
Page 1 4 5 I. II. A. B. Daniel Guggenhe im Airship Institute Akron, Ohio TAELE OF C01.TENT3 Object Test SetUp Water Tank Equipment Model Table I III . Test Procedure I .. T ~ . Test .:{esu 1 ts v. Discussion A. hCCuracy of Recording System B. Sources of Error in Test SetUp 6 Table Ill C. ~on1parison of Expe r imental with .,omputed Curves ? D. General VI. Conclusions Liet of Figures Table 11 Figures ero ·   ·r  ·~ Akron, Ohio REPORT 01. fEA UREN . 1· OP TriE BEf.DII~G Lu:ME1 l'S rr THE HULL OF AN IRSHIP .~ODEL BY ,'IATER 'l'Al;h 'l'ESTS (USING ARTIFJCIAL GUST JITH TRi SITION W.1E EQUAL TO ONE HALF HULL LEtGTli) Item 11 of Contract NOs47286 I. OBJECT The purpose of these tests was to measure the bending moment at four different crosssections along the hull of an airship tr.odel while the model was beir.g towed through an artificial water guEt. The tow carriage was desitned to permit the model to turn and drift i11 the gust . 'I'he n1omer:tsof inertia of the model about its center of buoyancy and of the separate parte about the four crosssections at Y1hich measurements were taken were made to scale with the full size ehip so that the motion and reaction of the model would be as nearly as possible comparable to that of the full scale ehip under the test conditions. II. TEST SETUP A. Water Tank Equipment The water tank equipment has been described in previous reports ( reports on Items 5a and 10 of this contract). The gust increasing from zero to maximum velocity in one half model length was used for these tests. The maximum velocity v1as 4 u; sec. B. wodel The model used in these tests was a 1/150 scale model of the Akron constructed of magnesium alloy castings . To provide for mea~uring bending mo~ents in the hull, the model was built up of five sections hinged together by pairs of ball bearing jointe located in a vertical plane . ~ig . 1 shows the location of the cross::ections at .·1hich the bending moments were measured. Deflections between the different sections were resisted by small coil springs under initial tension connected to adjacent sections of the model. These springs were placed ~ust inside the shell of the model in a horizontal plane through its axi$. The elongation oi' these springs, or the ang:ular dieplacement between two sections, was proportion~l to the bending moment at the hinged crosssection. Fig. 2 is a schematic drawing shcwnng a plan view through the three center sections of the model . This shows how the moments were recorded. A longitudinal pushpull tube runs from the end sections of the model to a lever mounted crosswise in the center section carrying a srrAll diamond point . The pushpull tube is connected near one of the restraining springs and so practically moves with the deflecting end of the spring. This motion is transmitted to the diamond pojnt multiplied 4sl by the lever system. The diamond point scratches on a glass cylinder wt1ich is turned at constant spee~ by a sJnchronous clock motor and so record~ the bending moment experienoed at the particula r crosase~tion as a function of time. The resulting scratch is magnified approximately 75 times, photographed, and rr.easured. The maximwr. spring eloncation recorded in these tests was approximately .Ol", at the crosssection just ahead of the center of buoyancy. The corresponding eidevrise deflection of tn~ nose of the model was approxirr:.ately 1/1611 •  2 uan1muuggenneim Alrsnip institute Akron, Ohio Using this simple type of recording system, it was impoesible to measure directly the bending moment~ nt all the crosssections simultaneously , as the records of the end sections would be affected by the deflections of the tv10 sections adjacent to •.. . '1e central part. Accordingly, lock screv1~ vrere proviaed so tnat all the hinged sections could be made rigid. Tests were first run with the two end sections free and the three in..~er sections locked together, ~nd were then repeated with the two inner crosssections free and the outer sections locked. Fig. 2 also s hows in detail how the hinged crosssections were made water tight. The clearance provided between sections was approximately 1/32". The brass shield rint;s were filed or built up with pltj,stilina as was necessary to confonn to the model contour. The Selsyn motor operating the rudder~ wae mounted in the extreme tail section. The wires leadi~g to it were small and flexible and plenty of slack was provided so tney would have ~o effect on the deflections of the hinged cr1Jsssections a.:ross which they passed. For balancing the model, the scaledda1m moment of inertia of the part of the model from each crosssection to the nose or tail, whichever was nearer, was co1nputed by the equation; se lcomputed • "' · " ·dx cross sectior no so ~1 I 4Tf • dX ·ross section lvhere S =crosssectional area at x (in2 ) x  distance of S from hinged crosssection (in . ) r radius of model at x (in.) f'  density of water (ffseo~/in~ ) l is tnen biven int  in.  sec 2 • Thie ussumes that the buoyancy and the load at any parti~ular crosssection are in equilibrium. The moments of indrtia of the various parts of the model were then adjusted by means of lead weights until the experimental values were very close to the computed values, that is, very close to scale witn the full size ship. Fig . 3 is a photograph showing the experimental method for determining I for each section. The section was supported at one and by a coil spring of known stiffnees and at t he other end by the ball bee.ring hinges. The perion of oscillation about the hinge axis was timed with a stop watch, and the moment of inertia found from the rela.tion T :: 2 idfr°'_ .fk.f '\vhere T = period (seconds) l·  spring constant (;r/in . ) 1 .,. distance from axis to spring (in . ) .i.  moment of inertia (=ff  in.  sec 2 ) 3 Daniel Guggenheim Airship Institute Akron, Ohio Table I gives the computed and experimenU..l moments of inertia for the different parts of the model . The moment of inertia of t.e complete model about its center of buoyancy was determined as before by suspending it from a va re e.nd comparing its frequency of torsional oscillation with that of a steel bar sus pended from the same wire. Table I Nose section Two nose sections locked together Two tail sections locked together Tail section Entire model about C.B. oment of Inertia, in 1f in.  sec 2 Computed 4.65 27.5 9. 41 1.24 69 .8 Experimental 4.67 27.6 9. 52 1.22 69 .2 In this connection it should be noted that the moment of inertia of the entire model about its center of buoyancy computed by the method outlined came out 10 °/0 higher than the value used ir. the previous tests . It is probable that in arriving at the value previously used t ~ e second term in the equat.ion for !computed was omitted . Adding this term to the old value of I brings it within 3 °/o of' the value used in these tests. Fig . 4. is a photograph of the intE:rior of the model. This shows the glass cylinder J.