Anda di halaman 1dari 22

Large Bolted Joints

TESTS OF A490 BOLTS


(Preliminary Report)

by
Gordon Sterling
John W. Fisher

March 1964
SYNO-PSIS

Presented in this report are the results of calibration tests

t, of individual ASTM A490 alloy steel bolts. The b'olts tested were 7/8 in.

nominal diameter. ' Results of'twenty direct tension and thirty torqued

tension tests are reported.

This is a preliminary report and will eventually be re-written

to include information on similar tests conducted on the same bolt lots

'at, the University of Illinois.

INTRODUCTION

At the February 14, 1963 meeting of Committee 15 of the Research

Council on Riveted and Bolted Structural Joints, it was recommended that

Lehigh University and the University of ~llinois conduct tests on bolts

from the same 'lot to determine if -testing procedures constitute a vari-

able. Each university was to test bolts in direct tension and torqued
tension using its own standard calibration procedures.

This report discusses the results of the tests conducted at

Lehigh. The bolts were purposely ordered near the minimum specified

tensile strength.

-1-
TEST PREPARATION

Two lots of 7/8 in. diameter, heavy hexagon head, A490 bolts

were tested.' B'efore testing', all bolts were st'arriped with a lot designation

and bolt nu~ber. In additio~, each ,bolt had ho~es ~ril1ed in the centers

,of the' head and nut 'ends ,to accommodate the C-frame extensometer.

~~lts designate~ as AB ~ot· were'9~ in. long wit~ 1~ in. of cut

thread. 'Those labelled' Lot' :U were 5~ in. long with 1% in. of rolled '

thread. In al~ tests, one hardened washer was used under .n A194 Grade

2Hnut· (heavy hexagon). Each lot w~s tested with 1/8 in. and 9/16 in.

of ~hread in ·the grip •. This gave grip lengths of 8\ and 8-11/16 in. for

Lot AB.. bolts, .and 4-1/8 in. and 4-9/16 in. grip for the, LI bolts.

A representative sample of twenty bolts and nuts from the AB

lot were'checked for tolerance by using the NC2A go and no-go ring gages

and the NC2B ~o and no-go plug gages. All specimens tested were accept-

able. The AB lot bolts were received with a light coating of shipping oil.

The L1 lot bolts were received with a greasy film, but were not

, as'well lubricated as the AB bolts. This lot of bolts was not checked

with the NC2A gages. 'The, same keg of nuts and washers was used in the

tests of both the AB and LI lots.

-2-
TEST PROCEDURE

a. ~ORQUED TENSION TESTS

Specimens from both lots were tested in the Model M Skidmore-

Wilhelm calibrating device. Ten bolts from the L1 lot were also tested

in a solid block of' A440 steel.

For all tests conducted in the Skidmore-Wilhelm calibrator

the nut was turned by using a hand wrench until a lI'snug" tension of

ten kips was reached in the bolt. Bolt load, elongation and turn-of-
the-nu,t (in degrees) measurements were taken at 5 and 10 kips. The nut

was then rotated in 45° increments (1/8 tur'n-of-the-nut) with a pneu-

matic impact wrench until failure oc'curred either by fracture or by

thread stripping. Bolt elongation and load data were taken at each

45° increment of nut rotation.

From the tests done in the Skidmore-Wilhelm device a mean

elongation at a "snug" lo'ad of ten kips was determined for the L1 lot

specimens. This mean elongation was applied to thos~ specimens being

tested to failure in the solid A440 steel block by using a hand wrench

to turn the nut. The bolt was then tested b~. turning the nut', in 45°

in~rements, using a pneumatic impact wrench. Bolt elongation measure-

ments were taken at -each posi tioIJ, of, nut rotation. This procedu're

al1owe~ a direct comparison of the nut rotation VB. elongation charac-

teristics of bolts tested by torquing in a block of A440 steel and in

the less rigid Skidmore-Wilhelm device.

-3-
-4..

, ,

b. DIRECT TENSION TESTS


As. a preliminary check the bolt being tested was loaded to the

specified proof load and then unloaded to check the ASTM requirement of

minimum 'permissible set (0.0005 in.). No bolts were rejected by this

test.• ,The 'bolt wa's ·then' reloaded and elongation readings were, taken

. at ten kip interva1s until the inelastic range was reached. At this

point load readings were taken for every 0.01 in. of elongation in the

. bolt •. The direct tension tests were conducted in a 300 kip Baldwin

hydraulic testing machine at a, strain rate of approximately 0.01 in. per

'minute.

