Our 1995 Comments to CPSC
on their draft standard
Summary: Mostly for historic interest, these are the comments BHSI sent to the Consumer Product Safety Commission on the first draft of their bicycle helmet standard.
Bicycle Helmet Safety Institute
4611 Seventh Street South
Arlington, VA 22204
March 1, 1995
Office of the Secretary
Consumer Product Safety Commission
Washington, DC 20207
Re: Proposed Rule: Safety Standard for Bicycle Helmets
Dear Sir or Madam:
We are pleased to have this opportunity to comment on the
proposed Safety Standard for Bicycle Helmets published in the
Federal Register of August 15, 1994.
We are impressed with the draft standard, including its specifications
for testing, its requirements for performance and the quality
of its drafting. The drafters have sought out the best provisions
of several standards, and have improved on all of them in some
respect. The document provides a good standard for today and a
basis for future improvement as technology and industry capabilities
We would particularly like to commend CPSC on the coverage
or test area provisions of this standard. For adults, the test
line in the rear is slightly lower than ASTM, but higher than
Snell's 1995 standard. A recent Technisearch study showed that
many rear impacts fall below the test lines of current standards.
This provision will not result in major redesign problems for
the industry. Although some current helmets must be modified to
increase their coverage, most helmet models have a half-life of
only a year or two, and retooling to meet this standard before
it becomes final should be commercially painless. The helmets
will in our estimation be just as acceptable to consumers. We
urge you to keep this provision intact. We also urge you to remove
the section providing coding of label information on dates and
names of foreign manufacturers. This information is needed by
consumers and permitting it to be coded is inconsistent with your
role as a consumer agency.
We are appending detailed comments to this letter.
Thank you for your efforts to provide protection for consumers.
BHSI Comments (cont.)
As we calculate the test line defining the limits of the area
to be tested, the CPSC proposed rule would compare to other standards
as follows for helmets tested on the medium size ISO J headform:
Standard Distance in mm of test line
above the Basic Plane
ANSI 85 35
ASTM (Current) 110 85
ASTM (Proposed) 85 65
CPSC draft 85 50
Snell B 1990 75 75
Snell B 1995 53 33
Snell N 1994 53 13
The chart shows that the proposed CPSC test line will be well
above that for the 1984 ASTM standard and Snell's 1994 multipurpose
standard. It will be above that specified in Snell's 1995 bicycle
helmet standard as well. But it will be below the test lines for
the ASTM standard and for Snell's current 1994 bicycle helmet
Virtually every helmet on the market today has a sticker in
it stating that the manufacturer certifies that the helmet passes
the ANSI standard. That standard tests down to a level only 35
mm above the basic plane for this size helmet. CPSC's specification
should therefore not be a major design problem even though the
CPSC test drops are at higher energy levels than the ANSI standard.
It is important for cyclists to be seen by other road users.
For that reason all helmets should have bright outside colors
for daytime conspicuity and reflective surfaces for nighttime.
There is no excuse for black, purple or camouflage helmet covers,
but all have been used by manufacturers, and 1995 helmet lines
are dominated by dark colors. New Zealand's bicycle helmet standard
actually requires a brightly colored outer cover.
Reflective trim adds only about 24 cents to the manufacturer's
cost per helmet, translating to perhaps 50 cents at the consumer
level. As helmet prices drop, this appears more and more reasonable.
The need for regulation here is evident. Consumers find it difficult
to judge which trim is reflective and which is not in a normally
lit store. We have received helmets in boxes which proclaimed
reflective trim, but which did not in fact have reflective
trim. Further, among the 1995 lines are helmets using a silver
tape which mimics silver reflective trim but is not reflective.
Consumers can be misled into thinking that they are more visible
than they actually are after dark.
We urge you to consider adding requirements to the CPSC standard
for both daytime and nighttime conspicuity.
Point Loading Test
There is no test in the current draft which requires that
the helmet should prevent localized loads or point loads from
exceeding a given level. Although this is not a factor when testing
with a magnesium headform, which of course bridges any local differences
in foam density, a high point load could potentially cause a human
skull to deflect and do unnecessary damage. At least two manufacturers
are now producing helmets with extra-hard foam around the vents in order to
open up larger vents. Although injury data will never be available at a
level of detail which could distinguish these few helmets' performance
in skull injuries, we believe that harder, more dense foam should
generally be discouraged. Australia's bicycle helmet standard
has a test for point loading. We urge the Commission to examine
that test and consider adding a point load test to the CPSC standard.
Hair Oil Test
It is not reasonable to sell a helmet which can be damaged
by oil or other preparations which consumers normally use on their
hair. The provisions of section 1203.5(d) provide a well worded
statement of intent, but an actual test would be better. Japan
has a provision in its standard for a hair oil test. We recommend
that the Commission explore this question further. We also recommend
removing the reference in section 1203.6(a)(4) where hair tonic
is listed among the examples of substances a manufacturer might
warn consumers about as a possible damager of helmets. If hair
tonic could damage the helmet, it is clearly not in compliance
with section 1203.5(d).
As written, the draft provides that retention systems are
tested prior to impact tests. This is not a realistic schedule,
since the worst bicycle crashes involve one impact with a car
and a second impact with the street. If the helmet disintegrates
in the first crash and the retention system fails, the rider will
be bare-headed for the second impact.
We recommend that the Commission consider at least one impact
test per sample prior to testing the retention system. This will
force manufacturers to reinforce their helmets to insure that
they do not fly apart in a first impact.
Although the draft uses the standard formula of 300 g peak
acceleration for the pass/fail impact testing, there is considerable
support among the testing and injury prevention community that
300 g's in a test lab is not an adequate threshold to prevent
some real-world injuries. Testing by Consumer Reports has shown
that at least some helmets on the market can achieve peak acceleration
readings of less than 200 g's with the crash energy of the test
set at the CPSC-specified level. We think the Commission should
consider specifying a lower threshold on the order of 250 g's
for at least the flat anvil tests. The difficult part is find
a scientific justification for 250 vs. some other number.
Visors, Mirrors and other Appendages
The draft has strong and appropriate language for testing
of visors and mirrors, providing that all helmets must pass the
tests with and without any accessories. But there may be a need
to specify other tests for such parameters as easy peel-off in
the event of a crash, and shatter resistance of visors. The Australian
standard has provisions for some of these tests, and the Snell
motorcycle standard has a shatter test for visors.
The provisions of 1203.34(6)(c) permitting coding of information
on helmet labels are shocking in a document produced by a consumer
protection agency. The draft permits the coding of the name of
a foreign manufacturer, but not the name of a U.S. manufacturer.
This is apparently to hide from the consumer the true name of
the foreign manufacturer, and we can find no valid reason for
it. The product lot and date of manufacture can also be coded.
This denies the consumer important information about how old their
helmet is, even though manufacturers commonly recommend replacing
a helmet after five years, and the consumer will usually not be
able to remember when they bought the helmet. Coding this information
would also prevent a consumer from identifying a recalled helmet
by the date. Much information on recalls is passed verbally, and
a rider or bicycle shop owner who sees someone wearing a potentially
recalled helmet will not be able to remember a coded lot number
to identify which lots were recalled. They are more likely to
remember that all helmets of that model manufactured in 1996 have
been recalled. There is no excuse for this provision, we are dismayed
to find it in the draft, and we urge that the Commission delete
Apart from the comments above we believe that this draft provides
the basis for a fine standard which will meet consumers' expectations
that the U.S. Government will not permit sale of a defective or
inadequate safety appliance.