•Jr ~rding apparatus in the center sE:iction of the model e.nd the lead weights required iu the two nose sectiona. The ball bearing housicgs for connecting the two parts of the model are ehown clearly on the nose Eections. III. TEeT PROCEDURE Test runs were :ma~e for three gust velocity to forward velocity ratios: 1/3.5, 1/4.5 and 1/5.5. Test runs were made at o0 initial yaw with rudders set at o0 and 'vith rudders r.1oved from o0 to ... 20° in approximately one hull lengtr. for all three velocity ratios . In addition, for Vgjv = 1/4.5 the following runs were rnnde : 1) o0 initial yaw, rudders held constant at ~2 0° , 2) o0 injtial yaw, rudder~ moved from  10° to +10° in one hull leng1h, 3) +5° initial yaw with Tl0° r udder setting and_200 gms . opposing side fore~, rudders moved to +20° in ~ne hull length and 4) 5° in~ tial ya\v \Vi th  10° rudder setting a~d 200 gms . opposing side foroe, rudders moved to rl0° in one hull length. The method of applying; the side force for the last_two cases hae been described in the report on Item 10 of this contract. As noted in the previous section , the testE were first run recordinb moments on tLe nose and tail section~ and were then duplicated to rGcord the moments on the inner crosssections. The bending moment recording system ~~s calibrated before test! were started and between the two sets of runs. This was done with the model submerged at its regular depth. A cord looped around the model led off horizontally to a pulley submerged at the model depth and then up and around a second pulley 4 Daniel Guggenheim Airship lnetitute Akron, Ohio above the surface. By hanging weights on t his cord, horizontal forces were applied at given points marked on the hull of the model. The force~ were applied in the same direction that the gust forces acted, and the friction in the pulleye wae minimized by jarrin{; the weight and then letting the reEulting vibrations dampen out. The forces were applied normal to the model axi~ at two separa.t.e measured distances frorn each hinge axis. The resultir.g calibration curves were consistent straight lines. ost of the records showed vibre.tions with a frequency of appr oximately l~ cycles per second. rhilo a mean curve could be drawn with fairly e.;ood accuracy, it 'WUS con~idered advisable to make three different runs for each condition and average the curves obtained. The maximum valuee and the general shape of the different curves were in good agreement in practically all cases. Figur es Sa and b are examples of the results obtained. ln addition to the test runs already outlined, runs were made with the cross carriage fixed and the model clamped at the following yaw angles: o0 , s0 , 10°, 15°, 20° and 25°. The elevator setting was o0 in all cases. Thea~ runs v1ere made pri11ci pally to perr:ii t a direct check between the complete apFBratus for measuring bending moments in tne water tank and available wind tunnel tests. IV . RESULTS Results are given in the form of curves showing the values of Cm (defined as noted on the curve sheets) for the four different crosesections plotted against the position of the nose of the model in the gust . The ten differert test conditions are presented as fo!lows : Fig . 6 Vg/V  1/3.5 00 initial yaw 00 rudder Fig . 7 vg;v 1/3.5 oo initial yaw oo to +20° rudder Fig . 8 Vg/V "' 1/4.5 00 tt II 00 r udder 1',ig. 9 Vg/v  l/4.5 00 n " o0 to..20° r udder Fig . 10 Vg/v  1/4.5 ao " " +20° r udder Fig . 11 Vgjv ,.. 1/4. 5 00 II II 10° to 110° rudder Vg/v • 1/4.5 +50 ,, Fig. 12 II +10 0 rudder, 200 grams side force, rudder moved to 120° in gust . Fig . 13 V g;lv = l/4.5 5 ° initial yaw 10° rudder J 200 grams side force, rudder moved to ~10° in gust. Fig. 14 Vg/v e 1/5.5 o0 initial yaw oo rudder Fig. 15 7ifv  l/s.s 00 II ,, oo to +20° rudder 1'"igurds 16, 17 and 18 give tt.e crosswise and an£ular dis t>'L:J.cernents and velocities f'or Vg/v = 1/3.5, 1/4.5 and 1/5.5, respectively. Results are summarized in Table II v1hich gives Cn:max for each crosssection and 6ach test condition, together with thE a~proximate angle of steady yawed flicht to give en equivalent Cm • The equivalent Cm val11e~ wer& obtained from Fig. 19 which gives the results of the steady yawed flight tests. Table II also lists the ~Aximum effective fin angles obtained from previous tests in the same gust conditions with the old model. (Report on Item 10 or this contract). 5 V . DISCUSSION A. Accuracy of Recording System Daniel GuGgenheim Airship Institute Akron, Ohio In Fig. 19 the results of the ste~dy yawed flight tests are compared with results computed f r om the pressure distribution measur ements reported in NAC.A. Technical Reports .:o. 443 and l~o . 604 . The computed Cm values were obtained for the two crosssections ahead of the center of buoyancy by the equation nose f/q · > • dx , where cross section r/q ~ transverse fo rce per rt. length of hull divided by impact pr essure (ft . ) . (Data given in NACA Report l!o . 443) x ~distance from crossaection to position where x/q is measured (ft.) Vol ~volume of ship (cu.ft.) For the two rear crossnections a similar integration of the transver se hull forces was carried out going from the crosssections to the tail of the ehip. To this was added the moment due to the fin forceE, computed from the data on the tratJSVerse force per ft . length of fin given in NACA Report No . 604 . The computations neglect the longitudinal components of the local press\tr es , but their effect is not of great importance . The close agreement between the measured and computed bending moments would seem to establish t~e accuracy of the method used to record tho moments in the water tank. It is believed that the CJJ\max values are accurate within! 5 °/o. B. Sources of Error in Test SotUp A casual inspection of Table II show3 that the coefficients here presented are in general quite high. Since the recording system is known to be reasonably accurate, it is nec~ssary to examine the test setup for ~ources of error. The fundamental aim of the testo is to duplicate as cloeely aE possible the condition~ which occur when an airship runs into a horizontal gust or wind shift. ne principal defect of the setup used at present is that the thrust rema.·ns in the direction of the original flight path. In the actual case the direction of the thrust changes with the angular displacement of the ship. This has the effect of gradually applyjng a side force at the center of buoyancy equal to T · sin cX. , where T .,.. thrust aua ol ~ angular displacement. In addition, the. dra0 01 the model ir. the direcio of motion of the tow carriage increases with the angular displacement. This puts greater loads on t he croQs carriage vtheels and bearings. Hence it ie quite probable that the friction of the cross carriage varies throughout the run, increasing 'vith the angular displacement. This frictional force ~s e1uivalent to an equal force applied at the center of buoyancy and directed i~to the &urt. Both of these effects tend to increase the oending momont beyond the value that would result from true free flight conditions . These sources of e rror are both a maxi~lll'l'l at the maximum angle of attack. It is interesting to note that for >l,l ~ . 