Cause of weld failure?

admin

I'm not a welder, I'm a mountain biker with a broken bike frame. I've heard that the weld, if done correctly, will be stronger than the metal it joins. The weld on my frame appears to have failed instead of the aluminum. The break is just above the pedals which tends to get a lot of back and forth side forces. Can you tell by looking at the weld if there is something wrong with it? The frame is being replaced under warranty, I'm just curious what caused the break. Any opinions?
Reply:Hi Flenser,The short answer is yes, you can learn quite a bit from visual analysis of fractures.  There's not a whole lot I can tell you from the photo you provided.  The fracture does follow the toe-line of the weld.  Which certainly makes me suspicious that something isn't right.In order to make a more thorough evaluation, you'd have to completely break the fracture open, using a swift sharp blow.  I'm not sure you can or want to do this in these circumstances, as it might interfere with your warranty replacement.  If, by chance you do, don't let the fracture surfaces revealed touch anything before you try to photograph them.  Don't put the part back together, don't let any curious by-standers touch the fracture surfaces.  Any contact could alter the surface and the evidence revealed.Last comment, unless you've got a 5-50X microscope, there's a limit to what you can see with the naked eye.  Visual inspection might be able to determine if the fracture was due to a massive overload or a slow accumulation of fatigue damage.  But, there's no guarantee that the evidence will be visible to the naked eye.
Reply:Its hard to tell by the angle of the picture but it looks to me that the weld failed because it did not tie into both pieces being welded.  It was like the weld just laid on the surface of the metal like bubble gum and didn't bond to both metals.  Under normal use the weld would have never reared its head but I suppose the frame sees its share of rough terrain!NickoliIf I can't fix it, Its probably not broke.... Yeah Right!~ Damn I let out the factory installed smoke again ! ~
Reply:You need a gussett between the upright tube from upabove where the pedals are to the down tube from the "A" frame mount is located.....zap!I am not completely insane..Some parts are missing Professional Driver on a closed course....Do not attempt.Just because I'm a  dumbass don't mean that you can be too.So DON'T try any of this **** l do at home.
Reply:as said, the joint would need to be seperated to stand a chance of fully identifing the cause of failure- doing so would probably complicate (maybe even invalidate) the warrentythis sort of failure isn't that uncommon in the cycle industry (the bottom bracket cluster and head/down tube being favorite spots), especially with aluminium frames. failures can be caused by a one time overload (say, a heavy impact) or from fatigue (applying a cyclical load over time)i would guess that we're looking at fatigue failure here, although it's likely that a flaw reduced the fatigue strength locally. the weld is (was  ) joining the seat tube to the machined bottom bracket/swingarm mount. i would suspect there may be poor fusion to the machined part (it's larger mass requiring more heat than the seat tube), possibly even an area of incomplete root fusion (a void under the weld where the two pieces meet).
Reply:That's about as obvious a failure due to lack of fusion as I've ever seen.
Reply:Thanks for the replies. It sounds like poor fusion is the most likely cause? It's not obvious in the picture, but the weld is completely separated. I can raise it up a couple of millimeters. Unfortunately, it's been that way for a while, so analyzing the break surface would probably be useless. I already returned it, as well.I have a history of breaking bikes, but I have been "reasonably" gentle with this one - at least when you consider what it's designed for.  I'm pretty sure I would remember a ride that exceeded this frames design specs  The break must be from fatigue.Here's two more pics that might provide some more information. The first is my frame cleaned and ready to return to the manufacturer. The second is a different bike of the same make with a very similar break.
