1.4i 16 - nasa · pdf filetable i lists the material weight ... inflation 3.5 x rated...

28
_ 1.4i 16 https://ntrs.nasa.gov/search.jsp?R=19740004392 2018-05-09T00:53:44+00:00Z

Upload: doanthuan

Post on 06-Mar-2018

221 views

Category:

Documents


2 download

TRANSCRIPT

_ 1.4i 16

https://ntrs.nasa.gov/search.jsp?R=19740004392 2018-05-09T00:53:44+00:00Z

N7&-1250!

INASA Contract No. NAS-9-12040

DRL No. Exhibit"A"

Line Item No. 9

DRD No. MA 129T

dt_i ' 7

ENGINEERING REPORT NO 4230

PART II

NASA Wheel and Brake Material

Tradeoff Study for Space Shuttle

Type Environmental Requirements

May 8, 1973

B.F.GOODRICH AEA£.SPACE & DEFENSE PRODUCTS

Wheei and Brake Plant

T roy, Ohio

E:

_._._ •

_b

Chief Engineer

i

III

I

ER - 4239

FSC 97153

['ITLE PAGE

TAB i.E OF" CO?J i'I£?<TS

DISTRIBUTION I,IST

TABLE OF CONTENTS

_E) a_ e

i

iv

AB,qT R A C T

m_

IN TRODUC TION

RESULTS

Brake Material Trade-Off Study

Wheel Material Trade-Off Study

Wheel Design StudyC ONC LUSIONS

BRAKE MATERIAL TRADE-OFF

Brake Calculations

Loads

Materials

Analysis

Results

thru 5

and 3

7 thru 11

7 and 8

7

7

9 and I 0

It

WHEEL MATERIAL TRADE-OFF STUDY

Wheel Calculations

Part I Material Trade-Off

Loads

Materials

Analysis

R esults

Part II Loads Trade-Off

Loads

Anal,(sis

Results

l 2 thru *_

12

12

12

13

14 thr_ 17

lg

19

19

20

21

i

I

REFERENCES

Written by:

L. D. Bok

Checked by:iii

23

t,_.'.__,:'.,,._ _u_ BLANK NOT PIL_IP_/)

ER-4239

FSC 97153

i

iI

III

I

I

I

I

I

I

I

ABSTR AC T

NASA Wheel and Brake Material Trade-Off Study

This investigation covers material selection and trade-off for

the structural components of the wheel and brake optimizing

weight vs cost and feasibility for the space shuttle type ap-

plication. Analytical methods were used to determine section

thickness for various materials ending with a table showing

weight vs. cost trade-off. The wheel and brake were further

optimized by considering design philosophies that deviate from

standard aircraft specifications and designs that best utilize

the materials being considered.

V

ER -4Z39

FSC 97153

r_

INTR ODUC TION:

The purpose of this study was to design a lightweight wheel and

brake which would be lighter than the latest state-of-the-art

designs. To insure optimization between weight and cost, five

basic metals were considered: Steel, titanium, aluminum,

magnesium, and beryllium. For the wheel, the optimum

material was selected and the design was subjected to four

additional loading philisophies.

1

!

ER -4239

FSC 97153

RESULTS:

BRAKE MATERIAL TRADE-OFF STUDY

dlr"

Table I lists the material weight cost trade-off for the structural

components of the brake as shown by 14-1493-2. These components

when combined with the heat sink material, make up the total brake

weigat. There are two lightweight and heat sink materials under

consideration for the space shuttle type application; one bein_

structural carbon and the other being carbon or sintered iron lined

beryllium. The heat sink evaluation and study is reported in

Engineering Report No. 4239, Part III.

TABLE I

BRAKE MATERIAL TRADE-OFF STUDY

.m

WEIGHT WEIGH f BUDGE TA R YATEBIAL LBS SAVED (LBS) PRICES **

Piston Housing

Alun_inum 15.4 .... $ 340. O0 ....

• cosT/ ]

LBS SAVED/

tMa gne si.urn

Torque TubeQ

9.6 5.8 $ 400.00 $10. Z0

Steel 34.6 .... $ 525.00 ....

Titanium 19.7 15.0 $I, 878.00 90. Z

Beryllium lZ. 7 21.9 $6,600.00 $277.69

Back Plate

Fitanit_m 7. 769 .... $ 800.00 .... II

* COST/iBS SAVED = Difference in Price ** Based on 100 Piece Order

Difference in Weight

Z

ER-4239

FSC 97153

- _. j

I

I

I

I

i

|

I

I

I

I

I

I

I

RESULTS: (continued)

The magnesium piston housing offers a cheap weight saving

but presents a fire hazard. The fire hazard would be more

probable with the carbon heat sink since it must operate ap-

proximately I('0("(;V hotter than the bcry[liun_ heat sink to be

weight effective.

Three materials were considered for the torque tube. Steel

being the state-of-the-art for the high performance carbon

heat sink and titanium state-of-the art for the beryllium heat

sink. The titanium torque would need to be proven compatibl_

with the carbon heat sink designed to the space shuttle require-

merits. Beryllium, the lightest torque tube has not been tried

in a torque tube application although it has proven successful

in other structural applications. The high heat capacity and

conductivity of the beryllium torque tube could make it com-

patible with both the carbon and beryllium heat sinks.

Both beryllium and titanium back plates are in production on

high-performance military aircraft. The material selection

for the back plate depends only on the cost trade-off.

WHEEL MATERIAL TRADE-OlaF STUDY

Table II lists the material weight cost trade-off for the wheel.

The table lists the aluminum and beryllium wheel weights

based on the same design, and a beryllium wheel with the

design optimized for the beryllium material.

