Hello JM;
Aluminum is a strange and wonderful material. One of my friends calls it "almost metal" because of its unique properties.
The weld area needs to be cleaned just before welding the joint. I recommend using a stiff wire brush with austenitic stainless steel bristles. Push the brush in one direction, toward the edge to be welded. Do not use a push pull motion, push only. Then draw file the groove face. Again, use a pulling motion in one direction.
All your cleaning implements need to be soaked and then flushed with 90% isopropyl alcohol before use (even new tools usually have an oil coating on them). Tools used to clean aluminum should only be used on aluminum to prevent cross contamination. Iron and aluminum do not get along very well. Flush the weld preparation with isopropyl alcohol after wire brushing and draw filing. Let the alcohol evaporate. Don't forget to clean any fixturing before using it.
DO NOT TOUCH THE CLEANED SURFACES WITH YOU BARE SKIN!!!!!!!!!!! That is the kiss of death when welding aluminum, yet at least 90% of the welders will draw their finger across the cleaned surface to see that it is free of burrs and is smooth. Silly is as silly does!
Even the filler metal used for GTAW must be cleaned. I suggest stainless steel wool (available at most marinas and marine supply stores) to remove heavy oxides followed by a flush with isopropyl alcohol. Let the alcohol evaporate and "keepa you fingers off".
Handle the cleaned parts and filler metal while wearing clean gloves. Gloves that look like they were used to change the oil in your motor vehicle aren't the best practice. Bare hands are worse in most cases. If you have any doubts, look at your thumb nail, it should have a dull shine to it. Now rub it beside your nose, now it will look shiny and oily. That's the same oil (from your skin) you will introduce into the weld if you don't wear gloves.
You mention the main problems encountered are porosity and W inclusions. Porosity in aluminum can usually be traced back to hydrogen sources (see cleaning precautions above). The W inclusions can be minimized by using the proper electrode. I suggest EWZr with alternating current. The diameter can be rather small since you should only need 30 to 40 amps to weld 0.032 inch thick aluminum. Use straight argon with alternating current. You can use direct current EP with EWTh-2, but that usually requires the use of helium and is typically reserved for use on thicker materials.
As for using alcohol versus acetone, I prefer to use acetone to remove heavy oils, grease, cutting fluids, etc., followed by an alcohol flush. Notice I didn't mention wiping with a saturated wiping cloth. Acetone leaves the surface of the aluminum dull, where as the alcohol leaves it with a shine, i.e., no residue. Cleaning by wiping (with no flush) has a potential of leaving pieces of lint on the surface of the metal. The lint is cellulose in the case of cotton cloth or some oil based synthetic material in the case of polyester based cloth. Not a good thing if you are trying to prevent porosity.
NEVER, NEVER use compressed air to dry or cool aluminum. Compressors use oil for lubrication. Some of the oil can and will be in the air stream.
Good luck.
Al
I apologize, I should have been more clear. It's aircraft grade titanium that I am working with, which the same thing may apply.
As far as cleaning the wire we use Methyl Ethyl Ketone, aka MEK.
I work for an aerospace company, we are nadcap approved, but we had gained some work that requires Class A radiography per AWS D17.1 specs. Any radiography failures we are allowed to repair 1 time, making these intricate parts challenging.
But you were on the right track with the aluminum content in the alloy i'm talking about. Roughly around 6% Aluminum, 4% Vanadium, 90% commercially pure titanium.
I'ts mainly the edges that i'm concerned with, to cut the material we have a laser and waterjet, then the details get chemically cleaned in hydroflouric acid and alkaline cleaned. We just need a higher success rate on the radiography to minimize the repairs because of failure.
By Lawrence 
Date 04-17-2007 05:17
Edited 04-17-2007 05:26
0.032 titanium test coupons either sheared or watercut should not need to be filed anyhow. If you must file use a dedicated tool, however I advise against.
I suggest a 1 1/2" 120 grit silicon carbide disc driven by a pnumatic die grinder with rear exhaust @12,000 rpm for pre weld oxide removal and surface prep followed by isopropal or acetone... NO TRICHLOR for Titanium.
