Butt Joints For Hull Plating

Discussion in 'Metal Boat Building' started by bobk, Jul 16, 2010.

  1. bobk
    Joined: Apr 2009
    Posts: 8
    Likes: 1, Points: 3, Legacy Rep: 22
    Location: Big Bend Wisconsin

    bobk Junior Member

    Hi, I'm getting close to tack welding on the hull plating for a 28' aluminum boat. I'd be curious to see what other people think is the best method to butt joint two pieces of hull plating. The material is 3/16" and 1/4" thick 5086 aluminum. The completed sections would be 28' feet long. Currently, I've been bolting a stainless steel bar to the backside of the the butt joint to control distortion during welding. After the first side side is welded I grind the weld flat then install the plate on that side. The other side is cut about 1/2 way through the butt weld seam with a skill saw then welded. This techinque seems to be producing very good joins with a minimum of distortion. Under ideal conditions would it be better to do the butt welds with the hull plating on or off of the boat? I'm using a Millermatic 350P with an AlumaPro gun. Also, I'm using 3/64" 5356 wire, is that what you would use? Thanks for any good advice.

    Bob
     
    1 person likes this.
  2. Ad Hoc
    Joined: Oct 2008
    Posts: 7,774
    Likes: 1,679, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Im sorry, but I don’t fully understand, bit confused.

    Is the framing already up, and you are offering up plates, then tack weld the plate to the hull, then butt them?

    Or are you doing it another way?...pictures and/or more words would help.
     
  3. bobk
    Joined: Apr 2009
    Posts: 8
    Likes: 1, Points: 3, Legacy Rep: 22
    Location: Big Bend Wisconsin

    bobk Junior Member

    Framing Already Up

    Hi, yes the framing is already up and fair. I simply wanted to know what the group thought was the best way to butt join hull plating under ideal conditions. For me, it would probably be easier to join the plates on saw horses off of the boat. What do you think?

    Bob
     
  4. Ad Hoc
    Joined: Oct 2008
    Posts: 7,774
    Likes: 1,679, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Well, as for the 3/8" to 1/4", always use a 1:4 taper to smooth out the transition.

    I would never recommend restraining the joint. It must be free to take up a natural shape.

    Aluminium has a high thermal conductivity (twice that of steel), as such it moves when welded, a lot! You need to understand the implications of restraining aluminium when welding, especially the plating. This can cause solidification cracking.

    The reason i suspect you are doing it the way you described is one or more of the following:-
    1) the plates are not cut to shape
    2) the plates are not rolled to shape
    3) the plate have too much 'gap' for their thickness
    4) incorrect weld sequence
    5) too much heat

    Not sure what you mean by 'saw horses'.
     
  5. mark775

    mark775 Guest

    He meant "sawhorses".
     
  6. tazmann
    Joined: Aug 2005
    Posts: 329
    Likes: 17, Points: 18, Legacy Rep: 215
    Location: California

    tazmann Senior Member

    welding sheets together

    Hello
    to give you some ideas here is a pic of a clamping fixture to weld together .100" sheets. Took a few differnt trys at the proccedures to finaly get good results. It did come out ok but it's best not to have the Buts.
    Tom
     

    Attached Files:

  7. alangluyas
    Joined: Aug 2008
    Posts: 14
    Likes: 1, Points: 3, Legacy Rep: 28
    Location: Kettering Tasmania

    alangluyas Junior Member

    Are the plates precut? NC plasma cut or otherwise already cut out?

    If not, how do you intend to template the plate shapes?

    If the plates are pre-calculated and you intend draping the finished plates on the hull frames, then I would be tempted to weld them on the floor (or on the sawhorses), welding downhand.

    If you are going to have to template the plate shapes to the hull frames (the conventional way) then I would probably weld them on the boat.

    I normally tack short (500mm or so) strongbacks across the weld at 90 degrees prior to tacking and welding.

