Announcement

Collapse
No announcement yet.

Welding advice needed.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Welding advice needed.

    The time to weld the diagonal braces onto the vertical segments of my CNC bridge-mill project is approaching & some advice from you experienced welder/fabricators would be greatly appreciated.

    It's really nothing fancy, just a 3" square tube (1/4" wall) welded to a 3" square tube (.290" wall). The welds will be made with the parts flat on a table and then the 2 assemblies aligned & bolted to the base. I'm hoping the bolt together plan will allow it be assembled without having to have them stress relieved.

    I have a 200 amp, AC-DC stick welder that has worked well for me so far but never anything this big. I have a few typical rod sizes & types on hand but am happy to pick up what is best for this job.

    Any suggestions as to amperage & rod sizes? Also, to minimize distortion, I won't be welding it solid all the way around. I'm thinking a series of short welds will be plenty strong enough with a cool-off period after each weld. How many, how long, where & what order should the welds be made to keep the warping to a minimum? Any other tips are welcomed.

    Milton

    "Accuracy is the sum total of your compensating mistakes."

    "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

  • #2
    Welding is an art so the best advice I could give you is get plenty of extra 3" 1/4" wall square tubing for practice welding. Keep practice welding on your spare tubing until you feel you're ready.

    Comment


    • #3
      Start by grinding a bevel on the piece that is cut. A 1/8" deep will do just fine. Start with a 1/8" 6013 stick at 90 amps. Weld a substantial tack at the upper end of the mate, then quickly flip it over and make the same tack at the lower end of the mate. Let cool and examine for alignment. If okay, repeat the tacks at the other ends of the previous tacks. Keep laying down these tacks in pairs, keeping the pairs at exact opposite of each other to keep distortion at check. Keep an eye on the straight piece, as a lot of heat can pull it into a curve.

      Comment


      • #4
        It's hard to say exactly. I always make a few practice runs on scrap before I proceed to the actual work.
        Some manuals like from Lincoln Electric may give you more information on where to start as far as rod selection and amperage, but in the end you have to fine tune.
        I always run a few test passes on scrap before I do anything.

        JL................

        Comment


        • #5
          Welding flat on the table, no problem. Try a 1/8" 6013 for a nice looking bead, or 6011, at around 140 amps, you are welding fairly thick steel. Adjust your speed so the weld forms nice and smooth. Too slow and it will build up and be convex. Too fast and you will leave gaps. You want it to be slightly concave or flat making a 45 deg angle between the two pieces. Hold the rod at about 45 degrees to everything. The thin edge of the diagonal piece may require less amps. To minimize distortion tack it all together with short say 1/4" long tacks then weld on opposite sides a little at a time. You probably want 3/4 to 1" long welds. Possibly you could weld only the 45 deg. cut sides too.
          Last edited by wdtom44; 04-18-2018, 09:36 AM.

          Comment


          • #6
            I would suggest that you bevel the cut piece all around, and grind all the rust and mill scale of mating piece. 6013 is a poor choice in my opinion, very minimal penetration. If you choose to go with short tacks, each one should be ground before adding the next. Not really a technique for 1/4" material and a stick welder, more for sheet metal/mig. I would use 1/8" 7018 100 amps or so, on DC. Leave a gap, maybe 1/8" or a little less between pieces, line up, clamp everything down. Make 2 tacks on wide angle, check for alignment both ways then tack 2 corners on narrow side. Clean up tacks, then weld an inch from the corner out both ways. That's how I would do it, but I'm not a pro welder so YMMV.
            Good luck.
            Be Safe.

            Comment


            • #7
              Post a picture of the whole piece to be welded and I will tell you where to start and in which direction to weld. Use some NEW 1/8 " 7018 rod o at least a newly opened box. Or bake the rod a couple of hours at 300 degrees. 90 amps maybe 95 will be just about right.
              Location: The Black Forest in Germany

              How to become a millionaire: Start out with 10 million and take up machining as a hobby!

              Comment


              • #8
                Originally posted by Captain K View Post

                I would suggest that you bevel the cut piece all around, and grind all the rust and mill scale of mating piece.

                6013 is a poor choice in my opinion, very minimal penetration.

