Announcement

Collapse
No announcement yet.

Using water submersion to keep work cool/straight.

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

  • Using water submersion to keep work cool/straight.

    Not sure what brought the thought on - perhaps it was the youtube'r who was failing at keeping work from warping on long welds in thick materials.

    On an automotive forum I saw a guy who installed a roots supercharger on his BMW inline 6 shooter. To get the crank hub/pulley combo to work, he simply parted off the crank hub from the donor car/supercharger (wanna say it was an Eaton style from a Bonneville SSC). He them dropped the BMW crank hub in a 5 gal bucket of water, only exposing the small area to be welded before rotating it to weld more (having trouble finding pics). He did this because of the rubber ring in a crank hub, as to not burn it though. The crank hub outer ring has to stay in proper alignment since it doubles as a reluctor wheel for the crank sensor.

    Anyways, was watching a YT video to get ideas for the belt grinder I want to build. Something similar to this (what got me thinking):

    https://www.youtube.com/watch?v=a_RlL1O-bK4 Skip to 3:20

    As you can see, he is using solid 1.5" square stock and making a thick wall box for it to slide in - of course he welded it up and it warped. For those who don't watch the video, he then makes another box and stitches the welds to keep it straight.

    Has anyone tried submersing work to keep warpage down instead? I know it can be stitched, but it seems like a similar job would go much quicker if the entire length could be welded in one shot. In my mind, if only the corner that was being welded could be sticking up out of the water to localize the heat.

    I have submerged little things like A/C lines that still had rubber sections crimped on when swapping late model engines into old cars - now curious if that would help any warping issues.

    Hopefully get some insight from you folks.

    Thanks
    Last edited by Forced_Firebird; 09-05-2019, 08:51 AM.
    J.Caraher
    Wide Open Throttle Technologies (WOT-Tech), Pompano Beach FL
    Miller Sync 300,Hobart 190
    RogueFab pneumatic, Hossfeld Manual
    Kitamura CNC, Bridgeport 2j
    TunerPRO, HPTuners, AEM, Megasquirt, DynoJet
    NASA Racing Official/Driver

    YouTube Link, Instagram Link, FaceBook Link

  • #2
    You certainly have my interest, but I never thought about it. Hope someone can shed some more light on this.

    Comment


    • #3
      I've done this kind of thing a few times - Once, had to weld near a spring and couldn't get the spring hot so I setup a little cup to hold the spring compressed, dropped the ring on the shaft, filled it up with water to submerge the spring, but not the part I was welding then melted it together. Kind of weird watching the steam and water boiling as you're welding and the shielding gas blowing it away. In the picture where it's still glowing, you can see the water still boiling. This was a proof of concept prototype - NOT a production process . Probably did 30-40 of these that went through millions of durability cycles without failure.

      Click image for larger version

Name:	retaining-ring-welding-1-spring-free.jpg?resize=850%2C850.jpg
Views:	207
Size:	58.4 KB
ID:	601668

      Click image for larger version

Name:	retaining-ring-welding-2-spring-compressed.jpg?resize=850%2C850.jpg
Views:	207
Size:	52.8 KB
ID:	601669

      Click image for larger version

Name:	retaining-ring-welding-3-ring-installed.jpg?resize=850%2C850.jpg
Views:	207
Size:	53.8 KB
ID:	601670

      Click image for larger version

Name:	retaining-ring-welding-4-welded-still-red.jpg?resize=850%2C850.jpg
Views:	209
Size:	48.0 KB
ID:	601671

      Click image for larger version

Name:	retaining-ring-welding-5-welded-cooled.jpg?resize=850%2C850.jpg
Views:	212
Size:	47.3 KB
ID:	601672

      Comment


      • #4
        Sounds like it could work.

