Search Interpretations

Interpretation ID: nht90-4.30

TYPE: Interpretation-NHTSA

DATE: October 2, 1990

FROM: Paul Jackson Rice -- Chief Counsel, NHTSA

TO: S. Kadoya -- Manager, Safety and Technology, Mazda Research and Development of North America, Inc.

TITLE: None

ATTACHMT: Attached to letter dated 11-7-89 from S. Kadoya to S.P. Wood

TEXT:

This responds to your request for interpretations of several safety standards and the Bumper Standard, in connection with a planned "active" suspension system. I regret the delay in responding to your letter. Your questions are responded to below.

By way of background information, the National Highway Traffic Safety Administration (NHTSA) does not provide approvals of motor vehicles or motor vehicle equipment. Under the statutes administered by this agency, it is the responsibility of the manufac turer to ensure that its vehicles and equipment comply with applicable standards. The following represents our opinion based on the facts provided in your letter.

According to your letter, Mazda is concerned about the protocol of compliance testing of vehicles equipped with an active suspension system. This concern arises because many standards do not specify a suspension height that is to be used during complianc e testing. As you noted, this has not been a concern for conventional suspension systems, since they do not provide for variable height.

Mazda's planned active suspension system would be actuated by hydraulic fluid or compressed air, with control pressure being developed by a hydraulic pump or air compressor driven off the engine. Consequently, the active suspension system would be opera tional only when the vehicle's engine is operating. At vehicle speeds in excess of "z" mph, where z is greater than 35 mph, the suspension height would be lowered by "x" mm from the nominal or design position for vehicle operation. If the engine/vehicl e were not used for several consecutive days, pressure in the control system would fall such that the supension height may be lowered from the nominal or design position for vehicle operation by "y" mm, where y is greater than x. The suspension height w ould return to the nominal or design position for vehicle operation after such an extended period of inoperation almost immediately after starting the vehicle's engine.

Before discussing your specific questions, I would like to discuss more generally the issue of how compliance is determined in situations where a standard does not specify a particular test condition. In issuing Federal motor vehicle safety standards, N HTSA attempts to specify all relevant test conditions. The agency does this as part of ensuring that its standards are objective and practicable. As a practical matter, however, it is not possible to specify every conceivable test condition. This is p articularly true for ones which may only be relevant to

as-yet-undeveloped technologies.

In cases where a standard does not specify a particular test condition, we believe there are several relevant factors to consider in interpreting the standard. First, in the absence of specification of a particular test condition, we believe there is a presumption that the requirements need to be met regardless of such test condition, since the standard does not include any language which specifically limits applicability of its requirements to such test condition. For example, where a standard does n ot specify suspension height, its requirements may need to be met at all heights to which the suspension can be adjusted. Before reaching such a conclusion, however, we also consider the language of the standard as a whole and its purposes. Even if a s tandard is silent as to a particular test condition, the language of the standard or its purposes may indicate limitations on such test condition. Finally, in situations where a limitation on a particular test condition may appear to be appropriate, we also must consider whether the limitation is sufficiently clear, both with respect to justification and specificity, to be appropriate for interpretation. For example, in a situation where it may appear to be reasonable to limit a particular test condit ion but it is not obvious what particular limitation should be adopted, it would be inappropriate to select a particular limitation by interpretation. Instead, such a decision should be reached in rulemaking.

I will now address the specific questions asked in your letter.

Standard No. 108, Lamps, Reflective Devices, and Associated Equipment

In asking about Standard No. 108, you stated the following:

NHTSA has previously issued an interpretation of the requirements of FMVSS No. 108; at the request of a confidential applicant and dated February 12, 1985, with respect to active suspension equipped vehicles. This interpretation stated that the requirem ents of FMVSS 108 must be meet (sic),"...at any time in which..." lamps, reflective devices, and associated equipment are to be,"...operated for its intended purpose." Consequently, headlamps, tailamps, stoplamps, the license plate lamp, and side marker lamps, must comply with the location requirements of FMVSS No. 108 whenever the vehicle's ignition is in the "on" postiion. Conversely, reflex reflectors, and turn signal lamps that also function as hazard warning signal flashers must comply with the lo cation requirements when the vehicle's ignition is in either the "on" or "off" position. However, it is Mazda's interpretation that hazard warning flashers are not intended to be operational for a period of days, but rather for a period of hours, at max imum, only.

