RB No. 4B18

if q

NATIONAL ADVISORY COMMITTEE FOR AERO NA UTICS

c! 7

"Tl'll.rl'I)II~ 11.1~1)()ll.rl' ORIGINALLY ISSUED

Restricted Bulletin 4B18

COMPARATIVE TESTS OF THE STRENGTH AND TIGH'INESS OF

Cot-1MERCIAL FWSH RIVETS OF OOE TYPE PND NACA

FLUSH RIVErS IN MACHmE-CotJNTERSUNK

AND COUN TERPUN CHED JO:rnTS

By Merven W. Mandel

Langley Memorial Aeronautical Laboratory Langley Field) Va.

WASHINGTON

NACA WARTIME REPORTS are reprints of papers originally issued to provide rapid distribution of advance research results to an authorized gr oup r equir ing them for the war effort. They were pre­Viously held under a security status but are now unclassified. Some of these reports were not tech­nically edited. All have been reproduced without change in order to expedite general distribution.

L - 297

https://ntrs.nasa.gov/search.jsp?R=19930093010 2020-01-24T07:21:06+00:00Z

NATImTAL ADVISORY Cm1~:! IT'rES FOR AERONA UTICS

R"SS'T'R ICTED BUI,LETIN

C0 :PARATIVE TESTS OF THB STRTo.;N(!TH ArID TI0-HT1IJESS OF

COMMERCIAL FLUSH RIVETS OF ONE TYPE AND NACA

FL USE RIVETS I N 1\llACBlNE - C OUNIJ'E:1SUNK

AND COUNTERPUNCHED ·TOIHTS

By Me rve n ~ . Mandel

An investigation was conducted to coupare the strength CiTld tifhtness of 1l:ach5.ne - co1Ll'ltersunk flush­r i veted jo~nts assembled ~ith ~ffiCA flvsh rivets and one type cf cor,rrnercial flus h rivet and also to co.npare the strength 8.j:ld tightness of cou:'1.terpu ... '1ched flush-riveted j oints asser:lbled wi~h the eame types of rivet . The re s~lts of the inv7.3tig~tion ~re p~e e~ted in the form of loed-displacemeYlt cL~ves, vvhl ch indicate t hat the NACA flush-ri·,reted join-t,s tended to be somewhat stronger and tigh.::er tllBn t ~le corresponding com.'11ercial flush­riveted j~int8o The t~st results also show tha t both the com:nerc- ial and the r"\CL counterplllc:"1ed flush - ri ve t specimen q ha~ consiaerabl~ gr eater strength than the machine ·,col1ll"C~rsunl: ~pecJ.mens of c orresponding sheet t hickr..ess.

LJTRODUCTION

'The studies o£' ti[!htness and flushness of machine ­countersm:k flu.sh 1"1 vots fo.7 aircraft de scribed in referADces 1 to 3 indicate~ t~at t~e ~ffiCA flu3h - riveting pr occ":''..c";"·u I'TiYl11Csd "re L.ghtes"C joints ~Ol. the t pes of fl'Jst ct·v f.'l·.:t:c·q =n·r·~fj.1..tg--t'C('c. . As a result 0: this con­clu,j:>~n., r.:1J. :!.:-:.v2s'Cig&tion W?S cOldueted -:0 de~ermine the com:r:.;-n·at.:L ~;p ~; j c1J.tnlJ.ss CJ:~ r:Flci::.in8· C0..;;.ritersunk flus h­ri i;'l3v~ci ~'-;.-:..r:+~2 ElE.s3.:1~)1'3d 1u5.t,h ~qC.~ Tl.ush rivec:s a.nd co~r;lr!;·:,::_ .. ~~.~ flT'u ~'~."0C5 0:' cn? type /?,nd alsL- t te COIn­

par' ut:1':.7S t~.g:!t....1883 u.:' ':o.lr,·:3r.L:)ll~ch(:'d flush - rivete d joint s 8ssp.;!!b:~.r~tl VT~1J. the ,.:Pme :;:'-1'(;"3 ;-.1' 2lvet . Thirt~T- six

s pec i.:r.6r.:3 ~ prep8re5 by 11..1 a.il:'p:alW ::rlf'I:ufac turer, were te sted in the pre30nt. illveotigdtion.

- - .---.----

2

SPECIMENS

Machine counters .Ulk . - Ea ch machine - countersunk specinlcn-Cor.sistud of t j O sheet s of 2hs - T allliTIinl~ alloy assemb l ed in t he for:n of a lap joint with two aluminum-

. t ., r . 1 f " . . 1 mh 1. h d- ' t a.LJ..0Y rlve 's, as SdOwn 1 _J.[Ul"8 __ . J _ e trJ.nc - lame er

rivets v:ere of r\ 17S- T and the ~- inch- diameter rive ts , of

17s- rr al u.rninum alloy .

