10mr98r (1)

8/9/2019 10mr98r (1)

1/37

f e d

e r a

l r e g i s t e r

11711

TuesdayMarch 10, 1998

Part II

Consumer ProductSafety Commission16 CFR Part 1203

Safety Standard for Bicycle Helmets;Final Rule

8/9/2019 10mr98r (1)

2/37

11712 Federal Register / Vol. 63, No. 46/ Tuesday, March 10, 1998/ Rules and Regulations

1

The standard was approved by the Commissionunanimously, by a vote of 3–0. Chairman AnneBrown, Commissioner Mary S. Gall, andCommissioner Thomas Moore each issued aseparate statement concerning the vote. Copies of these statements are available from the Office of theSecretary.

2Sacks, Jeffrey, J., MPH; Holmgreen, Patricia, MS;Smith, Suzanne M., MD; Sosin, Daniel M., MD.‘‘Bicycle-Associated Head Injuries and Deaths inthe United States from 1984 through 1988,’’ Journalof the American Medical Association 266(December 1991): 3016–3018. Sosin, Daniel M., MD,MPH; Sacks, Jeffrey J., MD, MPH; and Webb, KevinW., ‘‘Pediatric Head Injuries and Deaths fromBicycling in the United States,’’ Pediatrics 98(November 1996): 868–870.

CONSUMER PRODUCT SAFETYCOMMISSION

16 CFR Part 1203

Safety Standard for Bicycle Helmets

AGENCY: Consumer Product SafetyCommission.

ACTION: Final rule.SUMMARY: Pursuant to the Children’sBicycle Helmet Safety Act of 1994, theCommission is issuing a safety standardthat will require all bicycle helmets tomeet impact-attenuation and otherrequirements.

The standard establishes requirementsderived from one or more of thevoluntary standards applicable tobicycle helmets. In addition, thestandard includes requirementsspecifically applicable to children’shelmets and requirements to preventhelmets from coming off during an

accident. The standard also containstesting and recordkeeping requirementsto ensure that bicycle helmets meet thestandard’s requirements.

DATES: Effecti ve Date: This rule iseffective March 10, 1999.A ppl icabil i ty Dates: This rule applies

to bicycle helmets manufactured afterMarch 10, 1999. Interim mandatorystandards that went into effect on March17, 1995, wi ll continue to apply tobicycle helmets manufactured fromMarch 17, 1995, until March 10, 1999,inclusive. In addition, as of March 10,1998, firms wil l have the option of

marketing helmets meeting the standardin this final rule before its effective date.Incor porat ion by Reference: The

incorporation by reference of certainpublications listed in the rule isapproved by the Director of the FederalRegister as of March 10, 1999.

FOR FURTHER INFORMATION CONTACT:Frank Krivda, Office of Compliance,Consumer Product Safety Commission,Washington, D.C. 20207; telephone (301) 504–0400 ext. 1372.

SUPPLEMENTARY INFORMATION:

Outline of Contents

A. Introduction and Background1. Introduction.2. Injury and death data.3. The Children’s Bicycle Helmet Safety

Act of 1994.4. The current rulemaking proceeding.

B. Overall Description of Standard1. Impact attenuation.2. Children’s helmets: head coverage.3. Retention system.4. Peripheral vision.5. Labels and instructions.6. Positional stability (roll off).7. Certification labels and testing program.8. Recordkeeping.

9. Interim standards.C. The Final Standard— Comments,

Responses, and Other Changes1. Accident scenarios.2. Future revisions.3. Compliance with third-party standards

as compliance with the rule.4. Scope of the standard.a. Definition of ‘‘bicycle helmet.’’b. Multi-activity helmets.

5. Projections.6. Requirements for qualities of fittingpads.

7. Impact attenuation criteria.a. Extent of protection.b. Distance between impacts.c. Impact velocity tolerance.d. Other children’s requirements: peak g-

value and drop mass.8. Impact attenuation test rig.a. Type of test rig.b. Accuracy check.c. Test headform characteristics.d. Alignment of anvils.e. Definition of ‘‘spherical impactor.’’9. Impact attenuation test procedure.a. Anvil test schedule and use of curbstone

anvil.b. Definition of ‘‘comfort padding.’’c. Testing on more than one headform.d. Number of helmets required for testing.10. Helmet conditioning.a. Low-temperature environment:

temperature range.b. Water immersion environment.c. Reconditioning time.11. Labels.a. Label format and content.b. Use label.c. Labeling for cleaning products.d. Warning to replace after impact.e. Durability of labels.f. Labels on both helmets and boxes.12. Instructions for fitting children’s

helmets.

13. Retention system strength test.14. Positional stability test.15. Vertical vision.16. Reflectivity.17. Hard-shell requirements.

D. Certification Testing and Labeling1. General.2. The certification rule.3. Reasonable testing program.a. Changes in materials or vendors.b. Pre-market clearance and market

surveillance.4. Certificate of compliance.a. Coding date of manufacture.b. Telephone number on label.c. Certification label on children’s helmets.d. Minimum age on labels for children’s

helmets.e. Identifying the Commission.f. Certification label on packaging.

E. Recordkeeping1. General.2. Location of test records— time for

production.3. Length of records retention.

F. Regulatory Flexibility Act CertificationG. Environmental ConsiderationsH. Paperwork Reduction ActI. Executive OrdersList of Subjects in 16 CFR Part 1203Part 1203— Safety Standard for Bicycle

Helmets

A. Introduction and Background

1. Introdu ction

In this notice, the United StatesConsumer Product Safety Commission(‘‘Commission’’ or ‘‘CPSC’’) issues amandatory safety standard for bicyclehelmets.1

2. Injury and Death Data

Data from the National Center forHealth Statistics (‘‘NCHS’’) indicatedthat in 1993 there were 907 pedalcyclist(primarily bicycle-related) deaths in theUnited States. Of these, 17 (about 2%)were of children under the age of 5years. Research has shown thatapproximately 60% of all bicycle-related deaths involved head injury. Forchildren under age 5, about 64%involved head injury.2 Information onthe impact forces involved in these fatalincidents was not available, althoughabout 90% of the pedalcyclist deaths,including those of children under age 5,involved collisions with motor vehicles.

Based on data from CPSC’s NationalElectronic Injury Surveillance System(‘‘NEISS’’), there were an estimated566,400 bicycle-related injuries treatedin U.S. hospital emergency rooms in1996. Of these, approximately 30%involved the head and face. A higherproportion of head injuries and facialinjuries occurred to young children thanto older victims.

CPSC’s NEISS data showed that thetypes of injuries to young children weresomewhat different from those to olderchildren and adults. Younger children

had a smaller proportion of concussionsand internal injuries to the head thandid older victims, as well as a largerproportion of relatively minor headinjuries (i.e., lacerations, contusions,and abrasions). The extent to whichthese differences can be attributed to theuse of helmets, other aspects of thehazard scenario, or the physiology of young children, is not known. It is alsopossible that caregivers are more likelyto bring young children to theemergency room for relatively minorinjuries.

8/9/2019 10mr98r (1)

3/37

11713Federal Register / Vol. 63, No. 46/ Tuesday, March 10, 1998/ Rules and Regulations

3 Tinsworth, Deborah K., MS; Polen, Curtis; andCassidy, Suzanne. ‘‘Bicycle-Related Injuries: Injury,Hazard, and Risk Patterns,’’ International Journalfor Consumer Safety I (December 1994): 207–220.

4Rogers, Gregory B. ‘‘The Characteristics and UsePatterns of Bicycle Riders in the United States,’’ Journal of Safety Research 25 (1994): 83–96.

5 Thompson, Diane C., MS; Rivara, Frederick P.,MD, MPH; and Thompson, Robert S., MD.‘‘Effectiveness of Bicycle Safety Helmets inPreventing Head Injuries,’’ Journal of the AmericanMedical Association 276 (December 1996): 1968–1973.

6 The estimated reduction in risk for children 6–12 years of age was 70%.

7 Thompson, Robert S., MD; Rivara, Frederick P.,MD, MPH; and Thompson, Diane C., MS. ‘‘A CaseControl Study of the Effectiveness of Bicycle SafetyHelmets,’’ The New England Journal of Medicine320 (May 1989): 1361–1367.

8Recent research indicated that helmets reducedthe risk of serious injury to the upper and middleface by about 65%, but had no significant effect onserious injury to the lower face. Thompson, DianeC., MS; Nunn, Martha E., DDS; Thompson, RobertS., MD; and Rivara, Frederick P., MD, M PH.‘‘Effectiveness of Bicycle Safety Helmets inPreventing Serious Facial Injury.’’ Journal of theAmerican Medical Association 276 (December1996): 1974–1975.

A 1993 Commission staff study of bicycle hazards indicated that whenother factors were held constantstatistically, the injury risk for childrenunder age 15 was over five times therisk for older riders.3 This study alsoindicated that children were atparticular risk of head injury. Aboutone-half of the injuries to children

under age 10 involved the head,compared to one-fifth of the injuries toolder riders. This may have been in partbecause children were significantly lesslikely to have been wearing a helmetthan were older victims (5% of victimsyounger than 15 were wearing a helmet,compared to 30% of those 15 and older).However, detailed information relatingthe type of helmet, age of user, andother aspects of the hazard scenario tohead injury severity was not availablefrom that study. A Commission study onbicycle and helmet usage patterns foundthat in 1993 about 18% of bicyclists

wore helmets.4

A 1996 study of about 3,400 injuredbicyclists in the Seattle, Washington,area included an evaluation of theprotective effectiveness of helmets indifferent age groups.5 When bicycliststreated in hospital emergency rooms forhead injuries were compared tobicyclists who sought care for othertypes of injuries at the same emergencyrooms, helmet use was associated witha reduction in the risk of any headinjury by 69%, brain injury by 65%, andsevere brain injury by 74%.