You then asked two questions, (1) whether Mazda's understanding of the subject NHTSA interpretation is accurate, and (2) whether Mazda's interpretation of the maximum intended operating duration of hazard warning signal flashers is correct.

I note that the February 1985 interpretation was written in the context of a vehicle with a variable height system actuated by hydraulic fluid. In that particular system, the hydraulic pressure relaxed over a period of

about three hours after the ignition was turned off, with the result that the vehicle assumed a lower height than it would have during driving. NHTSA stated the following:

We believe that the minimum height requirement should be met for any lamp at any time in which it is operated for its intended purpose. Since vehicles at rest do not require use of headlamps, the minimum height requirement would be measured at the point after the ignition is on and when the car begins to travel (your letter implies that the time lag between turning on the ignition and restoration of a complying mounting height is a matter of seconds). On the other hand, the hazard warning signal lamps are frequently operated when the vehicle is stopped, and therefore the minimum mounting height of turn signal lamps, through which they operate, must be met with the ignition off, even if the system requires three hours to deplete itself and lower the v ehicle to its minimum height.

With respect to your question of whether Mazda's understanding of the interpretation is correct, I would like to note two points. First, while you state that "the requirements of FMVSS 108" must be met at any time in which lamps, reflective devices, and associated equipment are to be operated for their intended purpose, our interpretation was limited to standard's minimum height requirement. While we are prepared, if asked, to address other requirements, our interpretations should be understood to be limited to their specific facts and conclusions. Second, while our interpretation only addressed headlamps and hazard warning signal lamps, you applied the interpretation for headlamps to taillamps, stoplamps, the license plate lamp, and side marker lam ps, and the interpretation for hazard warning signal lamps to reflex reflectors. We concur with this application, with respect to Standard No. 108's minimum height requirement.

We do not agree with Mazda's suggested interpretation of the maximum intended operating duration of hazard warning signal flashers. You would apparently like us to conclude that Standard No. 108's minimum height requirement for hazard warning signal fla shers does not apply after a vehicle's ignition has been turned off for a matter of days.

In addressing how Standard No. 108 applies in the absence of a specification for vehicle height, our February 1985 interpretation differentiates between situations where the vehicle is operating and where it is not. Looking at the purpose of the require ments in question, we believe it is obvious that the minimum height requirement for headlamps is only relevant in situations where the vehicle is operating, while the minimum height for hazard warning signal lamps is also relevant to situations where the vehicle is stopped and the ignition turned off. However, we believe that any determination that Standard No. 108's minimum height requirement for hazard warning signal flashers should not apply after a specified number of hours after the ignition has be en turned off is one that would need to be addressed in rulemaking.

It is therefore my opinion that the minimum mounting height of hazard warning signal lamps must be met at all heights with the ignition off, even if the system requires days to deplete itself and lower the vehicle

to its minimum height. If you believe that a time limitation should be placed on this requirement, I note that you can submit a petition for rulemaking requesting such a change.

Standard No. 111, Rearview Mirrors

You requested an interpretation of section S5.1.1 of Standard No. 111, which generally requires a passenger car's rearview mirror to "provide a field of view with an included horizontal angle measured from the projected eye point of at least 20 degrees, and sufficient angle to provide a view of level road surface extending to the horizon beginning at a point not greater than 200 feet to the rear of the vehicle...." You noted that since the specified procedures for determining the location of the driver 's eye reference points are made referenced to points with the vehicle's cabin, your active suspension system would not affect these measurements. However, different vehicle heights would be relevant to whether there is a view of level road surface exte nding to the horizon beginning at a point not greater than 200 feet to the rear of the vehicle. You stated that, based on "intended purpose," Mazda's interpretation of Standard No. 111 is that the requirements of this standard are to be met when the vehi cle's ignition is in the "on" position as rearview mirrors are not intended to be used when the vehicle's engine is not operating.

You then asked two questions, (1) whether Mazda's interpretation of the requirements of FMVSS No. 111 with respect to the state of the vehicle's ignition switch is correct, and (2) for the purposes of compliance testing to the requirements of FMVSS No. 1 11, what means of maintaining the intended suspension height for a given speed and operating condition would be satisfactory to NHTSA.

We agree that the field of view requirement specified in S5.1.1 for rearview mirrors need not be met for vehicle heights that only occur when the engine is not on, since the requirement is only relevant in situations where the vehicle is operating. Howe ver, the requirement would need to be at all vehicle heights that occur during vehicle operation, under the loading conditions specified in S5.1.1.