A detailed description of the procedures used, in the calibrating

of bolts is given in Reference 1. The reader is referred specifically to

Figures 1 and 2 of this ,report for the specific set ups used in the

direct and torqued tension tests$

liEST RESULTS

Table 1 and Figures 1 throughU summarize the test results. The

actual test points are given on most figures, and, therefore, Table 1

merely indicates the significant mean values determined from all the tests.

The 'specified proof load'for these 7/8 in. diameter, A490 bolts is ~5.45

1) John L. Rumpf ,and John W. Fisher


"Calibration of A325 Bolts", Journal of the Structural Division, ASC~,
Va •• 89, No. ST6, Prec. Paper 3731, December, 1963, pp. 215-234.
-5-

kips, (i,e. the load to which the bolt may be loaded so that after un-

loadi.ng there is no more than 0.0005 i~. permanent set), and the speci-

fied minimum ultimate load is 69.3 kips.

Four failure 'modes were noticed. All the ,AB lot bolts failed

in the direct tension tests on a jagged diagonai through the threads.

Two of the AB lot bolts tested in torqued tension with 1/8 in. thread in

grip failed by thread stripping as did two L1 bolts in direct tension

'with the same amount 0,£ thread under the nut. In all four cases, the

nut and bolt thread was so badly damaged that it was impossible to tell

which thread had originally failed. The LI specimens tested in direct

tension with 9/16 in. thread in grip failed on a jagged diagonal as

noticed with the AB direct tension tests. However, three bolts with

1/8 in. thread in the grip, failed on a level plane, through the threads,

at the juncture of the thread ·runout and bolt shank. The other two LI

bolts failed in the diagonal manner previously noted. All bolts tested

in torqued tension (with the exception of the two AB lot bolts that

stripped) failed by "twisting off" on' a level plane through the first

thread under the nut at the time of failure.

Figures 1 through 4 relate the load-elongation characteristics

of the bolts tested in direct and torqued tension. In all cases bolts

tested with 1/8" thread in grip gave.greater ultimate load than bolts

from the same lot tested in the same manner with 9/16" thread' in grip;

however, the specimens with 9/16" thread in grip sustained greater deforma-

tions before failure.


-6-

Figure 5 'shqws a typical relationship between the direct and

torqued tension tests.' Bo'1ts tested in direct tension gave greater ulti~

mate loads and sustained greater deformations to failure than did those

tested in torqued tension.

Bolts tested' in the' solid A440 steel blo.ck gave greater loads

for, fewe"r turns" as, is indicated in Fig. 6. Although there is, a broad

scatter of the data associated with each test there is a definite separa-

tion of' 'the two mean curves.

The bolt load is related to the number of turns from a "snug"

bolt load of 10 kips in Figs. 7, 8, 9 and 10. The data s,hown in Figse

7 and 8 refers to tests conducted in the Skidmore-Wilhelm calibrator.

Figure 8 presents the results of tests done in a solid A440 steel block.

The mean ~urves from ;Figs. 8 and 9 (with 1/8" thread in grip) are plotted

in Fig. 10 to compare with results of tests on A325 bolts(l). The "snug"

load for the' A325 bolts was 8 kips.

The load-elongation relationship of A490 and A325 bolts torqued

in the Skidmore-Wilhelm calibrator are co~pared in Fig. 11.. For "the

7/8" x 9-!" bolts there is practically no difference at ~ turn from snug o

(Again, for the A325 bolts "snug" was defined as 8 kips whereas for the

A490 "snug" was 10 kips).

The effect of exposed thread under the nut is given in Fig. 12

and compared with results obtained on A325 bolts(l). This plot shows

that the A325 bolts ,sustain slightly more rotation to failure than do the

"A490 bolts, especially at 9/16" thread in. grip.


CONCLUSIONS

The following conclusions are based on the results of 50 tests

of individual bolts. (Twenty 7/8" x 9~" bolts and thirty 7/8" x 5~ bolts).

These results and conclusions are ,not greatly different from those'

reported in a more comprehensive study of A354 alloy steel bolts(2).

Similar results and conclusions, based on a limited number of tests,

were also ~eported in Ref. 3.'

1. Bolts from the, same lot, and' with the' same amount of ,thread

in grip, gave B' greater ultimate strength in direct tension .tests than

was achieved in torqued tension tests. For the AB l?t specimens this

increase was 10 to 15 percent of the ultimate torqued, tension value, and

for the L1 lot bolts the increase was about 26 '·percent.'

