Weldment design....need help

admin

I need to make a bearing bracket for a milling machine.  It's basically a 6" x 7" x 1"thk A36 flat plate with a hole in the middle.  A 3" od x 1.5" id x 3" long A36 cylinder is inserted into the hole and welded (both sides) to the plate.Then, the weldment will be machined (faced/bored/c'bored) to size.Plan is to 45 deg bevel both sides of the hole in the plate as shown in the pics for full pen welds.Then, tack both sides 4 places followed by "root" pass both sides then stringer beads to finish. Equipment/consumables: MillerMatic 211, 75/25 ar/co2 mix, .035 ER70S6 wire. (I also have a Miller 200DX but I'm thinking the 211 would be the weapon of choice here...maybe...maybe not?)Me: Novice welder (as you probably guessed)Question 1 - How would you design this weldment?Question 2 - The depth of the bevel is about 1/2" and the gap at the top of the bevel is about 1/2" so, figuring a gun angle of 45 deg I'm going to have say 3/4" stick-out.  Is this too much?Question 3 - Is the MillerMatic 211 up to the task amp-wise or if not can pre-heat help.I'd appreciate any advice you guys might have.1st pic...cylinder inserted into hole in plate2nd pic...view from other side3rd pic...cross-section of assembly4th pic...one side welded5th pic...other side welded Attached ImagesMiller 211 w/ spool gunMiller Dynasty 200DXLongevity 60i IGBT plasmaO/A w/ crappy chinese torch/gaugesSouth Bend 10K latheGrizzly 4029 10x54 millGrizzly 7x12 hor bandsawangle grnders, bench grnder, bench belt sndr7.5 hp 80gal cmprsor
Reply:The problem I see is you keeping all of this square as you go to weld it the way you have this laid out.My 1st thought would be to machine a step in the plate and cylinder so the two pieces automatically want to square up, then use fillet welds. If you wanted more penetration, I'd weld a filet on the short side to lock everything in place, then toss it in the lathe and machine the bevel on the opposite side and then fill the bevel in. That way the original material locks the cylinder as square as possible. If you need the full diameter on the "small side", over weld the area and then machine down the extra material.Not sure way you need full pen welds on something like this. I'm betting you could probably set up an interference fit and  heat shrink the two pieces, similar to what Zap often does with roller flanges, and end up with hardly needing any weld at all. Attached Images.No government ever voluntarily reduces itself in size. Government programs, once launched, never disappear. Actually, a government bureau is the nearest thing to eternal life we'll ever see on this earth! Ronald Reagan
Reply:I would call what DSW posted a double step.  I personally like a single step.  Then press together and a small bead on each side to make sure it stays there.My name's not Jim....
Reply:I agree, machine a step in it and forget the full-pen welds.  What kind of forces is this piece going to see?Have we all gone mad?
Reply:I like everyone else's suggestions for altering the joint to be a self aligning fitup with a simple fillet all the way around on one or both sides.Once you get the components machined and fit together for welding, you'll definitely want to pre-heat both pieces.  You've got a huge heat sink you're working on.  If you can, pop the whole assembly into an oven and soak it for an hour at 350°F. This will give you thorough, even pre-heat, and time to get back to the shop and setup for welding, Soaking the whole piece will give gradual, even cooling that will minimize distortion from welding and residual stress in the weldments.While it's soaking, setup your Dynasty 200DX with either stick(E7018) or TIG.  I'd probably stick weld this for speed and conienence, but TIG would work also.I don't recommend your Millermatic211 and 0.035" solid wire, even with pre-heat.  If you had a roll of 0.045" flux-cored wire(E71T-1), you might try it, but you'll be maxing the machine out.  With solid wire, that machine is only capable of short circuit transfer.  Miller rates that machine as capable of single pass welds on 3/8" plate, and you're way, way thicker than this.  I think all the equipment manufacturer's recomendations are generous in this regard.  I've got a 211, and I don't use it on anything thicker than 1/4" material.  If you had a MIG machine capable of spray transfer, say 26V and 500IPM with 0.035" solid wire and 90/10 gas, you might be able to MIG weld this and make a sound weld with good fusion into the base metal everywhere.  But to garantee a sound weld with solid wire, you'll looking at 0.045" or thicker wire and spray transfer at ~400 amps.Last edited by A_DAB_will_do; 02-24-2012 at 08:31 AM.Benson's Mobile Welding - Dayton, OH metro area - AWS Certified Welding Inspector
Reply:Originally Posted by BoostinjdmI would call what DSW posted a double step.  I personally like a single step.  Then press together and a small bead on each side to make sure it stays there.