ER -4239

FSC 97153

R ESULTS (continued)

TABLE II

WHEEL WEIGHT - COS _ TRADE-OFF

'Q

Material

Aluminum

Beryllium

Optimized

Beryllium

WeightLBS

116.5

79.9

70.4

Budgetary

Cost _-

$ 1325.00

$7200.00

Cost/

Lbs/Saree

$1530.00

* Based on 100-Piece Order

WHEEL DESIGN STUDY:

The design philosophy and specifications used for commercial

aircraft wheels were reviewed for possible changes that wouId

allow additional weight reduction. Table III shows a comparison

of design conditions vs weight for the optimized beryllium

wheel design.

4

IIIl

ER- 4239

FSC 97153

RESULTS: (continued)

TAB [.F III

DESIGN CONDITION VS WEIGHT

g,

!.

{

_v

=t

1

!

!

!

'Load

Condition *

1 **

3

4

Ti.re

Deflection

@ Rotor Load

@ -55°F

37%

3z%

40°70

40%

40°70

Burst

Pre s sure

t. 5 x Bottoming

Pre ssure

3.5 x Rated

Inflation

I. 5 x Bottoming

Pre ssure

3. 5 x Rated

Inflation

3.5 x Rated

Inflation

Design

Load

Limi.t

Comb.

Ultimate

Comb.

Limit

Comb.

Ultimate

Comb.

Limit

Comb.

** Load Condition used in the weight cost trade-off study.

* Roll Life 100 miles for all conditions.

\Veigh_

70.4

91.6

68.6

84.7

77.2

!

ER -4Z39

FSC 97153

C ONC LUSICNS:

1

I .

.

.

The lightest brake would consist of a magnesium piston housing,

beryllium torque tube and beryllium back plate.

The magnesium piston housing and beryllium back plate

are cu:rrent state-of-the art designs and need no develop-

ment. The beryllium torque tube would require development,

but would offer' a low cost weight saving even when compared

to the titanium torque tube.

The beryllium wheel offers the greatest potential weight

saving in current lightweight wheel and brake designs. The

development of the wheel would be expensive and the appli-

cation would have to justify $1,500.00 cost per pound weightsaved.

The wheel design study indicated that design philosophies and

specifications used for commercial and military applications

could unnecessarily penalize the space shuttle, weight wise.

Careful consideration should be given to tire deflection,

wheel burst pressure, and trade-off in designing the wheel

to limit or ultimate load conditions.

I

I

I

I

I

I

I

i

!

[

[

1

I

I

I

I

I

ER -4239

FFC 97153

BRAKE MATERI%L TRADE-OFF STUDY

BRAKE CALCULATIONS

Loads

The brake structural components were designed to the following load

conditions:

Condition Load _ rot_gth (" ,,; to I-i:,

Normal Operating Pressure 1000 psi

Maximum Operating Pressure 2000 psi

Proof Pressure 3000 psi

Burst Pressure 4000 psi

Max. Static Torque (0.55 Gnd. Coef.) 54700 lb-ft

Hot Ult. Structural Torque (0.80 Gnd. Coef.) 79564 lb-ft

Cold Ult. Structural Torque {1.20 Gnd. Coef.) 119400 lb-ft

N_ Yield

No Failure

No Yield

No Failure

No Failure

Mate rials

The following materials were considered for the various components of the

brake. Aluminum and magnesium were not evaluated on the components "n

direct contact with the heat sink because of their temperature limitation.

See T_ble IV for material properties:

A. Piston Housing Assembly

1. 7175-T66 Forged Aluminum

Z. Ti-7A1-4mo Forged Titanium

3. AZ80A-T5 Forged Magnesium

4. S-240 Beryllium

B. Torque Tube

1. 4140 Forged Steel

2. Ti-7A l-4mo Forged Titanium

3. S-240 Beryllium

C. Backplate

1. 4140 Forged Steel

2. Ti-7AI-4mo Forged Titanium

3. S-240 Beryllium

ER-4239

FSC 97 153

Maximum operating temperature for each material was limited

to that temperature which produced a loss of 40 percent of room

temperature properties.

Steel 950°F

Titanium 900°F

Beryllium 825°F

The piston housing was analyzed for roor_ temperature only be-

cause it is insulated from the heat sink.

TABLE IV

MA TERIAL PR OPER T_ES

MATERIAL

7175-T66

Aluminum

Forging

AZ80 A-T5

Magnesiurn

Forging

S-240

Beryllium

Ti-7A 1-4Me

ROOM TEMPERATURE STRENGTH

UltimateTensile

(p,i)

q

86,000

YieldTensile

(psi)

76,000

50,000 34,000

,,, , .., , ,

40,000 30. 000

, , |1

Ultimate YieldShear Shear

{psi) (psi)

,', , _ i _' " '

24,000 18.000

MAX. OPEBATING TEMP.

Ultimate

Tensile

(psi)

24,000

(825°F)

ie ld

Tensile

(psi}

18,000

(825°F)

Titanium 160,000

, , i

4140

Steel 180, 000

Forging

[50,000

163,000

96,000

108,000

90,000

98,000

96,000

(900°F)

108,000(gS0°F)

90,000

(go@F)

98,000

(950°F)

STREINGTH

Ultimate Yield

Shear Shear

(psi) (psl)

.... :_

14,4000

(SZS°F)

57,600

(900°F)

64.800(950°F)

11,9000

(825°F)

54,000

[900°F)

58,800(950°F)

I

I

I

I

I

I

i

!!