By age group, this study showed thatthe reduction in the risk of head injuryranged from 73% for children under 6years to 59% for teens in the 13–19year-old age group.6 Based on the resultsof their study, the authors concludedthat helmets were effective for allbicyclists, regardless of age, and thatthere was no evidence that childrenyounger than 6 years need a differenttype of helmet. However, for childrenyounger than 6 years, there was onlyone helmeted child with a brain injury(a concussion), and no helmetedchildren with severe brain injuries. Thus, the protective effects of helmetson brain injuries and severe brain

injuries were not calculated for this agegroup.

A widely-cited 1989 study, publishedby the same authors, found that riderswith helmets had an 85% reduction intheir risk of head injury, and an 88%reduction in their risk of brain injury,when compared to cyclists withouthelmets.7 These results were found

when patients who sought emergencyroom care for bicycle-related headinjuries were compared to bicyclists inthe community who had crashes,regardless of injury or medical care. Arecent study indicated that helmets mayprotect more against head injuries thanagainst some facial injuries.8

3. The Chil dren’s Bicycle Helmet Safety

A ct of 1994

On June 16, 1994, the Children’sBicycle Helmet Safety Act of 1994 (the‘‘Act’’ or ‘‘the Bicycle Helmet SafetyAct’’) became law. 15 U.S.C. 6001–6006. The Act provides that bicycle helmetsmanufactured after March 16, 1995,conform to at least one of the followinginterim safety standards: (1) TheAmerican National Standards Institute(ANSI) standard designated as Z90.4–1984, (2) the Snell MemorialFoundation standard designated as B–90, (3) the ASTM (formerly theAmerican Society for Testing andMaterials) standard designated as F1447, or (4) any other standard that theCommission determines is appropriate.15 U.S.C. 6004(a)–(b). On March 23,1995, the Commission published itsdetermination that five additional

voluntary safety standards for bicyclehelmets are appropriate as interimmandatory standards. 60 FR 15,231. These standards are ASTM F 1447–1994; Snell B–90S, N–94, and B–95; andthe Canadian voluntary standard CAN/CSA –D113.2–M89. In that notice, theCommission also clarified that theASTM standard F 1447 referred to in theAct is the 1993 version of that standard. The interim standards are codified at 16CFR 1203.

The Act directed the ConsumerProduct Safety Commission to begin aproceeding under the AdministrativeProcedure Act, 5 U.S.C. 553, to:

a. Review the requirements of theinterim standards described above andestablish a final standard based on suchrequirements;

b. Include in the final standard aprovision to protect against the risk of helmets coming off the heads of bicycleriders;

c. Include in the final standard

provisions that address the risk of injuryto children; andd. Include additional provisions as

appropriate. 15 U.S.C. 6004(c). The Act provides that the final

standard shall take effect 1 year from thedate it is issued. 15 U.S.C. 6004(c). TheAct further provides that the finalstandard shall be considered to be aconsumer product safety standardissued under the CPSA. Section 9(g)(1)of the CPSA provides that a ‘‘consumerproduct safety standard shall beapplicable only to consumer productsmanufactured after the effective date.’’ Thus, the final standard, which theCommission is issuing in this notice,will become effective March 10, 1999, asto products manufactured after thatdate. The Act also provides that failureto conform to an interim standard shallbe considered a violation of a consumerproduct safety standard issued underthe Consumer Product Safety Act(‘‘CPSA’’), 15 U.S.C. 2051–2084.

The Act states that the CPSA’sprovisions regarding rulemakingprocedures, statutory findings, and judicial review (15 U.S.C. 2056, 2058,2060, and 2079(d)) shall not apply to thefinal standard or its rulemaking

proceeding. 15 U.S.C. 6004(c). The final rule is codified at 16 CFR

1203 and will replace the interimstandards as to bicycle helmetsmanufactured on or after March 11,1999. 15 U.S.C. 6004(d). In addition, thefinal standard is also being designatedan interim standard, so that firms willhave the option of marketing helmetsmeeting CPSC’s final standard before itseffective date. Because providing thisadditional interim standard is asubstantive rule that grants anexemption or relieves a restriction, the30-day delay of an effective date

otherwise required by 5 U.S.C. 553(d) isinapplicable, and this designation iseffective March 10, 1998.

4. The Current Rulemaki ng Proceedin g

The Commission reviewed the bicyclehelmet standards identified in the Act(ANSI, ASTM, and Snell), as well asinternational bicycle helmet standardsand draft revisions of the ANSI, ASTM,and Snell standards that were thenunder consideration. Based on thisreview, the Commission developed aproposed safety standard for bicycle

8/9/2019 10mr98r (1)

4/37


9Heh, S., Log of ASTM FO8.53 HeadgearSubcommittee meeting held May 21, 1992, date of entry June 17, 1992. Office of the Secretary, U.S.Consumer Product Safety Commission, Washington,DC 20207.

helmets. 59 FR 41,719 (August 15,1994).

The Commission received 37comments on that proposed bicyclehelmet standard from 30 individualsand organizations. A fter consideringthese comments and other availableinformation, the Commission proposedcertain revisions to the originally

proposed standard. 60 FR 62662(December 6, 1995).

In response to the second proposal,the Commission received 31 comments. These comments, and additional datathat have been received by theCommission since the second proposal,are discussed in Sections C–E of thisnotice.

B. Overall Description of the Standard

The major features of the standardissued in this notice are describedbelow.

1. Imp act A ttenuation

The standard establishes aperformance test to ensure that helmetswill adequately protect the head in acollision. This test involves securing thehelmet on a headform and dropping thehelmet/headform assembly to achievespecified velocities so that the helmetimpacts a fixed steel anvil. The helmetmust provide protection at all pointsabove a line on the helmet that has aspecified relation to the headform.

Under the standard, the helmet istested with three types of anvils (flat,hemispherical, and ‘‘curbstone,’’ asshown in Figures 11, 12, and 13 of the

standard). These anvils represent shapesof surfaces that may be encountered inactual riding conditions.Instrumentation within the headformrecords the headform’s impact inmultiples of the acceleration due togravity (‘‘g’’). Impact tests are performedon different helmets, each of which hasbeen subjected to one of fourenvironmental conditions. Theseenvironments are: ambient (roomtemperature), high temperature (117–127°F), low temperature (1–9°F), andimmersion in water for 4–24 hours.

Impacts are specified on a flat anvilfrom a height of 2 meters and onhemispherical and curbstone anvilsfrom a height of 1.2 meters. Consistentwith the requirements of the ANSI,Snell, and ASTM standards, the peakheadform acceleration of any impactshall not exceed 300 g for an adulthelmet, the value originally proposedfor both adult and child helmets. In therevised proposed standard, theacceptable g value for children’shelmets was reduced to 250 g and alower headform drop mass than that foradults was specified (3.90 kg). As

explained in section C of this notice,however, the final rule specifies that the5-kg headform mass and the 300-g peakacceleration criterion will apply to allhelmets subject to the standard, asspecified in the original proposal.

The standard provides that a helmetfails the performance test if a failure canbe induced under any combination of

impact site, anvil type, anvil impactorder, or conditioning environmentpermissible under the standard. Thus,the Commission will test for a ‘‘worstcase’’ combination of test parameters.What constitutes a worst case may vary,depending on the particular helmetinvolved.

2. Chi ld ren’s Helm ets: Head Coverage

The standard specifies that helmetsfor small children (under age 5) mustcover a larger portion of the head thanmust helmets for older persons. A studyby Biokinetics & Associates Ltd. found

differences in anthropometriccharacteristics between youngchildren’s heads and older children’sand adult’s heads.9

3. Retent io n System

The standard requires that helmets beable to meet a test of the dynamicstrength of the retention system. Thistest ensures that the chin strap is strongenough to prevent breakage or excessiveelongation of the strap that could allowa helmet to come off during an accident.

The test requires that the chin strapremain intact and not elongate morethan 30 mm (1.2 in) when subjected to

a ‘‘shock load’’ of a 4-kg (8.8-lb) weightfalling a distance of 0.6 m (2 ft) onto asteel stop anvil (see Figure 8). This testis performed on one helmet underambient conditions and on three otherhelmets after each is subjected to one of the different hot, cold, and wetenvironments.

4. Peripheral V ision

Section 1203.14 of the standardrequires that a helmet shall allow a fieldof vision of 105 degrees to both the leftand right of straight ahead. Thisrequirement is consistent with theANSI, ASTM, and Snell standards.

5. Labels and Instruct ions

Section 1203.6 of the standardrequires certain labels on the helmet. These labels provide the modeldesignation and warnings regarding theprotective limitations of the helmet. Thelabels also provide instructions

regarding how to care for the helmet andwhat to do if the helmet receives animpact. The labels also must carry awarning that for maximum protectionthe helmet must be fitted and attachedproperly to the wearer’s head inaccordance with the manufacturer’sfitting instructions.