567 jr. particular, the ~Axi~um cm generally occurs p1actically at tl wllTne time that the angular displacement is m.aximwn. Table III illustrates this fact . 1/3.5 0° rudder QO to ~20° rudder Vu_/·V e 1/4 .5 0° rudder OC to +200 rudder +20° rudder 10° to +10° rudder +5 ° initial yaw 5f'l initial yaw v 1v • i; E:v" 5.5 o0 rudder o0 to t20° rudder 6 Table III Daniel Guggenheim Airship Institute Akron, Ohio Position of nose of model acros~ gust in hull lengths (~/..! ) for Cn:.max (x/£ ,. .567) for ol max 1. 45 1.5 1.4 l . 45 1.4 1.5 1 . 45 1.3 1.6 1.3 1.5 1.4 1.6 1 . 35 1.4 1.45 1.45 1.55 1.8 1. 45 Again, with the model set at ~5° initial yaw the errors woulu be still further increased. In all cases Cm...__ for x/.L c .567 occurs v1hen the model is almost entirely .n:a.:x in the fully developed gust. Since the leading ed~e of the fin is approximately 0 . 2 of the hull length from the tail in all oases the fins ·would be at lea.st purti&lly in the fully developed gust ,,r.en the m.axi~um bending moment occurs . (The gust transition zone ends at ~/.f  .5) In some cases Cnlmax does not occur until the entire model is corpletely in the fully developed gust. For the e.ctual free fli e;ht case it aprcars somewhat doubtful th~t the maximum bending moment at Eections in the forebody of the ship should occur at this time. The above considerations indicate that the tests here roported probably overestimate the hull bendinf; momentt:. C. Comparison of Experimental with Computed Cur~e~ In Fig . 20a the bendinb moments measured at x/,/  . f.67 and • 736 for Vg/v • l/S:5 and o0 rudder e.rtt compared ·with the computed coefficients for the same case as given in the report on Item 10 of this contract. Although ~he computed coefficien~s a: ply to a cro~ssection lying between the two experin1ent&.l curves, the gererb.l r.ature of the computed curv6 is entirelJ different from the experimental ones. A comparison betwe~n the oricinal displacement records for the two cases revealed considerable differences, especially in the angular displacement. This is shown in Fig. 20b . Accordingly , comp·1ted co~ff'icients were v;orked out from the new displacement record for x/,,/ s .567 and . 736. These new comput~d curves are sl~o given ' 7 Danjel Guggenheim Airship Insti tute Akron, uhio in Fig . 20a. The differcnces between the old and new displacement recor ds are no doubt due to the higher moment of inert.ia of the ~resent model . Since the model is unstable for o0 rudder up to about 14 yaw, the tur ning moment about the ceni:;er of buoyancy on the old model vrith th'=' sms.ller moment of i 1 urtia v:ou" • .. rease faster than for the present model, as is sho,·m by the curves of oL/dt2 in Fig . 20b. Ylhile the curves computed from the present tests ere somewhat in line with the experimental values, the agre rront is not very satisfactor y . It appea r s that the computing of the bending moments b&.S"ed on analysis of the airPhip moveme1,ts is not practical due to already previousl~ suspected inherent weakr1esser:: of the necessary mathematical appcoa ch . D. Gen~ral In Fi gu re~ 6 , 7, 8 and 12 the experimdntal curves are somewhat irrefular from the origin to the mo. ~r~L value. In generRl, a bump in the curvts occurs at approximatet~ I,! ~ . 5. This irrecularity ie absent in the other curves. It is ez.L5 ,;_,, that this bump is due to some irregule.ri tj in the mo~ion of the model rather than to Gust forces. Possible causes are 1) oscillation of the model in the direction of the towing force ca.used either by a jerk at the start of the run or by variations in the towing force, 2) e. small irl·dzularity in one of the v;heels or cross rails, and 3) oscillation of the model in the direction of motion of the crosB carriage . Th~ relative magnitude of these irreGularities wa~ not suspectec f r om the visual examination of' the scratches made iI!Ullediately followint; the test runs and when the photographs v1ere meai;ured it was too late to investigate thi~ point. lt is quit;., probable that tho e.ctua.l development of tht; bending moment in the gust is more regular, as shown by the light line in Fig . 6. It should be noted from Table II that the equivalent angles of yawed steady flight are in general higher for the two cross sections behin·i the center of buoyancy than for those in the forward part of the ship and are also higher than the maximum effective fin angles taken from the report on Item 10 of this contract. The maximulTl effective fin angles, however, v;ere figured for the model with smaller moment of inertia and generally occurred about .2 or . 3 hull lengths before the maximum Cm v&lues obtained from these tests. VI . COllCLUSIO. S .n.. The results here presented do not apply exactly to the pr oblc:n of an air~hip in free flight encountering a gust, due to the direction of thr ust and to the friction in the c r oss carriage . The effect of these errors is to incroaso the b~nding m~~ent~ over free flight conditions . B. The check between the experimental and computed montents is not good and tho bendi11g moment coefficients computed frorn previous t ests shoul d be dis regarded . C. Further research vii th a selfpropelled rnodel should result in more accurate knowledge of gust bending moments . TABLE II l.aximum Cm  1) x!t fj = • 234 vdv  1/3.5 'Jo rudder .040 )0 to +20° rudder . 046 Vg/V = 1/4.5 oo rudder .026 po to t20° r udder .032 ..20° rudder .037  10° to +10° r udder .032 +50 in:itial yaw , 200 ~ms . .056 side fo rce _50 initial yaw, 200 gms .  .030 side force Vg/V ,.. l/s .s 00 rudder . 030 oo to +20° r udder .036 l) x  distance of section from tail end. ~  l ength of airship model . . 387 . 567 .116 . 158 . 117 . 15G . 078 . 124 . 077 . 12~ .067 .102 .090 . 152 . 140 . 250  .053 i. 095 .060 . 101 . 069 . 104 •' . 736 . 065 . 063 . 038 . 048 . 04:6 . 048 . 084 .030 .038 . 041 Approximate angle of steady pitched ..cil )('. • flight to give equivalent Cm oleff l~ . 234 . 387 . 567 .736 of fins 15° 15° 11° 12° 13 . 1° 16° 15° 11° 1 1~0 . 9. 3° 12° 11~0 .... 80 70 11.4° 13° i1J...0 80 91.,_...,. 0 6. e0 14° 10~0 70 80 7 .o0 13° 12~0 11° 8.~... 0  18° 17° 17?0 .., 15° 15 . 3° lo 12~ _90 _70 _50 5 .6°  13° 10° 70 70 I 10 . 7° 14° 11° 7~... 0 70 I 4 . 9° I FIGURES Daniel Guggenheim Ji.irsh · p lnsti tute Alpron, Ohio 1. Outline diagram showing cro ssections at wl~ch bending moments ;ore measured. 2. Photograph of schematic drawing of bending moment model . 3. Photot;raph showing method of' determining the moment of inertia or the model sections. 