                If you choose to go with short tacks, each one should be ground before adding the next. Not really a technique for 1/4" material and a stick welder, more for sheet metal/mig.

                I would use 1/8" 7018 100 amps or so, on DC. Leave a gap, maybe 1/8" or a little less between pieces
                .
                Captain K has it right.

                Slight bevel, you must grind it clean, 7018 is the proper rod, size is up to you but I would go with 1/8, you could go smaller if you are not confident it is easier to run, 100 amps DC approx. but every machine is different so it could be more or less to get the right heat. Trial and error. If you are welding on the flat then you can go hotter (125 maybe) for a second pass.

                Forget welding with short tacks, it can lead to inferior welds unless you are really good. A pass down each side, grind the starts and stops then weld across both ends. Dont cheap out and try and run the rods down to the stub. Sometimes the flux fails near the end and you get porosity. After you run weld down the first side you will need to reset the gap on the other, let it cool first if you want.

                Leaving a small gap 1/16" will help keep deformation (the contraction has somewhere to go so it help minimize the bending of the tubing) of the sections to a minimum but there will be some. The long piece that the brace is being welded to will pull towards the brace, nothing to do but try and minimize it. Post weld you can heat it red hot and then bend it back straight. Don't bend it cold. A welding engineer could calculate the amount of pre-bend, in the opposite direction, that should be used so that after welding the piece comes out straight but you don`t have that luxury. You could temp weld, at the ends of the tubing, a strong back (another piece of metal like an H beam that is less likely to bend then your tubing) on to the side opposite the brace with a piece of bar about one eight inch thick in between so that there is some pre-bend in the tube, this will help counteract the warping but is extra work and costs more.

                Weld it all around so it will be strong enough, 2 passes if necessary. Fix the warping later you cant stop it anyway.
                Last edited by loose nut; 04-18-2018, 11:10 AM.
                The shortest distance between two points is a circle of infinite diameter.

                Bluewater Model Engineering Society at https://sites.google.com/site/bluewatermes/

                Southwestern Ontario. Canada

                Comment


                • #9
                  I suggest 1/8" 7014 rod. Do all the welds in the horizontal position.
                  No matter what you do unless you grossly overweld it you won't get a lot of distortion. 3x3x1/4" tube is mighty stout stuff.
                  How long you weld vs how long you wait is not determined by metallurgy, it is determined by the way your buzzbox is built. Read the manual and it will tell you a duty cycle. This is one spec not to be exceeded! Many small welders have their current capacity determined by marketing people. They get away with it by saying of course you hardly ever get to weld because it always has to be cooling down. A Miller Thunderbolt AC/DC, for example (I've owned 2) has a 20% duty cycle. Weld for two minutes, sit around for eight minutes.

                  Anyway, if you want zero distortion you will need to prestress the joint the opposite way from where it will bend. Which is complicated. You can minimize distortion by only welding on the sides, not top or bottom.

                  And finally, it looks like you might have designed a piece with fully enclosed dead spaces. If your part will ever be outside, you should put in weep holes before you weld. You don't want to see what happens when water slowly leaks in and fills a space and then freezes.

                  metalmagpie

                  Comment


                  • #10
                    Originally posted by Captain K View Post
                    I would suggest that you bevel the cut piece all around, and grind all the rust and mill scale of mating piece. 6013 is a poor choice in my opinion, very minimal penetration. If you choose to go with short tacks, each one should be ground before adding the next. Not really a technique for 1/4" material and a stick welder, more for sheet metal/mig. I would use 1/8" 7018 100 amps or so, on DC. Leave a gap, maybe 1/8" or a little less between pieces, line up, clamp everything down. Make 2 tacks on wide angle, check for alignment both ways then tack 2 corners on narrow side. Clean up tacks, then weld an inch from the corner out both ways. That's how I would do it, but I'm not a pro welder so YMMV.
                    Good luck.
                    Be Safe.
                    My thoughts on rod selection exactly. I have never had good experiences with either 6011 or 6013 compared to 7018, with perhaps vertical and overhead being exceptions.

