weld FAIL!

[mike13u]

These are shots of a rear-wheel hub for my '52 HD panhead chopper. The wheel is several years old (maybe 6 or 7) but should not have failed here. I'm lucky this didnt end badly as I was doing about 60mph when the weld failed.
Basically, the rear hub has a flange that is welded on. That flange has a bolt-hole pattern that accepts the sprocket for the chain final drive.
I weld, but Im not a pro. I cant tell you why it failed but it does look to me like they didnt weld right at the seam. Maybe the set-up or jig wasnt properly aligned when the weld pass was made?

Also, If anyone knows how to make the pics larger and in one frame so people dont have to click each picture, feel free to help out. Thanks.

 

[hunter1151]

Do you have an extra spacer in there?? I can't tell by your spocket, did you have a good chain alignment?? Wondering why the weld is rusty but not the flange?? Just asking questions...... Glad you are ok that could have f----- some **** up!

 

[t100]

that's not all the way fail because you have gone somewhere with it.

look at this famous Lincoln Locker fail:

 

[472scout]

that's not all the way fail because you have gone somewhere with it.

look at this famous Lincoln Locker fail:

I was gonna say that nobody is that stupid but obviously someone is.

 

[mike13u]

Do you have an extra spacer in there?? I can't tell by your spocket, did you have a good chain alignment?? Wondering why the weld is rusty but not the flange?? Just asking questions...... Glad you are ok that could have f----- some **** up!

Yeah, everything was lined up fine. A friend told me that that rust line is from the chroming process. I dont know anything about that really.
Everything was laid out, aligned and put together properly. Its just a a weld failure of the part.

 

[Neppo1345]

The rust shows that this was not a sudden failure. That weld failed a while ago...it was probably a slight press fit keeping things together.

But then again, I'm not a forensic engineer. This is only my opinion.

 

[mike13u]

that's not all the way fail because you have gone somewhere with it.

look at this famous Lincoln Locker fail:

Not all the way fail? Yeah, yours is perhaps..well...more stupid.

BUT, You cant die on that fail. No matter what, death is the ultimate FAIL!!!

That is a beauty though....

 

[CatCow]

that's not all the way fail because you have gone somewhere with it.

look at this famous Lincoln Locker fail:

I could probably do that

This is what my harbor freight welding looks like... At least it held for my project

 

[gorilla]

Was that part chromed by the factory? Chrome plating can cause hydrogen embritlment in a weld if it's not properly baked after welding. That weld failure is not in the weld but in the heat affected zone. Not sure why but the penetration looks poor.

 

[ibedayank]

Mike

I would say you got lucky. that had been cracked for awhile new breaks dont rust

 

[mike13u]

Was that part chromed by the factory? Chrome plating can cause hydrogen embritlment in a weld if it's not properly baked after welding. That weld failure is not in the weld but in the heat affected zone. Not sure why but the penetration looks poor.

Yes, factory chrome.

 

[Defender Chassis]

Was that part chromed by the factory? Chrome plating can cause hydrogen embritlment in a weld if it's not properly baked after welding. That weld failure is not in the weld but in the heat affected zone. Not sure why but the penetration looks poor.

Are you saying that chrome plating after a weld is completed can effect the strength of a weld? If so, can you explain further or provide a link to an article that describes this phenomenon? Please do not take this wrong, I have just not ever heard of this an am interested in how this happens.

 

[NASTYZEN]

Hmmm. A bit hard to tell for sure from these pic's but it looks like there was more heat on one side of the weld. Full penetration one side none on the other.
I think your right about the jigging malfunction at the time.It takes lot's of load to shear something like that off.
Could lead to serious underpants malfunction when she let go!

 

[Olafur]

IMHO this started as fracture years ago. I can not see penetration as the problem. Sloopy design, high stress joint.. resulting in metal fatique.