2. A lesser amount of thread in grip gave an in~rease in the

ultimate bolt strength in both the torqu~d and direct tension tests.

(Sei Figures 1 to 4).

3. For specimens with 1/8 in,. thread in grip the LI lot bolts

required in average of 1~ turns from "snug" to failure; the AB lot bo1ts

needed an average of 1-3/8 turns. With 9/16 in. thread under the nut

this requirements increased to 1-5/8 and 1-3/4 turns for the LI and All

lot bolts respectively. Thus an increase from 1/8 in. to 9/16 in. thread

in grip required an additional 3/8 turn' to failure for both bolt lots.

(2) Richard J. Ch+istopher and John W. Fisher


"Calibration of A354 Bolts", Fritz' Engineering Laboratory Report No 0

288.9, March 1963.


(3) E. Chesson, Jr. and W. H. Munse .
"St\ld'ies of the B~haVior of High Strength Bolts and Bolted Joints",
Dept. ·of Civil Engineering, University of Illinois, Urbana, Illinois,
Dec. 1963 '

-7-
-8-

4. The 'ave·r~ge load at ~ turn from "snug" was less than the

specified proof load for both lots tested in the Skidmore-Wilhelm device.

Three o~ the ten 5~, in. long LI lot bolts tested did give loads greater

than. the ·specified proof load (55.45 kips) at % turn from snug. None of

the AB lot bolts reached proof load at ~ turn from snug e (See Figs~ 2
arid 4, ,and 7 and 8).

5. Figure 6 indicates that the LI specimens, with 1/8" thread in

grip, torqu'ed in solid steel did reach proof load at ~ turn from "snug".

,Also this plot shows that, to· achieve the same bolt elongation) fewer turns

of the nut are required in specimens torqued in the solid steel block

than is required for t~ose tested in the less rigid Skidmore-Wilhelm device.

If one assume~ that the load vs'. elongation characteristics of the bolts,

tested in torqued tension, are not' influenced by the material gripped,

then (from Fig. 6) one observes that the bolt load at ~ tu~n is indeed

above proof load for those tested in the solid steel block. Figure 8

shows that proof load was also reached at % turn for most specimens tor-
qued in solid steel with 9/16" t~read ~n grip.

6. The A325 and A490 bolts tested in the Skidmore-Wilhelm gave

substantially the same load-turns relationship up to the elastic limit

of the A325 bolts. (With such a broad scatter of data (see Figs. 7 and 8)

it is strictly fortuitous that the mean curves fallon the exact same

line ••• However, in the" elastic range of the bolts one would expect

the mean curves to be very close~

7. The· 7/8 x 9~" A490 and A325 bolts (with similar amounts of

exposed thread under the nut,) behaved similarly in to~qued tension tests, with

the A490 bolts merely going to a higher "plateau". However, ,for the shorter

bolts the A490 tests indicate a relatively quick load drop off as compared

to the A325 bolts. Note that at ~ turn from snug the load in the long bolts

was approximately the same regardless of typee (See Fig. 11).


- 9-

TABLE 1

A490' BOLTS - 7/8" DIAMETER

DIRECT TENSION TORQUED TENSION (in


Skidmore Wilhelm)
Bolt Lot AB AB L1 L1 AB AB L1 LI
Nominal Grip 'ins. 8% 8-11/16 4-1/8 4-9/16 8t 8-1~ 4-1/8 4-9/16
Thread in Grip ins~ . 1/8 9/16 1/8 9/16 1/8 .9/16 1/8 9/16
.Noe of Specimens Tested 5 5 5 5 555 5
Mean Ult. Load kips 73.2 70~8 76.0 72.1 65.4 61.8 61.1 58.4
Std. Dev. from Ult. Load kips 1.59 ;1.69 0.54 0.17 2.80 2.18 2.80 3.00
Mean Eloug. at Ult. ins. 0.0779 0.0846 0.051 0.065 0.053 0.070 0.026 0.031
Mean Rupture Load kips 65 61 67 59 52 50 40 34
Mean Elong. after rupture ins. 0.12 0.18 0.137 0.245 .0.08 0.114 0.075 0.11
Mean Elong. at Proof Load ins. 0.028 0.029 0.015 0.017 0.028 0.031 0.016 0.018
% Min. Spec. Ult. Load 106 102 110 104
Mean Load at ~ Turn kips 48.8 41.1 53.4 50.0
Ave. Turns to Failure 1-3/8 1-3/4 1% ' 1-5/8
(Torqued Tension ·Ult)'
(Direct Tension UIto)
0.90 0.87 0.80 0.77
No. of Stripping Failures o o 2 0 2 0 0 0

