"Frosted" Aluminum welds?

admin

Teacher gave me a challenge which made me go 1" square aluminum with 1/4" walls and 1/8" flat piece. Weld it all around in only 2 welds. Never had to do anything like this before so I sat there for a long time trying to figure out how to wrap the corner and weld have the damn box at once. First attempt was thinner and didn't work out so well, this one came out ok at best.I stood up looking down at the box with the torch facing down as well.Lincoln 185 tigAround 110ish amps, don't recall exactly.3/32 gold tungstenGas lens#20 water cooled torch#7 cup20cfh 100% argon4043 filler (1/16th or 3/32, can't remember)All wire brushed prior to weldingThis is the first section of the weldAnd the other half of the first weld looks ok, but it's not frosted. Teacher looked a little confused and said usually it's a gas problems, but it usually turns black. Trying to figure this out, because 20cfh, gas lens, I'm thinking I didn't have a gas issue. So maybe he hasn't seen this before?I did the second half of the box and go figure, my first section of that weld as well was frosted and the rest looked normal.Any thoughts/ideas?Torchmate 2x2 CNC with Flashcut CNC controlsHypertherm Powermax45 Esab ET220i Razorweld 195 MigRazorweld 200ac/dc TigTormach 770, Tormach xstechRazorweld, Vipercut/Vipermig, SSC Foot Pedal Dealer
Reply:For something like that, I would reach around and weld the farthest side from me and then the side to the left.  Also, weld one of the sides closest to the edge of your flat piece first.  Those edges will get hot the fastest so it helps to start them when the piece is cold.My name's not Jim....
Reply:Residue left on the metal that reacted to heat? Its looks like the frosting ends where the discoloration would end with a TIG on stainless.Millermatic250Miller Spoolmate 200Miller Spoolmatic 1Lincoln tombstone with century DC converter1945 K.O. Lee company stick welder (looks like R2D2)Miller 30E feederHobart AH27 FeederMiller Thunderbolt AC/DC
Reply:Usually it looks frosted if you move too slow (low amps), after one half it heated up and you were able to move faster.Airco Ac/Dc 300 HeliwelderMillerMatic 200 (stolen)Miller Maxstar 150STLMiller AEAD200LE (welding and generating power) Hobart MIG
Reply:Originally Posted by SquirmyPugUsually it looks frosted if you move too slow (low amps), after one half it heated up and you were able to move faster.
Reply:Originally Posted by GambleHow about structural? Does it make any difference?
Reply:Originally Posted by SquirmyPugUsually it looks frosted if you move too slow (low amps), after one half it heated up and you were able to move faster.
Reply:Damn you are pretty good. It took awhile to make that tack weld.Book?? What book? LolTeacher doesn't have us do book work. Or anything on paper. It's 99% welding. I had a safety test, midterm and 1 final test.Torchmate 2x2 CNC with Flashcut CNC controlsHypertherm Powermax45 Esab ET220i Razorweld 195 MigRazorweld 200ac/dc TigTormach 770, Tormach xstechRazorweld, Vipercut/Vipermig, SSC Foot Pedal Dealer
Reply:Originally Posted by GambleBook?? What book? LolTeacher doesn't have us do book work. Or anything on paper. It's 99% welding. I had a safety test, midterm and 1 final test.
Reply:Originally Posted by GambleDamn you are pretty good. It took awhile to make that tack weld.