4. hotograph showin~ recording mechanism and lead weights inside model . Su. Photogr ph shov1ing typical record obtained at ./' ~ . 567 . 5b. 6. 7. s. 9. 10. 11. 12. !g/v ~ 1/4.5 , o0 initial yavr, rudders moved i.. >0 to 20° in approxirn tely one hull length. Photograph ehowin~ typical record obtained at ~ ~ . 387. Vgjv • 1/4.5 , o0 initial yav, rudders moved ir J 0 to+20° in approximately one hu'l length. Cm • r ( s/ / ) , v g/v  1/3.5 00 initial y 00 rudder. ' ' Cm z: r(s/ j ), Vgjv  1,13 .5 00 " II 00 to +20° rudder. ' ' Cm z: r(s/J ), Vg/v !Z 1,/ 4.5 ' 00 " II ' 00 rudder. Cm  i'(E 1,1 , "'lg/v r 1/4 .5, 00 II II 00 to +20° rudder. ' Cm f(S/j), VyV : 1/4.5 oo II " + 20° rudder ' , Cm  f{S/ /)I Vg/v  1/4.5 00 " 11 10° to 110° rudder. ' ' Cm f'(S/ .J), Vg/v .. 1/4.5 +50 " II t10° rudder, ' ' +200 grams side force, r udders moved to 120°. 13. Cm !'! I.'(S/j), Vg/v = 1/4.5 , 5° initial ya"'• 10° rudder, 200 grams side force, rudders moved to +10°. 14. Cm  f(S/~), !g/v  1/5.5 , o0 initial yaw, o0 rudder. 15. r(s/~), Vg/v  l,ls .5 II " , o0 to +20° rudder. 16. Crosswiae and angular displacements and velocities for Vg/v = 1/3 .5 • 17 •• Crosswise and angular displacements and velocities for Vgjv = 1/4.5 . 18. Cross 11ise and ungular di"spla.cements and velocities for V g/v = 1/5 .5 • 19. Co~ ri~on of measured and computed bending moment coefficients for steady ya~ed flight, o0 rudder. 20a. Daniel Guggenheim Airship Institute Akron, Ohio FIGURES (Cont.) Comparison between experimental Vg/v = 1/5.5 , o0 initial yaw, and computed values of Cm , OO rudder. 20b. Comparieon of crosswiae and angular displacements, velocities and accelerations for old and new models under the same test conditions. Vgjv = l/s.5 , o0 inibial yaw, o0 rudder. • V NlE:G""GU"GGEl'T.BJli.ILm::~ AIRSHIP INSTITUTJD AKRON, omo • Daniel G ugg enhe i~ AirEhip Institute Akron, Ohio s t DIN t. TRIC MO TOR F R ROTA7 ING GL .4  CYl IND£R ~~~·~o~ .:_/__ ST~£L COIL SP /f\, FL£Jl18l t. L UOP£D RUBBER SHl ET JOIN I ~ DIAMOND POINTS /'/OTOR UNITS 5' TIN RL ODER i  ' J . ' ' I ' ' l ' I I I '\ L£V£R S 'S TEM BALL BEARING CARRYING 01 fOND POINT Fr>R RECORDIN  BENDING MOMENT • Fig . 2 . Photog r ~ 1h of schematic dra ing of bending moment mode l. \ • Daniel Gue;genheim Airship I=istitute Akron, Ohio ' • Fig . 3. Photo~raph showing method of deter:r.ining moment of inerti~ of model sections . • • Daniel Guggenheim Airship Instjtute Akron, Ohio Fig . ~ . Photo~raph ~howine: reco rding mechanism and lead weights insjde model , / \ Fig. 5a. ' Photograph shov1ing typical record obtained Vg/l/ = 1/4.5 • o0 initial yaw. o0 rudder • •  .567 • ,. ~=· • • • \ \ \ , Fig . Sb . Photograph showing typical record obtained at x/t, • . 387 Vgjv • 1/4.5 , 0° initial yaw, 0° r udder • • l/J6,4 l:t/ • (J ... •03. ''Z. ·~~ ·O.~t ·1'8 0¥ ~ .. .,,z • DANIEL GUGGENDlll AIRSHIP INST1'1'01'B AKRO J::; "387 ~: ;l.d' ':. .z~s4 ~t:4~~~~~~4+!7f~~~;ri ~D ~..No/ JtE ~.s. ..Aceo.s li.osr IN. ill4a{~ENfYlfJl DANIEL GUGGENHEIM AIRSHIP INSTITUTE ...~_,.. . .....,~~~~~ ....,..._...,.....,... ri~~....,..,................ ........ AKRON omo •zo ..............~ . .. c>...·1 oe~a~  _ ._,,_ ~ Z3'f. ~  'h /;o Z·o  fi fi~.tr11£N t'J.F. Nd.st£. .0£ MPPel, A&e~.r. ,.i.....,...~ ~ .. " _· j_: !fv.sT 1N flt1a ltN4.TH~ ~r~+~~4~Ba~r..._+..!_..,__.,~ ..., .0..r.1 ..Lill.a., \:l'U l:ttl'J!;.ft ti BUii. AIRSHIP INSTITOTJI ~, AKRON .. uz.., .;" ~ 41 .,0,q, .~~"'"'" ~ • (}  ~  f () ~ • •Id a ... #I~ ()2. (}t • .. at~ :4J_ rO¥ • ·llJ " .u .... .. DANIEL GUGGENHE AIRSHIP INSTITOT AKRON OHIO ~O~~~ • • ~ l)o 2'·b ~ = · t 'i ~ PA41.rl<'4 b.F ~SE llE h/@£1. ~ ~ &t!sr tl)f H~G f.8.:tqrN,s NJ .. yi{ !Uv ~ ~~~" MtrrBf ~J.v ~do~.p6& t • .10 . .,. .r DI . '"' •• P_w,rf"1< o, ND.SE 4"' .HiUJ~.L A<,ltlA 6v.1l" N ~' 1.#'.N'~  1 I r • , __ .... , / •    • • c . / ' CORR :CTIOl to Report on e surementE of tho Bending komente in the Rull of n Airship odel by later T n T ts (using rtificial 6uat th tr n ition zone equ l to on h lf hull 1 n th) It m 11. Contract NO 47286 i . G nd 7 t Ch n e C " nd Ill 3 le~ to agree 1th th followiDC ro.tios: c " .. , 1. 75 c .438 Cm The equiv lent vuluo of Cm for tho oorr•ot valu in tho follcnrln tt'lblee: ,, .10 .20 .so .40 qui lent Cm .057 .114 elr/l .228 t ,,, .02 .04 .06 .08 .10 of '' nd bqui lent Cm .046 .091 .157 .185 .228 ,,, Fig • l and 15: Ch ng 'TI" • nd ,,, ·oalc to gree \rl th the tollowin Th ivcn in " '" 2. 75 Cm . 686 c equivalent v lue of Cm for th the following blee: II quive.l nt om .10 .056 .20 .073 .so .109 .40 .145 correct valuee of r •• nd "' t •" Equiv l nt Cm .02 .029 .04 .058 • .06 • 087 .OB .116 re given tiosa re Daniel Guggenheim Airship In~titute Akron, Ohio CORRECT101 SHE~T for Report on ~easurement of the Bending Uoments in the Hull of an Airship 1.odel by iia ter Tank Tests (Using Artificial Gust with Tr11nsitlon Zone Lqual to OneHalf' Hull Length) . Item 11 of Contract NOc47286 Fig . 16. Change Vgjv = 1/3 •5 to Vgjv a l/s.5 Change fi&ure number to read i 0 • 18 Fig. 18. Change Vgfl = 1/s.s to Vg/v  1113•5 Change figure number to read Fig. 1€ . " LJ Nl1'iL <iUGGE.Nl:i!tiUA AIRSHIP INSTITUTll AKRON OHIO 0 ~ / ' I I ~ \ ' \ t i I \ I \ \ J N(J. 20b c;,N OF: ~"'¢ "" AN<W4AR DLSPUICJ!HGNT.S, Vi<«trT&J "'" A.C.tl~.S FOK 4,D ~ N£W ~ l/NIM/l TNli &.i'JH£ ~ CCl'/l)ITIDNS Vth = ~  Q:.U.JTIAL ~ ... O"K"5 UI> ,,. ;" ,,,. ~ ·· I# 60 .... ~ .~ ... • ,. ,,,..,, 0 I " · oll!J.., UUU<JVNRalll ,i.1aan lP IN8TITOTll AKRON OR1 /;d ZOii ~ i. ~ ~ "~* \ ~~ ' I ~ \ ~ ' \ ~ \ ~ \ ,. \ .. ~ ~ ~ ~ ~ N I..: ~ ~ ~~ ,,, s I ' ~ ~ I ~ l ~ J ~ I
Click tabs to swap between content that is broken into logical sections.
Title  Report on Measurement of the Bending Moments in the Hull of an Airship Model by Water Tank Tests (Using Artificial Gust with Transition Zone Equal to onehalf Hull Length) 
Creator  Daniel Guggenheim Airship Institute 
Date Original  19390121 
Date Notes  19390121 
Description  This report presents the findings of tests conducted at the Daniel Guggeheim Airship Institute in Akron, Ohio to measure the bending moment at four different crosssections along the hull of an airship model while the model was being towed through an artificial water gust. 
Subject Terms 
Daniel Guggenheim Airship Institute University of Akron. College of Engineering AirshipsModels AirshipsTesting Aerodynamics 
Location  Akron (Ohio) 
Type  Text 
Publisher  Daniel Guggenheim Airship Institute 
Digital Publisher  University of Akron. Archival Services 
Copyright Statement  This publication is protected by copyright law of the United States (Title 17, United States Code). Copyright to this publication lies with The University of Akron which makes it available for personal use for private study, scholarship, or research. Any other use of this publication including publications, exhibitions, or productions is prohibited without written permission of The University of Akron Archival Services. Please contact Archival Services at archives@uakron.edu for more information. 