[

[

[

[

ER -4239

ESC 97153

ANALYSIS

The methods of anaysis are shown here and not the actual values

for each component of the brake:

Piston Housing:

d = 14.5

d e = 10.0

1.5Moment at A =

fPT

d I

_'B

PT = total thrusl load= 19.8p

(d- d A) (d I + d2)2 12, d (6co + d I - c,2)o03 - a (),a 3 -_,a, _ b 3/k I) 1 )11"3

1.5 (d, +d2)2/6_*d , -d2)6o3 +a(Xa 2 - _a, Xb 2 /Xb,)It :_

where: _a 1 =

Ab I =

Aa 2 =

Ab 2 =

_,a 3 :

Xb 3 :

b =

24/l(a/b) - I]

IS.6l(a/b) 2 +.S3911a/bl(alb) 2 - II

15.6[.539(alb) 2 + 1]l[(alb)2. I]

24albl[ (a/b) 2 - I ]

{(2.483 Inalbl[(alb) 2- 1]} + .668

I(2.483 alb In a/b)l[(alb) _- I] } + .6681alb

dAl2

dB/2

l

[l

I

i, I!l

Moment at B = (Xb3P T + Xb2M A)/Ab,

and stressesat A & B are:

aA = 6MAItA2

a s : 6M B/t 2

ER -4239

FSC 97153

Mo & Vo are determined from the deflections & rotations of the cylinder - headjunction - Ref. 1.

vot o

t c

D

0 c

6Mo Vo

tc 2 tc

O D =

6Mo pr+

tD2 2t D

Torque Tube:

..p....,_ T t

Torsional Shear Stress: r I z

T = total brake torque

16-T- D O

11"(O o4 D|4)

Backplate:

where:

PT

t

= fn | ro/r I }

po= PT/lr (ro= -rlZ )

r o

ri

Ref. 2

10

")

/

I

I

I

I

!

ee

I

I

1

I

_R -4239

FSC 97153

RESULTS

MA TER IAL

Aluminum

Magne sium

Beryllium

Titanium

PISTON HOUSING

15. 347 Ibs.

9.619 Ibs.

lO 127 Ibs.

23. 050 Ibs.

TOIR QUE TUBE

12.745 Ibs.

19.699 lbs.

BACK t_I.A i /

Steel 34.317 lbs. It. 794 lbs.

The following materials are recommended for the various structural

members of the brake assembly based on weight saved only:

Pi ston Housing Magnesium A _ 80A - T 5

Torque Tube Beryllium S-Z40

Back Plate Beryllium S-240

J

r

.

!

l

!11

ER-4Z39

FSC 97153

WHEEL MATERIAL TRADE-OFF STUDY

WHEEL CALCULATIONS:

PART I: MATERIAL TR%DE-OFF

Loads

?

9_

The loads shown assume a 49 x 17 tire loaded to 37% deflection:

Inflation Pressure

Burst Pressure

Limit Combined Load

Roll Load

Z93 psi @ R T

224 psi @ -55°F

510 psi @ RT

168, 000 lb. radial

+ 59,0001b. side

@ -55°F

60,000 lb. for 100 miles

@RT

lZ

I

I

I

I

I

I

I

I

1

I

'I

r_- r

_T

I

I

!

1

I

'!l

FR-4239

FSC 97153

MA TER IA LS

All materials are forged except for beryllium which is hot pressed.

See Table V.

TAB LF V

MATFRIAI. PROPERTIES (Values are in ksi)

Material

7049-T73 Alum.

2014-T6 Alum.

7075-T73 Alum.

7175-T66 Alum.

ZK60A-T6 Mag.

FK31A-T6 Mag.

Ti-6A l-4V-Tit.

-- -L

Ti-7A 1 -4Mo Tit.

4340 Steel

tu

F

72.0

65.0

66.0

86.0

42.0

45.0

160.0

170.0

180.0

4140 Steel

m

I.,

.,,I

i i

62.0

55.0

56.0

,,,

76.0

26.0

26.0

150.0

160.0

163. 0

163.0

tu

74.0

67.0

68.0

88.5

45.0

48.0

181.0

192.0

186.0

186.0

71.0

ty i

57.5

58.5

79.5

28.5

28. 5

171.0

182.5

170.0

170. I)

i

1

60.0

F ', 1 ttl_ :-"

10.

c).o

105

5. 0

5, I)

180.0

(Brush S-350) 66.0

Beryllium

55.0

38.0

38.0

!13

ER-4239

FSC 97153

ANALYSIS

The wheel configuration used for this part was a split wheel held

together with bolts and one bearing in each wheel half.

The methods used include analysis for rings, cylinders, and circular

plates. The bearing reactions are calculated using the methods given in

the "Timken rngineering Journal" (Reference 3). A computer program

was used to calculate the section thicknesses required. This program uses

the thickness as a variable and iterates until a specified margin of safety

is reached. The results of the computer are shown in the following pages.

14

I

!

I

!

