Aluminum boat welding

Discussion in 'Metal Boat Building' started by adriano, Dec 30, 2021.

  1. adriano
    Joined: Aug 2009
    Posts: 107
    Likes: 4, Points: 18, Legacy Rep: 47
    Location: italy

    adriano Senior Member

    Hi guys,
    any expert on MIG aluminum welding among you?
    We have to MIG weld aluminum grade (5083H116) 4 and 6 mm hull plates by butt joint.
    Now we have to pass butt joint Mig welding test, tried several times but fails!
    test result:
    Radiography test – which passes without porosity., sometimes small porosity shows.
    Bending test – no crakes observed – result satisfactory
    Tensile test – min required 305 Mpa – Result obtained 280 to 285 Mpa
    Failure at welded Area.

    We use 99,9% Argon gas, 1,2 mm ER5183 filler wire, we use Aceton cleaner before
    60 degree beveling sanding (30 degree on both sides)and welding "upwards".
    Key factors, believe, are suitable heat and speed, gun inclination push or drag?
    Is anybody able to give us any guide line ?
    Will appreciate any help
    Thanks
    Adriano
     
  2. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    when you do your test runs, what voltage, amps and wire speed do you show?
     
  3. Barry
    Joined: Mar 2002
    Posts: 1,851
    Likes: 505, Points: 113, Legacy Rep: 158

    Barry Senior Member

    Send some pictures of finished beads as well as a few pictures of the failed weld area from the tensile test.
     
  4. Scuff
    Joined: Nov 2016
    Posts: 285
    Likes: 44, Points: 28, Legacy Rep: 10
    Location: Richmond VA

    Scuff Senior Member

  5. Barry
    Joined: Mar 2002
    Posts: 1,851
    Likes: 505, Points: 113, Legacy Rep: 158

    Barry Senior Member

    What is the best filler alloy for 5083 base alloy? https://www.esabna.com/us/en/education/blog/what-is-the-best-filler-alloy-for-5083-base-alloy.cfm
    Some notes re filler rod above


    Some notes re 5183 filler below
    http://www.alcotec.com/us/en/support/upload/a5183tds.pdf

    Note that the UTS of the "as welded joint" for 5083-0 is 40ksi or 275 mpa which limits you are achieving. Why do you expect more than this?
    Normally, and a discussion on the forum several times, is that you normally do not expect the weld bead strength to be the same strength as the hardened or heat treated
    parent material due to the effects of the HAZ edit Unless it is 5xxx-0
     
    Last edited: Dec 30, 2021
  6. baeckmo
    Joined: Jun 2009
    Posts: 1,654
    Likes: 670, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    I think Barry "nailed it" here; but I'd like to comment on the cleaning and beveling procedure. We have found denaturated alcohol better than acetone for getting rid of oil residues on the plates, and we avoid sanding wherever possible. Edge preparation always by cutting methods; a router or in "rough" places a carbide tipped disc. And finally, use rotating SS wire brush along the edges just before welding.
     
  7. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Indeed.
    As noted by LR, for example:
    upload_2021-12-31_9-42-45.png

    To be clear, your joint prep is which of the following, for your 4 to 6mm butt:
    upload_2021-12-31_9-43-40.png

    Other factors...
    How long between grinding off the oxide layer to the actual weld, if too much time has passed, this may be a factor.
    Max limit is 4 hours between grinding and welding.

    As Baeckmo notes, correct grinding and s/s wire brushing eliminates most issues from the outset.

    You could also try using 5356 filler wire. We use this for 5xxx to 5 xxx series alloy plates.
    One gets a better cleaner weld.

    One assumes your MIG machines are Direct Current Reverse Polarity (DCRP)?

    Following on from my first comment - regarding the settings.
    You could, if you have available to hand, try helishield gas (25%ar + 75% He). This burns hotter and is used for thick plates.
    But you could use this, to see if you get any differences - if so, it would suggest some prep set-up/containment issues.

    Also, is the plate Class certified, if so, what grade of cert, we've had issues in the past of mills covering their arse and giving us inferior grade certified alloys, and when investigated (after we found out), came from a totally unreliable source. We had to test each plate for approval.
    This also goes for the filler wire, is this Class certified?
     
  8. ExileMoon
    Joined: Oct 2009
    Posts: 67
    Likes: 21, Points: 8, Legacy Rep: 45
    Location: Shanghai

    ExileMoon Junior Member

    upload_2021-12-31_10-30-10.png
    I think the data in this table is a bit wrong, the yield strength should be 125 instead of 215.

    The use of acetone can only dissolve some grease, and will not be all right, especially for removing the oxide layer on the aluminum surface. Moreover, when acetone is used, the cooling effect occurs due to the rapid evaporation of acetone, and the water in the air may condense again on the cooled aluminum plate. The remaining water may produce some tiny pores during the welding process.

    It is recommended to do a wire brushing treatment on the surface of the weld before welding (the shorter the time, the better) to remove the oxide layer.
     
