McLaren’s use of a secret wind-tunnel programme for the MP4 has stayed under wraps for four and a half decades. At the centre of the tale is the once-imposing National Maritime Institute in Teddington, which was part of the National Physical Laboratory (NPL). Within a sprawling industrial site were housed several wind tunnels, but it was Tunnel 5 and Tunnel 6 that caught the eyes of Ron Dennis, John Barnard and Alan Jenkins when they visited early in 1980 while scoping out the transition of Project 4 into Marlboro McLaren International.

Dennis, whose Project 4 team was enjoying success in Formula 2 and F3, was still working on a deal to take over McLaren which, with the assistance of Marlboro’s John Hogan, was completed in September 1980.

At that time, the NPL side of operations, from a fluid and aerodynamics standpoint, included a youngster who was cutting his professional teeth. Chris Saunders went on to hold senior aerodynamic positions at Williams, McLaren, Lola and Red Bull, but in 1980 he was a self-confessed “cocky kid know-it-all” who secretly “didn’t know it all but was willing to learn how to”.

“If Ron Dennis, who was impressively persuasive, says ‘we need to do this now’ then you just do it”

Saunders was one of a number of people who moved across from NPL and the British Maritime Institute into F1. Others include Alan Stovold, who worked at McLaren for 38 years before recently retiring; Williams, Mercedes and Ineos America’s Cup designer Geoff Willis; Laurie Cole, who went on to work for McLaren and Mercedes; Lotus engineer Jeremy Bliss; and Nigel Beresford, who engineered Jean Alesi at Tyrrell before becoming a key member of Penske’s IndyCar success in the 1990s and beyond.

“Tunnel 5 had a beautiful overhead balance in it, and we were charged to design an installation to go testing with McLaren,” recollects Saunders. “There’s a lot of work in getting the rolling road design and all the suction systems and computer software and it was quite challenging. But I guess at that age you’re not afraid of anything, and if somebody like Ron Dennis, who even then was impressively persuasive, says ‘we need to do this now’ then you just do it.”

What was the first concentrated long-term programme for an F1 constructor at a commercially available rolling-road wind tunnel had begun amid great secrecy. London’s Imperial College, where a rolling road had been installed by John Stollery, aerodynamicist for legendary land and water speed king Donald Campbell, had been used by Peter Wright and Lotus in the 1970s. But the McLaren/NPL partnership was a much longer-lasting arrangement, one that went on until the early 1990s.

“The team on the MP4/1 involved unique characters such as ‘Harry the Box’ and ‘Forklift’”

McLaren provided the scale model and its model makers supported the programme. As well as McLaren ‘lifer’ Peter ‘Poosey’ Hodgman, there was another young and ambitious ‘racer’ also starting a long career in F1 – Joan Villadelprat. Saunders recalls the Spaniard as a “really hard grafter and great fun”. A friend of Barnard’s, Villadelprat moved with him to become chief mechanic for Ferrari in 1987, a position he held until moving to take his first team manager job at Tyrrell, and then to Benetton in a similar role, where he was part of the team’s mid-1990s glory years.

As well as Villadelprat, the team on the MP4/1 involved unique characters such as ‘Harry the Box’ and ‘Forklift’. “They used to come over with Poosey and Joan to bring the equipment, have a laugh and smoke a lot of fags,” laughs Saunders. “We didn’t know their real names, but boy we had some fun.”

McLaren’s use of the tunnels alternated with visits from GEC-Marconi, which was testing weapons including the secret Sting Ray torpedo, which was loaded on to HMS Ardent in the 1982 Falklands conflict. While Barnard and Jenkins looked for every pound of downforce to aid cornering speed one week, Saunders was working with something altogether quicker and more combustible the next. If that were not enough, next door to Tunnels 5 and 6 was, according to Saunders, “a particle accelerator that plenty of my colleagues mucked about in!”

Testing had started by late summer. While Barnard was majoring on the carbon-fibre innovation at Hercules in Salt Lake City, Jenkins and Saunders were getting the MP4/1 aero programme underway. “Alan was a very good graphic artist, had recently been race engineering and spent most of his time doing the aero design work,” recalls Saunders. “We went through different designs for the rolling road belt systems, tensioning steering just to get a good belt, and get it to stay flat. With a big underbody like F1 had then, the car would snap under high downforce at low ride heights. In a wind tunnel, it sucks the belt off the rolling road, it won’t track and then everything turns into a shitfight. Luckily, we worked most of that out beforehand.”

‘Most’ is the operative word. These were the pioneering days of F1 tunnel testing and there were still a few leaps of faith involved. One came particularly early, and Barnard happened to be right behind Saunders’ shoulder when it occurred. “We were running the car in anger for probably one of the first runs,” says Saunders. “It was all mechanical balances and strain gauges then and not a lot of data acquisition stuff like you have today. It was pretty much an analogue world we lived in. So, I’m looking at some of the data, we’ve got a big meter, and it suddenly just swings off to the left, and it all goes BAM! I swung my head around to look in the tunnel and the bloody model is starting to bounce around and porpoise. It got violent and we thought we’ve got an instability problem with the belt.”

Then the numbers shifted. Saunders and others “sort of pulled it back into a zone where we could make a measurement. There was a huge difference, between these two numbers, one where the car was obviously making downforce and one where the underbody was stalled. That’s ride-height-dependent clearly.

“John is looking at the model, then at me, then at the data. ‘What the hell is going on, Chris?’”

“We changed the ride height of the car, and all the data shot to the other end of the scale meter – just like that. And the model’s there, banging around. John is looking at the model, then at me and then at the data. ‘What’s the hell is going on, Chris?’ he says. A bit flippantly, but genuinely, I replied, ‘How long’s a piece of string, John?’

“He didn’t really have much of a sense of humour, and at the time rightly so. You have to remember that this was groundbreaking stuff, just the fact that in real time we were seeing this porpoising. We genuinely didn’t know what it was. It was totally fascinating for all of us.”

The suck-it-and-see philosophy went through many stages. One was to try to pressure-tap the underbody – drilling small perpendicular holes into its surface to measure local static air pressure. “We were running several pressure taps up the underbody to find out what ride height the underbody let go,” explains Saunders. “We found out that this was dependent on skirt height, as well as ride height. But the issue was ‘how are we going to display this?’

“So, I wrote a piece of software to plot all this data on an ‘x/y co-ordinate plotter’. It probably took me a week or something like that, just using the raw data straight off the pressure gauges. It was all great fun, wasn’t work at all to me. Then you’d start trying to put the underbody fillers in, going nose up, so you could then plot the pressure distribution along the underside of the car and see what was going on. Really, we were just making it all up as we went along.”

By the end of 1980, most of the initial work was completed and the team readied to debut the MP4 at the Argentinian Grand Prix – the third round of the ’81 season. Soon the car showed a lot of promise. In its fifth race, the Spanish GP at Jarama, John Watson was third – part of the leading train in an event unforgettably won by Gilles Villeneuve’s agricultural Ferrari. Next time out, Wattie went one better: second to Alain Prost’s Renault in the French GP at Dijon.

“Back then you were lucky if you had 15 engineers in the company”

Two weeks later at Silverstone, Watson claimed his second F1 victory and the MP4’s first in a dramatic British GP. He had to navigate his way through the chaos of the early accident that accounted for Villeneuve, Alan Jones and McLaren team-mate Andrea de Cesaris, before the leading Renaults of René Arnoux and Prost faltered. Watson was free to sail to victory, 40sec ahead of the Williams of Carlos Reutemann.

“I was there, and Wattie went and won it – what a joy,” reminisces Saunders, who was sitting in the grandstand at Woodcote.