Reply:Aluminum will just not stand up to these stresses as well as steel. I hear the bikers talk about going through 2 or 3 aluminum frames a year. I own a bike framed in aluminum also, but rarely ride it and my offroad biking days are over. Actually, we were checking out the carbon-fiber framed bikes today at the races...like I asked Ed, "Can you weld those?".Last edited by tanglediver; 11-04-2007 at 10:01 PM.City of L.A. Structural; Manual & Semi-Automatic;"Surely there is a mine for silver, and a place where gold is refined. Iron is taken from the earth, and copper is smelted from ore."Job 28:1,2Lincoln, Miller, Victor & ISV BibleDanny
Reply:That was an outstanding last photo...I have never seen an aluminum weld break like that.  WOW!!!I would love to hear more about how to prevent such a fracture w/ aluminum.Great post ......kind of makes  me humble now w/ what little aluminum I've done. weld it like you own it
Reply:with aluminium fatigue failure is inevitable at some point as the material has no definite fatigue (or endurance) limit unlike steel- flex it enough times and it will crack. have a read here,... http://web.mit.edu/course/3/3.11/www...es/fatigue.pdf  googling 'S-N curve' or 'endurance limit' will provide moreany stress risers will accelerate the process causing premature failure. stress risers can be any flaw that disrupts the 'flow of stress' through the structure- sharp edges, craters in welds, widely spaced 'roll of dimes' welds, incomplete penetration/lack of fusion etc.
Reply:Tanglediver, personally i'm not a fan of carbon fibre bikes, especially parts like bars/seatpoststhe nature of metal means that giant overloads (big crashes) aside failure is usually fatigue type in nature- regular inspection will find cracks before they grow large enough to cause catastrophic failure.re-inforced fibre materials tend to let go suddenly when they've had 'enough'.
Reply:The problem with Aluminum is that it has no endurance limit. It will always fail under a fatigue stress. Steel and Titanium alloys have endurance limits. --Gol'
Reply:Flenser,The second photo you posted was very helpful.  It wasn't obvious at all that the fracture rain through the middle of the weld bead in the first photo.  My initial assumption that the fracture occurred at the bottom toe line of the weld was incorrect.The second photo clearly shows that the fracture is through the middle of the weld.  Additionally, there's a pit/void or crater that is visible right along the fracture line.  This could very well be the stress riser or concentrator that someone mentioned earlier.It concerns me greatly to hear that you've seen other identical failures.  That suggests either a design flaw, which Zap proposed potential fix to.  (Not sure if a gusset is the right solution Zap, as I've heard that bike frames are supposed to have some flex to them.  I wonder if a gusset might create a new problem while solving this one.)The other possibility is a manufacturing defect; the crater left in the weld bead.  I lean in this direction myself.  Could be the maker produced a number of bikes with imperfect welds; a crater at the end of the bead.  If it were my bike, I'd check the new one for a similar appearance/defect in the same location before accepting delivery.  You might not be able to reject the replacement bike if you do see something, as there's no definitive proof that a defect at the weld crater is the root cause of the problem.But if you break the replacement, at least you've mentioned the possibility of a defect to your merchant/manufacturing rep....-Dave
Reply:this sort of failure is very common in the cycling world. while gussets are often used by many manufacturers (typically on the underside of the head/down tube joint) one shouldn't be needed here. the cause of premature failure is a welding defect- the majority of frames will survive years of (ab)usegussets did indeed cause as many problems as they solved when first introduced. the head/down tube gusset was often fully welded which simply moved the failure point to the weld running across the tube at the rear end of the gusset. eliminating this weld solved that problem- a better solution would be to taper the wall thickness of the gusset and blend the weld but that takes time and moneygussets add weight too. downhill racing aside, manufacturers are fighting in a very competitive market to produced the lightest, stiffest best equipped bike for a given price point- high end bikes are using asymmetrically butted, shaped tubing made from proprietary alloys to shave every last gram without sacrificing stiffness. i saw/autopsied hundreds of similar failures while working in the cycle industry- invariably there was clear evidence of incomplete root fusion, often a noticeable void at the rootwith less material available at the join (any join is a stress riser) the fatigue life is reduced. if there are other defects (undercut, crater etc) at the surface of the weld then a crack will likely propagate from one of these (highest stress occurs at the surface) and then grow along the line of least resistance until there is insufficient area available to resist the stress
Reply:I agree with Dab.  It looks a little like the break might have started in a crater at the end of the bead...The hole that's dropped out in the center is pretty characteristic of a bad finish.Smithboy...if it ain't broke, you ain't tryin'.