Reply:For tig the "Hobart Institute of Welding Technology- Gas Tungsten Arc Welding" (EW369GTAW) is a good workbook. Hobart has a book for each welding process and even some for learning pipe. Their "Welding Guide" book is really good.The book I have used the most to find info is "Modern Welding Technology" it is a little expensive but a great book to have with tons of info on welding.If you are learning blueprint reading then "Blueprint Reading for Welders" is great, again lots on info and blueprint reading exercises.Airco Ac/Dc 300 HeliwelderMillerMatic 200 (stolen)Miller Maxstar 150STLMiller AEAD200LE (welding and generating power) Hobart MIG
Reply:Originally Posted by SquirmyPugFor tig the "Hobart Institute of Welding Technology- Gas Tungsten Arc Welding" (EW369GTAW) is a good workbook. Hobart has a book for each welding process and even some for learning pipe. Their "Welding Guide" book is really good.The book I have used the most to find info is "Modern Welding Technology" it is a little expensive but a great book to have with tons of info on welding.If you are learning blueprint reading then "Blueprint Reading for Welders" is great, again lots on info and blueprint reading exercises.
Reply:"My teacher says he doesn't teach book work because in the field you need to know how to weld vs read."  What a bunch of crap.Just think about the fact that "doing a weld right" is just as important as the technical aspect of "why did it work."  Without that technical background you are severely limiting yourself for future welding situations.Example of something you would read: using too little amperage and traveling too slow causes excess overall heat input to the workpiece.I could go on with numerous examples of questions people ask that are in standard textbooks.End of rant.Dave J.Dave J.Beware of false knowledge; it is more dangerous than ignorance. ~George Bernard Shaw~ Syncro 350Invertec v250-sThermal Arc 161 and 300MM210DialarcTried being normal once, didn't take....I think it was a Tuesday.
Reply:Everyone has their own style of teaching. Like I said maybe some of this was in the first class.Torchmate 2x2 CNC with Flashcut CNC controlsHypertherm Powermax45 Esab ET220i Razorweld 195 MigRazorweld 200ac/dc TigTormach 770, Tormach xstechRazorweld, Vipercut/Vipermig, SSC Foot Pedal Dealer
Reply:Sounds like he is training guys to be grunt welders, someone tells them what to do then they weld it.Airco Ac/Dc 300 HeliwelderMillerMatic 200 (stolen)Miller Maxstar 150STLMiller AEAD200LE (welding and generating power) Hobart MIG
Reply:Originally Posted by SquirmyPugSounds like he is training guys to be grunt welders, someone tells them what to do then they weld it.
Reply:I will try not to get involved in the comments regarding the instructor not utilizing books.  To each their own. The "frosting" you're referring to in the picture above is from the cleaning action of the machine.  Pure Aluminum reacts with the atmosphere to create a very thin Aluminum Oxide layer on the surface.  Aluminum oxide is some tough stuff.  Almost as hard as diamonds.  It has a melting point of ~3700°F while pure Aluminum has a melting point of ~1200°F.  To give you an idea, they make sandpaper and sanding belts out of it.  This is the stuff you're removing when you brush Aluminum.  When using AC current, the wave shifts from positive to negative (duh).  Depending on the machine (newer technology vs older technology), the freedom you have to adjust specific parameters with newer inverter machines is insane.  Older machines you aren't so lucky with.  Older machines run at 60 Hz which means the wave will start, follow the path and peak up to the positive end, head back down and peak on the negative end and return to 0, 60 times in one second.  Again, duh.Visual:When the wave is in the positive, the electrons are bombarding the plate knocking off the Aluminum Oxide layer, and when you're in the negative, you're welding.  On the old machines, you can't change much of the wave characteristics.  Typically you're stuck at 60 Hz, welding with a "balled" pure Tungsten.  Now a days, with inverters, you can adjust how much time you spend on each side of the wave (Wave Balance), how many times the cycle will repeat in one second (Frequency), and on some newer machines you can even set peak amperage on the negative and positive side.  I sure hope this is ALL review.The white area on each side of each toe of the weld is where the "cleaning" action took place.  Honestly, you look like you have quite a bit of cleaning.  