The standard also requires that

helmets be accompanied by fitting andpositioning instructions, including agraphic representation of properpositioning. As noted above, thestandard has performance criteria forthe effectiveness of the retention systemin keeping a helmet on the wearer’shead. However, these criteria may notbe effective if the helmet is not wellmatched to the wearer’s head andcarefully adjusted to obtain the best fit.

To avoid damaging the helmet bycontacting it with harmful commonsubstances, the helmet must be labeledwith any recommended cleaning agents,

a list of any known common substancesthat will cause damage, and instructionsto avoid contact between suchsubstances and the helmet.

6. Posit i onal Stabil i ty (Roll Off)

The standard specifies a testprocedure and requirement for theretention system’s effectiveness inpreventing a helmet from ‘‘rolling off’’ ahead. The procedure specifies adynamic impact load of a 4-kg (8.8-lb)weight dropped from a height of 0.6 m(2 ft) to impact a steel stop anvil. Thisload is appl ied to the edge of a helmetthat is placed on a headform on a

support stand (see Figure 7). The helmetfails if i t comes off the headform duringthe test.

The safety requirements discussed inparagraphs (1)–(6) above are issuedpursuant to the Bicycle Helmet SafetyAct and are codified as Subpart A of theSafety Standard for Bicycle Helmets.

7. Cert if i cat ion Labels and Testin g

Program

Under the authority of section 14(a) of the CPSA, the Commission is alsoissuing certification testing and labelingrequirements to ensure that bicycle

helmets meet the standard’s safetyrequirements. These certificationrequirements are in Subpart B of theSafety Standard for Bicycle Helmets andare discussed in section D of this notice.

8. Recordkeepin g

Under the authority of section 16(b) of the CPSA, the Commission is issuingrequirements that manufacturers(including importers) maintain recordsof the required certification testing. These recordkeeping requirements arefound in Subpart C of the Safety

8/9/2019 10mr98r (1)

5/37


Standard for Bicycle Helmets and arediscussed in section E of this notice.

9. Interim Standards

The interim standards, which arecurrently codified as 16 CFR 1203, wi llcontinue to apply to bicycle helmetsmanufactured from March 16, 1995, toMarch 11, 1999. Accordingly, the

interim standards will continue to becodified, as Subpart D of the standard.Also, Subparts A–C of the standard arebeing added as an interim standard, sothat firms wil l have the option of marketing helmets meeting CPSC’s finalstandard before its effective date.

C. The Final Standard— Comments,Responses, and Other Changes

This section discusses comments onthe second proposal, as well as otherissues that were dealt with in decidingthe requirements of the final rule.Numbers in brackets refer to the numberassigned by the Commission’s Office of the Secretary to a comment on thesecond proposal.

1. Accid ent Scenari os

Mr. Frank Sabatano [14], President of the London Bridge BMX Association,recommended that bike helmets beconstructed so as to accommodate moreserious accidents that might result froma child bicycle racing or jumping ratherthan merely riding on a path or street.

Whi le no helmet can protect againstevery conceivable impact, the availableevidence supports the conclusion thathelmets designed to meet the CPSC

standard will be very effective inprotecting against serious injury withina wide range of common bicycle ridingconditions. This would include many of the impact conditions that could occurduring racing or jumping. Furthermore,a standard for all bicycle helmets has tobalance the benefits of more protectivehelmets against the additional cost,weight, bulk, and discomfort that moreprotection may impose. Suchundesirable qualities may discouragemany users from wearing helmetsdesigned to protect against very severeimpacts, which could more than cancel

the effects of the additional protectivequalities. Thus, the force with which thehelmets are impacted in the standard’sperformance test has not been increased.

2. Future Revision s

Randy Swart, Director of the BicycleHelmet Safety Institute [16], suggestedthat the following items be consideredas future revisions to the CPSC standardas progress in head protection researchcontinues:

a. A test that requires the retentionsystem to be easily adjusted for good fit.

b. A test for protection againstrotational injury.

c. A test to limit localized loads or‘‘point loading.’’

d. A test for damage to the helmet byhair oil or other common consumerpreparations.

e. A test of the retention system afterimpact to simulate field conditions.

f. A test to ensure that visors andmirrors are shatter-resistant and easilypeel off in a crash.

The Commission agrees that it isimportant to periodically reviewresearch related to improvements inhead protection to determine if revisions should be considered for theCPSC bicycle helmet standard.

3. Compli ance With T hir d-Party

Standards as Compli ance With th e Rule

Jane McCormack [7] requested thatthe Commission ensure that bikehelmets meet the Snell requirements.Norte Vista Medical Center [15]

requested that helmets certified to theSnell B–95 or Snell N–94 standards beconsidered to be in compliance with themandatory standard.

The Commission declines to makethese changes. One of the objectives of the Bicycle Helmet Safety Act is toestablish a unified bicycle helmetstandard that is recognized nationallyby all manufacturers and consumers. Itwould defeat Congress’ intent to addlanguage to the regulation stating thatcertified conformance to any existingvoluntary standard satisfies compliancewith the mandatory rule.

4. Scope of the Stand ard

a. Definition of ‘‘Bicycle Helmet’’

The original proposal defined bicyclehelmet as ‘‘any headgear marketed assuitable for providing protection fromhead injuries while riding a bicycle.’’ The definition of bicycle helmet in thesecond proposal included not onlyproducts specifically marketed for useas a bicycle helmet but also thoseproducts that can be reasonably foreseento be used for that purpose.

Bell Sports [12] suggested that thedefinition of bicycle helmet should not

include all products with a reasonablyforeseeable use as a device intended toprovide protection from head injurieswhile riding a bicycle. Bell maintainsthere are many helmets that have aforeseeable use by bike riders thatshould not have to be certified to a bikehelmet standard (e.g., baseball and rollerhockey helmets).

The respondent suggested thatfootball helmets, baseball battinghelmets, and motorcycle helmets willalso have ‘‘easily foreseeable’’ uses asbicycle helmets.

The Commission did not intend forthe definition of bicycle helmet toinclude football helmets, baseballbatting helmets, and motorcycle helmetsthat are not marketed for use whi lebicycling. It seems unlikely that ahelmet that is not marketed or promotedfor bicycle use will have a reasonablyforeseeable use as a bicycle helmet.

Thus, the ‘‘reasonably foreseeable’’language is unnecessary. Therefore, inorder for the definition to provide moreguidance, the ‘‘reasonably foreseeable’’language has been deleted, and thedefinition of bicycle helmet has beenchanged to read: ‘‘Bicycle helmet meansany headgear that either is specificallymarketed as, or implied throughmarketing or promotion to be, a deviceintended to provide protection fromhead injuries while riding a bicycle.’’

Helmets specifically marketed forexclusive use in a designated activitysuch as skateboarding, rollerblading,

baseball, roller hockey, etc., would beexcluded from this definition becausethe specific focus of their marketingmakes it unlikely that such helmetswould be purchased for other than theirstated use. However, a multi-purposehelmet— one marketed or represented asproviding protection either duringgeneral use or in a variety of specificactivities other than bicycling— wouldfall within the definition of bicyclehelmet if a reasonable consumer couldconclude, based on the helmet’smarketing or representations, thatbicycling is among the activities inwhich the helmet is intended to be

used.In making this determination, the

Commission wi ll consider the types of specific activities, if any, for which thehelmet is marketed, the similarity of theappearance, design, and construction of the helmet to other helmets marketed orrecognized as bicycle helmets, and thepresence, prominence, and clarity of any warnings, on the helmet or itspackaging or promotional materials,against the use of the helmet as a bicyclehelmet. The presence of warnings ordisclaimers advising against the use of a multi-purpose helmet during bicycling

is a relevant, but not necessarilycontrolling, factor in the determinationof whether a multi-purpose helmet is abicycle helmet. A multi-purpose helmetmarketed without specific reference tothe activities in which the helmet is tobe used will be presumed to be a bicyclehelmet.

b. Multiple-Activity Helmets

Some commenters on the originalproposal recommended that the CPSCinclude provisions for children’s bicyclehelmets to provide protection in

8/9/2019 10mr98r (1)

6/37


activities in addition to bicycling, suchas skateboarding, skating, sledding, andthe like. Two commentersrecommended that the CPSC bikehelmet standard also apply to helmetsmarketed for roller skating and in-lineskating. Other comments stated that theCommission should not delaypromulgation of the bike helmet

standard while multi-activity issues areexplored.

The Commission did not propose thatthe standard address activities otherthan bicycling, because the CPSC’sauthority under the Bicycle HelmetSafety Act is to set mandatoryrequirements for bicycle helmets.Establishing criteria for products otherthan bicycle helmets would require theCommission to follow the proceduresand make the findings prescribed by theCPSA or the Federal HazardousSubstances Act (‘‘FHSA’’).

The National Safe Kids Campaign(‘‘NSKC’’) [22] and the ConsumerFederation of A merica (‘‘CFA’’) [23]recognized that the scope of the CPSCstandard must be for bicycle helmets,but requested the Commission to moveforward in investigating the issuesrelated to multi-activity helmets. In acomment on the revised proposal, Mr.Frank Sabatano, President of theLondon Bridge BMX Association [14],recommended that bicycle helmetsshould serve as multi-purposeprotective devices for various sportssuch as bicycle riding, bicycle racing,skateboarding, and in-line skating.