With respect to the issue of how suspension height should be maintained for purposes of compliance testing, you note early in your letter that, for reasons of practicality and safety, a vehicle's engine is not actually operational during compliance testi ng. However, since the active suspension system derives its power from the vehicle's engine, the system's ability to maintain and regulate suspension height is only possible during engine operation. You therefore indicated that Mazda is seeking guideli nes (for several standards) by which Mazda may be able to establish a means to maintain the intended suspension height for compliance testing purposes in the absence of engine operation.

We are not able, in an interpretation, to specify a particular means for maintaining suspension height for compliance testing in the absence of engine operation. However, the basic principle that should be followed in selecting a means for maintaining s uspension height is that it should not result in different test results than would occur if testing could be conducted with suspension height being maintained by engine operation,

i.e., what would happen in the real world. This should be relatively straightforward for section S5.1.1 of Standard No. 111, since the test is static. For a crash test, it is important that a vehicle not be altered in any way that would change the vehi cle's crash performance relevant to the aspect of performance being tested.

Standard No. 204, Steering Control Rearward Displacement

In asking about Standard No. 204, you stated the following:

Section S4 of this standard specifies the compliance parameter for this standard. Section S5 specifies the testing conditions to determine compliance with this standard. Section S5.1 specifies that the vehicle be loaded to its unloaded vehicle weight. Section S5.5 specifies that the vehicles fuel tank be filled with Stoddard solvent to any capacity between 90 and 95 percent of the total capacity of the tank. Mazda's interpretation of the requirements of this standard is that they are to be met when the vehicle's ignition switch is in the "on" position only. Furthermore, Mazda interprets the vehicles suspension height pursuant to S5.1 and S5.5 to be the intended suspension height for the vehicle given the conditions of S4, i.e., 30 mph vehicle speed and steered wheels are positioned straight ahead.

You then asked whether Mazda's interpretation of the requirements of FMVSS No. 204 are correct. As discussed below, we agree that Standard No. 204's requirements need to be met only at the suspension height that occurs at a 30 mph vehicle speed and with steered wheels positioned straight ahead.

Standard No. 204 specifies requirements limiting the rearward displacement of the steering control into the passenger compartment to reduce the likelihood of chest, neck, or head injury. These requirements must be met in a 30 mile per hour perpendicular impact into a fixed collision barrier. While the standard specifies a number of test conditions, it does not specify suspension height.

Looking at the Standard No. 204 as a whole, we believe it is clear that NHTSA explicitly decided to limit the standard's evaluation of steering control rearward displacement to how vehicles perform in 30 mph perpendicular impacts, even though the require ments have relevance at lower and higher speeds. Therefore, we agree that the standard's requirements need to be met only at suspension heights that occur at a 30 mph vehicle speed and with steered wheels positioned straight ahead.

With respect to Mazda's question concerning means of maintaining intended suspension height for compliance testing, please see our discussion provided with respect to Standard No. 111.

Standard No. 208, Occupant Crash Protection

In asking about Standard No. 208, you stated the following:

Section S8.1.1(d), "Vehicle test attitude," specifies the procedure for determining the vehicle test attitude that is to be used for testing. Specifically, this section requires that the vehicle's pretest attitude,

"...shall be equal to either the as delivered or fully loaded attitude or between the as delivered and fully loaded attitude." The as delivered attitude is defined by S8.1.1(d) as being, "...the distance between a level surface and a standard reference point on the test vehicle's body, directly above each wheel opening, when the vehicle is in its "as delivered" condition. The "as delivered" condition is the vehicle as received at the test site..." Because it is highly likely that the test vehicle wil l not have been operated for a period of days prior to arriving at the test site, the suspension height may have fallen by "y" mm. The fully loaded attitude is defined as the attitude of the vehicle when loaded in accordance with S8.1.1(a) or (b) and a determination of the height of the suspension at the fully loaded condition is made from the same level surface, using the same standard reference points, as were used to determine the "as delivered" condition. The definition of the "as delivered" condi tion is quite clear. However, Mazda interprets the "fully loaded condition" of the vehicle to be the condition when the vehicle's ignition is "on." In this instance it is likely that the height of the standard reference points on the vehicles body when in the "fully loaded condition" relative to the level surface will be greater than for the "as delivered" condition. Conversely, conventional vehicle suspension systems will likely have an "as delivered" height greater than the "full loaded" height. H owever, this fact is of no importance as S8.1.1(d) states that the pretest vehicle attitude may be, "...between the as delivered and the fully loaded attitude." With respect to the injury criteria specified by section S6 of this standard, Mazda's interp retation is that these criteria must be met with the vehicle's ignition in the "on" position only.

