Roof Truss Tests

Publisher’s version / Version de l'éditeur:

Technical Note (National Research Council of Canada. Division of Building Research), 1961-08-01

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la

première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.

Questions? Contact the NRC Publications Archive team at

PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.

Archives des publications du CNRC

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

https://doi.org/10.4224/20338405

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Roof Truss Tests

Hansen, A. T.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site

LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

NRC Publications Record / Notice d'Archives des publications de CNRC: https://nrc-publications.canada.ca/eng/view/object/?id=eae82a78-d50a-4256-a77f-636dd7472970 https://publications-cnrc.canada.ca/fra/voir/objet/?id=eae82a78-d50a-4256-a77f-636dd7472970

DIVISION OF BUILDING RESEARCH

No.

NATIONAL RESEARCH COUNCIL OF CANADA

343

NOTE

'IfE

C

1HlN ][

CAlL

NOT FOR PUBLICATION _{FOR INTERNAL liSE}

PREPARED BY _{A. T. Hansen} CHECKED BY H.B.D. APPROVED By N.B.H.

pATE August 1961

/fYL.

Discussion with Central Mortgage and Housing Corporation and Forest Products Research Branch

ROOF TRUSS TESTS

SUBJECT

PREPARED FOR

As part of a long-term study of the strength of house frames, the Division has undertaken, in co-operation with the Forest Products Research Branch, the development of a series of standard roof truss designs for housing. These designs were for 4/12 and 5/12 roof slopes and spans ,of 24, 26 and

28 ft and were published in CMHC Builders Bulletin #109. '

Recently there has been an increasing demand for similar truss des igns for low slope roofs of 3/12 pitch and for an extension of the pre-viously developed designs to include spans of 30 and 32 ft. Accordingly, additional work has been planned and the details discussed with Forest Products Research Branch and CMHC as recorded in DBR Technical Note

#331. Subsequently a request was received from the builders through CMHC for the development 'of standard truss designs incorporating cantilevered details.

The present note contains a short review of the pUrpose of this current investigation and summarizes the results obtained to date on a pilot study of 7 designs, with recommendations for further development and acceptance testing of specific trusses. It has been prepared for review and further discussion with CMHC and Forest Products Research Branch to determine future action.

PURPOSE

The aims of the current project to investigate roof trusses are as follows:

1. To develop truss designs for a 3/12 roof slope and a range of spans 2. To develop truss designs for 30- and 32-ft spans for 4/12, and

3. To reduce the number of nails in truss designs included in CMHC Builders Bulletin #109 with the aim of providing trusses to meet the deflection and failure load criteria for SO, 40 and 30 psf snow load areas.

4. To investigate the possibility of providing a standard detail for cantilevering of W trusses over the exterior walls. PILOT TEST PROGRAM

In order to explore the above aims without becoming too involved at this stage with an extensive series of testing, a number of pilot tests were conducted. It was hoped that from these tests sufficient information could be collected to form a basis on which more extensive testing could be planned.

TEST PROCEDURE

Trusses were tested in pairs spaced 2 it o. c. and sheathed with plywood to prevent buckling. The first set of trusses (Test IP) were tested using concrete blocks to apply the roof load and lead-filled bags to apply the ceiling load. This method was used rather than the previous method using hydraulic jacks because the geometry of the trusses did not lend itself to a realistic loading using hydraulic jack loading. The method was too cumbersome, however, and was abandoned after the first test.

The ceiling load of 10 psi was applied first and deflection readings were taken after five minutes. The roof load was applied in 25-psf increments. De-flection readings were taken after five minutes after the first increment. five minutes and one hour after the second increment and five minutes after the third increment. It was thought to be unsafe to continue further loading by this method and the total roof load of 75 psf was maintained on the trus s.

Subsequent tests were conducted using hydraulic jacks since the uniform geometry of subsequent trusses made this possible. In these tests, the ceiling load was applied first be means of lead-filled bags and the roof load applied in increments. In all cases, readings were taken after five minutes of loading each increment.

In Test 2P, the trusses were loaded in 10-psf increments up to 40-psf total roof load. When the 40-psf roof load was reached, the roof load was removed and 20 per cent of the nails were reduced before reapplying the

load up to 40 psi. The load was again taken off and an additional 20 per cent of the nails removed before reloading to 40 psf. This procedure was repeated until only 40 per cent of the original nailing was left. In some cases, loading was not carried up to 40 psf due to excessive deflection. The trusses were then renailed to original nailing and loaded to failure.

.' ｾ

3

-In Test 3P, a similar procedure was followed except that the trusses were loaded to failure when only 20 per cent of the nails remained.

In subsequent tests, the trusses were loaded in 20 psf increments until failure.

INSTRUMENTATION

Deflection readings were taken to the nearest 0.01 in. by means of a wire strung along the bottom chord. Readings were taken at the panel points and mid-way between panel points along the bottom chords.

DES CRIPTION OF TRUSS DESIGNS

In all, a total of seven designs were tested as illustrated in Figures 1 to 7.