The Commission intends to monitor

developments relevant to the multi-activity issue. Wheeled recreationalactivities such as traditional rollerskating and in-line skating are typicallyconducted on the same surfaces asbicycling, and can generate speedssimilar to bicycling. Therefore, it isreasonable to assume that helmets thatmeet the requirements in the CPSC bikehelmet standard will also provide headprotection for roller/in-line skating andperhaps some other recreationalactivities. However, as discussed in theDecember 6, 1995, Federal Registernotice on the proposed rule, theCommission does not have sufficientdata on the benefits and costs of additional features directed at injuriesincurred in activities other thanbicycling to make the statutory findingsthat would be needed to issue arequirement for such features undereither the CPSA or FHSA. Also,procedures in addition to those requiredby the Bicycle Helmet Safety Act wouldhave to be followed. The Commissiondoes not want to delay establishment of a mandatory bicycle helmet standard inorder to pursue rulemaking for other

types of helmets. Accordingly, the finalstandard only addresses requirementsfor bicycle helmets. However, asdiscussed below, the Commission wi llexamine what actions it could take toencourage the use of bicycle helmets inactivities that present head injury riskssimilar to those in bicycling.

NSKC [22] also urged the CPSC to

work with community-basedorganizations to develop acomprehensive educational campaignregarding the importance of wearing afederally-approved bicycle helmet whenparticipating in non-motorized activitiesother than bicycling. The Commissionwill consider what activities areappropriate in this regard when settingits priorities for future activities.

5. Projection s

Projections on the inner or outersurface of a helmet can concentrateapplied forces and cause injuries. Therefore, the revised proposedstandard provided that projections onthe outer surface would not exceed 7mm (0.28 in) unless they break away orcollapse on impact and that projectionson the helmet’s interior not makecontact with the headform duringtesting.

NSKC [22] urged that the Commissionprohibit any external projections onhelmets intended for children. NSKCbelieves that external projections, suchas visors, are unnecessary componentsof helmets intended for children.

With regard to a possible hazard fromexternal projections on children’s

helmets, §1203.7 of the standardrequires that helmets must pass all tests,both with and without any attachmentsthat may be offered by the manufacturer. This provision, and the requirement thatany external projections shall breakaway or collapse, will address thepotential hazard of external projectionson helmets intended for riders of allages. The proposed language isconsistent with existing voluntarystandards, and no changes were made inresponse to this comment.

SwRI [2] remarked that the proposedstandard does not state how to

determine if an internal projectionmakes contact with the headform duringtesting. NSKC [22] also suggested thatinstead of requiring inner surfaceprojections to not exceed 2 mm, theinside of the helmet should contain nosharp edges or rigid internal projections.

After considering these comments, theCommission decided to revise thesection on internal projections toeliminate the requirement that internalprojections not make contact with theheadform during testing, while retainingthe requirement that such projection not

exceed 2 mm (0.08 in). The purpose of this section is to prohibit potentiallyhazardous projections but make someallowance for common helmetconstruction practices. The languageabove is consistent with Snell helmetstandards, and the Commission is notaware of safety problems associatedwith projections on helmets meeting

existing standards.

6. Requir ements for Qual it i es of Fit t in g

Pads

NSKC [22] urged the Commission toinclude safety requirements for fittingpads in the final standard. Thecommenter asserted that since fittingpads are often necessary to ensure asecure fit, the standard should addressthe integrity of the materials used toconstruct them, as well as theirthickness, durability, and adhesiveness.

CPSC staff has no information thatlong-term integrity of fitting pads is a

problem with helmets meeting existingstandards. The interim mandatorystandards have no provisions of the typesuggested by the commenter.Introducing new requirements for fi ttingpads is not essential at this time, and nochange to the proposed standard hasbeen made in response to this comment.

7. Imp act Attenuation Criteria

a. Extent of Protection

The originally proposed CPSCstandard, and current U.S. voluntarybicycle helmet standards, specified an

extent-of-protection boundary and animpact test line. The extent-of-protection boundary defines the area of the head that must be covered by thehelmet. The impact test line designatesthe lowest point on the helmet wherethe center of an anvil may be aligned fortesting. The second proposal specified asingle impact test line and no extent-of-protection boundary requirement. Notrequiring specific helmet coverageallows manufacturers the flexibility toinclude desirable features, such as acentral rear vent, provided the featuresdo not hinder the helmet’s ability tomeet the impact requirements if tested

anywhere on or above the impact testline. Accordingly, the Commissiondeleted the extent-of-protectionboundary from the revised proposedstandard.

In commenting on the latter proposal,Snell [28] discussed the practicalproblems in certifying helmets whenonly an impact test line is specified.Snell recommended that the standard beamended to require coverage below theimpact test line, particularly at the frontand rear of a helmet.

8/9/2019 10mr98r (1)

7/37


10Dean Fisher and Terry Stern, ‘‘Helmets Work!,’’Bell Sports, Inc., AAAM/IRCOBI Conference, Lyon,France (September 1994).

11Martin Wil li ams, ‘‘Test Line Requirements andSnell B–95 and N–94 Standards,’’ TechnisearchEngineering & Scientific Services (August 1994).

The Commission disagrees with thiscomment. Coverage does not implyimpact protection. The only area on thehelmet required to pass impactprotection requirements is the areaabove the impact test line. Therefore, itis unnecessary to specify additionalcoverage below the test line.

The manufacturers of the Protective

Headgear Manufacturing Association(‘‘PHMA’’) [29] reported that theybelieved the proposed CPSC standardrequires coverage at the rear of the headlower than any other standard. Theystated that they are not aware of anystudies indicating that lower coverage atthe rear is warranted. They also statedtheir concern that the helmet-wearingpublic wi ll not purchase helmets thatare perceived to be more ‘‘clunky’’ or‘‘bulbous,’’ and that helmets withextended coverage are likely be soperceived. Mr. Becker of Snell [28]stated that the CPSC-proposed coverages

are more extensive than any currentU.S. standard, except for Snell ’s B–95and N–94 helmet standards. He statedthat unless the CPSC coverage ischanged, many contemporary helmetmodels that have protected theirwearers from li fe-threatening injury wi lldisappear from the market. Snell urgedthat the CPSC adopt the coveragedescribed in the ASTM F1447–94 orSnell B–90 standards. According to thiscommenter, these coverages reflect thecurrent state of the industry and shouldbe expected of every bicycle helmet.

The proposed CPSC impact test line isnot lower at the rear of the helmet than

all other standards. The proposed CPSCimpact test line is somewhat lower atthe rear of the helmet than the impacttest lines in the Snell B–90 and ASTMF1447 standards. However, the CPSCline is higher at the rear of the helmetthan the impact test lines in thefollowing interim mandatory standards:Snell B–95 and N–94, CAN/CSA–D113.2, and ANSI Z90.4–1984.

CPSC is aware of two studies thatshow that it is not uncommon forhelmets involved in accidents to sufferimpacts at the rear portion of thehelmet. A Bell Sports study of 1100

helmets involved in accidents foundthat 26% of the impacts were at the rearof the helmet and that the majority of these rear impacts occurred within 50mm of the bottom edge of the helmet.10

Another study, by Technisearch of Australia, examined the effect of lowering the impact test line from theSnell B–90 standard to the impact testlines in the Snell B–95 and N–94

standards.11 The Technisearch studywas based on examinations of 104bicycle helmets whose wearerssustained impacts to the head duringaccidents. The study concluded that theB–90 standard test line would haveprovided coverage for 51% of theimpacts. The impact test line of the B–95 standard would provide coverage for

65% of the impacts. The increase from51% to 65% was represented by 20additional impact sites that would fallwithin the area of the B–95 coverage,including 8 impact sites at the rearportion of the helmet.

One of the directions of the Children’sBicycle Helmet Safety Act is to includeprovisions from existing appropriatestandards for adoption in the final CPSCstandard. The CPSC impact test line isa reasonable requirement that willimprove the protective characteristics of helmets overall, while falling within testlines of established North Americanbicycle helmet standards.

b. Distance Between Impacts

A commenter on the original proposalrecommended revising the minimumdistance between impact sites from theoriginally proposed ‘‘one fifth thecircumference of the helmet’’ to 120mm. The Commission believed that 120mm allows sufficient distance tominimize the effects of impact siteproximity and provides a morestraightforward measurement than theoriginal one-fifth circumference criteria.Accordingly, the Commission adoptedthis recommendation in the revisedproposal.

Two commenters on the revisedproposal [27 and 29] recommended aminimum distance between impacts of 150 mm, or about 6 inches. One of thesecommenters stated that the CPSC madethe minimum distance shorter thanthose in voluntary standards.

The Commission selected the 120-mmimpact spacing based on recentlyballoted ASTM headgear standards. TheSnell B–95 standard also specifies aminimum impact separation of 120 mm.

This distance is consistent with theSnell B–90 specification of 1⁄6th themaximum helmet circumference, if calculated for smaller helmets. Aminimum impact spacing of 150 mmwould l imit flexibili ty in choosingimpact sites, especially on smallerhelmets. Therefore, no change to theproposed rule was made in response tothis comment.

c. Impact Velocity Tolerance

The University of SouthernCalifornia’s Head Protection ResearchLab (‘‘USC–HPRL’’) [8] suggested thatthe tolerance for the impact velocity bechanged from ±3% to¥0% to +5% toensure that impact testing is done at noless than the specified velocity.