 

[gorilla]

Defender Chassis, PM sent

 

[zmotorsports]

I would say that this weld broke a long time ago and moisture was allowed to get into the crack which results in the rust. There are about three areas that look to me like they were still holding and when those failed the sprocket let go from the hub.

As far as hydrogen embrittlement during the chroming process that is not as big an issue as it was a couple of decades ago. The chroming process has really become refined and has much less stress on weld joints than years ago.

It appears that upon the initial welding that much more heat/metal deposits were applied to the hub than the flange that holds the sprocket. Possibly when it was jigged it was not aligned correctly or there was a gap and it cooled at a different rate.

Mike.

 

[Olafur]

To add:
It looks like the "pipe" is supposed to be press fit into the flange. Well, if something went wrong in the machining process and it was not press fit this would result in additional stress on the weld resulting in high stress concentration area under cyclical load.

 

[Spam16v]

Harley, Harley, made of tin, ride it out, and push it in.

 

[Defender Chassis]

Wikipedia says the following about Hydrogen Embrittlement:

The mechanism starts with lone hydrogen atoms diffusing through the metal. At high temperatures, the elevated solubility of hydrogen allows hydrogen to diffuse into the metal (or the hydrogen can diffuse in at a low temperature, assisted by a concentration gradient).

I find it hard to believe that the chroming process would be described as being at "high temperatures". I am familiar with hydrogen embrittlement in fired heat exchangers and other process equipment but that stuff is up close to the melting point of the base metal. That only leaves the concentration gradient as a potential cause. Does chrome plating create a lot of hydrogen?

This is of particular concern to me because the front end on my dragster is chrome plated. The car is over 20 years old thought so I am sure a problem would have surfaced by now if it were going to.

 

[toolman1967]

I would check the sprocket for issues. I have seen cheap sprockets that were out of round causing a tight/loose chain issue throughout the rotation. I have also seen sprockets that would visibly wobble when viewed from the edge during rotation. The shop I used to work for had seen both of these issues with aftermarket sprockets that could possibly cause a hub flange failure like you show. Tens of thousands of these hubs have been chromed with no ill effects so I would hesitate to point at chroming as the cause of the failure.....But then again, anything is possible. I always look for the most simple cause and work from there.

 

[Frank The Plumber]

This is a hard tail or spring less rear assembly is it not. Is the rear assembly that is in there set up for that? Unsprung beats the heck out of this stuff and sometimes parts are not made for that.

 

[sberry]

Harley, Harley, made of tin, ride it out, and push it in.

My FIL has one of those, seems like after a while a guy would get tired of all the grief, you would have to be paying me by the hour to own or ride one.

 

[mike13u]

Appreciate the knocks on Harley, but that has nothing to do with this bike. The bike is custom. It is only a Harley powerplant. But, for what its worth, one that has been running strong since 1952. I have never owned a stock Harley but have been around bikes and customs pretty much all my life and can tell you I disagree with your observations about thier reliability and durability.
That said, if your talking about choppers, well then yes, its total madness. Steel banging, oil spilling, noise polluting madness. Much like hot roding is. And I wouldnt trade it for the world.

 

[nosnownogo]

Did you know that 85% of harleys made are still on the road today!!!


















And the other 15% made it home!!!!

 

[Boiler]

I had some ideas on your problem and took them to our CWI (certified weld inspector) to discuss. He confirmed what I was thinking.

The rust indicates that the failure occured at least some amount of time before it came apart. The CWI agreed and noted the gouging / scratching that appears to have taken place in the press-fit joint. It looks like it had spun and marred it all up. My guess is that it was shifting / spinning over some span of time after the failure, and eventually worked out enough to fail.

But why did it fail in the first place? From what we see, there is plenty of penetration here. What is amazing to me is that it appears that both base metals and the filler metal mixed up in a nice bead which seems to penetrate about 1/8" into the base metal on the sprocket flange. You can't tell how deep into the hub but it didn't release from the hub so its less of an issue. Then the entire portion of mixed base metal and filler metal broke loose of the sprocket flange base metal. It broke loose very cleanly. This is notable because if it would have taken out chunks of one base metal or the other, then you would have to assume at least some areas had fusion strong enough to overstress the base metal (or even the filler...).