0.032 needs no root gap. The coupons should be a 100% perfect flush fit with absolutely no mismatch and restrained in in a backpurge fixture.
The filler wire (as Al mentioned) should also be cleaned with acetone right before welding and always kept under the argon shield for 100% of your arc on time... If you must restart, clip the end.
Just about the only way to get tungsten inclusions with 0.032 6Al4V is to stick the tungsten.... The only exception would be if you have one of those Lincoln Invertecs with the recalled reverse polarity pulse sequence at arc start. I suspect your "tungsten inclusions" are in reality debris imported by the file.
0.040 or 0.063 dia. tungstens will do. A diamond wheel is best but at least have a look at your typical tip prep under at least 10X to be sure your not bringing any uninvited guests to the party at the end of your electrode. If no diamond wheel I like a 2 inch 120 grit belt sander worn flat to put on a nice smooth finish.
Acetone or Isopropal Alcohol leave less residue and evaporate quicker than MEK
That's what happens when I read and respond to threads at midnight.
About the only thing I would change is; weld with DCEP and use EWTh-2 with a tapered end prep. Other than those two things, all else remains the same. Lawrence has the right info.
Make sure you have proper shielding, which doesn't sound like a problem from your inquiry.
Best regards - Al
By jwm80
Date 04-17-2007 17:20
Edited 04-18-2007 00:28
Thanks alot guys for the info. Another question was that the argon flow might be the culprit also. We run the argon very very low on the root side and use a #12 gas lens cup on the face side with alot more pressure to keep alpha casing off of the parts.
If you have any links with some information about titanium x-ray would be great.
Another thing is we use very secure tools to purge and hold down the parts.
Btw, Lawrence I remember some of your posts quite a ways back when I frequented these boards, I learned alot and I had forgotten my old user name so I re-registered.
I doubt shield/trail/backing gas are the culpret here......
This is why. For porosity or inclusion indications on RT to appear due to gas issues there would surely be visible discontinuities at the face and root. Now I understand that the amount of Oxygen/Nitrogen contamination required to form alpha case (alpha case is not detectable via RT) may be less than the amount of Oxygen/Nitrogen required to cause surface discoloration..... This being said, the gas would have to be terribly wrong to cause inclusions or porosity, you would be seeing greens, blacks, whites and chalky yellows and I bet your not.
But I have been thinking about your situation and several things have come to mind.
1. Laser cutting (in air) is going to leave an alpha case and heat affected zone. Both the face and top of the sheet at the edge of the kerf would need special attention and mechanical treatment...(I still say the 120 grit discs) in order to remove stuff.
2. Waterjet cutting usually requires some form of abrazive media too, could some silica or whatever that media is be leaving traces on the faces of your coupons...
It is easy to forget to pay attention to the faces of coupons when they are as thin as the stuff you are working on. This is true for Lasers, Plasma, Oxy-Fuel, Waterjet, Shears (who knows what that shear has been cutting eh?) They just blend off the oxides from the top and bottom and tack it up. I wonder if the faces of your butt joints are where the source of contamination lies?
And for your Titanium trivia nugget of the day......... Titanium cuts very nicely with Oxy-Fuel cutting torches. Edges of course need attention after the cut but the process works just nicely.
By jwm80
Date 04-18-2007 00:26
Edited 04-18-2007 00:45
Yes, the laser puts a slight discolor approximately .125 from the edge of cut. The waterjet puts a burr on the bottom side of the cut and has a sandblasted looking edge on the seam. Then, it would go to the blend/deburr department, they use 3M 120 garnet 2 inch roloc discs using a 90 degree die grinder. The edges are broken to remove the slag from the cut, or the sharp edge from the waterjet.
I'm beginning to think, as Al has suggested, most of the porosity I am seeing in the film may be caused by oil in the air from the waterjet intensifier that is located in our welding department, which is climate controlled.
Not to mention cleaning off debri by blowing parts off with compressed air.
Another issue may be the rags, which are cloth and DO leave fine hairs behind after a MEK wipe prior to weld. Do you think denatured alcohol might be a better solvent to clean the edges before welding?
