Following our discussion about the strength of my two fillet welds, I decided to put them to the test.
First I experimented with my technique a little. As Roy said that I was using too much heat if the tip of the rod was red hot when I'd finished a weld, I reduced the current from 145 amps to 115 amps (the next setting down on my welding transformer). I had problems. First the rod kept sticking to the work, then I had trouble keeping the weld pool in contact with both the horizontal and vertical surfaces. I had to weave up and down a little. And of course I ended up with some slag inclusions, like this:After I while I thought I was getting used to working with less heat, so I tried making some test pieces. I cut some 50 mm lengths of 50 mm x 50 mm x 3 mm angle and welded them back-to-back: I welded one using the drag rod technique (at 145 amps) and one manipulating the rod by hand (at 115 amps). The one where I manipulated the rod by hand ended up with a slag inclusion at the start, but I decided to test their strength anyway: I tested their strength by clamping one angle section in a vice, then holding the other section in a mole wrench (vise grips if you're American) and bending it back and forth. I counted the number of complete cycles of bending each weld could stand before it broke: The first sample (welded using the drag rod technique) took 13 complete cycles to break it. When it broke, it was actually the parent metal that broke rather than the weld: The second sample (welded manipulating the rod by hand) took just one full cycle of bending to break it. This time it was the weld that broke: I was a bit disappointed with this result, because apart from the slag inclusion this weld looked okay, and I had hoped it would perform better. So I made another test piece. This time I turned up the current to 145 amps, and again manipulated the rod by hand. I got a better looking weld without the slag inclusion. On the whole it felt better welding at a higher current: This one took 6 complete cycles of bending to break, but again it was the weld that failed: Although the test was crude, I think the result is pretty conclusive. The drag rod weld is stronger, despite the fact that it uses about half the amount of metal. It would also appear that the penetration of the drag rod weld is better. The penetration of the hand manipulated weld is pretty well zero, and it seems to have a tiny slag inclusion at the root which can only be seen when the weld is broken open.
I'm a bit discouraged because I can't seem to reproduce the drag rod weld when manipulating the rod by hand. Even if I turn down the current, I can't get the weld pool small enough, and I have problems with slag inclusions. Maybe I need a rod with a thinner flux coating which is more penetrating? Someone mentioned that 6013 is meant to be used as a drag rod, and someone else said that 6013 gives pretty bad problems with slag inclusions. Does anyone think I should try a different rod, and if so, which one?
Any suggestions would be much appreciated.