Original Article Assessment of second-order clearances between orthodontic archwires and bracket slots via the critical contact angle for binding Robert P. Kusy, BS, MS, PhD; John Q. Whitley, BS Abstract: Twenty-six archwires and 24 brackets were selected from among the hundreds of products available that nominally have from18t022 mil bracketslots and 14, 16,17, 18, 19,and/or21 milarchwiresizes. After the archwires and brackets were dimensioned, ‘a minimization- maximization algorithm was applied to the measurements in order to establish the likely boundaries ofthe critical ‘contact angle for binding (0,) a8 defined by the presence and absence of second-order clearance. From among the myriad archwire- bracket permutations possible, 64 combinations were identified —20 using thebracket slots the controlling dimensionand 44 using the bracket width. Usinga previously derived mathematical expression that relates the dimensions of each archwire-bracket couple toitscalculated 8, the corresponding sets of indices were plotted. The results show that the maximum value of thecalculated 8, can never exceed aboiit 5”, or else sliding mechanics will always be hampered. Other outcomes were validated experimentally using 5 of the 64 archwire-bracket couples by measuring the resistance to sliding (RS) at 15 different contact angles (@) ranging from O=0" to 2° and by subsequently determining a measured 0. These values agreed with the calculated 0, values, When the practitioner knowsthe 8, treatment timemightbe reduced because the tecth donot need tobe over-aligned prior toemploying sling mechanics (ce, by not making #<<8,) and, further, because the contact angle beyond which the binding phenomenon retards or halts tooth movement does not need tobe excecded (ic. by not making #20.) These results underscore the importance of exact wire and bracket dimensions on packaging; otherwise, sliding mechanics can be compromised by miscaleulating 8, Key Words: Archwires, Brackets, Clearances, Contact angles, Sliding mechanics thin the past 30 years, sev- eral investigators have shown that resistance to sliding increases substantially as the contact angle (0) between an archwire and a bracket slot in- creases.'* From such results, Frank and Nikolai? and others'§”* showed that as 0 increased, resistance to slid- ing was greater with stainless steel than with titanium alloy wires. Al- though other experimental observ. tions have been documented, no rigorous mathematical approach or scientific principle has been forth- coming that relates the specific di- mensions of _archwire-bracket couples and their angulations to slid- ing performance, In 1998, a theory was reported that was based on the relative geometry of the archwire-bracket couple.” This theory showed that the @ at which opposing bracket tie-wings engaged the archwire and the second-order clearance reduced to nil could be de- scribed by just three geometric pa- rameters (Figure 1): the dimension of the archwire that engages the floor of the slot (“size”), the corresponding bracket dimension at the floor of the slot (“stor”), and the mesiodistal width of the bracket (“wort”). The closed-form solution for this critical contact angle for binding (@,) could be expressed as ow cog LSE) (worn)? : (ze) (uor) = @P in which Z = (wom? [(s2e) + (suor} + (woral) @) A mathematically simpler, but a) Author Address Professor Robert P. Kusy University of North Carolina DRC Building 210K cB 7455 Chapel Hill, NC 275 emai: rkusy@bmeuncedu open-form solution of eqns. (1) and (2) was subsequently derived as SE. Going) +c080, (in the closed-form solution, 8, can be obtained directly by substituting the values of the three geometric pa- rameters into eqns. 1 and 2. In con- trast, in the open-form solution, only trial and error can be used to solve for 8,in eqn. 3, Fither solution is ex- act, however.) From eqns. 1 and 2 or eqn.3, 8, was, plotted as a function of two dimen- RP. Kusy, Derital Research Conter, Schoo of Dentistry: Department of Orthodontics, Schoo! of Dentistry: Department of Biomedical Engineering, Schoo of Medicine: Curricu- umm in Applied and Materials Sciences, University of North Carolina, Chapel Hill, N.C John Q. Whitley, Dental Research Center, School of Dentistry, University of North