Stack data-loger


Racing drivers are employed to drive, not communicate. Thus a race engineer’s lot was not always a happy one – until, that is, stack developed a device that could communicate precisely what the car was doing in ever corner. By Keith Howard

There is no mega-theory of race-car design, no means whereby a supercomputer can pen a winning car from the ground up. So when the design team has done all it can theoretically, there is no alternative to bolting everything together and seeing how the car works in testing. All being well, you have a platform good enough in the essentials to respond to further development.

Even 20 years ago, that honing process was largely one of skilled inference from driver feedback. But in the second half of the 1980s, data-loggers arrived. The development of increasingly powerful microprocessors and more compact, more affordable solid-state memory made it practicable for the first time to measure in detail how a car was behaving, not only during development but even during races. Right at the forefront of this new technology was Stack Ltd, based in Bicester.

Stack had been formed in 1984 by Alan Rock and Trevor Tapping, initially to design and manufacture EPROM (electronically programmable read only memory) programmers. But being located in Oxfordshire, the heartland of Britain’s motorsport industry, it wasn’t long before the young company found itself working as a consultant to race-car constructors. Persuading electronics to work alongside the electrically ‘noisy’ ignition system was considered a black art at the time, and Stack was hired to help solve interference problems with electronic tachometers.

However, there were more fundamental problems with conventional tachometer technology, as Rock explains. “Because normal meter movements weren’t perfectly balanced, when a racing car went around a corner the g-forces would cause the needle to move. The tacho would read 500rpm too high or too low, depending on whether you were turning left or right, and in Formula One they were trying to rev the engines to within 200rpm of the pistons hitting the valves.

So this tachometer vagueness was a problem. “The breakthrough came in 1986 when we designed a tachometer based around a stepper motor for driving the needle, controlled by a microprocessor. There was absolutely no needle waver, and as we didn’t have to damp the needle movement, we had a very fast response. It was the first all-digital tachometer, even though it still had the analogue movement.

“It also became the first data-logging system that people could race in their cars. It only recorded engine rpm but that was a major step forward. You could download the information on to a laptop, if you had one — these were the earliest days of laptop computers — or you could print it out as a graph on a very long roll of paper. People would compare laps by holding two lengths of paper up to the sun, one behind the other, and overlay the data that way. “Our very first customers for the system were Rak and March in Formula 3000. It then spread like wildfire. Virtually all the Fl teams were soon using it, starting with Williams. Motorsport always works as a great grapevine, so we accumulated enquiries from all over the world.”

The System 600 data-logger, launched in late 1987, grew out of the tachometer experience, but it was much more than a means of storing a wider range of sensor readings.

“The 600 System was an integrated display and instrumentation system as well as a data-logger. The MRG data-logging system that had been used for Indycar testing was huge — about 12 by 14 inches by about 211 inches thick — and much too heavy to race. With the System 600, we had a solution that was lighter to begin with because it replaced conventional instruments, so weight-wise the logger came for free.” Thanks in significant part to the use of surface-mount electronics technology, the System 600’s controller/data-logger box measured only about 170 x 100 x 50mm. “It was a modular system so it communicated with dials and/or one or more LCD displays in any combination. In Fl they just had a tachometer and the LCD display. In Indy they obviously had a boost gauge as a second dial. Ray Mallock ran one of his Group C cars with a second LCD display showing tyre temperatures.

“The LCD display was two lines by 20 characters, and the driver had four switches that could be programmed with different functions — it was completely user-configurable. The system had eight input channels, each of which could be analogue or digital. The display could be configured to show any of that information, four parameters at a time. You could then press one of the buttons to show another four parameters, or some the same and some different.

“Primarily it monitored the engine — vehicle speed, engine rpm; oil pressure, oil temperature, water temperature and fuel pressure. If it was a turbo then you’d also have boost pressure and perhaps inlet air temperature. Top teams then started using a second logger, which is when they got into measuring other things like throttle position, steering angle and suspension travel. The unit sampled at a maximum of only 50Hz [50 times a second] — enough to see suspension travel but not damper velocities.”

Key features of System 600 were its diagnostic capability and programmable alarm system. “At that time the quality and reliability of electrical harnessing was one of the biggest problems in racing. Constructors used to do a lot of their own harnesses, whereas now it’s accepted that specialist companies do it. The System 600 would allow you to wire the sensor in backwards, short the sensor to ground or the battery or whatever, and it would just tell you what you’d done wrong. It would continuously check the operation of all sensors and report when it found a fault, even if it was intermittent, so you could catch faults before they became catastrophic.” And the system could be configured so that if any parameter moved outside a predetermined range, an alarm would warn the driver.

Almost 25 years later, modem data-loggers can have over 100 input channels compared to the System 600’s eight, sample at a maximum rate of a few kilohertz rather than 50Hz, and carry hundreds of megabytes of on-board memory compared to its paltry 128 kilobytes. But Stack’s original data-logger, with its modular design and serial data bus, was a milestone product, the trendsetter. In the 1988 season, Penske was first to use it in ChampCars, and Williams the first in F1. QED.