RESULTS OF TESTS

The results of the tests may be summarized as follows: (1) ｾ･ｳｴ IP - 2a-ft span, 3/12 slope, W trus s (Figure 1)

(a) Defle ctions

Deflections at centre8 of Trusses

(in. ) Roof Load in Addition

to Ceiling Load Ceiling load only

25 50 75

Note: 1/360 of span

=

Trusses failed after lower chord broke in combined tension and bending near the heel gusset after being loaded with 75-psf roof load for three days.

(2) Test 2P - 28-ft span, (a) Defle cHons

Roof Load in Addition to

Ceiling Load

Deflections at centres of Trusses, in.

% of Nails Removed (based on nailing shown in Fig. 2)

0 20 40 60 Ceiling load 0.26 0.51

O.

Note: 1/360 of the span

=

When the trusses were renailed to original nailing and reloaded, the upper part of the top chord broke at a large knot at an applied load of 80 psf. ( 3) Test 3P - 28-ft span, 3/12 slope, king post truss (Fig. 3)

(a) Deflection

Deflection Ratio Between Panel Points in Lower' Chord for Various Reductions in Original

Roof Load in Nailing

Addition to

0% 20% 40% 600/0 80%

Ceiling Load

removed removed removed removed removed

Ceiling load only 1/480 1/380 1/380 1/340 1/280

10 1/330 1/280 1/280 1/260 1/210

20 1/260 1/240 1/240 1/220 1/170

30 1/210 1/210 1/200 1/190

40 1/180 1/180 1/180 1/170

(b) Failure

Failure occurred at 30 -psf roof load after three minutes when the heel joint gusset plate pulled away from the top chord. Eighty per cent of original nailing was removed when failure took place.

5

-(4) Test 4P - 28-ft span, 5/12 slope, W truss (Fig. 4) (a) Deflection

Roof Load in Addition to Ceiling Load

..

Ceiling load only 20

40 60 80

Deflections at centres of trusses, in.

0.13 0.53 0.91 1. 38 2.24

Note: 1/360 of the span

=

Failure occurred at heel joint gusset when nails pulled out after being loaded with a 100-psf load for about two minutes.

(5) Test 5P - 28-ft span, 3/12 slope, W truss (Fig. 5) (a) Defle ction

Roof Load in Addition fo Floor Load

Ceiling load only 20 40 50 60 80 100 120

Note: 1/360 of span

=

Deflections at centres of trusses, in.

0.16 0.47 0.71 0.82 0.93 1. 23 1. 59 2.18

(b) Failure

Failure occurred at 140 psf after three minutes of loading when upper part of top chord broke in bending.

(6) Test 6P - 28-ft span, 5/12 slope, W truss (Fig. 6) (a) Defle ction

Roof Load in Addition to Ceiling Load

Ceiling load only 20

40 60 80

Deflections at centres of trusses, in.

0.15 0.51 0.89 1. 36 2.32

Note: 1/360 of ｴｾ･ span = 0.93 in. (b) Failure

Failure occurred at 100 psf after the load had been applied for about three minu.tes, when the upper part of the top chord broke in bending. (7) Test 7P - 28-ft span, 5/12 slope, W truss cantilevered 4 to 8 ft out over

one end support (Fig. 7) (a) Defle ction

Roof Load in Addition to

Ceiling Load

Ceiling load only 20 40 60 80 Deflection Ratio Between Supports 1/1120 1/750 1/500 1/370 1/160

Deflections of cantilevered heel of truss, in., relative to support

0.02 0.13 0.26 0.42 0.90

7

-(b) Failure

Long diagonals on cantilevered side showed considerable buckling at 80-psf roof load and broke in bending as the load approached 100 psf. Load of 100 psf was held for three minutes when the top and bottom chords broke over the support on the cantilevered side.

DISCUSSION OF RESULTS

1. Tests IF, 2P, 3P and SP were concerned with the development of low slope truss designs. The geometry of lP differed slightly from 2P and 5P in that the upper part of the short diagonals was at right angles to the top chord rather than being located at the centre span of the top chord members as was the case with other W type trusses. In Test 2P, king post trusses were tested. The king post truss proved to be so weak and flimsy that further-development of this design was abandoned. Test IP indicated that the trusses tested should be adequate for snow loads of almost 50 psi. Although these trusses failed at a load of 75 psf after three days, it was noted that the lumber in this truss was of particularly poor quality and probably did not indicate a typical strength. Test 2P, on the other hand, while of slightly different geometry, should have performed similarly to IP. This test, however, showed the trusses to be unusually flimsy, even at full nailing. Because of the inconsistency in results, a third test was carried out on trusses of similar geometry as 2P but with slightly larger gusset plates and with slightly revised nailing. This test, 5P, indicated that the trusses should be satisfactory for 50-psi snow-load areas both in respect to deflection and failure load.

2. Tests 4P and 6P on 28-ft span 5/12 slope trusses were conducted to determine the effect of reduced nailing on truss stiffness.