You then asked three questions, (1) whether Mazda's interpretation of the definition of the "fully loaded condition" is correct with respect to the condition of the ignition switch, (2) whether Mazda's interpretation of the irrelevance of the relative re lationship between the "as delivered" and "fully loaded" conditions is correct, and (3) whether Mazda's interpretation of the meaning of "between the as delivered and the fully loaded attitude" is correct.

In addressing your questions, I will begin by noting that Standard No. 208 specifies occupant protection requirements which must be met in specified crash tests at any impact speed up to and including 30 mph. While the standard specifies a number of tes t conditions, it does not specify suspension height. However, the standard does specify vehicle attitude, which is closely related to suspension height. In addressing how Standard No. 208 applies in the absence of a specification for vehicle height, th e relationship between the standard's attitude specification and vehicle height must be considered.

Section S8.1.1(d) specifies the attitude of the vehicle during testing, i.e., the angle of the vehicle relative to the ground. This test condition ensures that the vehicle is not overly tilted toward the front or back, or to one side. The section accom plishes this purpose by specifying that, during a compliance test, the height of the vehicle at each wheel is within a specified range. This range, which may be somewhat different for each wheel, is determined by looking at the vehicle in the "as delive red" condition and the "fully loaded" condition. A vehicle must meet the requirements of Standard No. 208 when its height at each wheel is

anywhere within the specified ranges.

On first glance, one might read section S8.1.1(d) to create a height requirement, since ranges of height are determined under that section (at each wheel). This would be incorrect, except in a very narrow sense, since Standard No. 208 does not specify, for vehicles with variable height suspension systems, what suspension height should be used in the two conditions ("as delivered" and "fully loaded") where the specified ranges of height are determined under section S8.1.1(d).

Looking at the Standard No. 208 as a whole, we believe it is clear that NHTSA explicitly decided to limit the standard's evaluation of occupant crash protection in frontal impacts to how vehicles perform in impacts of 30 mph or less, even though the requ irements also have relevance at higher speeds. It is our interpretation that the frontal crash test requirements need to be met at all suspension heights that can occur at speeds of 30 mph or less, with the vehicle operational. It is also our interpret ation that the crash test requirements need to be met only at suspension heights that can occur at the speed used in the crash test.

A remaining issue is how section S8.1.1(d) applies for vehicles with variable height suspension systems. As discussed below, vehicle attitude should be determined under this section using the actual suspension setting (or equivalent, if the setting is a utomatic) to be used in a crash test.

For purposes of illustration, I will assume a vehicle with two very different suspension height settings. It would not be appropriate to conclude that the ranges of height determined under section S8.1.1(d) should simultaneously cover both suspension he ights. Such ranges would be very large, and would not ensure that the vehicle is not overly tilted toward the front or back, or to one side. Moreover, such ranges would not be relevant to the real world, with respect to vehicle attitude.

Traditional vehicles can be viewed as having a single suspension "setting." This single suspension condition is used in determining vehicle attitude under section S5.8.8.1. The ranges of height result from the differences in loading under the "as deliv ered" and "fully loaded" conditions.

A single suspension "setting" (or equivalent, if the setting is automatic) should similarly be used in determining vehicle attitude for vehicles with variable height suspension systems. The "setting" should be the one to be used in a crash test.

With respect to Mazda's question concerning means of maintaining intended suspension height for compliance testing, please see our discussion provided with respect to Standard No. 111.

You also asked for an interpretation of section S8.2.7 of Standard No. 208. That section specifies additional conditions to be used for lateral moving barrier crash testing. Section S8.2.7(a) states that the vehicle is at rest in its "normal attitude." You stated that Mazda interprets the meaning of "normal attitude" to be that vehicle attitude which is intended

when the vehicle's ignition is in the "on" condition, with the vehicle loaded pursuant to S8.1.1(a) or (b), and while the vehicle is at rest. Standard No. 208 provides manufacturers the option of either equipping their vehicles with safety belts or meeti ng certain alternative requirements, including lateral moving barrier crash test requirements. These requirements are relevant at all vehicle heights that can occur during vehicle operation, regardless of speed. Moreover, NHTSA has not decided to limit the standard's evaluation of this aspect of safety performance to how vehicles perform at certain limited speeds. It is our interpretation that the lateral moving barrier crash test requirements, if applicable, must be met at all suspension heights that can occur with the vehicle operational. "Normal attitude" is the attitude determined under section S8.1.1(d). As discussed above, attitude for vehicles equipped with variable height suspension systems is determined under section S8.1.1(d) using the ac tual suspension setting (or equivalent, if the setting is automatic) to be used in a crash test.

