Help!! Need Advice With Aluminum Welding Problem

I'm a first time boat builder who has tried to learn as much as possible about the process of constructing an aluminum boat. My boat is a 28 footer made of marine aluminum. I worked with Bill Lincoln (great guy, wonderful architect) and had my dream boat water jet cut from marine aluminum at Pierce aluminum. The building process has been going well (all the frames have been welded and are ready to install). Yesterday I misread the welding schedule for the two main girders/stringers which support the stern of the boat. Basically they are 16 feet long, 18" wide of 3/16" aluminum capped with a 1 1/2" x 3/16" flange on top. Each girder has 4 slots cut in it to receive the frames which intersect it. I should have known better but when I attached the flange the welds contracted enough to bend the girder approximately 5/8" along the top of the flange. In the attached picture you can see the straight girder placed on top of the bent girder to get an idea of the deflection. This may not sound like a lot but these girders define the sole for the boat and I'd like them perfectly flat. So here is my problem, I'd like any good advice you can provide to help me straighten the girder. So far, I've tried flame straightening, and mechanical means to bend the girder back straight but it refuses to yield, even a little. A boat builder suggested cutting the girder at the slots and rewelding it in the straight position. I'd be curious to hear anyone who has had to deal with a situation like this and how they corrected it. Thanks in advance,

 

after 30 years building alloy boats, I can say now there IS nothing you can do at all except,, TAKE OF THE FLANGE, straighten the girder, buy peining the top edge, then when straight, fit the flange, but wedge it up in the middle and weld the end first, then you have free length in the flange, in other wwords you make the flange a bit longer than needed, start it and finish it where it should be BUT have the middlew wedged up say 10mm you can then push it down to the girder edge and tack, After all the structure is done and bottom is plated you can then stitch the flange
Next time make a mylar pattern of the girder, and leave the bottom unshaped, either weld on the flange and cut the bottom to the pattern or do as I described before FREE LENGTH IN THE FLANGE STITCH ON AND IT,LL NOT PULL DOWN, or weld your flange, then put on a watewrline and mark the plate , then cut after all welding done
sUCH IS THE POWER OF THE SHRINKING WELD, STEEL IS WORSE
It is rather hard to describe, if you leave a msn , yahoo, or skype conntact in your profile I will talk you through it
Scuse caps unintended
dont forget to drill hole end those slots , it is sound eng practice
sometimes if you clamp the flange to a big steel beam and then weld it stops this happening
your weld throat and leg look heavy too, , sure you will win, you are making tidy job, good luck eh!!
could you drop the sole height the 8mm? and adjust all other to suit, I assume the sole sits upon the girder top

 

A butt weld will require a safety factor of about 10 to survive under any tensile fatigue loading. I doubt if the designer made such an allowance in his design.

The advice from whoosh will probably be the best you will get on this site - maybe anywhere.

If you do not take the time to correct this error properly you are likely to have compounding problems. A quick and nasty repair is something an unscrupulous builder would provide and will get out of the yard well hidden so becomes the buyers problem.

Rick W

 

bobk

As noted....take it off and start again. Your sequence is all important, as well as your heat input.

Also, never use heat to straight ally, nor over use mechanical means too. All these degrade the alloy in one form or another. Too long to go into on this short thread.

 

STOPPPPPPPPPPPPPPPPPPPPP
DARN bit rusty these days
lay the pice on a steel beam and pein with fairly large hammer ajacent to the welds, one inch each side of them so your peining will be abt 4 inches long, try this gently to start, watch the thing come stright, keep the hammer face flat or else you will bruise the stuff, all you are doing is stretching your shrunk areas , if you had stretching wheels you could do thsi in them, hope I am in time
cheers

 

Laying the ally over a steel beam and doing what you are suggesting will cause cross contamination and lead to more problems later..not recommended.

Difficult to say without seeing, but if you have 'over marked' your plate as you remove it, then i would not recommend reusing it. Again, asking for problems later on; principally cracking.

Correct sequencing and heat input, this should not be a problem. No need to "wedge up" or any other means of restraint. Restraining, in any form will just cause more problems later on such a solidification cracking. Seen it far too many times.

 

as usual you post the most neg stuff ad hoc, , you do not contaminate, you use clean steel and wire brush after Restraining is fine and works And what do you think we roll and fofm with? steel rollers in all my years nevber had problem steel contamination, you sir are a theory wallah
you have no hands on experience, you refuse to show who you are and whom you have worked for and I know that you never built anything ever in your life,, you just observed
Your name?
your company? your gallery, some references please, of the boats you have built?
the absurd post you made abt the Brit M.OD, who was your contact there?

 

here we are stretching( rolling pein)
maybe you can come over and show me your considerable skills?

 

You are suggesting to use a hammer on a steel surface.

Unless you are protecting the surface of the ally, every time you hit the ally any surface contaminants will be forced and imbedded into the ally. Small particulates can cause superficial micro-pitting when exposed to moisture. To wire brush the affected areas afterwards just demonstrates a lack of quality and understanding in the effects of what one is doing.

To form is a totally different process, as are the results on the ally surface. The dies are lubricated to prevent the aforementioned. To suggest otherwise again, just points to a lack of understanding in quality control.

As for posting negative stuff, only someone who feels threatened or exposed in their lack of quality control, or what they have actually 'designed' or their actual knowledge (as opposed to years of working which only indicates the number of years in the industry, nowt else), would feel threatened.

