Recent Research ch on Lightning, with Implications f s for Air Terminals William William am Rison am Rison Professor of Elect ectrical Engineering New Mexico Institute of of Mining and Technology Socorro, Ne , New Mexico
Recent Lightning ing Research with Implications for for Air Terminals � Computer Modeling of t f the Lightning Attachment Process � High Speed Videos of th f the Lightning Attachment Process Process
Video by Tom Warner, June 17, 2007, ne 17, 2007, near Devil’s Tower, Wyoming High Speed eed Video of Negative Cloud to G to Ground Discharge 7,100 Frames es per Second
Video by Tom Warner, June 17, 2007, ne 17, 2007, near Devil’s Tower, Wyoming
Video by Tom Warner, August 4, 2007, ne 4, 2007, near Rapid City, South Dakota High Speed eed Video of Attachment P t Process for a Negative Cloud to G to Ground Discharge 7,100 Frames es per Second 500 foot tall tower
History of Attempt pt to Pass a Standard for ESE Air ir Terminals � 1980’s --- Laboratory ex experiments indicate 100 µ s time advantage fo e for ESE air terminals � Early 1990’s --- Push by by ESE manufacturers for adoption of an ESE stand adoption of an ESE stand andard andard � 1990’s --- Prolonged pro procedural and legal fight over proposed NFPA sta standard � Early 2000’s --- NFPA d decides to reject ESE standard and retain NFP FPA 780 � 2005 through 2008 --- Fe Federal court rules against ESE manufacturers
History of Attempt pt to Pass Standard for ESE Air ir Terminals � ESE based on three assu sumptions antage of about 100 µ s will also Laboratory time advant 1) present in real lightning ing Initial early streamer w Initial early streamer w will develop into successful will develop into successful 2) 2) upward leader is about 1 x 10 6 m/s Upward leader speed is 3) � If these three assumption ions are true, they will give ESE terminals a 100 met eter capture radius � No input from scientific fic community on development of standard ard
History of Attempt pt to Pass Standard for ESE Air ir Terminals � No scientific consensus us on ESE assumptions at when the standard was p s proposed
Validity of ESE SE Assumptions 100 µ s also present in real lightning Laboratory time advantage of 100 1) No scientific research indica ndicating that this is the case • Golde (1941), “The Validity dity of Lightning Tests with Scale • Models” “there is a question regar garding the validity of any laboratory • test designed to simulate te the physics of the lightning stepped leader and the attachment ent process, since typical leader step lengths, tens of meters, e , exceed the size of the laboratory experiment” (Rakov & U Uman, 2003) Becerra and Cooray (2008), 2008), “Laboratory Experiments Cannot • be Utilized to Justify the Act ction of Early Streamer Emission Terminals ”
Validity of ESE SE Assumptions 2) Initial early streamer w r will develop into successful upward lead ader No research at time sta standard proposed • Moore et al (2000) “Me Moore et al (2000) “Me Measurements of Lightning Measurements of Lightning • • Rod Responses to Near earby Strikes” shows early streamers ineffective
Validity of ESE SE Assumptions Early streamers do not develop into leaders
Validity of ESE SE Assumptions d is about 1 x 10 6 m/s 3) Upward leader speed is Not many measuremen ents of upward leader speeds at • time standard proposed sed Now, several measurem Now, several measurem rements of upward leaders show rements of upward leaders show • • speed of about 1 x 10 5 m/s Tom Warner tower stri trike video: average speed of • 1.2 x 10 5 m/s Speed of 1 x 10 5 m/s gi gives only 10 meter capture • radius
Validity of ESE SE Assumptions Research since the ESE SE standard was proposed • shows all three ESE as assumptions are incorrect There have been no ind independent field studies • showing ESE terminals showing ESE terminals als are effective als are effective There is now scientific fic consensus that ESE • terminals are ineffectiv tive It was premature to ado adopt a standard for ESE • terminals in early 1990 90’s
CVM H History Idea initially proposed by y Eriksson (1979) • Theory was further develo eloped by D’Alessandro and • Gumley(2000) and D’Ales lessandro (2002) Tested by D’Alessandro in in Kuala Lumpur and Hong • Kong (Petrov & D’Alessan Kong (Petrov & D’Alessan sandro, 2001, D’Alessandro sandro, 2001, D’Alessandro and Petrov, 2006)
