Plasma Welding - Filler Wire

admin

Good afternoon! Please forgive my newbie-ness, but hopefully someone can shed my light on some things. Let me preface that I'm not very well versed in welding.We use a Plasma welding process on 436MT stainless steel for a high-volume automotive part we manufacture. We've been doing it for about 10 years now. During that time, we've been plagued with leaking parts (the part we make is a pipe.) Our process is mostly automated, with a production operator simply loading and unloading the parts. We use a 309L filler wire as well. We have recently discovered that if we don't use any filler wire, we actually get far fewer leakers - almost a 50% reduction in fact. (Not to mention saving $$ on wire).The blueprints we have do not specifically say "use filler wire" or anything like that - it simply says "Plasma Welding". We've had a lot of turnover over the years (go Michigan economy!) and no one remembers why we decided to use the filler wire in the first place!My question for the experts here is, do you feel that there would be any detriment to the "quality" of the weld? We've performed pull tests which actually exceed that with weld wire, but we're not sure if there would be any affects on corrosion resistance or if the weld would get brittle or anything like that.Thanks in advance for everyone's thoughts!
Reply:Do you have any info on the basic nature of the leaks?  Are the leaks from cracks in the base metal HAZ, or porosity in the weld, or ?  This info would help alot is solving your leak problems.  I think you should know why these leak as you begin seaching for process related solutions.How do you leak test these assemblies?  I would locate some typical leaks via bubble testing, mark the location, and have metallographic cross sections done to show the leak path.Cracking would indicate a metallurgical issue, maybe related to your process,  which could include things like excessive grain growth, excessive dilution and martensite formation, and hydrogen embrittlement.  Porosity would indicate hydrocarbon contamination or shielding problems.  There should not be any detrimental metallurgical effect of autogeneously plasma welding 436, but the joint properties may be different since you've eliminated the weld reinforcement (extra metal) and ductile nature provided by the 309.Where I worked with auto components, process changes like eliminating the filler wire would require weld develpment, testing and qualification, and approval from the customer.
Reply:This may be far fetched, but I worked with an "automotive component" weld that was leak tested and suffered a high reject rate.  It was a plasma weld between low carbon steel and Inconel 625, and was autogenous.  The problem was liquation cracking in the Inconel 625 HAZ due to a Niobium (Columbium) phase within the Inconel.  By coincidence, your 436 SS has a high Niobium content as well.  Maybe Nb liquation cracking is not an issue with 436, but who knows.This link mentions Nb additions in 436.http://www.spiusa.com/Ref001/ferritic1.html
Reply:Originally Posted by pulserDo you have any info on the basic nature of the leaks?  Are the leaks from cracks in the base metal HAZ, or porosity in the weld, or ?  This info would help alot is solving your leak problems.  I think you should know why these leak as you begin seaching for process related solutions.
Reply:Pin holes, probably easier to solve than cracks.Classically, pin holes are caused by hydrocarbon contamination or water.There are probably any number of contaminants, besides oils and water, that cause gas to be evolved into the weld pool and result in pin holes.First I would look at the plasma welding torch.  I assume you are using the very common Thermal Arc 3A or 4A torches.  These torches can easily leak water internally at a couple of O-ring seals.  Depending on the torch model, installing the O-rings can be tricky and requires a special tool.  You can get a clue as to whether this is an issue by looking at the tungsten electrode and copper tip of the plasma torch.  If the torch is not leaking water or getting contaminated shielding, the tungsten and tip will be clean, shiny, with no discoloration.  See this link for plasma torch exploded view and parts.http://198.170.235.20/catalog/web_st....html&cart_id=At the same time I would go through the welding process with a fine toothed comb, looking for causes of shielding gas contamination.  The system should be leak checked.  It may be counter intuitive, but leak in a pressurized gas line will allow air to enter.  Pressurize up to the end of the torch and bubble check absolutely every fitting/valve/connector and flow meter.  Check the assembly of the torch, are any "O-rings" worn out or missing, up front or on the back cap?There are different orifice sizes available for these torches, and the tunsten "set-back" distance can be adjusted.  Check these out.  Generally the tungsten would be run flush with the end of the copper orifice tip.  These things could affect gas flow/turbulence.Check gas flow rates, plasma and shield.  Too high can cause turbulence and introduce air into the arc.  Too low obviously is also bad.  You may be able to tell something about shielding problems by looking at the tungsten to see if it's discolored, and by looking at the production weld.Check for air leaks on nearby equiment/mechanisms, or other disturbances that could affect your shielding gas.This may sound far fetched, but what about your plasma torch electrodes, are they precision ground/machined or irregular ground by hand?  This could make a difference in turbulence in the plasma.Also ensure that there is nothing at all (contamination) on the parts.  Has the vendor changed to a new cutting fluid, or has the cleaning process changed, or does the operator eat pepperoni pizza prior to loading parts in the welder (you get the drift)?  Also, check the filler wire and the wire delivery system for contamination.  Is the spool covered/protected?  Maybe consider cleaning some test wire (and parts) with acetone/alcohol.In my past auto-component job, we fought a leaker problem that ended up being related to residual moisture left in the parts feeder when the machine operator would clean his machines in the morning.The metallurgist says "we appear to be applying too much heat"?  Ask him why he says that, what exactly are the indicators of "too much heat".  Is the weld discolored/oxidized?  If you are getting the required penetration, and it is not excessive, then the "heat is about right".
Reply:Great answers Pulser, keep up the good work. I'll reiterate - check for air/gas leaks in equipment, make sure material or weldment isn't contaminated, clean material! If you're getting minute pin holes, it sounds like an incomplete weld...maybe the automated process wasn't such a good idea if you're getting that high of a failure rate? Now I'm curious which automotive part! John -  fabricator extraordinaire, car nut!-  bleeding Miller blue! http://www.weldfabzone.com
Reply:Originally Posted by MicroZoneGreat answers Pulser, keep up the good work. I'll reiterate - check for air/gas leaks in equipment, make sure material or weldment isn't contaminated, clean material! If you're getting minute pin holes, it sounds like an incomplete weld...maybe the automated process wasn't such a good idea if you're getting that high of a failure rate? Now I'm curious which automotive part!
Reply:Welcome! Yep, that's why I was curious. John -  fabricator extraordinaire, car nut!-  bleeding Miller blue! http://www.weldfabzone.com
Reply:How about argon purging the inside of the tube during the weld?