Source Collection  Daniel Guggenheim Airship Institute Records 
Identifier  Report No. 02.46 
Medium  Document 
Collection Category  LighterThanAir Flight 
UA College  University Libraries 
UA Department  Archival Services 
Website  http://www.uakron.edu/libraries/archives/ 
Contact Information  The University of Akron, Archival Services, Polsky Building, Room LL10, 225 South Main Street, Akron, OH 443251702, Phone: 3309727670, Fax: 3309726170, Email: archives@uakron.edu 
transcript 
Page
1
4
5
I.
II.
A.
B.
Daniel Guggenhe im Airship Institute
Akron, Ohio
TAELE OF C01.TENT3
Object
Test SetUp
Water Tank Equipment
Model
Table I
III . Test Procedure
I .. T
~ . Test .:{esu 1 ts
v. Discussion
A. hCCuracy of Recording System
B. Sources of Error in Test SetUp
6 Table Ill
C. ~on1parison of Expe r imental with .,omputed Curves
? D. General
VI. Conclusions
Liet of Figures
Table 11
Figures
ero ·   ·r  ·~
Akron, Ohio
REPORT 01. fEA UREN . 1· OP TriE BEf.DII~G Lu:ME1 l'S rr
THE HULL OF AN IRSHIP .~ODEL BY ,'IATER 'l'Al;h 'l'ESTS (USING
ARTIFJCIAL GUST JITH TRi SITION W.1E EQUAL TO ONE HALF HULL LEtGTli)
Item 11 of Contract NOs47286
I. OBJECT
The purpose of these tests was to measure the bending moment at four different
crosssections along the hull of an airship tr.odel while the model was beir.g
towed through an artificial water guEt. The tow carriage was desitned to
permit the model to turn and drift i11 the gust . 'I'he n1omer:tsof inertia of the
model about its center of buoyancy and of the separate parte about the four
crosssections at Y1hich measurements were taken were made to scale with the
full size ehip so that the motion and reaction of the model would be as
nearly as possible comparable to that of the full scale ehip under the test
conditions.
II. TEST SETUP
A. Water Tank Equipment
The water tank equipment has been described in previous reports ( reports on
Items 5a and 10 of this contract). The gust increasing from zero to maximum
velocity in one half model length was used for these tests. The maximum
velocity v1as 4 u; sec.
B. wodel
The model used in these tests was a 1/150 scale model of the Akron constructed
of magnesium alloy castings . To provide for mea~uring bending mo~ents in the
hull, the model was built up of five sections hinged together by pairs of
ball bearing jointe located in a vertical plane . ~ig . 1 shows the location
of the cross::ections at .·1hich the bending moments were measured. Deflections
between the different sections were resisted by small coil springs under
initial tension connected to adjacent sections of the model. These springs
were placed ~ust inside the shell of the model in a horizontal plane through
its axi$. The elongation oi' these springs, or the ang:ular dieplacement
between two sections, was proportion~l to the bending moment at the hinged
crosssection. Fig. 2 is a schematic drawing shcwnng a plan view through
the three center sections of the model . This shows how the moments were
recorded. A longitudinal pushpull tube runs from the end sections of the
model to a lever mounted crosswise in the center section carrying a srrAll
diamond point . The pushpull tube is connected near one of the restraining
springs and so practically moves with the deflecting end of the spring. This
motion is transmitted to the diamond pojnt multiplied 4sl by the lever system.
The diamond point scratches on a glass cylinder wt1ich is turned at constant
spee~ by a sJnchronous clock motor and so record~ the bending moment experienoed
at the particula r crosase~tion as a function of time. The resulting
scratch is magnified approximately 75 times, photographed, and rr.easured.
The maximwr. spring eloncation recorded in these tests was approximately .Ol",
at the crosssection just ahead of the center of buoyancy. The corresponding
eidevrise deflection of tn~ nose of the model was approxirr:.ately 1/1611
•
 2
uan1muuggenneim Alrsnip institute
Akron, Ohio
Using this simple type of recording system, it was impoesible to measure
directly the bending moment~ nt all the crosssections simultaneously , as
the records of the end sections would be affected by the deflections of the
tv10 sections adjacent to •.. . '1e central part. Accordingly, lock screv1~ vrere
proviaed so tnat all the hinged sections could be made rigid. Tests were
first run with the two end sections free and the three in..~er sections locked
together, ~nd were then repeated with the two inner crosssections free and
the outer sections locked.
Fig. 2 also s hows in detail how the hinged crosssections were made water
tight. The clearance provided between sections was approximately 1/32".
The brass shield rint;s were filed or built up with pltj,stilina as was
necessary to confonn to the model contour.
The Selsyn motor operating the rudder~ wae mounted in the extreme tail
section. The wires leadi~g to it were small and flexible and plenty of slack
was provided so tney would have ~o effect on the deflections of the hinged
cr1Jsssections a.:ross which they passed.
For balancing the model, the scaledda1m moment of inertia of the part of
the model from each crosssection to the nose or tail, whichever was nearer,
was co1nputed by the equation;
se
lcomputed • "' · " ·dx
cross sectior
no so
~1
I 4Tf • dX
·ross section
lvhere S =crosssectional area at x (in2
)
x  distance of S from hinged crosssection (in . )
r radius of model at x (in.)
f'  density of water (ffseo~/in~ )
l is tnen biven int  in.  sec 2
•
Thie ussumes that the buoyancy and the load at any parti~ular crosssection
are in equilibrium. The moments of indrtia of the various parts of the model
were then adjusted by means of lead weights until the experimental values were
very close to the computed values, that is, very close to scale witn the full
size ship. Fig . 3 is a photograph showing the experimental method for determining
I for each section. The section was supported at one and by a coil
spring of known stiffnees and at t he other end by the ball bee.ring hinges.
The perion of oscillation about the hinge axis was timed with a stop watch,
and the moment of inertia found from the rela.tion
T :: 2 idfr°'_ .fk.f '\vhere
T = period (seconds)
l·  spring constant (;r/in . )
1 .,. distance from axis to spring (in . )
.i.  moment of inertia (=ff  in.  sec 2
)
3
Daniel Guggenheim Airship Institute
Akron, Ohio
Table I gives the computed and experimenU..l moments of inertia for the different
parts of the model . The moment of inertia of t.e complete model about its
center of buoyancy was determined as before by suspending it from a va re e.nd
comparing its frequency of torsional oscillation with that of a steel bar sus pended
from the same wire.
Table I
Nose section
Two nose sections locked together
Two tail sections locked together
Tail section
Entire model about C.B.
oment of Inertia, in 1f in.  sec 2
Computed
4.65
27.5
9. 41
1.24
69 .8
Experimental
4.67
27.6
9. 52
1.22
69 .2
In this connection it should be noted that the moment of inertia of the entire
model about its center of buoyancy computed by the method outlined came out
10 °/0 higher than the value used ir. the previous tests . It is probable that
in arriving at the value previously used t ~ e second term in the equat.ion for
!computed was omitted . Adding this term to the old value of I brings it
within 3 °/o of' the value used in these tests.
Fig . 4. is a photograph of the intE:rior of the model. This shows the glass
cylinder J.•Jr ~rding apparatus in the center sE:iction of the model e.nd the lead
weights required iu the two nose sectiona. The ball bearing housicgs for
connecting the two parts of the model are ehown clearly on the nose Eections.