Carolina, Chapel Hi, NC. Submitted: November1997; Angle Orthod 1999,69(1)71-80, Revised and accepted: une 1998 ‘The Angle Orthodontist Vol. 69 No, 1 1999 71Kusy; Whitley Tablet Archwires measured ‘Company (location) “Archwirename Materia” Nominal ‘Actual Tagged 2¢ (mis) ze (mils) archwires ‘American Orthos, Gold Tone Round ss 6 15.60:+0.08 c (Sheboygan, Wisc) Standard Edgewise ss 21x25 20.990.02 Standard Round ss 14 43.941 0.02 Dentaurum Remanium 88 16 18774003 (Plorzheim,Germany) Remanium $s 16x22 16204003, d Remanium $s 17x25 17,180.08 f Remanium ss 19x25 19184011 i Remanium ss. 21x25 21.7540.00 ' ac Nubryte Gold ss 4 14.10 0.03 » (Central Isip, NY) OrmeoiSybron Round 8s 4 13.9810.03, (Glendora, Cali) Found ss 16 16.20+0.03 Round ss 16 18.18+0.03, h MO Elgiloy—Blue Cocr 16x16 18.8540.11 (Denver, Colo) Elgloy—Be Co-Gr 21x25 20.8440.08 k Elgiloy—Green CoGr 1 17.60+0.04 9 Elgiloy—Fed CoGr 4 19.79+0.02 Elgiloy—Yellow Cocr 16x22 15.99+0.09 Elgiloy—Yellow CoCr 17x25 16.73+0.05 Eigloy—Yellow CoCr 18x25 17.88 +0.04 Eigiioy—Yellow CoCr 19x25 18.48 0.08 i Tru-chrome ss 4 13.77 40.03, a Tru-chrome ss 17x25 16.23 0.03 e UnitekioM StandardRectangular SS 17x25 16.90:0.04 (Monrovia, Calf StandardRectangular SS 16x25 15.98 0.15 StandardRectangular $$ 19x25 48.99 +0.06 Standard Square ss. 16x16 16.11 £0.08 * $8 =stainless steel iron, chromium, and nickel alloy) Go-Gr= cobalt-chromium alloy sionless parameters: the engagement index (s12#/stor)—that is, how well the wire size fits in the bracket stot; and the bracket index (wioT#/sto1)— that is, how much larger the mesio- distal bracket wior dimension is than its coronal-apical stor dimen- sion (Figure 2). For nominal archwire sizes (14, 16, 17, 18, and 19 mils), nominal bracket sLors (18 and 22 mils), and assumed bracket wir (125 mils for 3s, 4s, and 5s, and 250 mils for 1s), the limits of @, were cal- culated and ranged from 8,= 0° to 0 4.5°, This outcome indicated that if the practitioner wanted to use slid- ing mechanics without any binding, he or she always had to align and 72. The Angle Orthodontist Vol. 69 level teeth specifically within this aforementioned envelope. Moreover, if the practitioner exceeded the char- acteristic value of 8, for a given wire- bracket couple, sliding would be increasingly compromised as @ in- creasingly exceeded @. Having established the mathemati- cal relationship, the present effort seeks to determine the limits of 8. values from actual archwire-bracket couples of 3s, 45, and Ss. A surpris- ing shift is observed because, for whatever reason, the actual sizes and actual stors are not always what ven- dors label them to be. Ultimately, ex- perimental measurements from five ‘wire-bracket couples confirmed that No. 1 1999 these measured @, values are in agreement with the calculated @, val- ues that were obtained from the pre- viously derived eqns. 1-3. Materials and methods “Archwires and brackets ‘Twenty-six archwires were ran- domly selected from among six ma- jor vendors and two major alloy groups (Table 1), These selections in- cluded at least one wire from each of the nominal wire sizes. The actual sizes of the wires that engage any bracket were measured to the near- est 0.01 mil at six locations using a Sony \-Mate micrometer (Sony Magnescale America, Ine, Orange, Calif).Second-order clearances between archwires and bracket slots Table2 Brackets measured ‘Company (location) Bracketname Matera” Nominal Actual ‘Actual “Tagged stor (mils) sor (mils) wor (mils) __brackats “A"-Company Starfico sca 18 1a2202 121.000 (San Diego, Cali) Starfire SCA 2 22as02 121.0400 ‘American Orthod. 20120 PCA 18 1290200 — A(7.25)" (Sheboygan, Wisc) 20/20 PCA 2 4280400 (6.00) Dentaurum Fascination PCA 18 207405 — 1360+00 (Pforzheim, Germany) Fascination PCA 2 236405 137000 Rematitan (7°T""*) Ti 18 209403 1425405 8 (6.82) Rematitan (17°) Ti 18 201403 1430400 —O(7-11) Ustra-minitin ss 8 204402 105.0400 015.15) Uttra-minitvie ss 185 195401 148.3206 Uttra-minitrim (7°T) SS 185 186402 180.000 Uttra-minitim (17°T) SS 185 199401 151.500 Untra-minitin ss. 2 237402 146.000 (6.16) ac Allure i Poa. 22 228201 1485200 (Central isip, NY) Ormoa/Sybron Lumina Twin PCA 18 1277403 (Glendora, Calif Lumina Twin PCA 2 1315400 MO Quasar POA 18 128.5400 (Denver, Colo) Quasar PCA 2 