Standards No. 212, Windshield Mounting; No. 219, Windshield Zone Intrusion; No. 301, Fuel System Integrity

In asking about Standards No. 212, No. 219, and No. 301, you noted that NHTSA has previously issued an interpretation to Mazda about how these standards apply to adjustable height suspension systems. In a letter dated August 10, 1982, the agency address ed a vehicle equipped with a suspension system having two height positions, one for normal highway driving and another for off-road driving, which could be selected by the driver. NHTSA stated the following:

(Safety Standards No. 212, No. 219, and No. 301) do not specify a height adjustment because almost all vehicles have a single, set adjustment. . . After careful consideration, it is the agency's position that such a vehicle capable of variable height adj ustment would have to comply with the vehicle adjusted to any position that is possible. This is true because the vehicle could be driven on the highway, for example, even if it were adjusted to the off-road position. Consequently, it is important that the vehicle comply with the standards in all positions.

You noted that while suspension height could be adjusted by the driver for the system discussed in the agency's previous interpretation, the active suspension system you are currently considering would use an on-board electronic controller to select susp ension height, and suspension height would not be adjustable by the driver. Consequently, according to your letter, only one unique set of suspension height parameters is possible for a given vehicle speed and loading condition as is the case with conve ntional suspension systems. You stated that because it is possible to determine exactly what the intended suspension height should be for a given situation, it is Mazda's opinion that the test vehicle should be tested at the intended suspension height g iven the specified speed and loading conditions. You also stated that, using an "intended purpose" argument, Mazda concludes that the requirements of the three standards are to be met only when the vehicle's ignition is "on." You then asked whether thes e suggested interpretations are correct.

Standard No. 212 specifies windshield retention requirements that must be

met in a specified frontal crash test at any impact speed up to and including 30 mph. Similarly, Standard No. 219 specifies windshield zone intrusion requirements that must be met in a specified frontal crash test at any impact speed up to and including 30 mph. Standard No. 301 specifies fuel system integrity requirements for several specified crash tests. These include a frontal crash test similar to those in Standards No. 212 and No. 219. Requirements for this test must be met at any impact speed up to and including 30 mph. Other tests include a rear moving barrier crash test, a lateral moving barrier crash test, and a static rollover test.

We agree that the requirements of Standards No. 212, No. 219, and No. 301 need not be met for vehicle heights that only occur when the engine is not on, since the requirements are only relevant in situations where the vehicle is operating. Looking at th e three standards as a whole, we believe it is clear that, for the frontal tests specified by the three standards, NHTSA decided to limit the standards' evaluation of safety performance to how vehicles perform in impacts of 30 mph or less, even though th e requirements have relevance at higher speeds. It is our interpretation that the frontal crash test requirements specified by these standards need to be met at all suspension heights that can occur at speeds of 30 mph or less, with the vehicle operatio nal. It is also our interpretation that the crash test requirements need to be met only at suspension heights that can occur at the speed used in the crash test.

We reach a different conclusion for Standard No. 301's other crash test requirements. These requirements are relevant at all vehicle speeds and suspension heights. Moreover, NHTSA has not decided to limit the standard's evaluation of these aspects of s afety performance to how vehicles perform at certain limited speeds. It is our interpretation that these crash test requirements must be met at all suspension heights that can occur with the vehicle operational.

Part 581 Bumper Standard

In asking about the Part 581 Bumper Standard, you noted that NHTSA has previously issued several interpretations of how the standard applies to vehicles with adjustable height suspension systems. In a letter to Subaru dated May 6, 1986, NHTSA stated the following:

Given the absence of a specific test condition concerning suspension height, it is our interpretation that a vehicle must be capable of meeting the standard's damage criteria at any height position to which the suspension can be adjusted. There is no la nguage in the test requirements of the standard limiting their applicability to "the manufacturer's nominal design highway adjusted height position."