The parti c ular corr~erc ia l lush- riveting procedure used corr e s ponds to rive ting me t hod C of r efe r e nc e s 1 and 2 . The 1" ive t hole in t lle uppe:t' sheet was machine -count ersun for a 78 0 CO U-1'1tersLUlK - head rivet; as shown i n figurE; 2 (a) . Dr_II coizes and ang le of cOlmtersln..k wer e not spe c i fied by tb e manufactUl"'er . A 78 0 COU-1'1tersunk -head rivet v:as inserted in t1-;e rivet lio l e , and the manu ­fa c tured ~leo.d wa dr:l. ven with a vibr a t:'ng gU-1'1 while the shank end WetS bucked with a bar .

n:'he NACA flush-.riveting~::cocedure is riveting metho.d E of ref'o renc32 1 ane}. 2 . 'r'lle NAeA procedure involved the s ame preparat:'on of t~e rivet ho l e as the

COmlTIerC ia l nrocedure , but 8. l _ ineh-- diameter round - head - 8

(ANi-l-30 ) rivet or 8 1:.- 1nch- dia:neter braz ier - head (A N456 ) L~

rivet W2. S i nserted f!'Oir1 the back of the joint , and the manu.factw"'ed he ao. was driven with a vibrating gun while the shank end was b ucked intoche counters unk hole with a bar , as shown i n figure 2(b) . The p or tion of the formed head t ha t protruded above the skin surface was r emoved with a f l us h - rive t milling t ool similar to the one des cr ibed in r e f erence 4.

Counterpunchod. - The counterpu-1'1ched s pe cimens were of t he same fO::'"'inns' t he machine -c ountersunlc specimens , except for preparation of the rivet ho les. The lower sheets we r e rr..achine - conntel'sunk and the upper sheets were c ou-'1 t erpunched. into the lowe r sheets as shovm in fi gure 3.

Remarks. _. In only one of tlle nine l - inch- d i ame t e r ~-

ri vet s pecinens assembled ~'li th ei ther ~ila c hine - c ountersunk or counter J.:m{'lcheci NACA flush rive ts were the ri vets

3

driven sufficiently to fil l the ~olmterSUn{ hole com­plete l y . Figure 4 shows the gaps around the incompletely driven heads of four r epresentative specimens .

n ::'-":ST PH OCEDUHE

Load was applied to t he speclmens through 'remplin grips wi th a hydraulic testing mscrtin e accur&te to within one - ha lf of 1 percent . Dls~lacements of one sheet r e lative to the other were mea sured on the edges of the sheets at the rivet line by means of two I S - power optical mic r ometers . 30th the displa.cener..t und.er load and t he p e r manent displaceme n t remaining after removal of load were measured for E'uccerslvely increasing loads unti l fai lure occl.1I'red .

RI:SULT3 AND CCl'JCLU2I OJS

Typ ical spec in:e::.ls E fte fai lure are shown in figur e 5. All c..,e :.!uchine - countersu11.k specimens failed by s hear of t~e 1'1 ve ts, The co·,mterpu.r:.ched specimens fail ed. either by tension failure of t~le upper sheet across the rivet line or by ehear of thE rivets , except

for the t - inch-d iameter t;ACA counter punched ri.vet speci ­

mens, which failed by tension of the rivets . The manu­factured brazier heads cf the rive ts shearad parallel to the axes of the rivets, probably because of the tensile load imposed on the 1'1 ve ts by tl~e tende! cy of the dimple in t he upper shee t to ride l.lp t he side of the countersunk hole in the lower s heet .

'rhe results of the tests ar e presented in fi gures 6 to 9 as curves of load plotted against displacement under load and curves of load p lotted against permanent displacement after r emoval of load . 'rhe specimen num­bers on the curves are g iven in order to permit proper association of t hG curves of dis p lacement under load with the curves of permanent displacement . From these curves the following conclusions ma~ be drawn concerning the NACA flush rivets and the corr~ercial flush rivets of the type tested in the inves tigat ion :

4

1. When a machine -c ou..ntersunk o.oL~o - or 0 . 05 1- inch­thick sheet of 21.LS - T aluminUr.1 alloy W.J.S riveted to a

o . 12 5- inch':thick ~heet witil t - inch- diameter rivets of

A17S - T aluminum alloy , thei:e appeared to be , v:i thin the normal ecatter of data, no difference between t1-:le maximum strengt h or the tightness of the 5 0in~s as shown by the load.:.displacement curve for JlTACA and cor.1tl1e rcia.l flush rivets . (See fi g . 6. ) ~ ~nlen n. l~achine - couj.'1.tersunk 0 . 081 - or 0 . 091 - inch-thick s ~ee t "a s riveted to a 0 . 125 - inch- thick sheet with ! - inch- diamcter rivets of

- L. 17s- rr alUJ.'Tl inum alloy , the strength and t ig:'1tne S.'3 of the joints wet'e somevIhat bette~ for the l'nCA rivets than~or the cO!'1mercial rivets . ( See fig. 7.)

