Composites guru Paul Crosby: How to be an ace engineer

Race engineers are required to have a unique skillset. As standard, the engineer must possess an intricate knowledge of vehicle dynamics, but that's not much use without a flair for amateur psychology when instilling belief in the driver.

Having engineered Jean Alesi to the 1989 International Formula 3000 title and subsequently turned his hand to contesting regular rally events - with great success - in his 1969 Porsche 911, Paul Crosby knows this balancing act well.

"It's like being a tennis coach," he says. "You don't have to be able to hit the ball as well as them, but you still need to know the theory on what's good and what isn't."

But Crosby isn't your typical race engineer - and not just because he eschewed the now ubiquitous route of a mechanical engineering university degree for an apprenticeship at his local British Leyland garage, having left school with one O-Level in metalwork. He did it all while running his own business, an early mover in fiberglass and composites that meant race weekends became his fix of fun.

Starting off as a mechanic for "eccentric genius" Dr. Joseph Ehrlich, Crosby rose through the ranks to become the lead engineer on March's Formula 3 operation, running Nigel Mansell in 1979 (pictured below at Oulton Park). He was briefly loaned to Ron Dennis' Project 4 operation to help Stefan Johansson win the 1980 British F3 title and was approached by Dennis to join his take-over of the McLaren Formula 1 team, but opted to stay at March until it canned F3 in favour of IndyCar in 1981.

"I had huge admiration for Ron because he was single-minded, he just wanted to win," says Crosby. "That was one of the T-junctions in my life where I didn't know what to do, and I still wonder what would have happened if I had gone with Ron to McLaren and not stayed at March - who knows, there's no point in speculating."

Having decided that he was "fed up with counting calipers and roll-hoops" in his short spell as March's IndyCar production manager, he set up on his own in a 3,000-square-foot industrial unit at Silverstone and, with the full blessing of March boss Robin Herd, spent his notice period working in March's fiberglass shop in order to better understand the then-new process.

PLUS: The enduring legacy of a motorsport engineering giant

"I left March in February '81 with an order from March to produce airboxes, headrests, sidepods and bits and pieces, so they kicked me off and kept me fed with quite a few jobs over the years," he says.

"I put my name down for a unit that was 1500-square-foot and George Smith - who was running Silverstone at the time - phoned me up and said, 'you know, you've got to wait another month and a half to get the 1500-square-foot but, I've just got a 3,000-square-foot that's become available if you want it'. I said 'of course I do, I'll fill that up' and doubled the overheads in one conversation without really thinking it through properly!"

"Within a year I got to maybe two or three people, and within a couple of years I had Pricey phoning me up saying 'come and engineer a race car for me'. I said 'I can't, I've got this business' but I did it" Paul Crosby

Times were tough at first - Crosby worked 100 hours per week and at times had to sleep in the office overnight before carrying on early in the morning.

"I worked on my own for probably six or eight months, it was a hard slog," he says. "But I was so determined I was going to make a success of it."

He started out doing repairs for local teams Alan Docking and Dave Price Racing - "an old mate of mine from the Mansell days" - and made "a few pirate bits" using a mould of a Ralt F3 nose that allowed him to cater for teams further afield.

"I had the discussion with Ron Tauranac some years later when I started doing work properly for Ralt," says Crosby. "He said, 'back in the day you bloody copied my noses didn't you?' and I said, 'I'm really sorry Ron, I did'. You do what you think you have to do to make enough money to pay the rent.

"It took time and it took a huge amount of effort but gradually we got there year on year. The most difficult thing was to employ the first person. Within a year I got to maybe two or three people, and within a couple of years I had Pricey phoning me up saying 'come and engineer a race car for me'. I said 'I can't, I've got this business' but I did it."

Unable to resist the "drug" of trackside engineering, Crosby relented to Price and combined the two past-times. He would enjoy his greatest success in F3000 with Eddie Jordan Racing, first with Martin Donnelly's remarkable part-season effort in 1988, then winning the title with Alesi the following year.

PLUS: Martin Donnelly's F3000 gamble, 30 years on

From there, he went on to run Allan McNish at DAMS in 1990 - the Scot bouncing back from a frightening crash at Donington in which a spectator was killed to win the very next race at Silverstone - and had a four-year stint at Paul Stewart Racing running the likes of Marco Apicella, David Coulthard and Gil De Ferran.

He even briefly operated his own F3000 team, Alpha Plus, had a spell in Formula Nippon running Ralph Firman and sampled A1GP with Scott Speed, only stepping back when he bought the old Reynard composite factory in Brackley and day-to-day business demanded his full attention.

By the time he sold up in 2015, Crosby Composites was producing intricately machined parts for the majority of F1 teams, including the first all-in-one brake duct for McLaren subsequently coined the 'mega-duct', which involved experimenting with a metal called bismuth that melts at a relatively low 130 degrees using a pyrex jug borrowed from his wife's cooking cupboard.