I

I

I

[iI

[

I

[

_L

EI_ -4239

FSC 97153_s_},:r[L ol_:tr,ll.Paoor_i

k_4[IL - rlASA SPACE $ItU_TLI "" "

/;_T'L -7:,+;-_;; r_P,Y.:_ :,L;'|

DE$1f+I; Lno','+$ -

| B_rtST S]O.PSI

7 ULT CO_!T_ IG'_O(_0.P,;.ntAL 29690.SInF

• --- $ ROLl. {_'UOD.q_I,_L Ftlrt 10G.tllLIrS - "

SURF CO_PF.E,%$1V[ ST_rS_, 1$_._q.P$l

S[CTiOti X THICKr:E$$ i:$-r, UP,$T _S-ULT PfS-COLL

COMPUTER SOLUTION

IU_IC I._llffL _[Slttt pROC.fl^tl

I/If(EL - NASA '_P,_C[" StlUTTC[

I1_1'L -:=::,-T_ All:l: f()_r, tt_

I1($1(;!i LO,'m_ -

I [WIL, T 516.PS1

2 L_L'T (',',,':_ _C:;9"0.r_'_lt_L $9000..51_[

} ROLL _C_:_Y_h'_,_I,_L F_':l l_O,t'lL£$

SUrF CO!IP_ESSI'/r C.T_*rs$ 150C0.P_'1

FLAtIG[ O.5'_C0 59.1 36.1 1.O

Ill z O._,_ 0.7£."0 55.1 36.7 0.1

R _ 0.$1 0.7C.0 53._" 36.2 0.0--- Rh. %.11 O. {; 3.._ 5_.; 3_.2 C.S

Rill 1.56 0.55C:_ 55.$ }G.2 0.7

-- RI_I 1.%;, 0._. ,_ O _.7 36.1 C.7

RI_ 2.31 0.3w_0 SS,9 35.3 0.9

i;l_l 2.6_' 0.1770 5:.7 36.1 0.7

Rill ].OG O.Z_O 57._ ]5.t 0.0

---RIM - 2. _.;, 0.2270 5_.6 36.0 0,6

RUt 3.S1 0 .... 5_.5 35.9 0.5

RP4 L19 O.215_ 5_.9 36.2 C.3

RIM I.SG 0.17:0 53.& 36.2 0.3

- - Rl_l i._:_. 0.1_O 53.0 35.5 G.2

RIl:tlEt'l E.$5 O.:bO 152.9 32.3 0.$

-- RII_IIE_ ).0_ 0.I$_0 151.7 ._2.6 0._

Rlt;:_8 7.22 0o16=_ 102 .2 3_*.2 0.9

RIt:;IrB 7._?, O.."3_,0 172.7 3_.k 0,6

RI#',;;E P 7.6_ 0.:=30 107.1 3L2 O.t,

Rib:lit..5 1.79 0.3:!5 ).22.7 _i._ 0.6

RII::IEC 7.97 0.3507 102.= 3_.2 0.9

--- RI_':;;[B 8.26 0,37C,8 131.7 32.6 O.i

It Ir'l:Ell 8.35 O. _'_73 152.9 32,3 O.S-- TIEBC_LT /J*Zl DO_ 0.37_ 23}.3 151.3 37_._

IlEB-l_O :.O700 11_0.9 0.I 100.0

gEB-RI 1. i600 111,_. C 0o0 100.0

HUts O. kOO.0 11¢,0.0 0.1 100.0

I/H£EL Dltfl_NSIO:IS -

.... 0'.20.0_0 X- 1.720 ;,_* 1.2C0 HC- 1.500

1_1-19.',30 02-16.750 H'" 1.853 T2" 1.250

- - tl" 9.60U kl= _,,7.50 P.O= -_.lO0

----TIRE OD-k7.9$O BOLT _,T_- 1,_K,,5,!

Irte- C_r - 0.$31_ ].7.3 0.6 ."7.'9

P,Itt 0.:*_ 0.CC_O 17.6 0._ 2_._*

_llt 0.:I _. {2.C 17._ _.7 7_.L

hilt 1.19 0.5_7_ 17.G _.9 2&.t

I1.111 1._b 0. _C,," _ 17.0 0._ 27.k

l*. l el 1.97. 0.;TCJ ZC.; 0.1 ."7.1

l/Ill 2.31 (_. : I,: LI _7.7 1.0 ._._

RI'I ,_.C9 0. _. C,', 7 I7.7 _..9 ?I.S

P.ttl 3.CL, _.l,:0 :[.7 _,.1 :7.c

_P.III . .2._:+ . 0.1970 _.7._. C.i¢ :7._

R_/I 3.;: O. _;<:-c ;.;.5 0.,; =_.-_

Rl_ Ll._] 0.1r_c 17.C 0.9 7_,I

i Pill _.:; 0.I]:_ 07._ :.7 ::.:

P. I_!,:._ B I;.7_ O. _t,O 10:,.5 ,_.3 :.7!

;'I'_,,E_ 7.5;. 0.1710 37.C 0.3 3,_._

RI _;.r'3 7.._: 0. :.526. 66.1 1.0 :_,3

I rll_;.[ 6 7.CD 0..%;.ll 71_.; 0.5 -r_ -,

I RI,",_;E C 7.7_ O.._i,},10 $1.I tl.J 36._f',l t_IzE_ 7._7 0.;:?I 66.1 _.0 35.3

._ lli::_;._9 3.16 0.35C_ 17.5 0.8 3C.?ill.'. [ fl 3.35 0.::1_ 1P, I.5 n.3 37.1

TIESCLT I;'18 i_O* 0.;7_ 71_,.1 ' 1=2.= 373.Clie E .,',O 1.1L_0 I_CO.O 0.u IC0.0

l_EB-rti I. _'?o0 ; _C,_.O _.7 1_0.0

IIUO O.GCO0 1000.0 0.9 lq0.O

t/tlEEL- DIt:-_rlsIO;'S -" ..............

_ P-._O.CCO H* 1.750 l:- 1.2r}0 t;r,- 1.260

01.15.:,36 r,._-IC._50 H'- 1.5_0 T.TM 1.250

I1_,. _.C00 Rl" k.7CO t10- _.z.CO

.___TIRE D:)-IC.9$O_ I;0LT 5=_- lS_i;SI ........