        Good idea.
        if there's a welder, there's a way

        Comment


        • #5
          With heat comes expansion, contraction comes with cooling. Limit the heat, limit the expansion, that limits contraction. While immersion seems like a workable solution in some respect and to some degree, it's not a very practical solution when other methods of control are better suited in my opinion.
          Quite frankly, the tube didn't need to be welded as completely as it was and I'm sure would have held up and functioned just fine for it's intended purpose with less weld being deposited.
          I'm not sure why, but a guy sees an open seam and thinks it needs filling just because it's there to be welded, failing to account for the strength of the deposited weld metal or design, purpose and function of the weldment.

          Comment


          • #6
            Originally posted by Noel View Post
            With heat comes expansion, contraction comes with cooling. Limit the heat, limit the expansion, that limits contraction. While immersion seems like a workable solution in some respect and to some degree, it's not a very practical solution when other methods of control are better suited in my opinion.
            Quite frankly, the tube didn't need to be welded as completely as it was and I'm sure would have held up and functioned just fine for it's intended purpose with less weld being deposited.
            I'm not sure why, but a guy sees an open seam and thinks it needs filling just because it's there to be welded, failing to account for the strength of the deposited weld metal or design, purpose and function of the weldment.
            You are correct. Especially on someone seeing an open seam and thinking it needs to be welded.
            Before I retired, there were a few occasions when I had to go back to a completed job and "weld up spots that I forgot to weld"!!
            My boss and I new better, but he said it would be quicker than arguing with the know-it-all that discovered the "missing welds"!
            They were usually wearing a necktie.

            Comment


            • #7
              I like the spring trick--great idea.

              Comment


              • #8
                I agree on the strength part of that particular project. Watching the video and thought to myself "just stick it under water". We have to have all oue welds fully closed in racing cages - irregardless of them being strong enough stitched. Can't tell you how many times race cars fail inspections for incomplete welds because they "couldn't get the welder in the corner" - then hole-sawing/patching exterior sheet metal to pass. All my cages are 360° welded, and never had to ruing the exterior sheet metal - just have to plan the order of operation, and use common sense.

                Let's say it had to be air/water tight with a similar weld for the purposes of this discussion.

                Might just try it to see when building the belt grinder for posterity.
                J.Caraher
                Wide Open Throttle Technologies (WOT-Tech), Pompano Beach FL
                Miller Sync 300,Hobart 190
                RogueFab pneumatic, Hossfeld Manual
                Kitamura CNC, Bridgeport 2j
                TunerPRO, HPTuners, AEM, Megasquirt, DynoJet
                NASA Racing Official/Driver

                YouTube Link, Instagram Link, FaceBook Link

                Comment


                • #9
                  I would think that, at some point, no matter how much water you surround something with you’ll have have a degree of warpage. It would be an interesting experiment though, building the fixture, making the welds and then comparison. If you can, make an identical test without the water, then you’ll have good data.

                  I’ve only done something very similar to what you’ve done with the spring issue.

                  Comment


                  • #10
                    Well, it works on the computerscreen.

                    I recall right ESAB University has 411 on this and underwater cutting 411 abounds.

                    It's a nice sounding thought,, but I strongly suggest looking at the outer wall of a Diesel engine sleeve to learn about wet + heat.
                    Conducting heat ain't as easy as it seems.

                    Comment


                    • #11
                      I think if the guy can afford 3/8" mild steel flat bar, a couple pieces of 3/8" aluminum flat bar as chill strips clamped to it would be a better and more chilling solution. Just saying... the water bath, the water will boils, turn to steam, the steam acting as an insulator preventing heat from escaping. That's why when you quench in water you shake like Chubby Checker's twist. To cool quickly.

                      Comment


                      • #12
                        Originally posted by Noel View Post
                        I think if the guy can afford 3/8" mild steel flat bar, a couple pieces of 3/8" aluminum flat bar as chill strips clamped to it would be a better and more chilling solution. Just saying... the water bath, the water will boils, turn to steam, the steam acting as an insulator preventing heat from escaping. That's why when you quench in water you shake like Chubby Checker's twist. To cool quickly.
                        Even with the "insulating" effect of steam, it's still a much more effective heat transfer medium than conduction via an aluminum bar. As that steam rises, it's replaced by water which absorbs orders of magnitude more energy as it changes state to steam.