The difference between tolerances of ±3% and¥0%, +5% has little practicalsignificance for a 300–g criterion. Sincethe commenter’s suggestion would notproduce a significant safety benefit, theCommission made no change to theproposed rule in this regard.

d. Other Requirements for Chi ldren’sHelmets: Peak-G Value and Drop Mass

One of the provisions of TheChildren’s Bicycle Helmet Safety Act of 1994 is that the Commission include inthe final CPSC standard provisions thataddress the risk of injury to children. This does not require that children’s

helmets be subject to requirements thatdiffer from those for adults’ helmets; itrequires only that the final standard beappropriate for children’s helmets. Theissue of whether special standardprovisions for young children’s helmetsare needed has been debated for severalyears by head protection experts.

A young child’s skull has differentmechanical properties than the skull of an older child or adult. Thesedifferences are especially evident forchildren under the age of 5 years. Theirskulls have a lower degree of calcification, making them more flexible

than adult skulls. During an impact tothe head, the increased skull flexibil ityresults in a greater transfer of kineticenergy from the impact site to the braintissue. Besides the different mechanicalproperties, the mass of a young child’shead is also different from that of a moremature person’s head. Studies show thatthe head mass of children under the ageof 5 years ranges from approximately 2.8to 3.9 kg. This mass is lower than the5-kg test headform mass specified incurrent U.S. bicycle helmet standards.

The Commission first proposed asafety standard for bicycle helmets onAugust 15, 1994. In that proposal, theonly special provision for helmets forchildren under 5 years was an increasedarea of head coverage. On December 6,1995, however, the Commissionproposed special provisions forheadform mass, peak-g limit, and headcoverage for bicycle helmets forchildren under 5 years. The specialchildren’s provisions were based on theongoing work of voluntary standardsorganizations and proposals at that timein the technical literature. The followingcomparison shows the CPSC-proposed

8/9/2019 10mr98r (1)

8/37


12 Thompson, Robert S., MD; Rivara, Frederick P,MD, MPH; and Thompson, Diane C., MS ‘‘A CaseControl Study of the Effectiveness of Bi cycle SafetyHelmets,’’ The New England Journal of Medicine320 [May 1989]: 1361–1367. Thompson, Diane C.,MS; Rivara, Frederick P, MD, MPH; and Thompson,Robert S., MD. ‘‘Effectiveness of Bicycle SafetyHelmets in Preventing Head Injuries,’’ Journal of theAmerican Medical Association 276 (December1996): 1968–1973.

13Myers, Barry, M .D., Ph.D. ‘‘An Evaluation of AHelmet Standard for Children,’’ Report to the U.S.Consumer Product Safety Commission (July 1997).

test parameters for helmets for childrenunder 5 years and for helmets for olderpersons.

Under 5 5 and

older

Mass of testheadform.

3.9 kg ...................... 5.0 kg

Peak-g limit ... 250-g ...................... 300-g

Head cov-erage.

More coverage atrear and sides ofhead.

The proposal for increased headcoverage of children’s helmets isrelatively uncontroversial, and the finalrule contains this requirement.However, the Commission hasreassessed the proposed headform massand peak-g requirements. TheCommission’s conclusions are discussedin detail below.

A few respondents to the proposedrule [8, 16] supported the lower massand lower peak-g provisions, believingthat they will lead to an improvementin head protection for small children.One of these respondents, however,urged the Commission to consider themost recent research on this subjectbefore including the special provisionsin a final standard. One respondent [12]favored a reduced headform massprovision, but did not recommend areduced peak-g provision, stating that itcould result in a helmet with a lowermargin of safety.

Several respondents [3, 4, 6, 9, 10, 13,15, 18, 19, 27, 28, 29, 30] questionedwhether it is advisable to move forward

with the provisions of a reduced-massheadform and a lower limit for peakacceleration. Some respondentssuggested that special children’sprovisions should not be adopted sincestudies show that children’s helmets asthey exist today provide excellentprotection.

Studies by researchers at theHarborview Injury Prevention andResearch Center have shown thatbicycle helmets that meet existingstandards are effective in protectingagainst serious head and braininjuries.12 One of the items analyzed inthe most recent Harborview study waswhether the protective effects of bicyclehelmets vary by the age of the user. Forfour age groups of riders, they estimatedthe protective effect of helmets against

three levels of injury l isted in order of increasing severity: (1) head injury, (2)brain injury, and (3) severe brain injury.

Due to the small number of helmetedcase subjects that suffered brain injuryand severe brain injury, Harborviewresearchers could not estimate theprotective effect of helmets against theseinjuries for the under 6-year-old age

group. Accordingly, the Commissionhas not relied on this study in itsconsideration of whether specialrequirements are needed for children’shelmets. However, one of Harborview’soverall conclusions was that helmets areeffective for all bicyclists, regardless of age, and that there is no evidence thatchildren younger than 6 years need adifferent type of helmet.

The Commission requested technicalviews on this issue from Barry M yers,M.D., Ph.D. Associate Professor,Department of Biomedical Engineering,Duke University. In his report,13 Dr.

Myers explains that such modificationsof the standard should be consideredonly if it can be shown to improve theprotective qualities of helmets.Improvements may be shown byepidemiological or biomechanicalevidence. However, considering thedegree of head injury protectionprovided by current helmets,incremental improvement would bedifficult to detect, even with a largeepidemiological study.

From a biomechanical perspective, itis important to assess how changes intest headform mass and peak-g criteria

would affect helmet design andprotective capability. This can be doneby examining how a helmet functions toprotect the head in an impact.

The helmet has a crushable linertypically made of expanded polystyrenefoam. If the liner is crushed as the headpresses against the inside of the helmetduring impact, the liner allows the headto stop over a longer distance and timethan would otherwise be the case. Thisreduces the transfer of energy to thehead, thereby reducing the risk of injury.

The degree to which the liner resists

being crushed also affects the helmet’sprotective qualities. For a given impact,a helmet liner that is too soft wil l‘‘bottom out,’’ thereby losing itsprotective abil ity to allow relativemovement between the head and theobject being impacted. Conversely, aliner that is too hard will not allowsufficient crushing to adequately protectthe head.

Proponents of special provisions foryoung children’s helmets believe thatthese helmets should be tested underdifferent test parameters than helmetsintended for older persons. The currenttest parameters are based primarily onadult head injury tolerance and on aheadform mass that is approximatelythat of an adult head. Supporters of

special provisions contend that theseadult test parameters result in a helmetwith a liner that is too stiff to optimallyprotect a young child’s head. By usinga headform weight that better representsa young child’s head (e.g., 3.9 kg), andreducing the allowable peak-g, helmetswould need to be designed with a lowerdensity (‘‘less stiff’’) liner to furtherlessen the impact transmitted to thehead.

A simple way to examine the effect of changing headform mass and the peak-g criterion is to model the helmet as aspring and apply the one-dimensional

spring-mass impact formulas shownbelow. This approach is discussed byboth Dr. Myers and by Mr. Jim Sundahl,Senior Engineer with Bell Sports, in hisresponse to the proposed rule [12].

a Vo k

m

x Vo m

k

peak

peak

=

=

( )

( )

1

2

Where:

apeak = peak acceleration (peak-g)Vo = impact velocity

k = liner stiffnessm = headform massxpeak = required stopping distance (liner

thickness)

If the value for headform mass m isreduced in Equation (1), the value forliner stiffness k must be reduced toachieve the same peak-g at the sameimpact velocity. This means that if ahelmet that meets the standard’s criteriawith a 5-kg headform did not meet thepeak-g requirement using a lighterheadform, the helmet liner would needto be made softer so more crushing of the liner could occur.

If the value for peak acceleration apeak is reduced in Equation (1), and the othervariables are held constant, the value forliner stiffness k again must be reduced. Thus, a helmet that could not complywith a reduced peak-g criterion alsowould need a softer l iner to allow morecrushing. Equation (2) shows that, witha decreased liner stiffness, a greaterpercentage of the available crushdistance wil l be used during impact.

The biomechanical analysis showsthat, for impact conditions that do notresult in complete compression of the

8/9/2019 10mr98r (1)

9/37


14Rivara, Frederick P., MD, MPH, Thompson,Diane C., MS, Thompson, Robert S., MD‘‘Circumstances and Severity of Bicycle Injuri es,’’Snell Memorial Foundation/Harborview InjuryPrevention and Research Center (1996).

helmet’s liner, it is possible to lessen theimpact energy transmitted to the head(and reduce the risk of injury) byreducing the stiffness of the liner.However as the impact energy increases,a helmet with a softer liner will bottomout (crush beyond its protectivecapacity) under less severe conditionsthan a helmet with a more rigid liner of

the same thickness. To compensate, thesofter helmet would have to be madethicker to prevent bottoming out.However, there is a limit to how thicka helmet can be before it is no longerpractical or appealing to the user. Therefore, the goal of helmet design isto optimize liner density and thicknessto protect against the widest range of impact conditions and still have aproduct people will use.

The biomechanical analysis suggeststhat reducing the liner stiffness couldhave both a positive and a negativeinfluence on the protection provided by

helmets under existing criteria. Therefore, it is necessary to alsoexamine available epidemiological datathat relate to this issue. Decreasing theliner stiffness would benefit those whoexperience injuries with minimal or noliner deformation of current helmets.However, a decrease in liner stiffnesscould increase the number of headinjuries that occur during more severeimpacts that cause the helmet liner tobottom out.

To learn the effect on the level of protection offered by softer helmetliners for children under 5, twoquestions would need to be answered:

1. Are children suffering head injurieswith minimal or no deformation of current helmet liners?

2. Are children suffering head injurieswith a bottomed-out liner?

Unfortunately, currently availableinformation does not answer either of these questions. Therefore, it isuncertain whether young childrenwould benefit from special provisionsfor headform mass and peak-g.