III. TEeT PROCEDURE
Test runs were :ma~e for three gust velocity to forward velocity ratios: 1/3.5,
1/4.5 and 1/5.5. Test runs were made at o0 initial yaw with rudders set at
o0 and 'vith rudders r.1oved from o0 to ... 20° in approximately one hull lengtr.
for all three velocity ratios . In addition, for Vgjv = 1/4.5 the following
runs were rnnde : 1) o0 initial yaw, rudders held constant at ~2 0° , 2) o0
injtial yaw, rudder~ moved from  10° to +10° in one hull leng1h, 3) +5°
initial yaw with Tl0° r udder setting and_200 gms . opposing side fore~, rudders
moved to +20° in ~ne hull length and 4) 5° in~ tial ya\v \Vi th  10° rudder
setting a~d 200 gms . opposing side foroe, rudders moved to rl0° in one hull
length. The method of applying; the side force for the last_two cases hae been
described in the report on Item 10 of this contract. As noted in the previous
section , the testE were first run recordinb moments on tLe nose and tail section~
and were then duplicated to rGcord the moments on the inner crosssections.
The bending moment recording system ~~s calibrated before test! were started
and between the two sets of runs. This was done with the model submerged at
its regular depth. A cord looped around the model led off horizontally to a
pulley submerged at the model depth and then up and around a second pulley
4
Daniel Guggenheim Airship lnetitute
Akron, Ohio
above the surface. By hanging weights on t his cord, horizontal forces were
applied at given points marked on the hull of the model. The force~ were
applied in the same direction that the gust forces acted, and the friction in
the pulleye wae minimized by jarrin{; the weight and then letting the reEulting
vibrations dampen out. The forces were applied normal to the model axi~ at
two separa.t.e measured distances frorn each hinge axis. The resultir.g calibration
curves were consistent straight lines.
ost of the records showed vibre.tions with a frequency of appr oximately l~
cycles per second. rhilo a mean curve could be drawn with fairly e.;ood accuracy,
it 'WUS con~idered advisable to make three different runs for each condition
and average the curves obtained. The maximum valuee and the general
shape of the different curves were in good agreement in practically all cases.
Figur es Sa and b are examples of the results obtained.
ln addition to the test runs already outlined, runs were made with the cross
carriage fixed and the model clamped at the following yaw angles: o0
, s0 ,
10°, 15°, 20° and 25°. The elevator setting was o0 in all cases. Thea~
runs v1ere made pri11ci pally to perr:ii t a direct check between the complete
apFBratus for measuring bending moments in tne water tank and available wind
tunnel tests.
IV . RESULTS
Results are given in the form of curves showing the values of Cm (defined as
noted on the curve sheets) for the four different crosesections plotted against
the position of the nose of the model in the gust . The ten differert test
conditions are presented as fo!lows :
Fig . 6 Vg/V  1/3.5 00 initial yaw 00 rudder
Fig . 7 vg;v 1/3.5 oo initial yaw oo to +20° rudder
Fig . 8 Vg/V "' 1/4.5 00 tt II 00 r udder
1',ig. 9 Vg/v  l/4.5 00 n " o0 to..20° r udder
Fig . 10 Vg/v  1/4.5 ao " " +20° r udder
Fig . 11 Vgjv ,.. 1/4. 5 00 II II 10° to 110° rudder
Vg/v • 1/4.5 +50 ,, Fig. 12 II +10 0 rudder, 200 grams side
force, rudder moved to 120° in gust .
Fig . 13 V g;lv = l/4.5 5 ° initial yaw 10° rudder J 200 grams side
force, rudder moved to ~10° in gust.
Fig. 14 Vg/v e 1/5.5 o0 initial yaw oo rudder
Fig. 15 7ifv  l/s.s 00 II ,, oo to +20° rudder
1'"igurds 16, 17 and 18 give tt.e crosswise and an£ular dis t>'L:J.cernents and velocities
f'or Vg/v = 1/3.5, 1/4.5 and 1/5.5, respectively.
Results are summarized in Table II v1hich gives Cn:max for each crosssection
and 6ach test condition, together with thE a~proximate angle of steady yawed
flicht to give en equivalent Cm • The equivalent Cm val11e~ wer& obtained
from Fig. 19 which gives the results of the steady yawed flight tests. Table II
also lists the ~Aximum effective fin angles obtained from previous tests in
the same gust conditions with the old model. (Report on Item 10 or this contract).
5
V . DISCUSSION
A. Accuracy of Recording System
Daniel GuGgenheim Airship Institute
Akron, Ohio
In Fig. 19 the results of the ste~dy yawed flight tests are compared with
results computed f r om the pressure distribution measur ements reported in
NAC.A. Technical Reports .:o. 443 and l~o . 604 . The computed Cm values were
obtained for the two crosssections ahead of the center of buoyancy by the
equation nose
f/q · > • dx , where
cross section
r/q ~ transverse fo rce per rt. length of hull divided by impact
pr essure (ft . ) . (Data given in NACA Report l!o . 443)
x ~distance from crossaection to position where x/q is measured (ft.)
Vol ~volume of ship (cu.ft.)
For the two rear crossnections a similar integration of the transver se hull
forces was carried out going from the crosssections to the tail of the ehip.
To this was added the moment due to the fin forceE, computed from the data
on the tratJSVerse force per ft . length of fin given in NACA Report No . 604 .
The computations neglect the longitudinal components of the local press\tr es ,
but their effect is not of great importance . The close agreement between the
measured and computed bending moments would seem to establish t~e accuracy
of the method used to record tho moments in the water tank. It is believed
that the CJJ\max values are accurate within! 5 °/o.
B. Sources of Error in Test SotUp
A casual inspection of Table II show3 that the coefficients here presented
are in general quite high. Since the recording system is known to be reasonably
accurate, it is nec~ssary to examine the test setup for ~ources of error.
The fundamental aim of the testo is to duplicate as cloeely aE possible
the condition~ which occur when an airship runs into a horizontal gust or
wind shift. ne principal defect of the setup used at present is that the
thrust rema.·ns in the direction of the original flight path. In the actual
case the direction of the thrust changes with the angular displacement of the
ship. This has the effect of gradually applyjng a side force at the center
of buoyancy equal to T · sin cX. , where T .,.. thrust aua ol ~ angular displacement.
In addition, the. dra0 01 the model ir. the direcio of motion of the
tow carriage increases with the angular displacement. This puts greater
loads on t he croQs carriage vtheels and bearings. Hence it ie quite probable
that the friction of the cross carriage varies throughout the run, increasing
'vith the angular displacement. This frictional force ~s e1uivalent to an
equal force applied at the center of buoyancy and directed i~to the &urt.
Both of these effects tend to increase the oending momont beyond the value
that would result from true free flight conditions .
These sources of e rror are both a maxi~lll'l'l at the maximum angle of attack.
It is interesting to note that for >l,l ~ . 567 jr. particular, the ~Axi~um
cm generally occurs p1actically at tl wllTne time that the angular displacement
is m.aximwn. Table III illustrates this fact .