.......... _AS|C Ilt_t_l_l. n£$1f',;I pR_r.P, AM .... " " .......... BASIC llllFE{ l)E511?l PI;OC,_,A:I

" I_;!_IEI. " IVC,_ 5,'V,M: $tlUT'rLI ............. Illi[[L - II°L';^ ._t'AI'E $1'U'_TL[

I',._TiL -707.5-77} AL F_GII!_ tim',IlL -7_75-T_ra FGr:; ,*,LUP.

_-DE";IO *_ LOA.qS - " .... DESI_,II Ln/_DS -

I BU,'IST _,l(?,PSI 1 BU_S'I' $10.P.51

-' 2 ULT Cr_,_ ZCCCOO.I!_IilM. 55001_.$10E - 2 ULT COtI_J !C_,O_;_.ilt, l'llAL 5g0DO.5I_F

ROLL C30('..._,_PIAL FOP. 1J0.f11LE"- 3 ROLL CCOC":._,;._I,',L FfI", ].=0.lllLI-S

SURF CU;IPI_[S_-I_i[ ._T_FS_ 150U0.P_I . SURF ('O_!PZ[_,SI'/P STq[_, 1500'}.nSI

15

--L

4;

+

ER-4239

FSC 97153CO:,_PU'fER SOLUTION

IIA$1t_ Itll[l'L U[_tGII PiiOr.ltAl|

I/lirtL - I;A_,A sr,,_rr P,liUTTLI" . o

I)[SINII l.lsA:_$ o -- ....

1 Run.&! _,10. r&!

t Utl Cot_r'. I_.',0_=.R,_nI,%L Sqt_o0.Sl.£

3 nOll COOCU.It,%_IAL Irort :."_.,II[$

SUIIF f(':iP_¢f_._lV£ ";Tt:f.%S I_)GUO.PSI

$1[cYlo:: X

IrLA_:GE 0.}_70 .'5.C, ?.7 0.$

_lr_ 0.4; o. ,_z,7o 2s.7 ?.1 0.0

I'Jll 0.31 0. _'_7t) :5._ ).G 0.S

I11H 2.29 O..?-.20 2._._ 7.5 C.3

R|II 1.Sn 0.170'9 25.0 ?._ 0._

Rlt', 1.90 0.07_a .%.6 ?.7 0.0

Rill 2.31 O.o'31_ Z5.5 7.7 O.E

I_ltl :.C_ O.LZ;C, :5.0 3.0 1.0

P, Itl ).C £ O._'IS +.S._ 7.5 O.b

Rill _;. k_. O.gr.SG ?5.0 7.1 0.7

11111 .3.._ C. 0:+_,++ .+5.3 ?.9 0.0

_!II k.19 I_. t-,30 . :5.6 T.7 O.&

Rift k.5G O. 0i%'* 9 :5.| ?.ej 0._

_. Rlrl,lEB 7.1.* . C.101_2 1.17.-' 0.. _ _ 5.0 ....