190.000 TP onhod, Advantedge PCA 2 228211 1883203 (6.86) (LaPorte, Ind) Unitek/ot Mini Uni, Twin ss 2 232400 121.0400 + -G(6.22) (Monrovia, Cali) New Ceramic. PCA 8 teor0s «© 1427206 Transcend6000 PCA 8 183205 141.0200 Transcend 6000 PCA 2 226101 — 139.940.0 Victory Mini Twin ss 18 183202 139.0200 . ingle crystal alumina, aka. sapphire elyerystaline alumina ‘T= commercially puro tkanium ‘8S = stainless stoo! ron, chromium, and nickel alloy) Bracket Index - wora/s.r, as determined by actual dimensions reported here. "T= pretorqued bracket To complement these archwires, 24 brackets were selected from among eight major vendors and four major material groups (Table 2). Both stan- dard bracket siots were included as well as three 18.5 stots, The actual suors of the brackets were measured to the nearest 0.1 mil three times on each side for a total of six measure- ments per bracket, using the optics of a Kentron microhardness tester (Kent Cliff Labs, Peekskill, NY). The actual wits of these brackets were mea- sured to the nearest 0.1 mil three times using Starrett calipers (LS Starrett Co, Athol, Mass) ‘Minimization-maximization algorithm Each size, sLot, and wiotit measure- ment was reported as a mean plus or-minus its standard deviation in mils, Having made the actual mea- surements (Tables 1 and 2, columns labeled “Actual”), minima and maxima were identified that corre- The Angle Orthodontist sponded for each nominal bracket stor (18 or 22 mils) to the nominal archwire siZés available (14, 16, 17, 18, 19 and/or 21 mils). These archwires and brackets are labeled “Tagged” in the right-hand margins of Tables 1 and 2 as lower case and upper case letters, respectively. Note that, with regard to archwires, six nominal wire sizes and hence six minima (labeled a, c, ¢, gi, and k) and six maxima (labeled b, d, f, hj, and 1) are possible. With regard to Vol. 69 No.1 1999 73Kusy: Whitley brackets, however, two minima (la- beled A and E) and two maxima (la- beled B and F) are possible for each bracket if the stor dimension con- trols; whereas two minima (labeled C and G) and two maxima (labeled D and H) are possible for each bracket if the wiotH dimension con- trols. Thereby, the number of permu- tations is eight, unless a minima or maxima stor dimension coinciden- tally corresponds to a minima or maxima wipri dimension. In Table 2, the actual bracket indices (i.e., the ‘wsmi/s.or ratios) are shown in pa- rentheses alongside the tagged brack- ets. Because the minima and maxima of actual stot and actual wiorit do not coincide, separate analyses of stot values A, B, E, and F are required, while allowing the winris to assume their corresponding actual values. Similarly, separate analyses of wits values C, D, G, and H are required while allowing the stots to assume their corresponding actual values. Discounting the eight combinations in which the wire size will not engage the bracket stor leaves 64 viable com- binations: Using stot values as the controlling dimension— *With the A brackets, b, d, and *With the B brackets, a, ¢ @, 8, i and k wires *With the E brackets, b, d,f,h, and With the F brackets, ac, and k wires Using wiors values as the controlling dimension— *With the C brackets, a,b, ¢ dye, f, gh. i, and j wires + With the D brackets, a, b,¢, def g-h,i, and j wires + With the G brackets, a,b, ¢ d, ghyi,j, k, and | wires + With the H brackets, 9, b,c d, ef g hyi, j,k, and | wires ‘The above combinations include all of the A, C, G, and H permutations that also represent the minima and 7” ‘The Angle Orthodontist Vol. 69 ‘SIZE’ ‘WIDTH’ Passive Active Configuration Configuration \ | lof Figure t Geometric parameters of importance duting sliding mechanics: suze of archwire, s.oT ‘and work of bracket, and contact angle between archwire and bracket (0). The left hhand tooth illustrates the passive configuration in which second-order clearance exists between opposing tie-wings: the right-hand tooth ilustrates the active configu ration in which clearance not only no longer exists but @ actually excoeds the extcal Contact angle for binding, 8, (not shown, but evidenced by the intemal angle of the Wire relative tothe bracket w-oTH being less than the external 8 shown). Note that in this case, the direction of tooth movement is