Pegged mortice and tenon connections in traditional UK green oak carpentry

1
Pegged mortice and tenon connections in
traditional UK green oak carpentry

• Jon Shanks
• University of Bath 7th September 2006

2
Current design practices
. it does not cover well tried and
traditional methods of timber construction which
have been employed successfully over a long
period of time.
BS 5268 section 1.1
3
Research aims

• This research aimed to
• Establish a better engineering understanding of
  mortice and tenon joints in green oak
• Work closely with craftsmen and timber frame
  practitioners
• Produce pragmatic tools for modelling timber
  pegged mortice and tenon connections and
  disseminate findings

4
Research form

• A three year EPSRC and Buro Happold funded PhD.
  project at Uni. of Bath
• Extensive experimental programme comprising
• - 240 peg material tests, 180 peg tests in a
  simulated mortice and tenon, 66 full-size joint
  pull-out tests, 23 joint rotation tests and four
  joint shear tests, five knee-braced sub-frame
  tests and four knee-braced cross-frame tests
• Theoretical analysis using finite element and
  ultimate collapse analysis

5
Tensile tests (double shear)

• Cleft tapered, cleft die driven and turned pegs
  tested
• Failure loads range from 600kg to 1100kg
• In every case the peg failed before any
  significant damage occurred to the Mortice or
  Tenon

6
Typical results
7
Rotational tests

• Connections tested in rotation
• Varying fit between tenon and mortice tested
• Accurate prediction of rotational stiffness
  possible

8
Typical results
9
Connection modelling
10
Connection STRENGTH model based upon yielding
hinge failure

Connections displaced, cast in resin and dissected
External work done equated with Internal energy
dissipated to determine failure load NB Assume
all energy dissipated in hinge formation
11
Connection STIFFNESS model based upon 4 point
bending
Mortice and tenon boundary
Applied Load, Pp
Peg
d
a
a
L
Observations from experiment and finite element
analysis led to simple 4 point bending
stiffness model
Peg rotating in peg hole. Peg tip not bearing on
peg hole bottom.
12
Frame tests
Fourteen FULL-SCALE FRAMES tested Five
knee-braced sub-frames (thanks to GOCCo.) Five
Arched-braced sub-frames (thanks to COWCo.) Four
knee-braced cross-frames (thanks to Oakwrights)
Frame modelling
STIFFNESS model based on multi-spring
connections Spring stiffness determined from
connection stiffness model and connection
geometry
STRENGTH model based on upper-bound collapse
model Connection strength taken from connection
model and frame geometry used
13
Arched braced frames
Tested as part of undergraduate dissertation
Tom Hill
Thanks to Mann Williams and COWCo.
14
Frames then repaired with GRP, threaded bar, and
straps Retested
15
Knee-braced cross-frame
Thanks to Oakwrights
16
Frames tested
17
Frame collapse analysis
Load displacement, ?
Joint rotation, ?j
Applied load
Peg yield force, Pp
18
(No Transcript)
19
Frame response
20
Thanks to the contributors
Pete Walker (UoB) Richard Harris (Buro
Happold) Christopher Mettem (TRADA) EPSRC Oakwrigh
ts Carpenter Oak and Woodland Ltd. The Green Oak
Carpentry Co.