TORQUED TENSION (in


solid block)
L1 LI
Thread in Grip in o 1/8 9/16
No. of Stripping ~ailures 0 0'
Mean Elong. after Rupture in. .09'6 .10
Ave. Turns to "Snug" from "Finger Tight" .20 .28
Ave. Turns,to Failure' from "Snug" 1.30 1.44
Ave. Turns to Proof Load from "Snug" 0.29 0.41

Proof Load = 55.4S k


Specified Minimum Ult. Load = 69.3 k
80 liS Thread in Grip

--t= -tf- ~_.... .-. ~~j~~in.Sp~~d Ultimate


60
Proof Load,
- -55.45 kips --
BOLT LOAD
(kips) _
40
I grip
- -wi.

~8x9~i bolts, CutThreads


20
-9- -+- Fracture of Bolt
L
1

J t= thread in· grip

o .02 .04 .06 .08 . .10 .12 .14 .16

BOLT ELONGATION (inches)

FIG o 1 DIRECT TENSION CALIBRATION AB-LOT-BOLTS


80
lIe Thread in Grip
o

60 1.. - '
. proof JD0..9,_
- ------ --.- -- - ..... - --Q-~-
~5.45 kips- -<?- 1- .
BOLT LOAD
(kips)
9"6T hreod in Grip
40
i'2 T;I urn- 0 f Nut f rom ..SnugII
I

'la x9 lti bolts, CutThreads


-¢-, -+- Fracture of Bolt
20

·.0 .02 .04 .06 .08 .10 .12 .14


BOL~ ELONGATION (inches)

Flf}g 2 TORQUED TE~SION CALIBRATION At1-LOT SOLTS


.-.--
. I

- .

80 aThread in Grip
l/
Ulti.mate·· Load.- -.-.'. .: .
69.3kips,M~n. Specified

__ ~ -, e _.. 1
. .

• . - II'_.-tZ£: - ' ,-
d
~..

-'-. -,
.. '--. _
... -

--- . ..
.

60
PrO~~t~ .9'r6Threod in Grip
BOLT LOAD 55.45 kips _
.. (kips)

40

20
7, X 5 ~2 bolts, Rolled Threads

o . .02 .0-4 .06 .06 !IQ .12. -.14

BOLT ELONGATION -(inches)

FIG. 3 DIRECT TENSION CALIBRATION LI-LOT BOLTS


80
· "i Thread in Grip

60
• • Proof Load ,55.4~ kips
--------
BOLT LOAD

"(kips)

-40
-+t-. + ~ -rl.:'~ ..

«
9· ++
116 Thread ~n "Grip
20
Y8 x 5~2· bolts, Rolled Threads
U

o .02 .04" .06 .08 .10 .12 .14 .

BOLT ELONGATION (inches)

FIG o "4 TORQUED TENSION CALIBRATION LI-LOT BOLTS

f
"'¥f
~r
++ •• _ - ~4-

• I

80 _Direct TniQl'

60: e
____e-- " e__ - . Proof Lood
~ • . -55.45 kips \-
BOLT LOAD
(kips)
40 . Torqued in -4-
Skidmore~Wilhelm

'l8X5~2 Bolts
20--
~8Thread in Grip

o .02 .04 .06 .08 .10 '~12 . " .14··'

BOLT ELONGATION (inches)


FIG. 5 COMPM-ISON OF LOAD-ELONGATION CHARACTERISTICS OF
LI-LOT BOLTS, FOR DIFFERENT LOADING METHODS
, 1.50

-1.25 o 0

TorquecJ in'
Colibroti ng
1.00 Device-'- -
~---Torqued in _.
Solid Steel
TURNS- Block
OF~THE 0'15
NUT ·
FROM, '
, "SNUG II

0.50

'0.25 ~ux5~"
8· 2 Bolts
I/s,Thread in Grip

60
- , ----
Proof Load
--,
55.45 kips
Mean Load Curve
, , From Fig. 4'~8 Thread
40
BOLT in Gr,ip
LOAD
.' (kips)

o .02 .04 .06 ~08 .10 .12


BOLT ELONGATION (inches)
FIG. 6 ELONGATION-NUT ROTATION CHARACTERISTICS
OF LI-LOT BOLTS
" -'

.r~ J. _ _ . ~~~.