The only known study to examine therelationship between helmet damageand head injury was completed in 1996by the Snell Memorial Foundation andthe Harborview Injury Prevention andResearch Center.14 Of those bicyclehelmets collected from individuals (of various ages) who went to a hospital,40% of the helmets had no deformation,14% had significant damage in whichthe helmet was approaching a bottomed-out condition, and 7% of the helmetshad catastrophic damage. The data were

not presented specifically for the under-5 age group or any other specific agegroup. The study showed that there wasa risk of head and brain injury evenwith no or minimal helmet damage. Therisk of injury increased moderately asthe severity of helmet damage increased,until catastrophic damage was reached.As expected, the risk of head and brain

injury jumped dramatically when ahelmet was damaged catastrophically. This study suggests that if helmets forall ages were designed with softer liners,there is a potential to both improve theprotection for lower-severity impactsand increase the risk of injury at thehigher-severity impacts.

Since the risk of injury risesdramatically with catastrophic helmetdamage, and current helmets areeffective in reducing the risk of headand brain injuries, it would beimprudent to require softer helmetliners for bicyclists of all ages. The

available data are insufficient todetermine that such a change wouldincrease overall protection. Whenfocusing on the age range of under 5years, currently available information iseven more sparse. Therefore, if helmetsfor children under age 5 were madewith softer l iners, there are insufficientdata to estimate either (1) the level of protection that might be gained at thelower-severity impacts or (2) theprotection that might be lost at thesevere impact conditions thatcompletely crush the liner.

For the reasons discussed above, theCommission did not include specialprovisions in the final standard forheadform mass and peak-g criteria foryoung children’s helmets. There areinsufficient data to justify the changes,and these changes could provide lessprotection in the most serious impacts.However, should future studies provideevidence that young children, orbicyclists of any age, could benefit fromdecreased liner stiffness, theCommission could consider revisions tothe bicycle helmet standard at that time.

8. Impact Attenu ation T est Rig

a. Type of Test Rig

The originally proposed CPSCstandard and the current interimmandatory standards allowed the use of either a wire- or rail-guided impact testrig. In the revised proposal, theCommission specified only the monorailtest rig, to avoid the possibility thatdifferent results would be obtained withthe two types of test rigs.

Some helmet manufacturers [5, 29,30], and the Snell Memorial Foundation[28], disagreed with the specification of the monorail type of impact test rig.

Commenters stated that guidewire rigswere more widely used in the industry.Some commenters claimed that sincethere is no evidence that directlycorrelates monorail with guidewire rigresults, many firms would be forced tobuy monorail rigs to address liabil ityconcerns. Trek [5] stated that the burdenof this expense may require additional

analysis of the financial impact to smallbusiness, as required by the RegulatoryFlexibil ity Act. Snell wrote thatguidewire rigs have proven reliable,efficient, and highly repeatable. Theyare less expensive to install thanmonorail devices, and they are easier tomaintain. Snell stated that there is nodemonstrated improvement associatedwith the monorail rig in testingreliability and capability. Mostcommenters suggested that theCommission allow both monorail andguidewire rigs.

To respond to this issue, the CPSC’s

staff initiated a seven-laboratorycomparison test program. The mainpurpose of the study was to determineif there are statistically significant meandifferences in test results when usingmonorail and guidewire test rigs understandardized testing conditions.

Seven laboratories participated in thetest program, including the CPSC lab.Five of the laboratories tested on bothmonorail and guidewire rigs. Twolaboratories only tested on monorailrigs. Three different helmet models wereused. Each helmet was impacted twice,once at the rear of the helmet and oncenear the crown. Tests were conducted

using flat and curbstone anvils, and alltesting was performed with ambient-conditioned helmets. This experimentallowed the analysis of the effect of thefollowing variables: rig type, anvil type,helmet model, laboratory, anvil impactsequence, and impact location.

The statistical analysis of theinterlaboratory results showed that forthe majority of variable combinations,the choice of test rig did not have anappreciable effect on test results.However, on the Model I helmets, andonly when the second impact was onthe curbstone anvil, the monorailshowed a significantly higher meanlogarithm for peak-g readings summedacross laboratories having both types of test rigs. For reasons completelyunrelated to these test results, acurbstone impact in combination withanother impact on any single test helmetis no longer permitted in the finalstandard. Since the interlaboratory data(summed across the laboratories thatused both types of test rigs) show nosignificant differences betweenguidewire and monorail rigs under testconditions within those allowed in the

8/9/2019 10mr98r (1)

10/37


15Although the draft I SO/DIS 6220–1983standard was never adopted as an internationalstandard, it has become a consensus nationalstandard because all recent major voluntarystandards used in the United States for testingbicycle helmets establish their headformdimensions by referring to the draft ISO standard.

final standard, the standard allowseither type of rig to be used for impactattenuation testing.

Over the last 15–20 years, voluntarystandards in the U.S. have allowed bothmonorail and guidewire types of testrigs. Both types of test rigs have beenused extensively in independent test

laboratories and in manufacturers’ in-house test facil ities. The Snell MemorialFoundation, one of the establishedhelmet test organizations in the U.S.,uses guidewire rigs to test conformanceto their standards. The Commission hasno evidence that the allowance of bothtypes of test rigs in voluntary standardshas resulted in a compromise of safetyfor bicycle helmet users.

For the reasons discussed above, theCommission concludes that both typesof rigs are suitable for impactattenuation testing. Therefore, the finalCPSC standard specifies that either a

monorail or a guidewire test rig may beused.

b. Accuracy Check

After evaluating the results of themulti-lab testing, the Commissionconcluded that the instrument systemcheck procedure should include aprocedure for calibrating the accuracy of a test rig. Therefore, the final ruleincludes a precision and accuracyprocedure, so that laboratories canverify that their test equipment isrecording accurately. The procedurerequires that an aluminum sphere

(spherical impactor) of a specifieddimension be dropped with a certainimpact velocity onto a ModularElastomer Programmer (MEP). A MEP isa cylindrical pad of polyurethane rubberthat is used as a consistent impactmedium for the systems checkprocedure. Pre-test and post-testimpacts on an MEP to verify systemrecording is a standard practice of bicycle helmet test labs. All recordedimpacts must fall within the range of 380 g to 425 g. In addition, thedifference between the high and lowvalues of the three recorded impacts

must not be greater than 20 g. The range of 380 g to 425 g represents

an allowable tolerance of about 10%. The interlaboratory testing showed thistolerance to be attainable betweenlaboratories. However, test experienceshows that even greater precision can beobtained for the systems checkprocedure within a given laboratory. The test data from the interlaboratorystudy show that a target range of 380 gto 425 g and a precision range of 20 gcan be achieved.

c. Test Headform Characteristics

SwRI [#2] suggested that a moreappropriate value for the lower limit onthe resonant frequency of the headformmaterial should be 2000 hz instead of 3000 hz.

The important conditions for the testheadforms are the material specificationand the dimensions defined by the draftISO/DIS 6220–1983 standard.15 Thisgoal is accomplished by stating that theheadforms shall be rigid and beconstructed of K–1A magnesium alloy. Test experience shows that headformsmeeting this description will not exhibitresonant frequencies that will interferewith proper data collection. Therefore,§1203.9 has been changed to deletereference to any lower limit onresonance frequencies. The proposalalso stated that another ‘‘functionallyequivalent’’ metal could be used as theheadform material. This alternative hasbeen eliminated in the final rule to

specify the headform apparatus asprecisely as possible and ensure againstthe use of materials that may influencethe test results.

Dr. Richard Snyder, President of theGeorge Snively Research Foundation[19], referenced two studies that relatedhelmet fit to head size and shape. Thefirst study was conducted by Dr. BruceBradtmiller of the AnthropometryResearch Project, Inc. Dr. Bradtmilleralso responded to the proposed rule[20]. He concluded that, for properchild-helmet sizing, head breadth andlength variables were more accurateguides than using age or headcircumference. Dr. Bradtmiller urgescaution in basing the CPSC’s rules forchildren’s helmets on the draft ISO DIS6220–1983 standard for test headforms. The study shows variation in the ratioof head length to head breadth. Thisratio was found to be the primedeterminant for helmet fit. The ISOstandard, however, maintains a constanthead breadth/length ratio. A secondstudy also concluded that headcircumference was not always a goodindicator for helmet fit.

ISO headforms are the establishednorm for headgear testing in the U.S.,

Canada, Europe, and Australia. No othersystem of headforms is currentlyavailable that can be shown to preventmore injuries. Therefore, theCommission is retaining the ISOheadform specification in the finalCPSC standard. However, the

Commission’s staff will stay current ondevelopments of test procedures andequipment that could lead toimprovements in general helmet fit andin improvements that make it easier tofit and adjust helmets, especially forchildren.

d. Alignment of Anvils

The Commission amended§1203.17(a) to specify that the center of the anvil must be aligned with thecenter vertical axis of the accelerometer. This describes the already standardoperating procedure for bicycle helmettesting and is meant to preventimpacting helmets on the ‘‘corners’’ of anvils.

e. Definition of ‘‘Spherical Impactor’’

SwRI [2] suggested that it is moreimportant to specify a 5-kg combineddrop mass for the spherical impactorand the drop assembly than to specify

a 4-kg mass for the impactor itself. The Commission has adopted thissuggestion. The more precisespecifications for a spherical impactorfor use as a system check device arenow in §1203.17(b)(1), under thesystems check procedure.