  9. Barry
    Joined: Mar 2002
    Posts: 1,851
    Likes: 505, Points: 113, Legacy Rep: 158

    Barry Senior Member

    It would make it much better if you begin to use complete engineering terminology when you put out numbers for comparison.
    Your original post said that you were achieving 280 MPA tensile strength/stress.
    When destructively testing welds you will take the specimen to Ultimate Tensile Stress UTS and the value that you stated indicated as such.
    If you are quoting a tensile stress whereby the specimen begins to permanently deform, then you need to use Yield Stress\
    And if you want to strain a piece of material below yield to a particular percentage, then you use Proof Stress. And this is what you have circled. Ie what stress does it take a specimen to stretch .2% of its original length

    As this material is quite thin some welding notes. If you bevel the edges of thin sheet particularly 4mm, on a butt weld, it is sometimes difficult to move a distinct molten puddle ahead of the arc due to the fast heat conduction of aluminum and the aluminum will just drop out, leaving a large gap to fill.
    On 4mm, we would wire brush the joint, push the edges together and run a bead. When complete, we would take a skilsaw and back cut the the bead (from the back of course) to the point that we could see that the weld was solid. This of course cleaned up any contaminants and enabled a larger pooled bead on the back side. We were not welding bureau certified but other large production aluminum boat builders in the pacific northwest were certified and they use this sequence and it meets the specs.
    Re the 6mm. We would provide bevels as Adhoc has shown. If we had access to the back, the bevel would not quite go in as deep as illustrated, to provide backing for a slightly larger molten pool which reduced voids. And we always ground and back welded.

    To Adhoc a question

    Occasionally with 4mm, when we could only access one side, we would sometimes tack in say a 25mm wide by 4mm strap below a butt weld, to use as a heat sink and backing for a slightly larger bead. These would be non water immersed welds
    when in service and mainly freshwater
    Can you offer a downside to this?
     
    Last edited: Jan 3, 2022
  10. adriano
    Joined: Aug 2009
    Posts: 107
    Likes: 4, Points: 18, Legacy Rep: 47
    Location: italy

    adriano Senior Member

    Yes definitely you need more details I updated my post as following with some pictures from which you can have amps and speed welding parameters:
    test result:
    6mm thickness - 300 mm long butt joint:
    Radiography test – which passes without porosity., sometimes small porosity shows.
    Bending test – no cracks observed – result satisfactory
    Tensile test – min required 305 Mpa – Result obtained 280 to 285 Mpa
    Failure at welded Area. (at moment failed testing piece not sent back from test lab!)

    We use 99,9% Argon gas, 1,2 mm ER5183 filler alloy wire (best suitable imported from USA), we use Aceton cleaner before
    60 degree beveling sanding (30 degree on both sides)and within 3 minutes welding from bottom "upwards" (standing pieces) .
    Key factors, believe, are suitable heat and speed?!
     

    Attached Files:

  11. adriano
    Joined: Aug 2009
    Posts: 107
    Likes: 4, Points: 18, Legacy Rep: 47
    Location: italy

    adriano Senior Member

    Thanks Barry
    Pls check my post to AdHoch you may find my answers
     
  12. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    You mean like this:

    upload_2021-12-31_18-39-43.png

    The red bar being your backing bar.

    No issues generally, this is rather standard practice. We use this often in similar situations.
    However, you should have a small saw cut down the centre of the b.bar, as this allows a path for the gas to escape when welding. Without it, is can lead to porosity issues.
    upload_2021-12-31_18-43-54.png

    If it is subjected to water/moister, you should be careful as the two surfaces and be site for corrosion.
     

    Attached Files:

    Barry likes this.
  13. Ad Hoc
    Joined: Oct 2008
    Posts: 7,773
    Likes: 1,678, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    Nope, the data is correct, and Barry nicely explain why.

    The welding test is to UTS, not yeild strength.
     
  14. baeckmo
    Joined: Jun 2009
    Posts: 1,654
    Likes: 670, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    The weld is "thin" in the lower part of the joint, if I interpret your middle picture correctly, and it looks like it has a slight "ridge" in the upper part. Might be too slow advance in the beginning, and too fast towards the end, or the feed is not stable, or.....?

    Btw, what does the gas flow value stand for, is it l/min or what? And what setting do you use for the line impedance, high or low?

    Edit: Come to think of it, I think you are starting with a high burner angle, and reducing it gradually as you move upwards; it is easy to sort-of use the elbow as a "center" for a sweeping/rotating movement when you make a short weld vertically upwards. Take care to move your hand with a constant angle along the joint. And I prefer a longer start tab; it makes it easier to "find the rythm" before you enter the test piece.
     
    Last edited: Dec 31, 2021

  15. adriano
    Joined: Aug 2009
    Posts: 107
    Likes: 4, Points: 18, Legacy Rep: 47
    Location: italy

    adriano Senior Member

    Th
    Thanks we're providing 60-90 degreeV- gap shape like yr detail C picture, but additionally we leave in between 1 mm gap from each hull plate to be welded!
    We are now discussing the issue with the Classification Society person about the 275 N/mm2 to clarify why he was requiring
    over 300?! Let me see what he answered we sent your diagram about this!
    I ll get back to you all as soon as possible I may assume this is the confusion apart from few matter like using denaturized alcohol instead of Acetone and after sanding, cleaning by wire brush I do not think we are doing much different from standards?!
     
Loading...
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.