Ford’s 289 V8 has been tuned for racing for more than 50 years, but as Alan Stoddart discovers, a combination of ingenuity and cutting edge techniques means that improvements are still being made.
FINISHING at the front of a historic race at the Goodwood Revival or the Le Mans Classic is no mean feat. Aside from the enormous skill (and what often seems like superhuman bravery from the drivers), the cars themselves have to be set up just so, with everything scrutinised and perfected to make sure that every ounce of performance can be eked out and depended upon on the final lap of a hardfought battle to the chequered flag.
Nowhere is this process of continual refinement more thorough than within the engine, and thanks to modern techniques and the seemingly neverending innovation of skilful engineers working at specialists, engineers are able to bring about improvements on engines that have already been honed for more than 50 years, even if the 50 years of fettling have been carried out at the same company. This has been the case for Mathwall, which has been race preparing the Ford 289 and Chevy V8 engines since it was first opened by the current owner’s father in 1967.
There have been some things that have changed over that time though. When the current owner, Mark Mathieson, who has a background in Formula 1, took over in 2007 he wanted to use the modern techniques from motorsport’s fastest category at Mathwall, and so developed a part of the business focused on engineering services. This side of the business works on everything from innovative oil systems, including one that received government funding through the Niche Vehicle Network , to supporting process efficiency improvements by equipment redesign. This division is able to adapt and use its testing equipment to seek out small gains in specific systems within an engine and test out new ideas to learn about places where future advantages might be found.
Mathwall is also able to use this engineering services side for its own gain, giving it an edge when it comes to tweaking those Ford 289 engines that it started with all that time ago. One of the machines that gives the company options is the SpinTron. “It’s basically a large electric motor,” explains Mark Webb, Mathwall’s chief engineer. “It tends to be used a lot for valvetrain development because if you get it wrong you are not going to trash a complete engine.
“This makes it ideal for the historic side. If we wanted to make a new cam or test out new springs here is a good place to do it. We can also strain gauge components and assess whether we are getting loft and separation from the cam.”
One of the other ways the SpinTron might prove useful to Mathwall in future is to carry out a detailed study of the frictional losses within its historic engines. The machine would allow the company to measure how much of the engine’s power is wasted system by system and to prioritise where the firm could find improvements. Part of the reason for focusing on friction rather than the valvetrain is that, aside from being confident that Mathwall already has good valve control, with the FIA imposing a rev limit of 7,000 rpm, the returns from focusing on the valves are unlikely to be significant, or as Mark puts it, “there is only so much you can do”.
This careful consideration means that Mathwall is able to focus its resources on components where the biggest gains can be found. With the Ford V8s, one of these areas is the cylinder head.
Using a flow bench to scrutinise a number of senior engine builder Mathew Smith’s hand fettled ports, Mathwall has been able to work out what is the absolute optimum port shape and port size for getting as much air into the engine as possible. As FIA rules require the engines to use original pre-1966 heads, Mathwall has to work within the confines of the original castings which were designed for a modest 280 bhp output; the removal of too much metal would leave the cooling jacket walls too thin. Having reverse engineered the perfect design from these handmade ports, it is then possible to subcontract the machining of the parts to a CNC specialist. Going through this process means that not only is the preparation of the heads more efficient, but it also means that every customer gets, to exacting tolerances, the best porting possible.
Mark says that this reverse engineering has been made far, far easier thanks to measuring tools that engine tuners in the sixties could only have dreamed of. “I can remember when the CMM (coordinate measuring machine) just used to touch on, so if you wanted to measure a port you would effectively go around each section and the CMM would touch-on maybe 20 times before going deeper and repeating the process. Now the probe can stay in contact at all times, and it’s much faster.
“I watched a guy the other day – in fact he did the inside of a sump for me: I’m trying to make baffles and gates and walls fit in there, so having a surface is key. Well, he just waved this FARO Arm scanner over it and up it came on the screen. That would have taken so long until recently. Setting it up on a CMM would have been very time consuming, and that would have only given me a point cloud anyway. I don’t have to try and fit surfaces through a point cloud any more, the machine does it.”
