|
Publications
| |
FD&E Weld
Challenge 4
At last, the new weld challenge is here!
This challenge differs from previous efforts in that this is a design challenge,
as opposed to a life prediction challenge. In this particular problem the
designer is asked to optimize three details welded to a tube subjected to 3
point bending and 0 - 78.4kN (max.) loading (18.6kip). The target life for
each of these details is 10,000 cycles. The base material for the tube and
attachments is SAE 1010 steel.
Attached is a drawing of the tube. Overall
the tube has a square section which measures 2.5" on a side and a
0.12" wall. The overall length of the tube is 12", but it
measures 10" between the outboard loading points. The tube size has
been selected to be the best trade-off between testing force (< 20kip loads)
and construction (a welder needs to be able to actually make the thing for the
followup challenge).
The three weld details are labelled as "left", "center" and
"right". Weld stops and starts are to be on the tube's neutral
axis, and none of the attachments are to come closer than two wall thicknesses (2t)
from either the top or bottom surface of the tube. No detail is to come
closer than 1" from any other detail. Again, the designer is to
target a life of 10,000 cycles to failure for each detail. The specific
design constraints for each detail is as follows:
Left Detail - Thick tube welded to tube surface.
As can be seen on the drawing a short tube is to be welded around its
outside circumference. The tube is centered on the neutral axis and is
removed 3" from the principal loading point. The tube has a wall
thickness of 2t (0.25"). The designer
may change both the outer diameter, D, of the
attachment and its height, h_l (although any height
less than 0.5" is considered poor form and, if you offer such a solution,
you will be required to buy beer for the entire committee).
Center Detail - Weld bead traverse (no attachments).
This detail is designed to simulate a weld line run out onto the
surface to dissipate stress and remove the weld start/stop from the high stress
regions.
The shape of this detail is envisioned as roughly trapezoidal, coming to within
2t of the horizontal beam surfaces and governed
by the lengths L_g and L_w,
but it can follow any path the designer wishes as long as it stops and starts at
the same location at the neutral axis. The weld bead below the neutral
axis is to be ground flush with the original tube surface while the weld bead
above the neutral axis is to be left in the as-welded state.
It is hoped that the designer can offer failure locations in both the ground and
as-welded portions of this detail.
Right Detail - Angled Flange.
The 2t (0.25") thickness plate is to be welded to the surface of the
tube around its perimeter. It is assumed that this operation will take two
separate welding passes for a human welder to complete a weld entirely around
the flange perimeter. Each pass is to start and stop on the neutral axis.
The plate is centered on the neutral axis, and is removed from the central
loading point by 3.5". The geometry of the plate is governed by the
plate length (L_r), the angle which the length of
the plate makes with the neutral axis, ø,
and its height (h_r). Once again, the 0.5"
beer rule applies to the flange height.
Solutions:
The solutions of this problem should not only include the geometrical
information outlined above but the techniques used in the solution, any
assumptions that you have made along the way, and the anticipated failure
locations on each of the three details.
Summary
I freely admit that there may not be a solution space for each detail as I have
defined them. Please contact me *immediately* should you determine that
this is the case. Together we will modify and publish new criteria for
that detail.
Please contact me by
Weds. March 16, 2005 if you wish to participate.
These happy few, their unindicted co-conspirators, and, quite frankly, the usual
suspects will be accorded a time slot in the Wednesday April 13th afternoon
FD&E session at the 2005 SAE Congress in Detroit to present their solutions
to the committee/audience. The committee will, at that time, review the
designs and select those that will be manufactured into specimens and tested for
the fall meeting. The presentation schedule will be published shortly in
an upcoming email.
Respectfully submitted,
John Bonnen
--
John J. Bonnen
ph. (313) 322-9127
fx. (313) 390-0514
email: bonnen(at)fmsrlg.srl.ford.com
mailing: Ford Motor Co.
MD3135 SRL
P.O. BOX 2053
Dearborn, MI
48121-2053
shipping: Ford Motor Co.
Rm. 2024B SRL
Bldg. R, Rec. R
2101 Village Road
Dearborn, MI 48124
|