                    Dan L
                    Salem, Oregon

                    Comment


                    • #11
                      A lot of people think 6011, 6010 are a good general purpose rod. Not really. They are a fast freeze rod used mostly for gap welding on pipe joints to be followed by 7018 filler and cap passes. And they don't make nearly as pretty a weld as 7018!

                      Milton, you will get a lot of distortion on your geometry if you don't weld in a certain sequence. When welding a joint the joint first expands and then contracts to smaller than original. To test make two practice pieces with two 45's mated to make a 90 degree joint. Tack all four corners and then weld one along the joint from the inside to the outside corner. Do the other one from the outside point to the inside point. Do both sides and put a square on the two joints. You will see which way the pull. They will be off 90 by quite a bit. Try it and you will see. There are very specific steps to follow to keep things square.
                      Location: The Black Forest in Germany

                      How to become a millionaire: Start out with 10 million and take up machining as a hobby!

                      Comment


                      • #12
                        If you have two of those frames to weld I would clamp the two straight pieces back to back with a strong clamp. Then after tacking, weld one side of the one diagonal, flip and weld the diagonally opposing joint on the other diagonal and repeat until you are done. Let it cool completely before unclamping. This way the two parts pull against one another and minimize the bend that occurs.

                        We used to weld long 1” stainless pipe booms with inlets all on one side. We welded up one sample and determined it would bend 1/8” per foot from welding. So we prebent them that amount in the other direction before welding and they came out nice and straight. We tried straightening after welding and the new bend was always away from the fitting and it was rippled.

                        Comment


                        • #13
                          Best piece of advice I saw above was from loose nut - leave a gap between the pieces. It always "seems right" to butt things up tight, but what happens is, the first weld shrinks and pulls the whole thing out of square. A gap of 1/16" is fine; it'll give the weld more penetration. It means that once the first 2 tacks are in place, you can make sure the whole thing is square, get a 3rd tack in, recheck, then weld it up.

                          3" box, especially in that wall thickness is a doddle to weld. The hardest weld will be the one tight inside the vee - the rest will be easy.

                          Ian
                          All of the gear, no idea...

                          Comment


                          • #14
                            Wow, fantastic advice gentlemen; thanks! Many factors mentioned here I never thought about. I've only burned 7018 on a few occasions but really liked it. Hopefully my local welding store will sell me fresh stock in small quantities.

                            Definitely planning to clamp the pieces back to back but I'm still looking around for a way to get the diagonals held down flat in relation to the long pieces. The only really-flat & sturdy surface I have around is the ground flat top on the Brute table the machine is being built on. Should'a done this little welding job before the linear slides & ballscrew for the first axis were mounted & aligned on the table. (Dumb-a$$!)

                            Tweaking & shimming the assembles to compensate for the inevitable distortion won't be a problem though as long as the axis' slides end up in proper alignment to the other.

                            Is there a favored choice of electrode polarity for this?
                            Milton

                            "Accuracy is the sum total of your compensating mistakes."

                            "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

                            Comment


                            • #15
                              Originally posted by Black Forest View Post
                              A lot of people think 6011, 6010 are a good general purpose rod. Not really. They are a fast freeze rod used mostly for gap welding on pipe joints to be followed by 7018 filler and cap passes. And they don't make nearly as pretty a weld as 7018!
                              .
                              6010 and 6011 are primarily used for open root welding where fast freeze is desirable, as you said. They are considerably weaker then 7018 welds because of hydrogen enbrittlement IE: not low hydrogen rods. 6013S will give you a nicer looking weld but it is still not as strong. It is generally referred to as a farmers rod because they it is used for repairs around the farm with a simple A/C buzzbox. Most 7018 rods do not work well with most A/C machines but if a DC machine is available then 7018 is the rod of choice. There are other rod which will work. Personally I prefer 6010 on open roots like pipe joints because the flux on 6011 tends to "toenail", burn more on one side then the other. Lincoln 5P or 5P+ is an excellent 6010 rod and our first choice at work, back when I did that sort of thing.
                              The shortest distance between two points is a circle of infinite diameter.

                              Bluewater Model Engineering Society at https://sites.google.com/site/bluewatermes/

                              Southwestern Ontario. Canada

                              Comment

                              Working...
                              X