Evaluation
of Surface Preparation Methods on Porosity Formation in Aluminum Gas Metal
Arc Welding
- F.J. Hettes, Weiler
Corporation fhettes@weilercorp.com
D.L. Ketron, EWI
- The use of aluminum for
fabricating large structures is increasing rapidly. Recent developments
in the transportation industry have spurred this demand. Some examples
are:
High
speed ferry ships
1500 ton structures built entirely of welded aluminum- Automobiles
development of chassis and body structures entirely of welded aluminum - High
speed rail transportation
construction of 140 mile per hour trains using welded aluminum extrusions to form the coaches
- The main joining method
used for these aluminum fabrications is gas metal arc welding (GMAW).
While GMAW is an economical manufacturing method, the quality of finished
welds is very difficult to control. Porosity caused by hydrogen contamination
is a major problem.
- Weiler Corporation supplies
products to clean the surface of the aluminum prior to welding. Through
working with numerous aluminum fabricators, Weiler learned that there
is much more art to the GMAW process than science. Most of the fabricators
experience problems on a day-to-day basis and many times cannot determine
the source of problems. Some of the factors that affect porosity are
moisture in the atmosphere, contamination of shielding gas, welding
position, welding parameters, contamination of base or filler metal,
and surface cleanliness.
- Each customer struggles
with the porosity problem and each customer has developed solutions
which vary in effectiveness. Some of the practices in the field are:
- No
cleaning prior to welding
Mass
cleaning of surfaces in the weeks prior to fabrication
Local
cleaning of weld areas immediately prior to welding with coated abrasives
Local
cleaning of weld areas immediately prior to welding with wire brushes
Strict
use of iron-free coated abrasives
Strict
use of fine wire brushes
Research - Weiler Corporation contracted
with the Edison Welding Institute (EWI) to research the effect of various
cleaning tools on the formation of porosity in aluminum GMAW. In order
to isolate the surface prep method as the only variable, all other welding
parameters were held constant. An optimized welding process and set-up
were used to enable the surface preparation tools to be measured. The
following procedure was used:
- Research
purity grade argon shielding gas (99.9995% argon, <1.0 PPM moisture)
Filler
material from the same heat of material
Base
material from the same heat of material
Optimized
welding procedure developed for robustness
Robotic
welding cell used for repeatability in producing the welding samples
Test
plates degreased with acetone prior to application of the surface preparation
tool
Electric
powered grinders used versus air powered grinders (lube oil contamination)
Surface
preparation applied to all surfaces relative to the welding zone
Figure
1
- The welding set-up is shown
in Figure 1. The welding parameters were:
- Welding
position: 1F
Lead
angle 15 degrees
Robotic
torch manipulation
Push-pull
torch
- The test welds were 5/16"
fillet welds. Each weld was 1' long. A total of 10 different surface
preparation tools were tested, and each tool was used to clean 10' of
total test material. A milled set of test welds was used as a control.
Approximately 1/32" of aluminum was removed by the milling process.
This ensured that the welding was done on oxide free surfaces. The tools
tested included:
- Dry
machining (control)
Flap
disc, 60 grit zirconia
Flap
disc, 36 grit zirconia
Resin
fiber disc, 24 grit aluminum oxide
Resin
fiber disc, 36 grit aluminum oxide
Resin
fiber disc, 36 grit aluminum oxide (iron free)
Wire
brush .020" stainless steel twisted knot
Wire
brush .014" stainless steel twisted knot
Wire
brush .0118" stainless steel twisted knot
Wire
brush .008" stainless steel crimped wire
- All welding was done in
a random order. Test welding was performed on 3 different days, and
all tools were used randomly on each day. All welding was performed
as soon as possible after surface cleaning, but always within 2 hours.
A set of dry machined tests was performed 2 days after cleaning, 4 days
after cleaning, 6 days after cleaning and 30 days after cleaning. These
tests were performed to evaluate the effect of time delay on the formation
of porosity.
- Results
- The test results were measured
by the use of radiography and the physical counting of the porosity
in each weld sample. The radiography set-up is shown in Figure 2. The
surface preparation methods were ranked based on the average porosity
of all 10 weld samples. The individual weld samples were broken into
(4) 3" intervals, and the porosity was evaluated in each interval.
The results are shown in Figure 3 and Figure 4. Figure 3 results exclude
the first 3" interval (weld start). Figure 4 results include all
4 intervals. This weld start area showed the majority of the porosity
formed in the entire weld length. There was some shifting of the ranking
with and without the weld start area.
Figure 2
Figure 3
Figure 4
Conclusions - All test results excluding
the weld start area showed porosity levels acceptable for AWS D1.2-90
Structural Aluminum Code. There were differences in the amount of
porosity present as a result of the different surface prep methods.
It can be seen that excluding the weld start zone, the dry machining
was the best method; however, several of the other tools were very close
to the same performance. The 60 grit zirconia flap disc and the .020"
stainless steel wire brush were both close to the performance of dry
machining. The results showed:
- No
particular advantage to using iron free resin fiber discs.
While
the results were inconclusive, the tests indicated that cleaning just
prior to welding is the best practice.
Wire
brushes and coated abrasives are equally effective for cleaning aluminum
prior to welding.
- It must be understood that surface cleaning is only one aspect of a total process control that must be in place to eliminate hydrogen from the weld area to ensure high quality welds.
-
For a full copy of this report, contact Weiler Corporation's Marketing Services Department at 570-595-7495 ext. 212. For technical questions, contact Frank Hettes, Weiler Vice President of Product Development, at 570-595-7495, ext. 238.