critical, because ifthe right-hand tooth hhad been moving in the opposite direction, then 8 would equal 0, 1.0 og 08 o7 06 2.5" Engagement Index, (SIZE/SLOT) 05 4 6 8 10 12 Bracket Index, (WIDTH/SLOT) 14 Figure 2 ‘Actual boundarias of commercial products (shaded area) as determined by the ‘minimization-maximization algorithm for the stor data (0) and the wiotx data (). n all ‘some 50 archwires and brackets were studied in 64 ofthe 40 possible combinations todetermine that the maximum 0, <5°, that the (sze/s.o1) > 0.56, and that the 4.5 < (worws.01) < 7.5, No. 1 1999‘maxima of the actual bracket indices, ie., the actual wiptk/stor ratios (Table 2). Thus, by evaluating only 64 combinations, the actual boundaries of 8can be delineated for 540 pos- sible combinations of commercial products (Tables 1 and 2) Computations All engagement index (sizz/stor) against bracket index (wioTh/stor) plots and their subsequent calculated 6, values were obtained via eqns. 1 and 2 or equivalently via eqn. 3 Experimental verification As a validation of these calcula- tions, five archwire-bracket couples were selected to represent a range of bracket and engagement indices: a 19x25 size in an 18 stor (ie, a 19x25/ 18), an 18/18, a 16/18, a 16x22/22, and a 14/22. The first represented the high end of the engagement indices (which, in this case, should have been impossible!), and the last represented the low end of the engagement indi- ces. To avoid confounding the data by varying the alloy, all couples were stainless steel (85). This decision not only made the stiffness (resistance to elastic deformation) very high but also placed the smoothest surfaces having the lowest frictional coeffi- cient against one another. The effi- ciency and reproducibility of such couples have been shown to be quite good when clearance exists, i.e, in the passive configuration (Figure pre Using the same instruments as above, archwire sizes were measured at six locations, bracket sors were measured five times on each side for a total of 10 measurements per bracket, and bracket wintns were measured three times at the widest point, Using a special frictional test- ing device" in which adjacent brackets represented by frictionless bearings were placed at a great dis- tance (16 mm) from the test bracket, the resistance to sliding (RS) was measured in the dry state at 34°C us- ing a normal force of 300 g, This nor- Second-order clearances between archwires and bracket slots ing, RS (kg) ° no a 2 2 2 £ B @ 3 « 0 0.2 4 6 8 10 Contact Angle, © (°) 1214 Figure Plot of contact angle (8) versus resistance to sliding (RS) for five stainlese steel on stainless steel (SS-SS) couples in which engagement indices (sze/s.or) and bracket indices (vacrw/s.07) differed greatly (Table 3). By inscrbing linear gression lines to the upper and lower parts ofall 6 vs, AS data, he average values of the measured ciiical contact angle (8,) were adduced. From actual dimensional measurements and ‘eqns. 1 and 2, the calculated values of @, may be compared (Table 3). When tested at a normal force (N) of 300 q in the dry state at 34°C, the average measured & values (8) ranged from a low of 8 = 0.5° for a 19x25/18 SS-SS couple (jC) to ahigh of 8. = 2.5" fora 14/22 $S-SS couple (aF) mal force was chosen for conve- nience, as @, is dependent only on geometry. The RS values of two test runs were made for cach of the five aforementioned archwire-bracket couples according to each of the fol- Towing two schemes. In the first, @ began at 0° by rotating the bracket until its wiovn dimension was paral- Tel to the long axis of the archwire and then proceeded in 0.5° incre- ments to 5°, in 2° increments from 6° ‘The Angle to 12°, and then rotated back to 0°. In the second, @ began at 0° and pro- ceeded in 2° increments from 12° to 6°, and in 015° increments from 5° to (0°. Because the RS values of each in- dependent test run were measured at 8 = 0° both at the beginning and at the end of each run, the measure. ments at 0° verified that no change in the bracket surface occurred as each archwire was drawn through its bracket at different values of @. Al- Orthodontist Vol. 69No.1 1999 75Kusy; Whitley Tables ‘Archwire-bracket couples selected for experimental verification ofthe calculated ‘tical contact angle