Reply:Originally Posted by tanglediverAluminum will just not stand up to these stresses as well as steel. I hear the bikers talk about going through 2 or 3 aluminum frames a year. I own a bike framed in aluminum also, but rarely ride it and my offroad biking days are over. Actually, we were checking out the carbon-fiber framed bikes today at the races...like I asked Ed, "Can you weld those?".
Reply:As mentioned throughout some of the above replies, there are several issues at work in the failure(s).- Aluminum does not have a fatigue or endurance limit.  That means that no matter what, an aluminum structure WILL fail eventually under fatigue loading.  The higher the loading the faster it will fail, but even under smaller loads an aluminum structure will eventually fail.  That is a different physical behavior than a steel structure, which -can- last forever, load-wise, if the cyclical stresses are under the fatigue limit stresses for the specific material.- Two, any structural discontinuity WILL act as a crack propagation or nucleation site.  - In addition, poor welds weaken the structure because they have not properly fused the structure into one-ness (tech speak: the structure has not achieved material homogeneity).  See #2 above, and also a "poor weld" could be incomplete fusion, or discontinuities, or not properly sized beads, or incorrect filler alloy selection, or improper prep like not cleaning the area first or trying to weld over anodizing or the usually-present aluminum oxide surface layer, or ... etc, etc.- Running on 'the edge' means things break.  If you are racing and shaving weight and "pushing the envelope", then things will break.  If the designers and engineers get it all -just- right, then it breaks just as you cross the finish line.  If they get it wrong, or something else goes different than planned or calculated, then it breaks before that.So, what does it all mean?  An aluminum bicycle frame WILL fail structurally if it is ridden.  How hard you ride it determines how soon it will fail.  Poor welds WILL cause weldments to fail sooner than good welds, for multiple reasons.Glad you didn't get hurt.
Reply:On the steel vs. aluminum debate I don't know if the reason is the difference in manufacturing costs, material suitability or something else, but very few full suspension bikes are being made with steel these days. I would think long durability would be a pretty low priority, though, since the overwhelming majority of the bikes sold see very little real use. Most end up collecting dust in the back of a garage somewhere. Even with 5 year warranties being pretty standard, it's probably cheaper to replace the few that break than to make them indestructable.Fortunately for me the spot where mine cracked provided a degree of stiffness, but didn't dramatically strengthen the frame. I was able to ride for a couple of months without quite figuring out what was wrong. The crack is clearly visible in the pics, but nearly impossible to spot on a dirty bike while wearing sweaty sunglasses. I even missed seeing it when I changed out the rear triangle - I thought the rear tri was somehow responsible for the loss of stiffness  The new frame is already on its way - their customer service is excellent. I'll post some shots of that weld when it arrives.
Reply:Mnt. Bike Frame-Metal ChoicesCity of L.A. Structural; Manual & Semi-Automatic;"Surely there is a mine for silver, and a place where gold is refined. Iron is taken from the earth, and copper is smelted from ore."Job 28:1,2Lincoln, Miller, Victor & ISV BibleDanny
Reply:the only way to join AL with success is to bolt it. unless you are well versed in welding it. imo
Reply:Any aluminum longitudinal cracks I've seen like the 2nd pic were either caused by wrong filler selection for the type of base, or insufficient throat size. In this case it looks like it could be throat size. There doesnt seem to be enough filler added, the weld it too flat.
Reply:Originally Posted by westcoast welderAny aluminum longitudinal cracks I've seen like the 2nd pic were either caused by wrong filler selection for the type of base, or insufficient throat size. In this case it looks like it could be throat size. There doesnt seem to be enough filler added, the weld it too flat.