FIA:n sivuilta (suoraa linkkiä ei ole)
TOLERANCES
--------------------------------------------------------------------------------
Many of the Formula One Technical Regulations depend upon dimensions which in engineering terms can never be absolute. They require tolerances to enable parts to be manufactured, assembled and for inspection measurements to be made. It is well established in engineering that imposing tighter tolerances than are necessary for the correct functioning of the part, simply increases the cost of manufacture and inspection, without other benefits.
Tolerances in the Technical Regulations are either stated in the form:
Xmm ± Ymm ;
or they are implied with phrases such as:
“…must not exceed Xmm”, “…not more than Xmm”, or “…maximum Xmm", which imply a tolerance of: + Zero, - Free ;
while phrases such as:
“…must be at least Xmm”, “…must be no less than Xmm”, and “…minimum Xmm” imply tolerances of: + Free, - Zero.
Certain phrases, such as “…visible from underneath..”, “…visible through…”, and “…vertically above…” (as used in Art.3.12), all rely on an engineering interpretation that involves perpendicular projection from a plane and appear to imply a zero tolerance. However, a tolerance is required to enable components to be manufactured, assembled and inspected, all of which are imprecise processes.
Until the Malaysian GP, the FIA had applied engineering sense in carrying out measurements, judging each case on the complexity and magnitude of the measurement, the rigidity of the components and the possibility of damage to the component. As a result it has not needed to use extensions to the FIA measurement rig in the locations in which the barge-boards lie. The judgement of the FIA scrutineers has, until now, been accepted, and any obviously unintentional discrepancies that extend beyond reasonable tolerances have been noted by the FIA, or indeed other teams, being pointed out to the offending team's engineers so that they can be corrected prior to the car running in an event.
This sensible and accepted approach worked until the race in Malaysia. Because initial measurements, subsequently deemed inaccurate, indicated a 10mm discrepancy in the shadow plates, this was considered unacceptable and led to the Stewards' decision to exclude the two Ferraris. The accepted method of dealing with such oversights having been bypassed, it was inevitable that Ferrari would examine the whole subject of measurement accuracy and permitted tolerances extremely closely, ceasing to rely on accepted interpretations.
Art.3.12.6 states:
“To help overcome any possible manufacturing problems, a tolerance of ±5mm is permissible across these surfaces” (emphasis added).
While it has been accepted that this tolerance applies to the flatness of the underside of the car and has done so since flat bottoms were made mandatory in 1983, Ferrari asserted in the FIA Appeal Court hearing, that it also applied to the dimensions across the surfaces, unless otherwise bound by an implied tolerance e.g. “…must not exceed Xmm”.
However good the manufacturing process, it is not possible to make a perfectly flat vertical panel, nor fit it to a car such that it is perfectly vertical. Nor is it possible to measure it perfectly, however precise the measuring equipment or competent the operator. There must be a tolerance to make the manufacture, assembly and measurement of it practical. When the ±5mm tolerance is applied to a “vertical” dimension, which can be at least 600mm, it equates to less than ±0.5°. This is not an unreasonable tolerance, considering composites are still partially hand made. The components are assembled to the car, as are the surfaces to which the “vertical” is referenced for measurement purposes.
To claim that Art.3.12.6 does not apply to the vertical panels is to claim that they have a zero tolerance, as no other tolerance is defined for them in the Technical Regulations. If this were so, then all cars with straight, vertically mounted barge-boards would be illegal, unless fitted with shadow plates to accommodate any manufacturing errors.
Geneva 28 October 1999
Tällä tarkistuksella saatiin toleranssin sisällä oleva poikkeama (itse toleranssiahan ei tässä ollut sallittu!!!).