2 . V,ne n a cC'unterpunched O.O~_0-inch-tI1j ck sheet was

riveted to a 0 . 125 - inch- t Di ck s1J.eet with ~- inch-diameter. 1"1 vets of Alr(S - T alumLl'm alloy, t.he average maximum strength was higher and the average tight ness was better for the NACA rive ts than for the commercial rivets , although one l:IiACA flu.::;h - ri -..re t specimen failed at a con­siderably lov:er load t han the other I·J!~CA rivet specimens in the e:roup . (See fig . 8 . ) 'Jhen a c ounterpuncheci . O.OBl- inch- thick sheet was riveted to a O. 125 - inch-thick

sheet with ! - inch- diame ter rivets of 17S- T aluminum ~-

alloy , the _TA CA rivets were sliGht ly stronger and tighter at higher loads than the commercial rivets. (Soe fig . 9 .)

3 . Both the comt'1erc ial and the [·rACA counterpunched flush - rivet specimens had higher ultimate loads than the machine -counters lnk specimens of corresponding sheet thickness - approximately 60 percent higher for the

l - inch - diameter rivet specimens and 38 percent higher 8 -for t h e ~- inch-diameter rivet spe cimens.

Langley Memorial Aeronautical Laboratory, lIrati onal Advisory Committee for Aeronautics ,

Langley Field, ·Va.

~ C\J

. I

H

1 .

RE F'ERE?-TCES

Lundquist, :ii:ngene E. , and. Gott lieb, of the Tig~tness and. Flushness of Count ersunk Rivets f or Aircraft . JFne 1942 .

5

Robert: A Stlldy Ma chine -

NACA R . B. ,

2 . Gott l ieb , Robe rt: Effec t of Coun tersul1.k Depth on t he 'Tightn() s~ o.r Two 'I'Y,(;Jes of' ~!}achi:ne -Countersunk

Rive t . !ACA H. B ., Oct . 19li.2 .

3 . Gottlieb, Rober t: Test Data on t he Shear strength of Fachine - Col.J.:.'1tE..rsunk Rive t ed .Jo i nts As s emb l ed b'Y' an NA CA }Ilush- Ri ve t lng Pr oceuure . NACA R.D. , Dec . J.942 .

Gott lieb, Robert, ~nd Handel s l"el"Vell Irrproved -r;J lush- Bi vet ;~ illi.ng Tool . No . ?FIB J' ~"ny 1943 .

"~J .: An NACA R. B.

"'-----

3~

5! ~

tL= 0.r25 inch

\f3 11

rr---------\ I ~ ... I 2...l . 2.

5l

tt.:: 0.125 inch

(0) ! -inch-diometu rivtt SP~CHnWl. (D) ~ - inch - diameter rivet spQcjmtn.

Figure I. - Test specimens.

L-~')7

:z: »­n »-

':I:J f-- .

OQ

.......

I

I)' » »>'

Hole preparation and rivets

Rivets before driving

~ ~Rivets

~ rProtruding portion i removed with flush­

r- -- -, rive.t milling tool

after driving ~

(a) Commercial flush rivet. (b) NACA flush rivet.

Figure 2..-Machine. -countersunk flush-riveted joints.

------

L-2'17

z > o >

'XJ f-­

CJQ

t\)

Hole preparation and rivets

Rivets before driving

(0) Commercial flush rivet .

Protruding portion removed with f lush­

;;---;, , rJy~t mifl ing t ool

(b) NACA flush r ivet.

Figure 3:--Counterpunched flush-riveted joints.

1>- 297

z :> (")

:>

"x;J ~.

ClQ

~

fi gure 4.- Specimens assembled by the NACA flush-riveting procedures with

-L -inch-diameter rivets showing insufficient driving. 4

L~ ~: 1 7

2:: > ("')

>

":j 1-'­

OQ

~

NACA Fig. 5

(a) Specimen wi~h NACA machine-countersunk rivets showing shear failure of rivets.