    I also always work on a weld procedure on scrap plate of the thickness to be welded and run a few test weld that I section and break. I would not normally acid etch them to see the weld structure but I do normally cut a 25mm test piece, clean up the edges, notch it and break it to see where it fails. The biggest thing I learned at welding school was that welds in alloy are not always what they seem. I write the procedure up in a note book with all the machine settings for that particular condition for future use. You should only need to do this once for any one welding machine in any particular weld situation.

    Cheers

    Alan
     
  8. bobk
    Joined: Apr 2009
    Posts: 8
    Likes: 1, Points: 3, Legacy Rep: 22
    Location: Big Bend Wisconsin

    bobk Junior Member

    The Plates Are Waterjet Cut

    Hi Alan, the plates are waterjet cut and all ready to go. I thought it would be better to assemble them (butt joints) off of the boat as all the welding can be done in a flat position and the plates could be kept from waping easier.

    Bob
     
  9. bobk
    Joined: Apr 2009
    Posts: 8
    Likes: 1, Points: 3, Legacy Rep: 22
    Location: Big Bend Wisconsin

    bobk Junior Member

    Nice Welding Fixture

    Tom, thanks for sending the picture of the fixture you use for welding the shell plating. I can make on like that and try it out on some test pieces.

    Bob
     
  10. baeckmo
    Joined: Jun 2009
    Posts: 1,654
    Likes: 670, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    We use a similar setup, with a groove (2 x 1 mm) along the weld line. In order to protect the back of the weld, argon gas is flushed at low pressure into the groove during one-pass welds.

    Perhaps a little off-topic, but try to get the plating "texture" (their "inherent" bending from the rolling process) the same way in all plates in one panel. Makes it easier to get the hull fair with thin plating.
     
  11. tazmann
    Joined: Aug 2005
    Posts: 329
    Likes: 17, Points: 18, Legacy Rep: 215
    Location: California

    tazmann Senior Member

    Hello Bob
    Your welcome, on the heavier plate you have a few options. Some folks just weld bars accross the seam at 6 to 10 inch intervals and cut big enough notches in the bars to give access for welding. In my case I didnt want to weld anything temporary because the sheets were so light it would leave ridges from the welds. I haven't welded on the 5086 yet, I have been using 5052 and 6061 and I use the 5356 wire because of it's strength. Acording to my lincoln chart 5356 is the one to use for 5086.
    Here is a link on 1/4" but joints that might help
    http://metalboatbuilding.org/phpBB3/viewtopic.php?f=34&t=418
    tom
     
  12. rugludallur
    Joined: Jan 2006
    Posts: 81
    Likes: 17, Points: 0, Legacy Rep: 249
    Location: Iceland

    rugludallur Rugludallur

    I just wanted to correct some errors stated earlier in this thread and offer an insight on distortion in welding.

    There is no such thing as natural shape with metals, their "natural shape" is as oxide minerals, some metals do however have a "memory effect" where they tend to revert back to their previous form when cut, welded or heated, this is due to internal distortion stesses "locked in" during fabrication.

    Thermal conductivity has nothing to do with the coefficient of thermal expansion which is 0.000023 per degree centigrade for aluminum but about 0.000012 for steel, this being said it does mean that aluminum has twice the tendancy of steel to distort during welding and since heat is transfered faster (aluminum is much better at conducting heat, see thermal conductivity) away from the weld area a larger area will be heated.

    This might not sound like much but for a 10m section it's 10000mm * 0.000023 = 0.23mm per degree centigrade. Another way to look at it is if you heat a 16mm area up to 500°C (just under melting point) it will expand by 10*0.000012*500 = 0.96 or 0.1mm

    For aluminum welding we will always need to heat an area up to 660°C, the only thing we can control is the size of the area affected, this translates to moving as fast as possible, even though the welder will be on a higher setting the total amount of energy/heat is reduced.

    I recommend using strong-backs (flat bars tacked across the seam at a 90° angle) for any unsupported section longer than 250mm, you don't need to use continous welds, just tacking them is fine.