Mathwall has worked on almost all the engine components over the years, everything from the connecting rods to the gudgeon pins, but even after a component has been redesigned, it isn’t forgotten by Mathwall’s engineers, who revisit individual parts and improve them when there is an opportunity. The pistons for example, were redesigned when new piston ring technology meant that they could be modified to take advantage of these new, thinner rings and offer a piston that had less ring friction. It was also a chance to tweak the rest of the component; making it lighter to reduce the reciprocating weight, and using smaller, racing skirts. This again is another area where the cutting edge engineering services side of the business could be put to good use for historic engines, with its system-level expertise being able to accurately identify the benefits of any individual changes, which could then go on to inform the design direction for the next development of pistons.
Another of the tools at Mathwall’s disposal to improve the performance of individual components is the ability to predict performance using 1D simulation software. The camshafts for example are one place that Mathwall is currently focusing on in parallel with their new optimised ports. “We have a good bank of historical data now, so we can use five years [worth of data] to validate the model, and then we could step through the last five years and see if the model predicts accurately what we have found on the dyno from each of the cams we have used,” says Mark.
“So we could go through a lot of cams in the 1D simulation world and have a reasonable level of confidence that some ideas are worth trying in hardware.”
Furthermore having success in one area of engine development often opens up new avenues for the team to pursue. The hard-earned improvements
that have been made to the port have enabled Mathwall to go back to the cam to increase lift, but “the next thing that might happen is that the cam to lifter interface may start to get into trouble,” says Mark. “Well, we’ve now got to solve this or we’ve got to go back to the cam and soften it a little bit. So it’s a constant battle, as soon as you make an enhancement somewhere you have to think about everything else.”
This constant process of improvement has meant that the level of expertise and refinement is now at a massively granular level. One of the current ideas is to slightly modify the shape of the combustion chamber after realising that the shape of the pistons that Mathwall was using could cause detonation to be focused on one particular area, which isn’t ideal. So it has got to the level now where the company is teasing out improvements by managing the precise location of detonation using subtle changes to the shape of the piston crown. “It’s getting quite detailed now,” says Mark dryly.
Of course, all this scrutiny in the workshop and on the in-house dyno is one thing, but it is quite another to see how it performs on the race track.
This is why there is often a member of Mathwall’s team trackside, when its clients’ cars are being raced. Not only does this reassure car owners that if there are any problems, they can likely be fixed pretty quickly without ruining a weekend’s racing, it also means that the engineers are able to talk to the drivers and get lots of great qualitative feedback about the character of the engines and how they actually work in the heat of battle.
This is particularly important given the strict rules on data logging imposed by the FIA on historic racers. Without the ability to pore over huge datasets to understand what the car is doing, the engineers must decipher the customers’ comments to find out what’s really going on, which can sometimes lead to interesting discoveries.
It was one such occasion which sparked the idea for the study of the water pump, recalls Mark. “We had a comment from a client, who said that his Cobra was running quite cool, but his Daytona Cobra wasn’t. So, you think, ‘ok it’s one engine spec to us, what is different?’” After thinking about this puzzle, Mark began to consider the differences in radiator entry, with the Daytona having a smaller radiator than a standard Cobra. “OK so that might mean we can get away with a different pump on the standard Cobra.”
Reassuringly all this attention to detail, the unbelievable level of scrutiny and refinement, as well as the innovation and insight offered by the engineers, pays off. Although the competition is fiercer than ever, Mathwall is more than holding its own, with one of its engines in a 1965 GT40 MK1 taking the overall honours in Plateau 4 at July’s Le Mans Classic and another in a 1969 Lola T70 Mk III B achieving fourth place overall in Plateau 5.
This is a trend that looks likely to continue with Mathwall having recently received planning permission to expand with new design space and workshop space, as well as an investment in two young staff members who are on day release to learn more skills to bring to the business.
“The historic racing will continue,” states the company’s general manager, Alex Marshall. “We just want to be a much bigger player in the field… but it is the other side of the business, the engineering services side, that is what will set us apart,” he concludes.
© Mathwall 2021