For those who consistently bleats "show me your company, show me your boats, show me your numbers ad nauseum ", do not practice what you preach. In addition to the "pictures" you show, they do not demonstrate any form of "quality" at all, it is just a picture of someone or something being done. Just like those who bleat, “show me your numbers..” this just means someone does not understand or been taught the theory because they use a computer program to do all their calculations, and as such require “numbers” to answer questions properly. A program cannot give you the reasons why,.. just numbers!

So, if I were to say I have just 1 boat designed/built in my ‘gallery’, you would feel instantly better in your knowledge and expertise compared to mine. Yet, if I said I have 100 boats designed/built, you would perhaps be more humble or acquiescent. All this demonstrates is arrogance. It is the behaviour of someone wanting or needing superiority over others, as well as constant ‘pats’ on the back.

If you are unable to answer a Q effectively, you resort to an attack on the questioner, not the question at hand. In doing so, just exposes arrogance as describe above. Since whether the Q comes from a newbie or a well-seasoned professional, you can either answer the Q or you cannot. The person asking the question is irrelevant, only the question is relevant. Your replies may be different from someone else’s and as such become exposed to scrutiny by others. Attacking the questioner, rather than the Q or reply clearly demonstrates an unwillingness to accept someone else’s point of view which may or may not better than your own; especially if it exposes poor QA.

The biggest problem with quality boats built in aluminium, are those fabricators who do not understand what quality means.

 

Beam Contraction

bobk,

my concern is nothing for your cabin sole but the buttlines of the bottom will rise toward the transom if I understand your description of the work? If my (assumptions0 understanding that you've bowed a long that should be straight is true, then the resulting downward bulb in the bottom looks to me more critical and error than the cabin sole framing?

I think there are a few solutions that might be considered which might not be too drastic.

I'd suggest that you consider re-cutting the bottom of the main longitudinal into a line. Forget the deflection/cup/bow/belly, just make sure you don't add a bowed long to your hull. I am assuming the long will straddle the frames (four of them) and weld to the hull plate?

At the frame point notches drill the notchs' top ends (especially the middle two) up deeper/higher/farther toward the waterline to make sure this newly fit lower edge will not be held up by shallowing the frame notches? If the cut a new line along the bottom of the curved plate then the original notch depth won't match the frame depth at that point.

When it comes time to fit the cabin sole framing, shim it to the amount of cup the main long has welded into it. The transverse frames of the cabin sole seem way less critical to me than the hull's main longs! The after bottom should be nice and clean and if the boat planes, its more important to have flat after butts than to have to fiddle with a few deck/sole frame elements.

I think its good policy in aluminum boats to wait to weld (weld out- not discussing tacking) until you've got as much of the metal in the final location as you can and still get the gun into the weld zone. I don't see why you couldn't tack the flange to get the gain of straightening the long in plan view while avoiding the weld heat to keep that same member straight in profile.

Of course I can't see her lines and the lower edge may not be a line, it may rise toward the bow with the rise of the butts? If that is the case then I'd try to lay the warped long on a factor sheet edge of metal (if you have any left after having the profiles cut?) use that as a base line to re-loft the butt line/longitudinal lower edge and re-fair the bottom edge with a saw/router/sander/Vixen file.

Taking the flange off and peening requires removing the flange and redoing the welds, I'd rather find a way to overcome the distortion's net negative shaping influence than to re-work the pieces. If the amount of curvature is 5/8" in 16' then the most you'd lower the long's ht (stiffness) is by that amount. If you ask Bill Lincoln if that is going to weaken the hull (?)- I think it would be less strength loss to loose/trim/cut-off the 5/8" in long depth than to weld, remove, peen and re-weld?

I don't know that to be fact, I would think it was the case, twice the heat in the same place can't be any better than just a trim job's slight loss of depth to that one longitudinal?

As to cutting the long apart and rewelding, at the frames, that sounds like it would leave HAZ over the frames, not somewhere I would think this "natural" weak spots were best tolerated?

If I had to, no choice, had to cut the long at the frames, I'd only agree to do that if I could double the butt weld zone on both sides of the original long with 3/16" or 1/4" 'fish plates' or web doublers that straddled the xverse frames AND the butt weld zone by at least two or three web depths in length.

Anyway you look at, it balancing the heat of this weld with the matching stitches on the hull plate, when the time comes (again, assuming I understand where this frame element will be placed in the boat's overall design) to weld both sets of stitches would keep the long a lot more the intended shape than welding one side all by itself.

In the worst case, the cost of a 16' x 12 or 16" piece of plate to replace this piece can't be worth the effort to spend more than a day or so fixing this weld sequence oversight, can it? The notches seem like they could be copied from the remaining piece and if the lower edge is curved forward instead of a line, then that too could be copied from the remaining pattern?

I'd consider trying to use the piece somewhere else, in smaller pieces, if it turns out you can't re-cut the hull's edge and use it as it is welded?

cheers

 

Thanks

I'd like to thank everyone for their help with this problem. Whoosh's solution worked well. I devised a special base for my router so I could mill off the welds without destroying the girder/beam. After removing the welds and the flange, the beam was noticably straighter. I placed the girder against a large stainless steel angle and peened the edge using a block of aluminum to minimize any kind of marking. The beam is now within 3/16" of being perfectly straight in 16 feet, not too bad since I'm not even finished. A little more work on the beam and it should be good as new. I've learned a lot in this process but the most important thing was not to finish weld the flanges until the beam has been installed and tacked to the hull plate.

Bob