    Make sure you are welding using backstep technique, try not to weld more than 50-100mm long welds depending on plate thickness and allow the area to cool to room temperature before putting down the next weld.

    Jarl
    http://dallur.com
     
  13. Ad Hoc
    Joined: Oct 2008
    Posts: 7,774
    Likes: 1,679, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    I think you have completely misunderstood the terminology ‘natural shape’.

    To take up a “natural” shape is referring to what occurs during/after welding. When aluminium is welded, it shall move/displace/distort, the amount is unpredictable. You can have exactly 10 joints and weld each one with the same welder in the same sequence yet each one will have a slightly different “end result” in respect to its final dimensions.

    The thermal conductivity is directly related to how the metal behaves ie transmits heat. The degree of behaviour is measured by its rate of thermal expansion. If the thermal conductivity was low, the heat would dissipate quickly, ie a high thermal gradient over a short distance. The thermal conductivity of aluminium is roughly 4 times that of steel; steel is roughly 60 Wm-1K-1, whereas ally is 240. If the heat is conducted over a large surface area, as it is with aluminium, what happens to this now heated up metal, it expands over the same amount of surface area, not just where it is welded. The transfer of heat is seen as strain/displacement especially when restrained. This can easily be seen by making a run of weld along the centre of one side of ally plate that is unrestrained. The shape once welded is not the same as the original. The thinner the metal the greater the effect.

    Thus, one can either restrain the joint, to prevent movement (or expansion/strain), or weld the joint unrestrained. Welding unrestrained is termed as “taking up its natural shape”. In other words, when the joint is welded, with no restraint, the final shape after cooling is what it is…ie a natural shape, of its own selection. But more importantly it also means that there is no residual stress.

    If you weld restrained, you’re asking for trouble, starting with solidification cracking as well as building up residual stresses.

    A welder will never be able to move as fast as the heat during welding. If you weld too fast there is a very high possibility of lack of fusion. The weld pool must fuse both sides of the joint into the weld pool. Increasing the speed of welding reduced the chances of this occurring.

    A back-step technique is ostensibly a steel welding procedure. This is done by steel welders who have retrained in aluminium and who do not appreciate the differences between welding steel and aluminium. Aluminium requires more attention to the correct welding procedure as a whole, not piecemeal. It is all about ensuring a balanced amount of heat input into the structure to prevent distortion beit linear or angular and shrinkage/expansion. Too many short runs will lead to voids/cracks left in the stop/start region if not properly addressed. Not to mention that stop/starts are serious fatigue raisers in aluminium.

    A strong back would be used to hold a large free edge when another section is welded. This is because of the high thermal conductivity of aluminium, as already noted. Without a temporary strong back, the free edge will take up its own ‘natural shape’ which may not align properly with the plate it is being butted into, thus making the butt much harder. This is one reason why Class/IACS etc require butts to be close to the main supporting structure it is being welded too, so that the free edge is "stiffened" by the proximity of main structure…ie the main frame is the strong back.

    5356 gives a better quality joint, but 5183 has greater strength, by approx 10%. The gas makes a difference too.
     
  14. rugludallur
    Joined: Jan 2006
    Posts: 81
    Likes: 17, Points: 0, Legacy Rep: 249
    Location: Iceland

    rugludallur Rugludallur


  15. Ad Hoc
    Joined: Oct 2008
    Posts: 7,774
    Likes: 1,679, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Jarl

    It is your prerogative to do the procedure as you suggested. It is not the procedure I recommend. Everyone has their own opinion.

    But we do at least agree about aluminium’s conductivity:

    “…However, when welding aluminum, compared to carbon steels, the effects of some of the main contributing factors for distortion are increased. Aluminum has high thermal conductivity, a property that substantially affects weldability…..There is a direct relationship between the amount of temperature change and the change in dimension of a material when heated…”
     
Loading...
Similar Threads
  1. jprevj
    Replies:
    4
    Views:
    2,820
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.