~ , :,....

-ao
I/a Thread. in Grip
60
_ Proof .!OO<1-
55".45 kips .
BOLT "LOAD
(kips)
40
916Thread in Grip

20 Comparison of Load-Turns Data for Different


Lengths of Thread in Grip. ~8 x 9~2 Bolts .
(Tests Conducted in Skidmore~Wilhelm')

o' ;;; :::n'a») ---- '" I , I If"' I '" , I I I " I I

0.5 . o "0.5 1.0 ·1-.5 2.0"

AVERAGE NUMBER OF TURNS FROM "SNUG" OF. 10 KIPS


" -

Fig o 7
-80
l/aThreod in Grip

60

BOLT LOAD
(kips) ~

40
9'16Thread in Grip
.:....... +.

~~- i!.--

20 Comparison of Load-Turns Data for D-ifferent -


Lengths of Thread in Grip.1 x a 5ri Bolts
(Tests Conducted in Skidmore-Wilhe-Im)
o F7'C"::_ &' > i'C c; ' " [ , , , , , , , . , , , , , ,

0.5 o .0.5 1~0 . 1.5 2.0.

AVERAGE NUMBER OF TURNS FROM "SNUG" OF, 10 KIPS

Fig. 8
80

9~; llwead in Grip - _- ·


60 .' Proof" "11VV4.· '.' "--
--- -¥.45iipr--~-""
BOLT LOAD
. (kipS)
40
~8 Thread in Grip

20.. Comparison of Load~Turns Data for Different


a
Lengths of Thread in Grip. J x 5~;Bolts .
(Tests Oondvcted in Solid Steel Block)
o' ,a=: 5 2 'FtC , , , , , , , , , , , , , , , , •

0.5 o '0.5 1.0 - 1.5 2.0'

AVERAGE -NUMBER OF -TURNS FROM ·SNUG ELONGATION OF ~


II

Fig. 9

I"
i'
.!I
80

60
A490 ~oof Loa~
- - --- -- - ---- - ----
...... ......
- ~- .........

BOLT- LOAD
55.45 kips
-- - -" - ~ ~~A325,8B LOT ........

(kips) .........

40 "",
-A325 Proof Load 1-
36.05 kips
- - - - -- - - - - - - - - - -" -, - - - --
- - - Torqued in S.-W. ""
- Torqued in" Solid Steel Block "
20
la' x 5 ~21 bolts, Rolled Threads
~8Thread in Grip

o· · -- « , I , , I , , , , , I , I , I I ,

0.5 o 0.5- 1.0 1.5 2.0

AVERAGE NU MBER OF TURNS -FROM "SNUG"

Fig. 10 COMPA.TZISON OF LOAD:-TVRt-rs DATA FOR


7/8" x 5~" A325 A:ND A490 BOLTS
I'

" 80

LOT
60
5~.45kips
·BOLT LOAD
: (kips) A325,8B LOT

40 II- Turn from


. liS
,.~
- nugII
_____ ~ 36.05 klEL

20 . ~8 II x5 Y2 at Bolts, Rolled Threads'


!I~Thread in Grip

o __ ~ ----..I~ _ _--.-IIooo.--_ _- - , - ,

Aa LOT'1 A490 BOLTS


9~~ Thread in Grip

60
l
1i
'!

BOLT LOAD H LOT 1 A325 BOLTS


(kips) 40 ~2 Thread in Grip
~2 Turn from _ -..- ,36.00 JUgs _
IISnug"

20

o .02 .04 .06 .08 .10 .12


BOLT ELONGATION (inches)
Fig. 11 COMPARISON OF LOAD VS. ELONGATION DATA FOL"',
A325 AND A490 'i~OLTS, TORQUED TENSION 'J:'ESTS
4
+ LI Lot ,Tested in Solid Block
• U Lot t i l ' II Skidmore-Wilhelm

Mean Curve --
TH READ LENGTH Establised' for A325
UNDER THE 2 . Bolts Showing ....".-~,-..-.. ..........,..

NUT, (inches) Scatter Band


--=.... .. ~.~.-:,.....: .--

o 2 3 4
TU·RNS TO FAILURE

Fig. 12 EFFECT· OF THREAD LENGTH ON


ROTATION CAPACITY