I did a quick Google search of your machine because I'm not familiar with it, and it looks like there is a dial on the front where you can adjust "cleaning and penetration."  I would have moved the knob more to the penetration side to reduce the amount of cleaning.  I personally prefer to have about 1/16" of a cleaning trail at each toe.  Typically adjusting the settings to achieve that results in a good weld for me.  What type of tungsten is the gold?  Is it sharpened or balled?Little black specks in the weld can mean tungsten contamination (from touching the tungsten to the weld, the filler to the tungsten etc), dirty filler rod (I like to wipe mine down with isopropyl alcohol before welding), dirty base material, or an issue with your shielding gas.  Judging by the picture, I'd say your gas is fine and it was more of a contamination issue.I don't pretend to be a welding engineer, or to know everything there is to know about AC welding.  Everything I stated above was taught to me by my welding instructor, who does use books and expects us to read them on top of his lectures.  If you have any questions, feel free to ask.  I hope this helps.  Scott
Reply:Thanks.Gold is 1.5% lanthanted. I just left the cleaning/penetration to auto. I don't know what the percentage is for that machine on auto. The tungsten does ball up, but it's a really tiny ball. It still doesn't make sense as to why half of my weld has more cleaning than the other half when I didn't stop and restart. I guess I'll try it again next week and see what happens.Torchmate 2x2 CNC with Flashcut CNC controlsHypertherm Powermax45 Esab ET220i Razorweld 195 MigRazorweld 200ac/dc TigTormach 770, Tormach xstechRazorweld, Vipercut/Vipermig, SSC Foot Pedal Dealer
Reply:Beautiful explanation of the AL GTAW process from Seiler.Syncrowave 300Maxtron 450, S-52E, 30A
Reply:Based on the parent metals being joined, the machine being used (and the current you said you were running on it), and the appearance of the joint, I'd say you had a slow travel speed.  Fix it by cranking up the current (going to a higher amp machine if necessary).  And if that doesn't do the trick, mixing in some Helium into your shielding gas sure will fix it.  Doing some pre-heating is yet another approach.It's not that it's necessarily a bad weld joint, but it may have poor penetration.  For the type of fillet joint that you are doing, you can just build up your reinforcement tall enough for the desired strength.  Even if it doesn't penetrate very deeply, it can be strong with adequate reinforcement on the front.From experience, a weldor will know how at least approximately how deeply it penetrated, and whether there may be inclusions in the weld joint (which if you melted your metals together, rather the dribbling molten drops onto a solid surface, shouldn't be an issue).  I don't see anything that suggests you have inclusions, but would say to work on your consistency.And if you don't have the machine maxxed out in terms of its output current on a joint like that and are traveling so slow, why don't you crank it up?  Learn the proper heat input.  1 amp per thousandth is a rule of thumb that gets you in the ballpark surprisingly often.  You are not in the ballpark, and that's why your traveling so slow.  Also run just enough EP% to get good cleaning.  Anything more than that just makes your machine and your tungsten tip work harder than it needs to (and it reduces the effective capacity of the machine.)Hopefully your instructor will be teaching you how to verify your welds integrity, such as by performing destructive bend tests?  That is the ultimate test of a welded joint's integrity.  Do it as much as you can on you test coupons until you develop confidence in how much strength a given weld joint is going to have.Just my 2 cents.  Best of luck!Last edited by jakeru; 03-25-2013 at 11:13 PM.
Reply:I have a PT 375 at work with an auto AC balance option, it may be giving you more cleaning at some times than others, if you dipped the tungsten or hit a speck of crud, the cleaning may have ramped up on it's own. If you're trying to pinpoint what the settings do, I'd avoid "auto". I like the gold tung, they don't split on AC like the reds or ball up completely like the pure green.You really don't want to be as slow on Alu as you can be with steel. Crank up those amps, just for the tacks and starts. If you can't form a puddle in about 2 seconds you don't have enough.Last edited by Brazin; 04-08-2013 at 09:53 PM.SqWave 200Millermatic 190Airco 200 ACHypertherm PM45Boice-Crane Band SawVictor O/A