CVM Assu ssumptions (D’Alessandro and and Gumley, 2000) Electric field intensificatio tion by structures affects 1) formation of upward leade ader Critical radius of 38 cm fr from laboratory measurements 2) “Any sharp geometrical fea “Any sharp geometrical fea features with a radius of less that of features with a radius of less that of a) a) the critical value must be “ be “rounded off” to this value” “[T]he criterion for the init nitiation of corona and consequently b) a stable upward leader is the s the attainment of 3.1 MV/m at the critical radius” Ratio of velocity of downw nward leader to upward leader 3) is about 1.2
CVM Assu ssumptions (D’Alessandro and and Gumley, 2000) Electric field intensificatio tion by structures affects 1) formation of upward leade ader This is correct, from basic ic freshman physics •
CVM Assu ssumptions (D’Alessandro and and Gumley, 2000) Critical radius of 38 cm fr from laboratory measurements 2) “Any sharp geometrical fea features with a radius of less that the a) critical value must be “rounde ounded off” to this value” Moore et al (2000) showed Moore et al (2000) showed ed air terminals with radius of ed air terminals with radius of • • much less than a few centi ntimeters behave quite differently from those with ith radius of a few centimeters
CVM Assu ssumptions (D’Alessandro and and Gumley, 2000) Critical radius of 38 cm fr from laboratory measurements 2) “[T]he criterion for the init nitiation of corona and consequently b) a stable upward leader is the s the attainment of 3.1 MV/m at the critical radius” It is necessary for the field bet It is necessary for the field bet between the “critical radius” and the between the “critical radius” and the • downward leader to be strong ong enough to sustain propagation No field measurements to confir firm this • Becerra & Cooray (2007) mode odeled development of upward • leader and found this not to be o be the case: “It is shown that the collection v n volume concept overestimates the • lightning protection areas of air air terminals placed on complex structures”
CVM Assu ssumptions (D’Alessandro and and Gumley, 2000) Ratio of velocity of downw nward leader to upward leader 3) about 1.2:1 • During initial upward leader, velocity ratio is about 10 to 1 about 10 to 1 • Overall, velocity ratio is about 3 to 1 • D’Alessandro: “[A] higher velocity ratio yields a smaller attractive radius
CVM Assu ssumptions (D’Alessandro and and Gumley, 2000) Ratio of velocity of downw nward leader to upward leader 3) is about 1.2 During critical initial leade ader development (first 100 feet • of upward leader) velocity ity ratio is about 10:1 This is when other objects ts on the structure are competing • with the air terminal D’Alessandro & Gumley ( y (2001) • “In fact, if one attempts to to use a ratio of 4 [or higher], Eq. (9) • does not have a real solution.” ution.”
Field Validat ation of CVM (Petrov & D’Ale lessandro, 2000) (D’Alessandro an and Petrov, 2006) Studies in Hong Kong and and Kuala Lumpur, Malaysia • Buildings protected by ER Buildings protected by ER ERICO Dynaspheres placed ERICO Dynaspheres placed • • according to CVM calcula ulations Claim to show protection on levels in the 85-98% range • 2 to 15% of low intensity f ty flashed under 10 kA could bypass • lightning protection system tem
Field Validat ation of CVM (Petrov & D’Ale lessandro, 2000) (D’Alessandro an and Petrov, 2006) Study in Kuala Lumpur • • Data in study has been disp • Data in study has been disp sputed by Hartono & Robiah of sputed by Hartono & Robiah of Lightning Research Pte. Ltd Ltd. of Kuala Lumpur “ This review shows that the fie field data given in the above study is invalid and should not be us used to validate the CVM method of air terminal placement.” • No independent verification of of CVM method
Field Validat ation of CVM (Petrov & D’Ale lessandro, 2000) (D’Alessandro an and Petrov, 2006) Two problems with methodolo ology of field study: Two different techniques Two different techniques es were tested with one study, es were tested with one study, 1) 1) with no way to differentia tiate between the two There was no control 2)
Field Validat ation of CVM (Petrov & D’Ale lessandro, 2000) (D’Alessandro an and Petrov, 2006) Two different techniques es tested with one study, with no 1) way to differentiate betwe ween the two Study used ERICO Dynas naspheres placed according to • CVM technique If there was an effect, was as it due to the Dynasphere or • to the CVM?
Recommend
More recommend