for binding (2) Rechwie-brackel Test AclualAclual Actual Measured” Engagomant Bracket Calculated couples ind sue sor mom 6°) Index, index, 8) evaluated (ris) (ils) (ls) saeisior _wornsior ic 7 1910207 980 03 ose art x0 (19925118) 2 1910 205; 980 08 cost 468 08 no 1 tere 2011000 14 0.908 497 44 (iar) 2 1818 2a 998 22 0902 = 494 12 c 1 1368204 991 24 076 485 27 (iene) 2 1368 200 99, 27 0781 495 25 oF 1 16223513909 28 06s 5.94 30 (16x22) 2 16223414 27 oss = 807 29 oF 1 1370233 taat 25 0588 6.08 38 (az) 2 137002331410 at 0588 = «6.05 38 * Measured by extrapolating the regression lines of individual 8 versus RS pot. “Calculated via Figure 2, or eans. 1 and, oF ean. 3 though RS values under both static and kinetic conditions were mea- sured, only the latter were reported because they represented the average of at least 500 data points for each test run. These RS values equaled one-half of the actual measured drawing forces; and when clear- ance exists in the so-called passive configuration (Figure 1), the RS val- tues equaled the frictional forces from which the frictional coefficients could be caleulated Statistical analyses For each of the 10 test runs (two for each of the five archwire-bracket couples evaluated), a regression line was computed for each plot of RS against @, when @< 0_and when @> 6, From the intersection of each pair of regression lines, a “Measured 8,” ‘was obtained. The data of each me: sured 8, was plotted against each “Calculated 0,” and the regression line was determined. This line was compared with a second line, which presumes that the calculated 8, (i, the mathematical relationship) and the measured 8. (ie, the experimen- tal determination) are equivalent. When the data of the two RS against Vol. 69 The Angle Orthodontist @ runs were combined for each couple and the same linear regres- sion analyses were made, an average value of measured @, (8) was ob- tained. Inal cases, the probability (p) of all regression lines were deter- mined from the correlation coeffi- cient (1) and the number of date points (n) Results Examination of archwire dimen- sions showed that the actual size of archwires can be not only undersized (as expected) but also oversized in 30% of the 26 wires investigated (Table 1). In about 15% of the bracket measurements reported, the sLors were smaller than that nominally stated (Table 2). In some of the re- maining brackets studied, the s.o7s exceeded the nominal value by as much as 16% and 8% in the cases of nominal 18 and 22 stots, respec- tively. OF course, bracket wioms are not Specified on labels, so no nomi- nal values exist. Suffice to say that the wiorus varied by more than 50% as they ranged from 99 mils to 156 zis. In one bracket, F, the wibTis ac~ tually increased 9 mils from the base of the stor to the top of the tie-wings. No.1 1999 The adduced boundaries of com- mercial products (Figure 2, shaded area) show that 8, ranged from 0° to about 5°. The corresponding engage- ment and bracket indices ranged from about 0.55 to 1.0 and from 4.5, t075, respectively. Inthe final analy- sis, the minimization-maximization algorithm for the stot values del cated the top, right, and bottom boundaries of the bracket index-en- gagement index plot. Only the left- hand boundary was delineated by the algorithm based upon the wibr# values, The regression lines of @ versus RS were highly significant (p < 0.001 in Figure 3). For replicate test runs of five archwire-bracket couples, these plots show that, as the stor was pro- gressively filled, the value of 8, oc curred earlier; consequently, the binding that ensued was more se vere. For example, for the couples aF and jC, a distinctive break occurred between 3.1° > 8, > 25° for the mea sured 0, of the 14/22 couple and be- tween 0.8° > 6.