                        Aluminum = 0.90 kJ/(kg C)
                        Liquid Water = 4.20 kJ/(kg C)
                        Changing liquid water into steam at atmospheric pressure = 2257 kJ/(kg C)

                        Just saying...it's really hard to beat a water bath for effectiveness of cooling something.

                        Comment


                        • #13
                          An ounce of caution beats the snot out of a month in the burn ward.
                          A drop of water hitting a molten CadWeld will erupt the pot in a manner you don't want to see.

                          As to quench tubs, do't forget the horse additive.

                          Comment


                          • #14
                            Originally posted by Fix Until Broke View Post

                            Even with the "insulating" effect of steam, it's still a much more effective heat transfer medium than conduction via an aluminum bar. As that steam rises, it's replaced by water which absorbs orders of magnitude more energy as it changes state to steam.

                            Aluminum = 0.90 kJ/(kg C)
                            Liquid Water = 4.20 kJ/(kg C)
                            Changing liquid water into steam at atmospheric pressure = 2257 kJ/(kg C)

                            Just saying...it's really hard to beat a water bath for effectiveness of cooling something.
                            I think your over thinking this with a plan on using a pretty big tub of water? If you don't get any more water then my aluminum bars have for surface area, still think you'd do ok?
                            I think that I could replicate that in a science experiment.

                            Comment


                            • #15
                              Originally posted by Noel View Post

                              I think your over thinking this with a plan on using a pretty big tub of water? If you don't get any more water then my aluminum bars have for surface area, still think you'd do ok?
                              I think that I could replicate that in a science experiment.
                              I think there's a few things to clear up here - First is my issue of fixing things that aren't broken (hence my name...). I was too focused on fixing the fact that water is a better conductor than aluminum that I forgot about the topic at hand (minimizing heat induced distortion from welding). Sorry about that.

                              Second - Since the distortion occurs right at the weld bead, as the molten metal cools and shrinks, thereby pulling the adjacent material with it, I'm not sure that submerging the part in water right up to, but not including the weld bead (or using an aluminum chill(s) right next to the weld bead) would help to minimize distortion at all.

                              A small tack weld (relative to the size of the material being fused) that is properly clamped/fixtured will yield itself as it cools instead of the mating material. A big long continuous weld has a lot of strength to it and will yield the mating material as it cools.
                              There may be some minor effects of the base material expanding as it warms up a few hundred degrees and then cools back down, but this feels like a minor contributor compared to the molten weld puddle (and could be taken care of by preheating).

                              Here's a thought for an experiment to try if you're so inclined. All you welding experts out there can chime in with suggestions as well - this is just an idea - discussion and modification is encouraged!

                              - 4 pair of coupons, say 1" x 1/8" x 6" long, mild steel A36. Nothing particular about the size, just so they're all the same.
                              - Tack all them at the ends and once in the middle so they're at 90 degrees to each other - Check with angle gauge
                              - Weld one set in open air - full pass
                              - Weld one set in open air - 1" stitch welds
                              - Weld one set in a tray with water up to 1/8" from the weld
                              - Weld one set in open air with a pair of aluminum chills clamped to each side
                              - All welded with the same welding parameters

                              When it's all done, compare how much each one distorted with the same angle gauge

                              I've heard that if you weld with higher current/wire feed/travel speed that it reduces heat input and thereby minimizes distortion. Maybe a couple more sets of coupons where you weld at at various parameters (from a long slow weave with 0.023 wire to a single fast 0.045 wire hot pass, maybe even throw a TIG pass in the mix?)

                              Thoughts?

                              Comment

                              Working...
                              X