(b) Specimen with commercial counterpunc hed rivets showing tension failure of sheet.

(c) Specimen with commercial counterpunched rivets showing s 'hear failure of rivets.

(dJ Specimen with NAC~ counterpunched rivets showing tension failure of rivets (ghear failure of manufactured brazier heads).

Figure 5.- Typical--1..-inc-h-diameter rivet specimens after failure. 4

NACA

500

400 ~

.-'l :---

, Y:: ,...' ;;...--

v~ ~ 300 // "/

/ V/ 0/ 11~ /

200 I~t V/

If/ V'

100 I / /

..

Specim en ber num

__ 18.1 7J'--

Ff

8 16 9

/ /,4-'1/1/ "If /

I , I

-- f-_18 8

~ ~ 7 ~---16

// I--::" ~ ..... i'f9 ~

1-'"- V ~

V7 ~ '

~ tu = 0.051 in.

"0 soc

8 ~

15 5 -

~ -- 14-

~ - 13 4 6 -~: ,...'

'I

400 15 5 p--

--~ ,~

.~ ~ I - ~- 13 4"""

300 ./--J~

~~

q

Corrmen:ial z.oo !11/

~/ I ---NACA

100

o o .010 .02.0 .0.30 a .010 .020 030 Displacement under load, in. Permanent displacement, in.

figure 6. - Load-displacement curves for commercial and NACA machine -countersunk flush-rivClt specimens. d=k in.

NACA

2 000

60C ~

~ ~ -LG ~ V eoo ~ --IP ~ /"

00 h V ll!

8

tu =0.0 81 in.

Speci men her 3.

35

num ,-

~ )j ;;;.-: F 26

2S

~

I~ ~.

£ V ~

~ ;;-'"

~ ;..- t:::: ::::::-v-::: ......-

Fi g . 7

-~ .-:J f ~

34 .35 ~:

2.7 26

25

.0 4 00 tv I ..

a

t u=O.o91 in.

2000

--~b lA 1600 --22

~..- T 23 .....:: ~- .....-~, 1---"2.4

~ "/ V lV 3.e

1200 /' ~ ~ f ~ r:: ------I'

if ~ ,/

800 II/~ Vj J fh

400 I!f/ V

o o .010 .020 ,030 Displacement under load, in.

.-# ~/

l~ Iij

V r y/ ~ ~ 11/ " '/

~(~

,.,-...-

,... ---v --::;;: ~

~ If-"-

{. ~,...

I-----

I -- '31 33 22--~

~; 23 2. .. 32.

Commercial ---NACA

I

o .010 .020 .030 Permanent displacement, in.

Figure 7. - Load-displacement curve s for commercial and NACA machine -countersunk flush - rivet specimens, d=.t in.

NACA Fig. 8

tu = 0.040 in.

800 Specifn en

er o

numbl 700

, t::--::.. - --I

"...-.,/ /,./'

2. -19

-::::::- -- - I- -12

/' 1--;:'

1/ / 600 1 2

1/ ./ /'

v 1 ~ ~ / -- -- II

I I /

2. ; I V

......-1

/ ~

I / 1-- -- I--II

" / / g ~--50C ,

;/ ;) V '/

1/ I <1 7'" J.L

IV 1/ If

400 f' /1 lj 1 I/j

'J /1 II ~// ~

I, I. /jj 300 II/

I~ V Ii/; \'1/

200 i1l ~

~

r; U Commercial

----NACA IOu

o o .010 .020 .030 o .010 .020 .030 Displacement under load, in. Permanent displacement, in.

3

Figure 5 ....... Load -displacement curves for commercial and NACA counterpunched flush - rivet specimens, d=.g in.

I .... NACA Fi IS. 9

tu = 0.081 in.

)0

~O 2

000 2 ...0 ..

'"0 o o

600

200

--'8 00

40 )0

(/

I~/j; ,fA V

1ft!! I r;

Ii; ~

-,// ~-- -::

~ ~ / V--

~ /-

S~ci~e n r numbe

~ [f

I

302 ~-2

8 I - ....

j' o 9

o o .010 .020 .030 Displacement under load, in.

, 30 28 --- f-- -- ---- 21 /' ..- -- -~ ,g20 ,/

1/' ~ /-!-

~19 -'/// -I----

/ ,:'1' ~ /-"

~ V

'0/

I Canmc2rcial

--r- -NACA

o .010 .020 .030 P<2rman<2nt displacement, in.

Figure 9. - Load - displacement curves for commercial and NACA counter punched flush - rivet specimens , d = ~ in.

J