nJt_.;EU 7.3. + 0._976 170.5 0.9 S.&

I.'llll_[5 7.7_ 0.07_i. • _27._ 0.6 2.2

Jt I':;_£11 ?.39 0.1600 2._3.." 1.0 1.4

P_IJ:I:[_ _,.07 0.?050 175.; 0._ S.k

P. li%.'r.B :.2G 0.2310 1.17.- _ 0.2 5.0 .....

RI:lZ:_B _.'+5 0..".C,50 2_=.9 0._ 5.$

TI(I_OLT 81-11 OO- 0.._?S _,Sn.S 89.4 5;,9.2

UC "_- i',O 0. G 7_.O .+_C_. 0 103 _33.0

IIEB-RI 3,]_'00 _0C0.0 0.k _$0.b

ItU[_ 0.15,30 2gc0.0 0.0 lkl.2

TIJII_I:J?E"-S I:S-_t;_I.%T I;S-UL1 llS*rtOLL

U'r_[EL OIt:ECSIOP:_ *

• 0-20.00_, H- 1.7S0 II* 1.2C0 t_,. 1.000

Dl-15.11r. 1,2-16.950 ,'- 1.950 T2- 1.150

Lie 9.600 I11- ".700 RO- ".400

|A$1C Utt_L DtSIIWI PftOCRJUt

$:lllrlL - llASA SPP, cr _IIUTTI_

pl_T _. -'_.,...:..,. TIT.y,j_-! + peP,_

I P+Ui: ,_ T $1,.P$1

) I:OLL C_SC_.;ZAnI_L ro,, lO_.ttlLr_

S[_TlOfi X TIIICI:III'SS tIS-LLrlST tl$-ULT it._*"tOLL

- ELAt_G[ - 0.3350 _2.._ |$.5 0.S

IIIPI O.;l_ 0.2'_++ 0 )._.? l._.¼ O.k

t:Irl 1.19 0.2.;_0 31.3 13.: 3.1

_.lil 1.5C, O. ;.c:0 _:.7 13.: 0.9

P, IJI 1.'_ 0.07_.0 )_.r, 1!,.7 0.I

Rltl 2.3I 0. _'J:5 _12.7 15.8 0._1

ffl_t .+.r,9 o.ro:o 3._._ 13.2 :.2

I,+I tt .;.CO 0.C:+15 32.1_ _)._ 0.5

i,+lit _. t+_. 0.0_[,0 3_.g 13.L 0.0

Jilt) _.3; 0. _1_: 52._ 2,5._ 0.$

trill _,.I_ O. _*::, 3:.7 13.3 0._

I1111 _.50 0. C%'.l, ._2.7 ;._.: O.cJ

l_. I .*1 k,.n:+ O. _+17 33._ :_.7 0._

I; IPll!EB 6.05 0.+.0_7 I')9.$ :.? 0.9

ItlPI;,EB 7.1.* 0.;_5 ;_';..* 2.L: 0.$

P.I_;[ C T.51 0.0_£0 ;65.5 _.5 0.k

Ili!txlES 7.70 0.(_;50 150.; 5.k 0.$

ptIJII:CI_ 7._g O.15r0 2_r.5 ?,G 0.4

IIIiil+£O 3._7 0.lgC0 _. Sll. 5 3., _ 0.Y

__ P,I"1%1[_ |.._G 0.-'._50 I09.10 2._ 0.:

i_ lrr, zr _ :. :,_, 0.:;_ _.; .%9 _._

TIE"_OLT rl-lS 00- G..175 161.5 , '_3.g 571.1IlE_-l:O O.GS_ 3100._1 0.2 251..1

I/E8-RI 1._360 31L'0.0 0.2 2"0.3

HUB 0.1k:0 3100.0 O.7 125.1

.._TI_[ OD-47.950 BOLT s'Te- 111011"-I

(_4E££ Dlr'Cr SlOt.') -

G-20.00_ i;- 1.750 II- 1..*UO l:P- 1.0C,0

D1"19.%._0 ,:-16.%O It'- 1.950 T1- 1.150

Lle 9.600 _Ii- k.700 RO- _._00

•TIOE. 00-:.?. ]r. 0 . BOLT ST".- 13012SI ......

IASlC Will[EL DESll_lll PIIOG_IUI_..

BASIC I/HEEL I_ESIr, il Pi_Or-_l

I:H/EL * PJA3A SPACE _JIUTTLE b'NErL - II,_SA SP_CE SJ+UTTL/

ItATeL - Zt:-G_,A I'..%r. FCR_ "- ............. t_&T'L -E;:;le,-T& ),_.,.r_ FDR_

OESIGI: LOA;)S * OESIGfl LO_e.OS -I B_RST S10.PS! ..... - I fl,U_S T 510.PSI

2 ULT CC:_0 16_0U0.RAI_IAL ))000.SIDE 2 ULT C_:_9 "-_-'_C._;,'_I_L _9:0O.Sli'_ir

3 le_LL G(_0CU._,e._IAL F.Otl 1Cg.J*.iLES 3 ROLL GC_0._,'_DI,*,L FO,_, _.HILE._

SUrF COl:;'_[S$1:'[ STi_ES$ G000.P$I SURF (O!!PRE._._JVE ':-T9£$5 60OO.PSI

$I[CTIO/! Z IHICKtlESS r+$*OURST ilS-ULT IIS*_0LL* $l[CTlOIl

FLAI:G£ 0.7;,30 17.S 0.l +$1.S

RIM O.&i 1.2200 17.1 0.t 31.O

lilt 0.11 1.1500 17.2 0.5 31.1

-- llJI 1.19 ] .0700 1C.7 0.1 30._

Rill 1.5G 0.9?_0 17.7 1.0 31.9

RIll l.tk 0.8950 17.6 O._J 31.7

RI_I 2.11 O. 7;;'_ 17.0 1.0 32._.

RIM" 2.00 O._C_O 17..$ 0.8 31.&

RIM 3.06 O.kG_O 1.'._ 0.6 31.S

--_. RIM J.ki 0.3C$0 17 3 0.1 Sl.k

RIll S.01 G, 3_*)0 17.6 0.5 31.8

RJI_ i.lt O. 3C10 17.3 0.1 31.6

lltl 4.30 0.)7'_C 17.5 0.l 31,l

Hi!4 4.i:, 0.17_0 17.7 1.0 31.9

l If.',_FO 0.13 0.$200 lll.i 0.I 70.I

Rift lEO 7.0k 0. II, O0 173.6 0.7 13.b

RII%IEI 1.22 0.2rkO _J.S 0.$ 2l.k

RI :'_H.8 7.;I 0.)110 1.7 0._ 25.0

RII,;IE_ ?.SO 0._,100 0.b _)._ 1£0.?