The crowning glory, he says, was an approach from Ferrari - but it would have meant a significant increase in production - "they offered us enough work to half-fill our factory" - that he wasn't prepared to take on at the expense of existing customers.

"Turning down Ferrari almost made me cry," Crosby admits.

But Crosby says the two sides of his career are not as different as one might expect, with both demanding constant improvement. Whether chasing the ideal car setup or pursuing increasingly intricate composite constructions, he was constantly motivated by the next big gain.

"The last bit of performance from the driver is when he feels really comfortable with the car and the decision making that's going on within the team and in a place where they have enough confidence to drive the car to the absolute limit," he says.

"I used to get really frustrated. We could win a race and in debrief the driver would say 'it was alright, it did the job, but I still had too much oversteer coming out of the hairpin and it understeered around that fast corner out the back'. I would feel like I had almost failed, because the car wasn't perfect. You have to keep having that mindset of 'I'm going to improve' and it was the same in the business."

"I enjoyed the technical problems that came from doing something that had never been done before and then improving on it. I wasn't that interested in making money" Paul Crosby

A details man who liked being involved "in every single aspect of everything and understanding how it works", even teaching himself to program in five-axis, Crosby made the decision at the turn of the millennium to invest in CNC machinery as current Racing Point technical director Andy Green - then at BAR - designed components to a tolerance of plus or minus .1 of a millimeter which couldn't be made by hand.

"The motivating force for me was making a part and it not being perfect," he says. "You can't have a perfect race car for example, there's always something you can do to improve it to make it faster or more economical or use less fuel, and it's the same when you're doing an engineering project.

"I enjoyed the technical problems that came from doing something that had never been done before and then improving on it. When you screw up the part that you've made and work all night making up another part because that one wasn't quite good enough, that's what I mean. I wasn't that interested in making money."

As an engineer, he gained most satisfaction from finding "that different approach that makes a difference." That lateral thinking was in evidence when he developed an auto-blip throttle from some bicycle brake cables to aid Madgwick F3000 driver Kenny Brack's problems with heeling and toeing in 1995, a necessary skill with a sequential gearbox.

The system "worked like a dream" on the down-change, but did not go unnoticed by other teams when they went to Pau, where it was protested for not being under the driver's control and, after consultation with Charlie Whiting, banned.

"That sort of thing, a bit of lateral thinking and solving the problem in a different way is a very satisfying aspect of engineering," he says.

Crosby then realised the long-held ambition of running his own team, which was intended as a vehicle to run Brack to the F3000 title in 1996. The Swedish rising star was keen and the Alpha Plus project was green-lit, but by the time his sponsors performed a U-turn and signed with Super Nova in January, it was too late: premises had already been acquired and personnel employed, leaving Crosby to scramble to sign whoever he could get with budget.

While Brack fought for the title, Stephen Watson and Carl Rosenblad were not the same elite calibre, and failed to score a point all season. Both were comfortably eclipsed by McNish in a mid-season test at Snetterton.

"That was when I thought 'I'm wasting my time here, here's a top driver going two seconds faster than the driver I've got, there's no chance of doing anything with these guys' so I thought quite negatively from then on about their abilities," Crosby says. "You would ask them 'what's wrong with the car' and they would say 'well, it's alright actually' and you'd think 'I'm wasting my life doing this'."

Crosby had pinned his hopes on signing talented British F3 racer Juan Pablo Montoya for 1997 and the Colombian duly shone in testing at Jerez. His performance attracted the attentions of Helmut Marko, who put a significantly cheaper deal on the table than Crosby could offer. Alpha Plus was shuttered soon afterwards.

• More engineering stories from Autosport
• F3 racer turned engineer Stefan de Groot
• Action Express technical director Iain Watt
• Ex-Tyrrell F1 man Brian Lisles
• RML founder Ray Mallock

"I wasn't going to do it again and just be 12th on the grid, that wasn't me," he says. "There's a massive conflict between the engineer who just wants the team to be good and the team owner who has got to balance the books at the end of the year. When you're wearing both caps, you're having constant arguments with yourself."

But the experience didn't dampen his enthusiasm one bit and he remains a passionate competitor, even if he admits to being "surprised at how difficult it was for me to replicate all of the things that I've been telling drivers for all these years". At his happiest tinkering with his classic Porsche, Crosby has no plans to end his motorsport love-affair anytime soon.

"I can't imagine not doing it," he says. "Even if I physically can't drive anymore, I will be involved somehow in some shape or form. It's a drug and I'm addicted, there's no way I can stop!"

Top tips for engineers from Paul Crosby

"Modern race engineers will understand engineering and understand vehicle dynamics very well, but shouldn't ignore the human side and how important it is to get the driver in the right frame of mind. Self-belief and the trust in the engineer is absolutely paramount in reaching success, so don't just work on the car, work on the driver as well."

"You might have 10 ideas and half of them don't work well, but you've got to keep thinking outside the box and challenge 'the conventional way'. Forget about what people say about how you should solve it, think of it as a completely fresh problem that has never been solved."