This interpretation is consistent with the purpose of the Bumper Standard, set forth in section 581.2, to reduce physical damage to the front and rear ends of a passenger motor vehicle from low speed collisions. If a vehicle's suspension could be adjust ed so that its bumper height resulted in bumper mismatch with other vehicles in the event of low speed collisions, the reduction in physical damage attributable to the Bumper Standard would be defeated in whole or part.

In another letter, dated February 12, 1985, NHTSA stated that a vehicle is "required to meet the pendulum test of Part 581 in any vehicle use scenario in which the vehicle operates, and the barrier test of Part 581 when the engine is idling."

You suggested, for the barrier test, that the agency's May 1986 interpretation may be inappropriate for your active suspension system, since your system provides for only one suspension height when the engine is idling. You also suggested, for the pendu lum test, that these interpretations seem to be in conflict with the Bumper Standard's stated purpose to reduce physical damage to motor vehicles in low speed collisions. We assume that you are referring to the fact that your suspension system has heigh ts that occur only at speeds greater than 35 mph. You then requested that NHTSA provide an interpretation of Part 581 with respect to your system.

In addressing how Part 581 applies to vehicles equipped with an active suspension system, I will address separately the standard's barrier and pendulum tests. For the barrier test, a vehicle must meet specified damage criteria after an impact into a fix ed barrier that is perpendicular to the line of travel of the vehicle, at 2.5 mph. Section 581.6 sets forth conditions applicable to bumper testing. Under section 581.6(c), at the onset of a barrier impact, the vehicle's engine is operating at idling s peed.

Looking at the Bumper Standard as a whole, we believe it is clear that NHTSA decided to limit the barrier test's evaluation of bumper performance to how vehicles perform in 2.5 mph frontal impacts, even though the requirements have relevance at lower and higher speeds and when the vehicle is nonoperational. It is our interpretation that the barrier test requirements specified by this standard need to be met at all suspension heights that can occur at 2.5 mph.

We reach a different conclusion for the pendulum test, which serves the purpose of creating a bumper height requirement. This requirement is relevant at all vehicle speeds and suspension heights, and when the vehicle is nonoperational. I note that whil e Mazda is correct that the Bumper Standard's stated purpose is to reduce physical damage to motor vehicles in low speed collisions, NHTSA has justified the bumper height requirement on safety concerns related to "higher speed collisions." In proposing Standard No. 215, the predecessor of Part 581, the agency stated:

. . . in higher speed collisions the tendency of a bumper to override another or to ride under or over a guardrail creates hazards for vehicle occupants. Vehicles with interlocking bumpers block traffic and expose their occupants to considerable danger, particularly if they attempt to get out to unlock bumpers. By overriding or underriding a guardrail, a bumper may strike a supporting post, or similar fixed object, with serious consequences for the vehicle and its occupants. 35 FR 17999, November 24, 1970.

The relevance of the bumper height requirement to nonoperational

situations is also clear, e.g., to help protect parked cars.

Moreover, NHTSA has not decided to limit the bumper height requirement to how vehicles perform at certain limited speeds. It is our interpretation that the pendulum test requirements must be met at all suspension heights that can occur, regardless of ve hicle speed or whether the ignition is turned on.

This interpretation is consistent with an October 18, 1978 letter to Nissan, in which NHTSA addressed how the pendulum test applies to vehicles equipped with height control systems, including automatic height control systems. Among other things, the age ncy stated the following:

. . . There is no language in the pendulum test requirements of the standard which would limit their applicability to only the ignition-on or ignition-off situation or to the recommended driving position for normal roadways. The vehicle must be capable of meeting the pendulum test requirements at all stable bumper heights possible at unloaded vehicle weight.

Thus, in the situations described in Question 1 and 2 of your letter, in which an automatic height control system is employed, the vehicle must comply with the pendulum test requirements in both the ignition-on and ignition-off positions . . . .

I note that one of our past letters, a December 24, 1984 letter addressed to Porsche, appears to suggest that the pendulum test must be met in any setting in which the system operates "when the engine is idling." This might be read to suggest that the p endulum test need not be met when the vehicle is nonoperational. However, this interpretation cited section 581.6(c) in concluding that the engine is idling during Part 581 testing. Section 581.6(c) applies only to the barrier test and not the pendulum test. We therefore consider this interpretation to be incorrect to the extent that it is inconsistent with the analysis presented above.