> 0.3° for the measured 0, of the 19x25/18 couple (Table 3). As the value of @increased to 12%, the values of RS increased fourfold, from 83 g for aF to 329 g forjC (Figure 3), From calculations of the engagement and bracket indices based on the actual measurements of these specific archwires and brackets (calculated 0.5 in Table 3), Figure 2 (which embodies eqns. 1 and 2 or ‘eqn. 3) indicated that the average ®. or 8, = 38° for the 14/22 couple and @, = 0.9° for the 19x25/18 couple. As Figure 4 shows, the linear regression (solid line) of the calcu- lated versus measured 0, values were significant (r = 0839 and n =10 for a pevalue < 0.01) and generally ap- proximated a 1:1 relationship (dashed line). Discussion Influence of archwire-bracket dimensions Wires and brackets should be de- jgned so that the wire of one vendor will universally fit into the bracket of another. To accomplish this, wire sizés should be, in engineering toler- ance notation, plus zero and minus some value. To complement the wires, the bracket stots should be minus zero and plus some value. In this way an interference fit would never occur between a wire size of the same dimension as a bracket stot, even though the stor is filled. Other engineering tolerance schemes will work, 100, but for all schemes the di- mensions should be such that the ‘mean plus three standard deviations of the archwire size does not exceed the mean minus three standard de- viations of the bracket stor. Such a standard would provide a 99% con- fidence level. Clearly this is not be- ing done today, as the sizss of some wires exceed the nominal stor di- mensions and the stors of some brackets are smaller than the nomi- nal sior dimensions. The stots of the foreign vendors may be somewhat larger than those of the domestic ven- dors, £00, perhaps because they are using metric tooling and targeting 05 mm (ie,,19.7 mils) for the 18 mil stor and 0.6 mm (ie,, 23.6 mils) for the 22 mil stot (Table 2). For all brackets, Second-order clearances between archwires and bracket slots Measured @, (°) Calculated 0, (°) 3 4 5 Figure 4 Linear regression of calculated 8, versus measured 8, values for replicate test runs of {iva SS-SS couples at N = 300 qin the dry state at 34°C: [C, NC, eC, dF, and aF (Table 3). From the correlation coefficient and the number of data points evaluated, p ‘<0.01, thus further substantiating the validity ofthe earlier derived results."" The dashed line (---) denotes the 1:1 relationship. wiotHs must be specified. From the perspective of sliding me- chanics and the present analysis, having a larger stor is not necessar- ily bad, providing that the strengths and stiffnesses of the tie-wings and retention pads remain adequate. As the engagement and bracket indices decrease in magnitude (Figure 2, shaded area), the allowable 0, creases. All other parameters being equal, increasing the stot actually ameliorates binding as the stor re- duces the engagement and bracket indices. Reducing the size and ywiorit cof wires and brackets, respectively, also provides some benefits in in- creasing the allowable value of 0, ‘The wire size can be reduced as long as the strength, resilience, and force per unit of deactivation are sufficient; the bracket wiori can be reduced as long as control of each tooth can be maintained. Therefore, these obser- vations—that vendors have not only made larger bracket stots but also have generally made smaller wire sizes and bracket wiorits—facilitates the initiation of sliding mechanics as they obviate the need for as precise initial leveling and alignment. This can be readily seen in Figure 5 by comparing how the boundaries of commercial products have generally expanded and shifted (the shaded area) relative to the nominal values calculated elsewhere" for 3s, 4s, and 5s (the inscribed area). Thus, shifting the nominal dimensions of an archwire-bracket couple from the in- scribed area to an actual location that is down and/or to the left (Figure 5) produces a couple that is capable of sliding at a higher @,. Based on the nominal values for size, stor, and \win (125 mils) of the five couples that were used for experimental veri- fication (Table 3), greater shifts were observed for those wires that were tested against 18 mil than 22 mil brackets. ‘The extreme case was jC, Which should be impossible with a nominal engagement index greater than one. The calculated @.s of each of these three couples (jC, hC, and eC) increased over 1°, Using the first run of the 16/18 couple (cC) as an ex- ample, the engagement and bracket ‘The Angle Orthodontist Vol. 69 No.1 1999 77Kusy; Whitley indices for nominal dimensions are 0.889 and 6.94, respectively. These would be plotted in the upper right corner of the inscribed area (Figure 5), and 8, would