A full report on the 24-hours of Daytona is given elsewhere in this issue and, as M. J. T. indicates, it was a lack-lustre meeting emasculated by the C.S.I.’s premature plunge into the 3-litre sports/prototype limit. Only 14 of the 63 starters were in the prototype category and of these only one, the Howmet Turbine Experimental which was described briefly last month, was classed as a 3-litre car. Although the TX failed to go the distance (it ran for 1 hr. 22 min.), Daytona was the car’s first race and the team learned a great deal in a short time. And since the car is now a confirmed entry in the B.O.A.C. 500 next month it may be of interest to examine it further.
Two versions of the Howmet TX were taken to Daytona: the original “show” car and a newer “race” car that had a 2½-in. longer wheelbase and an engine producing about 20 b.h.p. more than the show car’s 325 b.h.p. on a standard day at sea level. However, the team suffered a severe setback on the Wednesday before the race when a bearing broke up in the race car’s engine after only 19 laps of practice. This engine was known as a “short-life engine,” which means that it had not run long enough for the manufacturers, the Continental Aviation and Engineering Corp., to be certain that there were no human errors in its assembly. After a certain number of trouble-free hours an engine graduates to the “long-life” category and one can be reasonably certain that it will run for as long as 1,000 hours without a tear-down.
The show car’s engine had graduated to the long-life category, but unfortunately it is much more complicated to change turbine engines than it is to swap reciprocating units. The Continental engine expands about 60-thousandths of an inch when it is running and must be very accurately aligned in the chassis to prevent undue loads on the drive train and shaft bearings. The Howmet car utilises a 3-point engine mounting system. At the left rear there is a tripod which prevents movement in any direction. A bipod at the front restricts vertical and lateral movement but permits fore-and-aft movement. A monopod at the right rear restricts vertical movement but permits both lateral and fore-and-aft expansion. A somewhat exotic device is needed to align the engine accurately on its mounts, and when the team lost its first engine it was left with two choices: switch engines without the device and run with no guarantee of correct alignment, or use the show car. The second alternative was obviously preferable but even so an entire day’s practice was lost transferring a number of parts from the race car to the show car. The most important of these was a set of massive new Kelsey-Hayes ventilated discs (11½ in. in diameter, 1¼ in. thick, and fitted with Girling calipers) which were designed for the race car after the show car’s brakes proved inadequate. Despite the extra weight of the new brakes the show car weighed in at 1,507 lb., 170 lb. lighter than the race car and only 77 lb. over the 3-litre minimum weight.
Thompson then qualified the show car seventh fastest at 2 min. 1.14 sec., after only 24 laps, and some of the team’s confidence returned. This compared with Gurney’s qualifying time of 1 min. 55.10 sec. in a 7-litre Mk. 2 Ford last year and Ickx’s qualifying record of 1 min. 54.91 sec. this year in one of the J. W. Automotive 4.7-litre GT40s. Both Thompson and Lowther were agreeably surprised by the car’s braking and stability. On braking they said they could go as much as one marker deeper into the corners than any other car in the field. As for stability, Thompson said he could enter the banking at full chat and never had to move the wheel. Even coming off the banks the car required only a ¼-in. of wheel movement to straighten it up. Thompson, Lowther and co-driver/project engineer Heppenstall commented frequently that it was the only car on the track that could run and pass where they wanted it to on the 31º banking—whether up against the wall, in the middle of the banks, or down on the apron.
The car’s main problem throughout the week was a sticking throttle or, more accurately, a sticking butterfly valve in the waste gate, which is the device the driver uses to divert the expanding gasses from the power turbine. Here I should clarify a point in last month’s column in which it was said that “the waste gate never diverts all of the gasses from the power turbine because it is this turbine which drives the gas-generating turbine.” The first part of that statement is correct, but the second is not. The Continental TS325-1 is a “free” turbine, with the compressor turbines (one axial stage, one centrifugal stage) and the gas-generator turbines (two axial stages) mounted on a common shaft. The power turbine (one axial stage) is mounted on a separate shaft, “free” of the other two. It is the gas-generator turbines, which absorb over 500 of the engine’s 900 b.h.p., that drive the compressor, the starter/generator and the other accessories. The waste gate is installed between the gas-generator and power turbines.
The sticking throttle problem is caused by the number of functions it has to perform and by the design of the waste gate. The first one-third of throttle movement controls the fuel supply to the gas-generator turbines and the final two-thirds operates the waste gate. Whenever possible the car is driven “on the waste gate”—i.e., within the final two-thirds of throttle movement, because then the fuel flow is not reduced and the gas-generator turbines continue to spin at maximum r.p.m. This is the technique used on the faster parts of the track and on the faster corners. On very slow corners the driver may be forced to back off into the first one-third of the throttle movement, thus reducing the fuel flow and slowing down the entire engine. This is undesirable because when the driver wants to accelerate out of the corner there would be a lag as the gas generator built up to maximum speed. To overcome this problem a different technique is used on slower corners. The driver backs off the throttle early, both opening the waste gate and reducing the fuel flow. When he is deep in the corner—and still braking—he gets back on the throttle but must be careful only to open it one-third of the way. The fuel supply increases again and the gas generator turbines build up to maximum speed while the car is cornering. This eliminates the lag, so when the driver wants to accelerate out of the corner and depresses the throttle the final two-thirds of its movement, thus closing the waste gate, full power is delivered instantly to the power turbine.
As mentioned earlier, however, the design of the waste gate itself has presented some problems. The valve in the waste gate is a simple butterfly disc, not unlike the damper in a stovepipe. A groove is cut in the periphery of the disc and on the first engine a cast-iron ring was inserted in the groove to act as a seal. This didn’t prove entirely satisfactory and on the second engine a Stellite sealing ring was used. When the second engine seized in practice its entire waste gate was transferred to the first engine because the Stellite seal was considered more satisfactory. Unfortunately the Stellite ring also showed a tendency to jam, which meant that although the driver had lifted his foot the waste gate remained closed and full power was still being delivered to the power turbine.
It was the result of just such a jamming incident early in practice that Lowther found he could go into the banking at full chat. He lifted his foot coining off the back stretch but the waste gate jammed closed, so he was forced to go into the banking at full speed. Nothing untoward happened and his lap times improved a couple of seconds a lap.
This same problem occurred during the race, only it happened at a much more critical part, of the course and resulted in the car’s retirement. Thompson had advanced the car to fourth and eventually third overall when he made his first pit stop for fuel (which the car was consuming at a prodigious rate) after 22 laps. Lowther took over and was still in the top 10 after another 12 laps when he came into the tight 136º corner that leads from the infield road course back on to the banking. He lifted his foot but the waste gate butterfly jammed closed, and he was forced into the corner at full power. The car slid across the track and side-swiped the wall, breaking the right front brake disc and damaging the right rear suspension.
Heppenstall is already hard at work on a number of possible solutions to this waste gate problem and is also considering a more radical alternative that may do away with the waste gate altogether, or at least use part of the gas flow to slow the car down (as is done in aircraft applications). The changes will be incorporated in the car for the Sebring 12-hour race this month.—D. G.