P, I/%'E I: 7.7_ O.kl_OO 1.7 0.2 23.0

llli:t+fO 7.SI 0.71130 S.S 0.5 II.4

RI;_l:f_ 1.1£ i.$lnO 173.6 0.7 13.4

RI::;tO i..5 101._, 0.2 ?G.l

UfB-RO 1.7200 4.*,'1.0 1.0 iO0.O

UEB'RI _,9000 II00.0 0.1 100.0

NUI 1.0'200 /20.0 0.4 liS.O

X _ TtlICIll[_S JI$-BURST /IS-ULT ilS-ffOLL

FLAtJG[ O. 7130 17.$ 0.1 3].1

Rill O.kll 1.2200 17.4 0.7 31.8

RIM 0.31 1.11_0 17.7 1.0 32.2

• RIM 1.19 1.07C0 17.i 0.7 _3.$

RIM 1.50 0.9800 16.7 0.1 30.7

RIM 1.94 0. |_g0 17.6 0.9 32.0

Rill 2.31 0.7:_ 17.5 O.I J1.5

Rill 2.Et 0.6500 17.7 o.q 32.1

Rill ).06 0.25:,0 lr,.9 0.3 31.0

-- RiM 3. _,;_ O. )GCO 16.g 0.) 3].0

RIM 3._1 O. 3_50 17.2 0.0 31.S

RIH 1.I$ 0.3690 17.3 •0.6 31.[,

Rift 4.$_ 0.1774 17.5 0.| 31.9

RIll _,.gk 0. $70.0 17.6 0.9 .12.1

RIJ_Ur R 0.8') 0.110_ 177.2 0.7 75.3

Rlt;:;F.O ?.Oi 0.17_0 16(.k 0.1 £2.]_

R I I_;./.r _, 7.7l 0.20_.0 10.7 0.0 2q.b

ffl e'+;t{ R 7.kl 0.1170 3.S O.i 13._

rlle+l;Pi ;.{_ 0. _,0_3 0.G $0.0 110.O

R IJ_tl[R 7.71 O.&?/O ).S O.k 2_..1

Rll;il[O 7,*j7 0.5_00 |0._7 0.$ 19.b

Rlfil_F C I.|C 0.57&4 I_6.4 0.3 IZ.2

_lJ_t;tu I.)S O.&lqO ]7_._ 0.7 75.STI[_OLT +I-I_ 00- 0.?50 21£.S 130.') ._'I.-

11111",'10 I. 7_00 02,_. 0 1.0 10.2.O

U[O-RI 1.q900 I020.0 0./ 100.O

IIUI 1.0000 o &20.O O.i Ik+,.O

t_l[[L OIII_IISIDII_, -

O-_0.0UO li- 1,7_/_ t/e 1,200 IJP.o l.SOf_

111-|0.100 fl;-I{.910 II'- I.+:0 TI- I.I$0

L I- 1.6U0 l_i* k.700 RO* 8.k00

o TIR[ 00-47.9S0 OOLT ._TR,. 11010SI. * -

lilll(L OIt_;lIllOll._ -

O-2O.C_0 It. 1.750 II- 1.2110 Ill.- !.$OO"

n!-1113o ,2,,IL._$_ ii I_ 1._.',o TI,, 1.11o

LIo !.io0 I1,,, 1.100 Io,,, I.I00

TIU OD-lll.l_| OOtT STR- 11Oi;$1 .....

16

I

I

!

!

I

I

i

+!

I

|

I

I

II

!iI

ER=4239

FSC 97153

COMPUTER SOLUTION

BASIC IIPfEFI. IIESI_P! PRO_RIt*A

tJHEEL o tZASA SP#.CE StlUTTLE ......IqAT'L -170!.SI tfZrlO STEELDKSIP, r; LnM:S -

I PUP.ST 510.PSI2 ULT CO_:PLI_0t_0.Rr, DIAI 59P_U0._,IDE3 ROLL GOO0O.P,P_IAL rnrt lfl0.PIILES

SURF COttP+_ES$1,/E ._T|tE_S 300"OIl. PSI

:ECTIOI! X THLCIttlESS ttS_-P,uRST ft.'_-ULT ftS-RflLL"

. +FLAIIGE 0.3130Rill 0.1111 0.31;_0RIM g.31 0.2700

-- RIFt . 1.19 0.23.70RIH 1.56 0.1r+50Rill 1.9l+ 0.0723Rill 2.31 0.O=SI_Rill 2.G9 0.0_:00Rill 3.O6 0.0_3

__.Rill ._.k,k 0.0k00RII4 $.gl 0.0379RIP! 4.19 0.0300RIB koS6 0.0360

. Rill 4.9k 0.0tO0RII_IEB 7.10 0.0000

_ RlllttEB 7.29 0.1]�c.I)RII_£B 7.47 0.0910RII_CB 7.6[; 0.0813RII_ICB 7.85 0,007:RtI#JJPG . _.Ob 0.;3£0RII_EB 2.22 0.171+0

._ RIIt$IEC 3,111 0.1970RlllU(B ,). '_0 0. 2115+_TIFCOLT 11-18 I_0= 0._75I/(R-RO O. 64 tiffIJEB-RI 1.2200IlUB 0.1450

'UttEEL I_lr+rflSlOPZ$ 2+D..20. Ou_ I+= 1.7_0

D1o19.430 02*10. 950LI" 9.000 RIo k.700

............ 1_

17.7 0.9 111.017.3 0.6 13.61+,; 0.2 12.'3:..7. _ 0.9 13.9

17.$ 0.6 13,11 .17.1+ 0.7 13.617.6 0.9 14.01G.G 0.O 12.617,5 0._ 13,817.6 0.9 13.9 ....17.S 0.9 14.O17.3 0.8 13.517.0 0.3 13.116.7 0.1 12.7

177.0 0.5 29.117=.1t 0,7 29,3170.3 0.9 29.1150.1 0.8 27.R

5.9t++g:t t,,170.3 0.9 29.117P..k 0.7 29.3177.6 0.5 29.1151.0 07.0 951_. _

_IGO.O 0.8 100.03100.0 0.II 100.O3160.0 0.6 2782.6

t_- 1.200 tiC. 0.G00,',, 1.9_0 T2= I.O00RO- 8,£+00

.......... BASIC UHEEL DESIG'I PROP_RAPI

.... TIRE OD,k7.�SO

tIHEEL - tlASA SPACE S_UTTLEItATeL " 5-35_ _ERYLLIUt_DE$1G_J L_ADS -

1 BURST 51n. PSl2 ULT COflB I_000._A_IAL3 ROLL G0_00._I_L FO_

SURF COPtPRFSSIVE _T;:ESS

SECTIOII-X

+5900G.SIPE

100.PtlLEBO.PSI

TtlICKIIES_ HS-BURST PtS-ULT tt$-_OLL

FL_,flGE 0,5_80 17.3 0.6 31.3RII] O. 11_, 0.7250 17.7 1.0 32.3RIIt 0.£1 0,($670 17.7 1.0 82.2Rift . _. 1.19 0._000 17.G 0.8 81.3RIft 1._6 0.521_0 17.6 C.8 81._1Rift 1.9k 0._+330 17.5 0,8 81._flit] 2,31 0.23q0 1"7.7 0.9 82.1Rift ?.u9 O. 17U0 17.3 0.6 ;_1.2Rift 3.00 0.1950 17.O 0.5 g0.S