equal 0.9°. From the actual dimensions of this couple, the engagement and bracket indices are 0.765 and 4.85, respectively (Table 3), and the calculated 8, was 27° (Table 3). For the couples tested against the calculated 22 mil brackets (4F and ak), 0, actually increased less than 0.5° more than the nominal dimen- sions predicted. Nonetheless in all clinical cases, the values of 8, will never exceed a maximum of about independent of the available hard- ‘ware, the wire technique, or the prac- titioner Presence and absence of second- order clearance Because @, represents the demarea- tion line between facile sliding on ‘one hand and restricted sliding on the other, the data shown in Figure 3 ray be shifted with respect to 8..As- suming that the average 8. or 0, equals a demarcation point above Which relative angles represent the absence of second-order clearance (cc, the active configuration; Figure 1) and below which relative angles represent the presence of second-or- der clearance (i.e, the passive con- figuration; Figure 1), Figure 6 results in two distinctive zones, When 6 < 0, and the passive con- figuration exists (Figure 6), RS is fairly constant and low in magnitude. These values are equivalent to fric- tional forces previously reported for 5-55 couples in the literature.” In this specific case, when N = 300 g in the dry state at 34°C, RS= 3545 g for a kinetic coefficient of friction (i) = 35 5/300 g = 0.12. Here the wire and bracket geometry have little ef- fect as contact areas are established and adjusted in accordance with the second law of friction." Conse- quently, these five 55-55 couples be- have similarly. This experimental outcome confirms that classical fric- 78 The Angle Orthodontist Vol. 69 e 2 & 68 Engagement Index, (SIZE/SLOT) ° os’ co 07 15° 06 su e\s\ a syeenee 05 1 1 4 6 8 10 12 14 Bracket Index, (WIDTH/SLOT) Figures Comparison ofthe actual boundaries of commercial products (shaded area) with the nominal values that were calculated (inscribed area representing 3s, 4s, and Ss from Figute 3of ret. 10). in general when the nominal sizes hea been overestimated and nominal s.ors had been underestimated during manufacturing (Tables 1 and 2) the boundaries shifted down andlor towards the lt. When the measured 8. of two replicates of five archwire-bracket couples (Table 3) ae plotted against this back: ‘top, the shit ofthe 0, from that ofthe nominal , (the Os) to that ofthe measured 8,5 (the arrowheads) increased from less than 0.5 for the 22 mil sors to over 1" for the 18 mil stors. tion dominates as binding and its as sociated phenomena are nonexist- ent. Once the ® reduces the clearance ofthe archwire within the bracket to zero (ie, 8 = 8), RS be gins to increase because binding be- gins to contribute to classical friction ‘As @ > 6 (Figure 6), binding in- creasingly Contributes tothe value of RS. Thus, as binding progresses, it overwhelms any contribution from classical friction and approximates the total value of RS. For the same relative contact angle (0- 8.) then, RS is greatest (Figure 6) when the en- gagement index is greatest and the bracket index is smallest (the 19x23/ 18 $5.65 couple [}C} versus other archwire-bracket couples in Table 3 and Figure 5). Moreover when @ >> 6, more serious problems that are associated with binding can oc- cur—namely notching," which can stop sliding mechanics. From these experimental results then, full No. 1 1999 engagement carries with it an obliga tion to have precise initial leveling and alignment, whereas selecting geometric parameters to obtain a high value of 8, carries with it an ob- ligation to never have @ >> @.. In the first case, sliding would at the very least be impaired by binding; in the latter, sliding would likely cease al- together by notching Clinical recommendations In the previous paragraphs, the val- ues of 8, have been discussed in terms of their effects on classical frc tion and binding. There are other considerations that deserve mention- ing, foremost of which isthe effect 0, has on clinical practice. By knowing 6, treatment time may be decreased by not initially over-aligning teeth prior to using sliding mechanics. In the past, an intuitive concern that 8 should never exceed 8, sometimes prompted clinicians to over-align