In Test 4P, the nailing was reduced from the nailing for similar trusses shown in Builders' Bulletin #109 by 55 per cent. This was calculated on the basis of test results for the standard designs in Builders' Bulletin #109 to obtain the minimum amount of nailing to provide adequate stiffness for a 50-psf snow load area. The test results, howeverp showed the trusses to be less stiff than

predicted, although these trusses should be adequate for a 40-psf snow-load area. In Test 6p, similar trusses were tested except that in this case, the nailing was reduced only 40 per cent of that of the standard truss designs. These trusses were not improved over those with lesser nailing in 4P however. In Test 6P, relatively green lumber having a moisture content of between 18 and 20 per cent was used, and it is presumed that this contributed to excessive deflec-tions for this nailing by lowering the modulus of elasticity of the members and contributed to larger movements at the nailed connections. It is proposed, therefore, to retest this reduced nailing with air-dried lumber.

3. Test 7P was conducted to develop a standard method for cantilevering trusses. Although the cantilevering did not produce very serious deformations at lower loads, this method caused a reversal in stresses in the diagonal

members on the cantilevered side. Because these members are not designed to cope with these stresses, this method of ca...Tltilevering is not to be recom-mended, at least for the trusses using I-in. thick diagonal members. No simple solution to the cantilevering problem has been found. Increasing the depth of the bottom chord was suggested but this would mean the introduction of very deep members which makes this impracticable. The remaining solution, introducing a new member to transfer the load from the bearing point to a panel point, requires that the entire trus s be redesigned with regard to nailing, and that this nailing would change, depending on the amount of

cantilevering. Rather than introduce this complication in the standard designs, it is suggested that standard trus ses be supported only at the ends. Where cantilevering is necessary, the trusses should be supported on a post and beam arrangement.

4. : Time did not permit further pilot tests to be conducted to provide infor±nation on long span trusses. It is hoped that this will be done at a future date.

CONCLUSIONS AND RECOMMENDATIONS

1, Nailed truss designs for 3/12 slope of the type of trusses tested in Test 5P should be developed as standard trusses for spans up to 28 ft. 2. A substantial reduction of nailing in the standard designs for 5/12 slope trusses seems possible, for 30- 40- and 50-psf snow-load areas. Further testing will indicate whether similar reductions are possible for the 4/12 slopes as' well.

3. A simple standard cantilevering detail that can be applied to existing truss designs does not q.ppear feasible, and no further test work along this line is proposed.

4. Further testing should be conducted to provide truss designs for spans of up to 32 ft for 3/12, 4/12 and 5/12 slope ｴｲｵｳｳｾｳＮ

5, Truss designs for spans of less than 24 ft should be provided. No testing is contemplated for these designs but the nailing in these designs will be calculated from the longer spans which have beelIl tested.

6.

e

e

e

LUMBER - C. L.A. NO.1 E. SPRUCE - 2"x 4II

PLYWOOD - 1;2" D.F. SHEATHING GRADE - BOTH SIDES

NAILS - 3" COMMON 12

Ｓｾ

. . . .

...

.

.:--...:..-..

-:=::-:::;:.::1

FIGURE

e

LUMBER - C.L.A. NO. I E. SPRUCE - 2·x 4"

PLYWOOD - 1/2"D.F. SHEATHING GRADE - BOTH SIDES

NAILS - 3"COMMON 2' - 6"

FIGURE

2

Ｓｾ

e

-

_e

LUM BER - C.L.A. NO. I E. SPRUCE

PLYWOOD - '/2" D.F. SHEATHING GRADE - BOTH SIDES

NAILS - 311 COMMON 12

ﾷＳｾ

FIGURE

3

7'- 0"

8 - 2

liz"

H

9'- 4"

5-2h"NAILS FROM FAR SIDE

10-3" NAILS

("x 6" TIE (FAR SIDE)---..

12"

-LUMBER - C.L.A. NO. I E. SPRUCE PLYWOOD -

'12"

28' SPAN

e

e

e

LUMBER - C.L.A. NO. I E. SPRUCE - 2"x 4"

PLYWOOD -

V2

D.F. SHEATHING GRADE - BOTH SIDES

NAILS - 3II COMMON \I NAILS

ｾ

ＳｾＮ

a"

］］ＺｴＧｾｾＺｾＱＲＧＢ

t

ｾＺＺＺＺＺＺＺＺＺＺＮ｟ｪＺＭＺ［ＮｊＭＮｾ］ｲ］ｊ

1'

:.:".

:""1""

:

"·1

ｾＮＺ

Ｇｾ

I' ."

ＮｾＺ ｟ＧｾＧＬＺ

.. , ..

,i-.--__

-

FIGURE

5

-LUMBER - C.L.A. NO. I E.SPRUCE

PLYWOOD -

Y2"

NAILS ... COMMON

7'-0"

I"x 6" TIE ( FAR

5 -

21'z"

FIGURE

6

e

sp

e

_'/z"

'12"

I

..

.

I

e

e

e

LUMBER - C. L.A. NO. I E. SPRUCE

PLYWOOD - ｾｺＢ D. F. SHEATHING GRADE

NAILS - COMMON

NOTE:

RIGHT HAND DETAILS

SAME AS PE R FIG. 6 2\ 4" 5-2!tz"NAILS FROM FAR SIDE 12

Ｕｾ

FIGURE

7