9. Impact A ttenuation T est Procedur e

a. Anvil Test Schedule and Use of Curbstone Anvil

Six respondents [5, 12, 27, 29, 30, and31] submitted comments requestingchanges to the test schedule in §1203.13regarding the use of the curbstone anvil.

All of the respondents expressedconcern over using two curbstoneimpacts on a single helmet. Asproposed, §1203.3(d) and Table 1203.13did not define the conditions of thefourth impact on a helmet. The fourthimpact in the proposed standard wasleft to the discretion of test personnel,and thus could have been a secondcurbstone impact. One of thecommenters was also concerned aboutimpacting the helmet with the curbstoneanvil after the helmet was conditionedin a wet environment [12].

There also was concern about thecurbstone footprint overlapping otherimpact sites and violating the ‘‘singleimpact’’ principle of testing [27 and 31]. The length of the curbstone anvilrestricts the location of impact sites thatcan be used without overlap. The use of a second curbstone anvil, and thedamage caused by curbstone impacts,can restrict the selection of test sitesfurther, to the point where only threeimpacts without overlap may bepossible on a small helmet.

The Commission agrees that thepreviously proposed test schedule

8/9/2019 10mr98r (1)

11/37


should be revised to prevent thepossibil ity of striking a test helmet withmore than one curbstone impact. Thepotential for overlapping ‘‘footprints’’ of curbstone impacts combined with otherimpacts on a single test helmet goesbeyond the intended principle of asingle impact for a given area. TheCommission disagrees, however, with

those commenters who recommendedthat only ambient-conditioned helmetsbe subjected to a curbstone impact. Toensure adequate protection againstimpact against curbstone-type shapes,tests for that anvil, as well as the othertest anvils, should be carried out in allof the environmental conditionsprescribed by the standard.Accordingly, revised §1203.13 and Table 1203.13 contain a revised testschedule to incorporate a singlecurbstone impact on each of four‘‘clean’’ helmet samples, one from eachof the conditioning environments.

The Commission’s staff discoveredduring testing with the curbstone anvilthat severe physical damage— namelysplitting of the helmet from the impactpoint to the edge of the helmet— couldoccur even though the impact did notexceed the 300 g criterion. This led toconsideration of whether in such casesthe curbstone anvil test should berepeated on another sample to helpensure that other helmets will not failthis test.

The Commission acknowledges that,when marginal or unusual results occurin any of the standard’s tests, retestingmay be appropriate, even though the

300-g criterion is not exceeded. Otherconditions that may prompt theCommission to undertake verificationtesting include (but are not limited to)peak-g readings that are very close to the300-g failure criterion. However, sincethe option of additional testinginherently exists, it is not necessary toinclude a provision requiring suchretesting in the standard.

b. Definition of ‘‘Comfort Padding’’

The proposed definition of comfortpadding included the statement: ‘‘Thispadding has no significant effect on

impact attenuation.’’ SwRI [2]commented that fit padding may havesome influence on impactcharacteristics.

The Commission agrees with thiscommenter and deleted this statementfrom the definition.

c. Testing on More Than One Headform

In the revised proposal, the standardwould have tested a helmet on all sizesof headform on which it fit. ‘‘Fit’’ wasobtained if it was not difficult to put thehelmet on the headform and the

helmet’s comfort or fit padding waspartially compressed.

PHMA [29] recommended that thesituation where more than oneheadform will ‘‘fit’’ a helmet should beaddressed by specifying the use of thelargest headform that wi ll accommodatethe helmet, with comfort paddingadjusted to optimize the fit.

The Commission concludes that it isappropriate to simplify the testprocedure by testing on only one sizeheadform. This is consistent with thecurrent interim mandatory standards.However, in contrast to the commenter,the Commission believes that it is moreappropriate to test on the smallestheadform that is appropriate for the testsample. The Commission believes thatthe smaller headform will represent themore stringent test condition for thepositional stabil ity test. Testing on onlyone size headform will lessen thenumber of test samples needed to testcompliance to the standard.

Therefore, a helmet shall be tested onthe smallest of the headformsappropriate for the helmet sample. Thissize headform is the smallest headformon which all of the helmet’s sizing padsare partially compressed when thehelmet is equipped with its thickestsizing pads and positioned correctly onthe reference headform.

Bell Sports [12] remarked that, wherea helmet will ‘‘fit’’ more than oneheadform size, choosing theconditioning environment for testing onthe larger headform(s) that produced thehighest g-value in the test on the

smallest headform that the helmet fitsdoes not necessarily provide the worstcase. The commenter recommended thatthere be four impacts in anyconditioning environment chosen by thetest technician. As explained above, theCommission is not going to test a givensize helmet on more than one headformsize. Accordingly, this comment is nolonger applicable.

d. Number of Helmets Required for Testing

Four respondents commented on thenumber of helmets required for testing

when the helmet includes attachments,(e.g., removable visor, face shield) andpossible combinations of attachments[5, 12, 29, and 30]. They expressedconcern that the proposed standardrequires too many production helmetsamples to be tested. One respondent[12] offered suggested amending§1203.7(b) to include the statement that‘‘Helmets can be tested with anycombination of accessories.’’

Section 1203.7(a) of the proposedstandard requires helmets to be ‘‘testedin the condition in which they are

offered for sale.’’ Additionally, they arerequired to pass all tests both with andwithout any attachments that may beoffered. To adopt the suggested wordingwould not maintain the requirementthat helmets would meet the standardwith all combinations of accessories.However, the Commission agrees wi ththese commenters that it may be

impractical and unnecessary to specifyan additional set of eight test helmetsfor each added attachment and eachcombination of attachments in order totest for compliance with the standard.

To address this issue, the Commissiondecided to specify that attachmentsneed be tested only when they can affectthe test results, and that even then onlya ‘‘worst case’’ combination of attachments need be tested. See thechanges to §1203.7(b) and§1203.12(d)(1). For example, in the caseof a removable visor that has noinfluence on the retention system

strength test, it would be unnecessary totest four helmets (one for eachconditioning environment) to that testwith the visor attached and anadditional four helmets without thevisor. However, it may be possible forattachments such as visors orfaceshields to influence tests such asimpact attenuation or peripheral vision.

10. Helm et Condit i onin g

a. Low-Temperature Environment: Temperature Range

SwRI [#2] commented that theallowable temperature range in the low-

temperature environment shouldparallel the allowable temperatureranges in the other environments.

The Commission believes it is moreimportant for the low-temperatureenvironment range to be consistent withthe current interim standards than forthe range to parallel the toleranceallowed in the other environments. Thus, this comment was not adopted.However, the proposed temperaturerange contained a typographical error. The range should have been (¥17 to¥13°C). This range is consistent withANSI, ASTM, Snell 95 and CSA

standards. This typographical error hasbeen corrected.

b. Water-Immersion Environment

Paula Romeo [26] suggested that thewater-immersion environment wasunrealistic and recommended a sprayconditioning environment.

Commission testing of both immersedand water-sprayed helmets undervarious time durations showed noconsistent trend in resulting peakacceleration levels. The immersionenvironment has the advantages of

8/9/2019 10mr98r (1)

12/37


being easier to define and of subjectingthe helmet to a uniform conditioningexposure. Since testing showed thatthese commenters’ concerns wereunfounded, the immersion method of wet-conditioning is retained.

c. Reconditioning Time

The revised proposed standard

provided that a helmet that wasremoved from its conditioningenvironment for more than 3 minutesbefore testing would be reconditionedfor 5 minutes for each minute beyondthe allotted 3 minutes before testingcould be resumed. SwRI [2] noted thatthere would be potentially no upperlimit to the exposure time to reconditiona helmet once it is removed from theconditioning environment for more than3 minutes.

The Commission agrees with thiscomment and has added a 4-hour l imitto the reconditioning time in

§1203.13(c).11. Labels

a. Label Format and Content

Two respondents [22, 23] urged theCommission to require ‘‘an appropriatesymbol to appear adjacent to thestatement of compliance on the label’’and to add wording to warn that ‘‘failureto follow the warnings may result inserious injury or death.’’

The Commission agrees that moreemphasis should be placed on thewarning labels. Accordingly, the signalword ‘‘WARNING’’ is used with the

warnings required by §1203.6(a)(2)–(5).See §1203.6(a)(6). The Commissionconcludes that the signal word will bemore effective than a symbol, and thelimited size of the inside of a helmet,and the amount of information alreadyrequired on the labels, prevents the useof both a signal word and a symbol.

The limited space also prevents usingthe additional suggested language‘‘failure to follow the warnings mayresult in serious injury or death.’’ Inaddition, this language could possiblymislead some to conclude that properuse of a helmet will always prevent

serious injury or death. Accordingly, theCommission is not requiring a warningsymbol or the suggested language that‘‘failure to follow the warnings mayresult in serious injury or death.’’

b. Use Label

The proposed standard required alabel stating ‘‘Not for Motor VehicleUse.’’ Some comments addressed thischoice of language. [Comments 11, 13,22, 26.]