1/3.5
0° rudder
QO to ~20° rudder
Vu_/·V e 1/4 .5
0° rudder
OC to +200 rudder
+20° rudder
10° to +10° rudder
+5 ° initial yaw
5f'l initial yaw
v 1v • i; E:v" 5.5
o0 rudder
o0 to t20° rudder
6
Table III
Daniel Guggenheim Airship Institute
Akron, Ohio
Position of nose of model acros~ gust
in hull lengths (~/..! )
for Cn:.max (x/£ ,. .567) for ol max
1. 45
1.5
1.4
l . 45
1.4
1.5
1 . 45
1.3
1.6
1.3
1.5
1.4
1.6
1 . 35
1.4
1.45
1.45
1.55
1.8
1. 45
Again, with the model set at ~5° initial yaw the errors woulu be still
further increased.
In all cases Cm...__ for x/.L c .567 occurs v1hen the model is almost entirely
.n:a.:x
in the fully developed gust. Since the leading ed~e of the fin is approximately
0 . 2 of the hull length from the tail in all oases the fins ·would be
at lea.st purti&lly in the fully developed gust ,,r.en the m.axi~um bending moment
occurs . (The gust transition zone ends at ~/.f  .5) In some cases
Cnlmax does not occur until the entire model is corpletely in the fully developed
gust. For the e.ctual free fli e;ht case it aprcars somewhat doubtful
th~t the maximum bending moment at Eections in the forebody of the ship
should occur at this time.
The above considerations indicate that the tests here roported probably
overestimate the hull bendinf; momentt:.
C. Comparison of Experimental with Computed Cur~e~
In Fig . 20a the bendinb moments measured at x/,/  . f.67 and • 736 for
Vg/v • l/S:5 and o0 rudder e.rtt compared ·with the computed coefficients
for the same case as given in the report on Item 10 of this contract. Although
~he computed coefficien~s a: ply to a cro~ssection lying between the
two experin1ent&.l curves, the gererb.l r.ature of the computed curv6 is entirelJ
different from the experimental ones. A comparison betwe~n the oricinal
displacement records for the two cases revealed considerable differences,
especially in the angular displacement. This is shown in Fig. 20b . Accordingly
, comp·1ted co~ff'icients were v;orked out from the new displacement
record for x/,,/ s .567 and . 736. These new comput~d curves are sl~o given
'
7
Danjel Guggenheim Airship Insti tute
Akron, uhio
in Fig . 20a. The differcnces between the old and new displacement recor ds
are no doubt due to the higher moment of inert.ia of the ~resent model .
Since the model is unstable for o0 rudder up to about 14 yaw, the tur ning
moment about the ceni:;er of buoyancy on the old model vrith th'=' sms.ller moment
of i 1 urtia v:ou" • .. rease faster than for the present model, as is sho,·m by
the curves of oL/dt2 in Fig . 20b.
Ylhile the curves computed from the present tests ere somewhat in line with
the experimental values, the agre rront is not very satisfactor y . It appea r s
that the computing of the bending moments b&.S"ed on analysis of the airPhip
moveme1,ts is not practical due to already previousl~ suspected inherent
weakr1esser:: of the necessary mathematical appcoa ch .
D. Gen~ral
In Fi gu re~ 6 , 7, 8 and 12 the experimdntal curves are somewhat irrefular
from the origin to the mo. ~r~L value. In generRl, a bump in the curvts
occurs at approximatet~ I,! ~ . 5. This irrecularity ie absent in the
other curves. It is ez.L5 ,;_,, that this bump is due to some irregule.ri tj in
the mo~ion of the model rather than to Gust forces. Possible causes are
1) oscillation of the model in the direction of the towing force ca.used
either by a jerk at the start of the run or by variations in the towing
force, 2) e. small irl·dzularity in one of the v;heels or cross rails, and 3)
oscillation of the model in the direction of motion of the crosB carriage .
Th~ relative magnitude of these irreGularities wa~ not suspectec f r om the
visual examination of' the scratches made iI!Ullediately followint; the test runs
and when the photographs v1ere meai;ured it was too late to investigate thi~
point. lt is quit;., probable that tho e.ctua.l development of tht; bending moment
in the gust is more regular, as shown by the light line in Fig . 6.
It should be noted from Table II that the equivalent angles of yawed steady
flight are in general higher for the two cross sections behin·i the center of
buoyancy than for those in the forward part of the ship and are also higher
than the maximum effective fin angles taken from the report on Item 10 of
this contract. The maximulTl effective fin angles, however, v;ere figured for
the model with smaller moment of inertia and generally occurred about .2 or
. 3 hull lengths before the maximum Cm v&lues obtained from these tests.
VI . COllCLUSIO. S
.n.. The results here presented do not apply exactly to the pr oblc:n of an
air~hip in free flight encountering a gust, due to the direction of thr ust
and to the friction in the c r oss carriage . The effect of these errors is to
incroaso the b~nding m~~ent~ over free flight conditions .
B. The check between the experimental and computed montents is not good and
tho bendi11g moment coefficients computed frorn previous t ests shoul d be dis regarded
.
C. Further research vii th a selfpropelled rnodel should result in more
accurate knowledge of gust bending moments .
TABLE II
l.aximum Cm
 1) x!t fj = • 234
vdv  1/3.5
'Jo rudder .040
)0 to +20° rudder . 046
Vg/V = 1/4.5
oo rudder .026
po to t20° r udder .032
..20° rudder .037
 10° to +10° r udder .032
+50 in:itial yaw , 200 ~ms . .056
side fo rce
_50 initial yaw, 200 gms .  .030
side force
Vg/V ,.. l/s .s
00 rudder . 030
oo to +20° r udder .036
l) x  distance of section from tail end.
~  l ength of airship model .
. 387 . 567
.116 . 158
. 117 . 15G
. 078 . 124
. 077 . 12~
.067 .102
.090 . 152
. 140 . 250
 .053 i. 095
.060 . 101
. 069 . 104
•'
. 736
. 065
. 063
. 038
. 048
. 04:6
. 048
. 084
.030
.038
. 041
Approximate angle of steady pitched ..cil )('. •
flight to give equivalent Cm
oleff
l~ . 234 . 387 . 567 .736 of fins
15° 15° 11° 12° 13 . 1°
16° 15° 11° 1 1~0 . 9. 3°
12° 11~0 .... 80 70 11.4°
13° i1J...0 80 91.,_...,. 0 6. e0
14° 10~0 70 80 7 .o0
13° 12~0 11° 8.~... 0 
18° 17° 17?0
.., 15° 15 . 3°
lo
12~ _90 _70 _50 5 .6°

13° 10° 70 70
I 10 . 7°
14° 11° 7~... 0 70 I 4 . 9°
I
FIGURES
Daniel Guggenheim Ji.irsh · p lnsti tute
Alpron, Ohio
1. Outline diagram showing cro ssections at wl~ch bending moments
;ore measured.
2. Photograph of schematic drawing of bending moment model .
3. Photot;raph showing method of' determining the moment of inertia
or the model sections.
4. hotograph showin~ recording mechanism and lead weights inside
model .
Su. Photogr ph shov1ing typical record obtained at ./' ~ . 567 .
5b.
6.
7.
s.
9.
10.
11.
12.
!g/v ~ 1/4.5 , o0 initial yavr, rudders moved i.. >0 to 20° in
approxirn tely one hull length.
Photograph ehowin~ typical record obtained at ~ ~ . 387.
Vgjv • 1/4.5 , o0 initial yav, rudders moved ir J 0 to+20° in
approximately one hu'l length.
Cm • r ( s/ / ) , v g/v  1/3.5 00 initial y 00 rudder. ' '
Cm z: r(s/ j ), Vgjv  1,13 .5 00 " II 00 to +20° rudder. ' '
Cm z: r(s/J ), Vg/v !Z 1,/ 4.5 ' 00 " II ' 00 rudder.