Reply:Guys,Thank you all for your advice.I'm redesigning this as follows..I'll machine a 45 bevel on one end of the cylinder then center the cylinder over plate.  Both the cylinder and the plate will have a 3/4" dia hole.  I'll use a 3/4 bar to align the two pieces ....pic 1Then weld out...then knock out the pin....pic 2 and 3.Then face both sides of the plate and bore / c'bore the pocket for the bearings....pic 4So it's not really a full pen weld until it's machined.  What I want is a smooth, continuous bore.I'm not too worried about maintaining perfect alignment or warping as both the cylinder and the plate are oversized to allow imperfections.  Facing/boring should bring it all back to square.  BTW lotechman....you're absolutely correct.  A bearing bracket on a small mill better not see high impact loads or I'll have far worse problems than weld integrity BUT I'm always guilty of over-designing things and also the bearings will have 400 lbs of pre-load so I don't want to take any chances. Attached Imagesbracket weldment1.pdf (56.0 KB, 53 views)bracket weldment2.pdf (75.9 KB, 50 views)bracket weldment3.pdf (9.0 KB, 48 views)bracket weldment5.pdf (113.9 KB, 36 views)Miller 211 w/ spool gunMiller Dynasty 200DXLongevity 60i IGBT plasmaO/A w/ crappy chinese torch/gaugesSouth Bend 10K latheGrizzly 4029 10x54 millGrizzly 7x12 hor bandsawangle grnders, bench grnder, bench belt sndr7.5 hp 80gal cmprsor
Reply:Ok last thing.  I will continue to suggest that overwelding and overdesigning will not be wise.   When you weld onto a thick plate like that you can create laminar (sp) tearing.  Imagine you "glued" that boss to a one inch thick piece of plywood.  When the joint fails it will tear out the laminations of the plywood rather than failing at the glued joint.  The added shrinkage stresses from a large fillet will create shrinkage cracks in the grain of the plate below.  Now I am going off on wild speculation and for your application it will hold... but if it was something that was on a heavy piece of equipment it would fail as I described.   I have had to do repairs on something like that and the only thing that worked was to actually dig down into the plate with a arc air gouge and replace some of the plate metal with weld so that the forces reached deeper into the laminations of the plate.  Same fillet size on the surface as original.
Reply:Originally Posted by BTDGuys,Thank you all for your advice.I'm redesigning this as follows..I'll machine a 45 bevel on one end of the cylinder then center the cylinder over plate.  Both the cylinder and the plate will have a 3/4" dia hole.  I'll use a 3/4 bar to align the two pieces ....pic 1Then weld out...then knock out the pin....pic 2 and 3.Then face both sides of the plate and bore / c'bore the pocket for the bearings....pic 4So it's not really a full pen weld until it's machined.  What I want is a smooth, continuous bore.I'm not too worried about maintaining perfect alignment or warping as both the cylinder and the plate are oversized to allow imperfections.  Facing/boring should bring it all back to square.  BTW lotechman....you're absolutely correct.  A bearing bracket on a small mill better not see high impact loads or I'll have far worse problems than weld integrity BUT I'm always guilty of over-designing things and also the bearings will have 400 lbs of pre-load so I don't want to take any chances.
Reply:Sorry to be off topic, but what program did you use to draw hat part up?TA Arcmaster 300CM3XMT 304S22P12 suitcase feederX-Treme 12VSOptima pulserTA161SMaxstar 150STLHypertherm PM45OP setupStihl 020AVP, 039, 066 Magnum