Two commenters stated that ‘‘Not forMotor Vehicle Use’’ wrongly suggested

the helmet was appropriate for any useother than motor vehicles. Anothercommenter felt that ‘‘Not for MotorVehicle Use’’ allows the helmet to beused for other activities similar tobicycle riding, where no alternativehelmet exists. A fourth commenterargued that ‘‘For Bicycle Use Only’’ wasa positive statement to which users are

more likely to respond.On reconsideration, the Commission

concludes that neither the ‘‘Not forMotor Vehicle Use’’ label nor the ‘‘ForBicycle Use Only’’ label adequatelyconveys the circumstances under whichhelmets that meet the CPSC standard areappropriate. It is reasonable to assumethat helmets that are certified to theCPSC standard will also provide headprotection for roller skaters, in-l ineskaters, and, perhaps, some otherrecreational activities. In-line skatersshould not be discouraged from wearinga helmet by a label stating ‘‘For Bicycle

Use Only.’’ The Commission also believes that

consumers understand both thedifferences between bicycle helmets andmotorcycle/motorsport helmets and thatbicycle helmets would not provideadequate protection for motorsportactivities. Therefore, the ‘‘Not for MotorVehicle Use’’ label is not a critical safetymessage that should be mandated in theCPSC standard. Therefore, the finalCPSC standard does not require a ‘‘use’’label, but maintains the requirement fora certification label that informs theconsumer that the helmet is certified to

the U.S. CPSC standard for bicyclehelmets.

c. Labeling for Cleaning Products

The second proposal required a labelwarning the user that the helmet can bedamaged by contact with commonsubstances (such as certain solvents,cleaners, etc.) and that this damage maynot be visible to the user. This label isalso required to state any recommendedcleaning agents and procedures, l ist anyknown common substances that damagethe helmet, and warn against contactingthe helmet with these substances.

Several respondents [2, 11, 12, 29]expressed concern that too muchinformation about cleaning productswould be needed on the label andargued that consumers should bedirected to the instruction manual forthe list of cleaning materials.

This label is not intended to list everypossible cleaning agent that can orshould not be used on the helmet. Sincethe consumer may not always have theowner’s manual, a label on the helmetshould provide some general cleaninginstructions and warnings. The language

of § 1203.6(a)(5) has been changed tomake this intent clear.

d. Warning To Replace After Impact

[Commenters 22, 23, 26.] Somerespondents agreed with the proposedstandard’s provision that the label onthe helmet should advise consumers todestroy the helmet or return it to the

manufacturer if it is involved in animpact. Others disagreed and requestedmore guidance on whether the helmet isimpaired before a consumer has toreturn the helmet.

The variety of factors (impact surface,impact location on helmet, impactspeed, etc.) that are involved in animpact to a helmet, and the level of interaction of each factor, are socomplex that it is inappropriate toaddress them in a label. It is to theconsumer’s overall safety benefit toreturn the helmet to the manufacturer ordestroy and replace it. Accordingly, the

proposed replacement warning is notchanged.

e. Durability of Labels

SwRI [2] remarked that a requirementfor labels to be likely to remain legiblethroughout the life of the helmet cannotbe tested and could lead to differencesbetween laboratories. The PHMA [29]also expressed concern about thisrequirement, stating that it was unawareof any technology that will ensure thata sticker will stand up under 5 years of the type of exposure that a helmetreceives.

The Commission shares these

commenters’ concerns. Currentvoluntary bicycle helmet standardsrequire ‘‘durable’’ labeling or labelingthat is ‘‘likely to remain legible for thelife of the helmet.’’ These conditions arenot quantified in current standards. TheCommission is not aware of any existingperformance test method that can beapplied in this circumstance. Since arequirement for legibil ity for the life of the helmet is vague and possiblyunattainable, the Commission haschanged the requirement to require‘‘durable’’ labels.

f. Labels on Both Helmets and Boxes

The American Society of SafetyEngineers (‘‘ASSE’’) [11] and the NSKC[22] suggested that ‘‘proper fit’’information should be on both thehelmet and the outside of the box.

The Commission does not believe it isnecessary to have the actual fittinginstructions on the box, because there isno information indicating that such alabel would be effective in assuringproper fi t. However, i t is important thatconsumers be aware that helmets docome in different sizes and that proper

8/9/2019 10mr98r (1)

13/37


fit is important. A label on the boxpromoting the need for proper fit couldinform parents, before they buy thehelmet, that they need to properly fit thehelmet to the child. Therefore, the finalstandard appl ies §1203.6(a)(3) to thehelmet’s packaging, as well as to thehelmet.

12. Instruct ions for Fit t ing Child ren’s Helmets

The NSKC [22] recommended that theproposed fitting instructions toaccompany children’s helmets be in age-specific language.

The Commission believes that age-specific instructions are unnecessary. The proposed standard requires both agraphic representation of properpositioning and written positioning andfitting directions. The graphics willreach more children than would age-specific instructions, because they allowchildren of all ages to compare the way

their helmet looks with the pictures. Inaddition, graphics convey the criticalinformation to non-English-readingindividuals and illiterates. Children andadults are li kely to be better able tounderstand and appreciate pictures thanage-specific instructions. This is morelikely to effectively deliver the message,allowing both parents and children tobecome aware of the proper fit.

13. Retenti on System Str ength Test

SwRI [2] asked whether both the peakand residual displacements in the test of the dynamic strength of the retentionsystem should be measured in order to

better describe the dynamics of thesystem.

Only the peak deflection reading isneeded to determine failure of theretention system. This is consistent withexisting U.S. bicycle helmet standards. Therefore, no change to the proposedrule was made in response to thiscomment.

USC–HPRL [8] suggested that theretention system test (§1203.13(d)) bedone after impact testing. Thecommenter reasons that an accident candamage a helmet and severelycompromise the retention system. The

retention system must ensure that thehelmet remain on the head during anaccident sequence.

After considering this comment, theCommission decided to make nochanges to the sequence for retentionsystem testing. Testing the retentionsystem prior to impact testing isconsistent with the ASTM and Snellstandards. The Commission has noevidence that the test sequence in theASTM and Snell standards allowshelmets that do not have adequateretention systems.

The commenter also recommends thatthe ‘‘zero’’ position for measuringelongation be established without theproposed step of pre-tensioning thestraps with a 4-kg mass.

There is no evidence that establishingthe ‘‘zero’’ position after pretensioningthe retention system, as proposed,would allow helmets that do not have

adequate retention systems to pass thetest. Therefore, the Commission madeno changes to the procedure forestablishing the pre-test ‘‘zero’’ position.

14. Positi onal Stabil ity Test

SwRI [2] remarked that the ASTMHeadgear Subcommittee is considering a7-kg preload to set the helmet duringtesting. SwRI also asked whether a thinrubber pad should be specified to softenhigh frequency impact noise.

Testing to support the development of the positional stabil ity test was withequipment specified as proposed in theCPSC standard. Subsequent to initialASTM discussions about possiblerevisions to the proposed testprocedure, the ASTM F8 HeadgearSubcommittee decided not to modifythe pre-load and not to specify a rubberimpact pad. Therefore, the Commissionmade no change to this section.

NSKC [22] also recommends that theCommission examine the potentialinfluence that fitting pads may have onthe helmet’s abil ity to comply with theretention system requirements.

When testing for positional stabil ity,the standard instructs testers to positionand fit the helmet on the test headform

according to the manufacturer’sinstructions. This procedure mayinvolve changing the size and positionof the fit pads in order to achieve asecure fit. A similar procedure isfollowed to fit a bicycle helmet to theuser. Although fitting a helmet to ametal headform will not account for allof the human elements involved whenconsumers fi t helmets to their heads, theproposed procedure is the mostpractical approach at this time andshould help keep the helmet secureduring an accident. Therefore, nochange to the proposed standard was

made in response to this comment.15. Vert ical V ision

One commenter on the originalproposal suggested that the Commissionadopt requirements for a vertical field of vision. The Commission declined to dothis because it had no information toindicate that bicycle helmets are posinga risk of injury due to inadequateupward or downward visual clearance.

In response to the second proposal,SwRI [2] suggested that requirements forvisual clearance at the brow be

considered and that this would beespecially important for racers who ridein the crouch position. However, a browclearance requirement might, in somecases, reduce the amount of headcoverage in the brow area. Further,CPSC has no information to indicatethat bicycle helmets meeting existingstandards are posing a risk of injury due

to inadequate ‘‘upward’’ visualclearance. Therefore, the Commissiondid not add a ‘‘brow’’ visual clearancerequirement to the final standard.

16. Reflectivi ty

Some comments on the originalproposal related to possiblerequirements for helmets to improve abicyclist’s conspicuity in nighttimeconditions. Data do show an increasedrisk of injury while bicycling duringnon-daylight hours. The Commissionindicated that it would study this issuefurther in conjunction with plannedwork on evaluating the bicycle reflectorrequirements of CPSC’s mandatoryrequirements for bicycles. 16 CFR part1512. The Commission stated that itwould decide whether to proposereflectivity requirements for bicyclehelmets under the authority of theBicycle Helmet Safety Act after thatwork is completed.

Several commenters on the revisedproposal [1, 7, 11, 13, 16, 17, 22, 23, 24,26] urged that the Commission notpostpone implementing bicycle helmetreflectivity requirements.

Since the revised proposal, theCommission conducted field testing on

bicycle reflectors and examined theissue of reflectivity on bicycle helmets.In the field testing, half (24/48) of thesubjects were tested using bicycle riderswith reflective helmets and the otherhalf were tested using riders wearingnon-reflective helmets. The reflectivetape used on the helmets met aproposed Standard on use of Retroreflective Materials

10mr98r (1)

Documents

Transcript of 10mr98r (1)