Cm  i'(E 1,1 , "'lg/v r 1/4 .5, 00 II II 00 to +20° rudder. '
Cm f(S/j), VyV : 1/4.5 oo II " + 20° rudder
' ,
Cm  f{S/ /)I Vg/v  1/4.5 00 " 11 10° to 110° rudder. ' '
Cm f'(S/ .J), Vg/v .. 1/4.5 +50 " II t10° rudder, ' '
+200 grams side force, r udders moved to 120°.
13. Cm !'! I.'(S/j), Vg/v = 1/4.5 , 5° initial ya"'• 10° rudder,
200 grams side force, rudders moved to +10°.
14. Cm  f(S/~), !g/v  1/5.5 , o0 initial yaw, o0 rudder.
15. r(s/~), Vg/v  l,ls .5 II " , o0 to +20° rudder.
16. Crosswiae and angular displacements and velocities for Vg/v = 1/3 .5 •
17 •• Crosswise and angular displacements and velocities for Vgjv = 1/4.5 .
18. Cross 11ise and ungular di"spla.cements and velocities for V g/v = 1/5 .5 •
19. Co~ ri~on of measured and computed bending moment coefficients for
steady ya~ed flight, o0 rudder.
20a.
Daniel Guggenheim Airship Institute
Akron, Ohio
FIGURES (Cont.)
Comparison between experimental
Vg/v = 1/5.5 , o0 initial yaw,
and computed values of Cm ,
OO rudder.
20b. Comparieon of crosswiae and angular displacements, velocities
and accelerations for old and new models under the same test
conditions. Vgjv = l/s.5 , o0 inibial yaw, o0 rudder.
•
V NlE:G""GU"GGEl'T.BJli.ILm::~
AIRSHIP INSTITUTJD
AKRON, omo
•
Daniel G ugg enhe i~ AirEhip Institute
Akron, Ohio
s t
DIN
t. TRIC MO TOR
F R ROTA7 ING
GL .4  CYl IND£R
~~~·~o~
.:_/__ ST~£L COIL SP /f\,
FL£Jl18l t. L UOP£D
RUBBER SHl ET JOIN I
~
DIAMOND POINTS
/'/OTOR UNITS
5' TIN RL ODER
i  ' J .
' ' I ' ' l '
I
I
I
'\ L£V£R S 'S TEM BALL BEARING
CARRYING 01 fOND POINT
Fr>R RECORDIN  BENDING MOMENT
•
Fig . 2 . Photog r ~ 1h of schematic dra ing of bending
moment mode l.
\
•
Daniel Gue;genheim Airship I=istitute
Akron, Ohio
'
•
Fig . 3. Photo~raph showing method of deter:r.ining
moment of inerti~ of model sections .
•
•
Daniel Guggenheim Airship Instjtute
Akron, Ohio
Fig . ~ . Photo~raph ~howine: reco rding mechanism and
lead weights insjde model ,
/
\
Fig. 5a.
'
Photograph shov1ing typical record obtained
Vg/l/ = 1/4.5 • o0 initial yaw. o0 rudder •
•
 .567
•
,.
~=· •
•
•
\
\
\
,
Fig . Sb . Photograph showing typical record obtained at x/t, • . 387
Vgjv • 1/4.5 , 0° initial yaw, 0° r udder •
•
l/J6,4 l:t/
• (J ...
•03. ''Z.
·~~
·O.~t ·1'8
0¥
~ ..
.,,z
•
DANIEL GUGGENDlll
AIRSHIP INST1'1'01'B
AKRO
J::; "387
~: ;l.d'
':. .z~s4
~t:4~~~~~~4+!7f~~~;ri
~D
~..No/ JtE ~.s. ..Aceo.s
li.osr IN. ill4a{~ENfYlfJl
DANIEL GUGGENHEIM
AIRSHIP INSTITUTE
...~_,.. . .....,~~~~~ ....,..._...,.....,... ri~~....,..,................ ........ AKRON omo
•zo
..............~ . .. c>...·1 oe~a~
 _ ._,,_ ~ Z3'f.
~  'h
/;o Z·o  fi fi~.tr11£N t'J.F. Nd.st£. .0£ MPPel, A&e~.r.
,.i.....,...~ ~ .. " _· j_: !fv.sT 1N flt1a ltN4.TH~
~r~+~~4~Ba~r..._+..!_..,__.,~
..., .0..r.1 ..Lill.a., \:l'U l:ttl'J!;.ft ti BUii.
AIRSHIP INSTITOTJI ~,
AKRON
.. uz.., .;" ~ 41 .,0,q, .~~"'"'"
~
• (}  ~  f () ~
•
•Id a ... #I~
()2.
(}t •
.. at~ :4J_ rO¥
• ·llJ "
.u ....
..
DANIEL GUGGENHE
AIRSHIP INSTITOT
AKRON OHIO
~O~~~
•
•
~ l)o 2'·b ~ = · t 'i
~ PA41.rl<'4 b.F ~SE llE h/@£1. ~ ~
&t!sr tl)f H~G f.8.:tqrN,s
NJ .. yi{ !Uv ~ ~~~"
MtrrBf ~J.v
~do~.p6&
t
•
.10
. .,.
.r
DI . '"' •• P_w,rf"1< o, ND.SE 4"' .HiUJ~.L
A<,ltlA 6v.1l" N ~' 1.#'.N'~

1 I
r •
, __ .... , / •
 
 •
•
c
. /
'
CORR :CTIOl
to Report on e surementE of tho Bending komente in the Rull of n
Airship odel by later T n T ts (using rtificial 6uat th
tr n ition zone equ l to on h lf hull 1 n th)
It m 11. Contract NO 47286
i . G nd 7 t Ch n e C " nd Ill
3 le~ to agree 1th th followiDC ro.tios:
c "
.. ,
1. 75 c
.438 Cm
The equiv lent vuluo of Cm for tho oorr•ot valu
in tho follcnrln tt'lblee:
,,
.10
.20
.so
.40
qui lent Cm
.057
.114
elr/l
.228
t ,,,
.02
.04
.06
.08
.10
of '' nd
bqui lent Cm
.046
.091
.157
.185
.228
,,,
Fig • l and 15: Ch ng 'TI" • nd
,,,
·oalc to gree \rl th the tollowin
Th
ivcn in
"
'"
2. 75 Cm
. 686 c
equivalent v lue of Cm for th
the following blee:
II quive.l nt om
.10 .056
.20 .073
.so .109
.40 .145
correct valuee of r •• nd "'
t •" Equiv l nt Cm
.02 .029
.04 .058 •
.06 • 087
.OB .116
re given
tiosa
re
Daniel Guggenheim Airship In~titute
Akron, Ohio
CORRECT101 SHE~T
for Report on ~easurement of the Bending Uoments in the Hull of an
Airship 1.odel by iia ter Tank Tests (Using Artificial Gust with
Tr11nsitlon Zone Lqual to OneHalf' Hull Length) .
Item 11 of Contract NOc47286
Fig . 16. Change Vgjv = 1/3 •5 to Vgjv a l/s.5
Change fi&ure number to read i 0 • 18
Fig. 18. Change Vgfl = 1/s.s to Vg/v  1113•5
Change figure number to read Fig. 1€ .
"
LJ Nl1'iL 



A 

C 

D 

I 

M 

S 

U 


