Cleaning - How much is needed.

[sschefer]

This isn't a question.

I see a lot of questions about arc starting, pops (inclusions), wandering arcs, etc. Everbody seems to be looking at torches, tungstens, gas and machines as the problem.

You can't troubleshoot equipment or operator problems if you don't eliminate the material itself first.

Aluminum is the worst so I conducted a few tests yesterday. The test was done on 1/4"x1" Chinese aluminum from HD. The test welds were Fillet and a beveled Butt weld. There were two pieces being welded together.

In the first test I simply wire brushed with a stainless steel brush dedicated to aluminum and welded. The black specks and spatters were present throughout the weld and the Fillet weld was difficult to keep moving. In otherwords, the weld looked lousy. Arc start was internittent and you could feel the transition from - to +. That was something I had gotten used to so I really didn't notice it until after the second test when it dissappeared.

In the second test, I washed the pieces in warm soapy water using Dawn dishwashing detergent. I followed that with a 3 minute soak in Alumaprep 33 and a rinse with warm water.

Next I dried the pieces with a heat gun and then used a right angle die grinder with a new green scotch brite disk to cut any remaining mill slag or oxidation. I then wiped them both down with Acetone.

The final prep was to preheat with my heat gun set at 600 degrees. I estimate that the metal was about 200 degrees when I finished. I followed this with a quick brushing with a new stainless steel brush that I had pre-cleaned with Acetone, (bet you never thought about the manufacturing oils on your brush).

Does that sound crazy? I thought so too until I stated tacking and was able to do so with nice small tacks. When I laid down the root pass on the beveled butt joint, it stayed right in the middle of the weld and I got 100% penetration without making a mess.

I let it cool for about 1/2 hour and then laid in the Fillet weld. It was one of the best Fillets that I have ever done. It had perfect penetration and was not undercut. The toes were evenly distributed on both pieces and the crown was perfect according to the pictures and charts that I was able to compare it with. Most notibly, the arc stayed right where I pointed it and the tungsten had just a very tiny ball. The tungsten was a 3/32 2% thoriated chinese brand.

For the tests I was using 80% argon and 20% helium. 175 amps with frequency set to approx 75 hz and DCEN at about 35%. The machine was my Everlast 250EX. Gas flow was approx 40 CFH because I was using He and it tends to drift up if you don't have enough flow. Pure Argon would have been around 15-18 CFH.

As a final test I bent both pieces nearly in half at the welds and the first piece broke, while bending it, almost immediately. Some of the breakage occured at the weld and others in the HAZ area.

The second welds did not break and there were no visible signs of material fatigue except those from the 3lb sledge that I was beating on it with. That tells me that the HAZ was very limited on the second piece.

I did not change the settings on any of my equipment during the tests.

It was proven to me, beyond a doubt, that the cleanliness of the material is critical to not only producing a good weld but also in determining if your skills and equipment are up to snuff.

Take it for what it is, just a test that proved something to me and made me a happy, complaint free welder.

Maybe this will help with some of the problems people are having.

Happy Holidays..

[WILD BILL]

I wish you could have posted this BEFORE I wasted half a tank of argon trying to practice on dirty metal


But you are right. Prep makes all the difference.


Merry Christmas

[geezer]

I have to agree with you , clean material is a must, that and a good ground clamp sure make a difference.

[sschefer]

I thought I'd post the results of the tested welds and finished welds. Mind you these are by no means perfect but they are a lot better that what I was getting.

This is of the Butt weld. The one on the left is ground flat and the one on the right is the piece that was bent in the vise then beat on with a single jack. The extra weld on the fill pass is so that the depression from welding is filled and the weld can be ground flush.



This is the fillet weld and the bent piece that show how much of a bend it took without a single crack anywhere.



Cleanning is a pain in the arse but the rewards are great! Well, in my mind anyway..LOL...

[performance]

Scheffer,

A front and back bend on a true test is in order. If you can, cut into 2, two inch strips, and bend one forward, the easy way and bend it backwards the hardway. That will tell you much more.

But it does look like it would have probably held.

[sschefer]

Mark, you're right but I don't really have the right equipment to test correctly. I should never settle for anything less than the best so maybe I'll go get that 20 ton press I've always been wanting.