.AIJI _ 3._k 0.2070 17.3 0.5 31.3Rift 3.81 0.2000 17.1 0.11 _0.3RI_I 4.19 O.1690 17.5 0._1 _1.+ +RIIt 4.56 0.10_;0 17.6 0.3 81._JRltt 4.9_+ o. 11._0 IC.B 0.2 _0.1RIf,qlEB 6.8.5 O. 19P,o 99.6 o.: 10_.2RIfflER 7.011 0.1uo0 87.8 O.11 lO0._It I Itt,,+rB 7.22 0.1570 G9.0 0.1 _G.1RII_IEC 7.1tl 0.1307 11:). 9 0.9 91._RIIIUEB 7.60 0.2._+0 73.2 0.3 1C _,,.9RII:IIEB 7.79 0.2030 k3,9 0.9 91.._R ItP, IEB 7.97 0.3220 K�.0 O.1 96.1Itlft;;EO 8.10 0.3tt70 87.8 0.11 1Cfl.3P,I t,_tE9 3.35 0.3710 99,6 0.8 lC:+. 2TIE�OLT 1t,,13 DO= 0.750 25_t.8 ln.2.11 01.P.IIEE-ItO 1.1110P 1000.0 0.0 197.7IIEB-RI 1.990U 1000.0 0.7 233.3IIUB 0.4600 1000.0 0.9 3_7.S

I/HEEL OI/IEtJSIOP_S -0=20.000 I1= 1.7S0 tl- 1.200 t+C,, 0.60t_

D1"19.1130 D2=16.9._0 I1'* 1._50 T2= 1.250LI" 9.600 RI- _.700 RO- _._00

BOLT STR- I_OKSI __ _IRE 00-k7.950 ROLT STR- I:0":SI

I17

ER _4Z39

FSC 97153

RESULTS

Mate rial

7049 Aluminum

of Basic Wheel

(Ib7')'

128.3

Weight

(Ib/in3)

• I01

_ , ,, m

• 101 116.52014 Aluminum

7075 Aluminum .101 116.5

L,

7175 Aluminum . I01 114.8

ZK60A Magnesium .065 IZ6.8

EK31A Magnesium .065 I26.8

Ti-6AI-4V . 163 134.8

Ti-7AI-4Mo . 163 134.3

4340 Steel .283 214.8

4140 Steel .283 214.8

S-350 Beryllium .066 79.9

Beryllium is recommended as the best choice of the eleven

materials studied•

18

I

l

l

t

I

l

I

!

!

I

I

I

I

I

I

I

ER -4239.

FSC 97153

WHEEL CALCULATIONS - PART If:

LOADS r__ ADK-OFF

From Part I, a beryllium wheel was chosen as being the best

possible choice of the eleven materials studied.

To make the best advantage of a beryllium billet, thereby nnini-

miz_ng cost, it was decided that a configuration change was needed.

The sketch below shows the wheel analyzed in this part:

Loads

The following loading conditions were analyzed in this part:

I

I

!

!

I

I

Tire Deflection @ -55°F

Burst Pressure

Load Type

Combined Load (kips)

Radial/Sidem m ,|,

Roll Load (kips) RadialMile_

Load 1n

Load 3Load Z Load 4 Load 5.... .

37% 32% 40_o 40,% 40%

510 psi

Limit

lZ36 psi

Ult

472 psi

Limit

/ 1oo

924 psi

Ult

1(30

924 psi

Limit

19

ER -4Z;9

FSC 97153

The philosophy of designing with these loads was to compare the weight

difference between the ultimate failure loading and the design limit load-

ing. The ultimate loads are designed to the ultimate material strength

and the limit loads are designed to the material yield strength.

Analysis

As in P_,rt I, the analysis includes ring, cylinder, and plate methods; the

results are shown in the Table VI below:

TAB LE VI

THICKNESSES AT VARIOUS LOCATIONS

,i

A

B

C

D

E

F

"G

H

I

.r

B

Load 1 Load 2 Load 3 Load 4 Load 5

.450 .631 .430 .546 .546

.360 .505 .344 •437 •437

• 500 •897 .444 .699 .699

• 395 .637 .351 .482 •482

• 850 1.100 .850 1•100 1.100,L, , • ,q

• 680 1.030 .640 .860 .640

.894 .934 •894 .934 .894

1.556 1•631 1.556 1.631 1.556

.738 .965 .738 .965 .738

.512 .690 ,512 .690 .512.o

2O

I

I

I

I

t

i

I

I

I

II

II

III

I

I

I

[

f

[

[

!,p+

[

!

III

ER-4_39

FSC 97153

RESULTS:

Load Wheel Weight*

1 70. 4 lbs.

2 91.6 lbs.

3 68.6 !bs.

4 84. 7 lbs.

5 77.2 lbs.

*The weights given do not include the bearings.

Loading philosophy number 3 gives the lightest wheel desigp.

This philosophy suggests designing a wheel to a 40% deflection,

aburst pressure 1-112 times the tire bottoming pressure, and

the limit combined load.

21

ER-4Z39

FSC 97153

REFERENCES

I °

_0

,

Roark, R. J. "Formulas for Stress and Strain", McGraw-Hill

Book Company, 1965 Table Mill - 30.

Roark, R. J. "Formulas for Stress and Strain", McGraw-Hill

Book Company, 1965, Page Z42.

The Timken Roller Bearing Company, "The T[mken Fngineering

Journal", 1967.

ZZ

1

1

I

1

I

1

I

1

I