Instead, it looks like a boundry layer between the mixed metals and the flange metal was maintained, even though the flange metal melted and mixed with the filler & hub base. He thought it might have occurred due to welding on a previously chromed part without grinding or machining the chrome away prior to welding. While this is a possibility, I found it unlikely as it appears that the weld bead's exposed face had the same finish as the hub, which I assume to be chrome (though very dirty...). I think it is a incompatability between the filler metal and the sprocket flange material. The flange could be some kind of hardened steel or alloy which required a special filler metal, or a high strength filler metal was used which caused residual stresses larger than the strength of the base metal. Even mixing with chrome, creating a weaker alloy or a chrystalline layer could be the culprit.

I think however, is that it is a heat affected zone crack, due to the inclusion of hydrogen, a sensitive material used for the sprocket flange, and high risidual stresses. (moisture, a non readily welded metal for the flanges, and a high strength metal for the filler).

You might check out the section on heat affected zone cracking in this pdf about weld cracks.

http://content.lincolnelectric.com/pdfs/knowledge/articles/content/weldcracking.pdf

Crack forms, then propogates along the HAZ due to cyclical loading (stress changes with every turn of the wheel). Eventually cracks through, and then rusts, while the pressed joint is all thats holding it together, it gradually (gradual enough to allow rusting completely) spins itself loose enough to come apart. Maybe under very fast take offs the hub spun a quarter turn at a time....or a twentieth.

Its hard to say exactly what happened but I believe this to be feasible.

 

[mike13u]

Boiler - In depth and well thought out. Thanks for your contribution and tell your CWI buddy thanks for his time also.

 

[ibedayank]

Most broken hubs are caused by potholes and curbs no matter what brand the cycle

 

[Frank The Plumber]

So what can be done now? Is that poor thing repairable or shot?

 

[mike13u]

So what can be done now? Is that poor thing repairable or shot?

Its shot. I could reweld, but I would have to give myself hours of headaches trueing the flange properly and being cautious of heat warpage and all the other headaches that its not worth it.

I'm currently trying to find someone local that will relace the wheel on a trueing stand and then I will just need to buy a hub at around $40-60.

If thats not doable, then its a whole new wheel at $250-400.

 

[hunter1151]

Hydrogen Embrittlement And Baking
What is hydrogen embrittlement and how is it prevented?

Hydrogen embrittlement results from the simultaneous codeposition of the primary metal and hydrogen on the surface of the work piece (cathode). The hydrogen is available from the water in aqueous plating bath chemistries. Because the hydrogen atom is much smaller in size than the metal of the deposit, it is able to migrate into the crystal lattice of the basis metal, and reside interstitially between the individual metal atoms. The interstitial hydrogen can greatly amplify the stress of applied forces within the basis metal which can produce catastrophic fracture at loadings much lower than the typical yield strength of the material. Hardened steels (> 40 Rc) are particularly susceptible to this phenomena and as such are generally required to be baked after plating to protect against hydrogen embrittlement.

Hydrogen codeposition can occur in the plating process either during the actual electrolytic deposition or during the cleaning and acid pickles preceding the plating bath. It is readily removed from the metal lattice by baking the product immediately after plating. The requirement for baking is a time-at-temperature cycle that is generally specified on the part print or within a plating specification. A typical cycle is to bake at 375ºF for 4 hours within 1 hour after plating.

I would bet that this is the culprit............My son was involved in racing go-karts for 15 years. You can pick any governing body for the sport, they all said the same thing: No chrome plating of any steering part, from the steering hub all the way to the front spindles. The reason has always been hydrogen imbrittlement.

 

[Defender Chassis]

^^^^^^Interesting

Thanks for the info. NHRA outlawed any new dragsters with chrome plated cages a few years back. I assumed it had to do with being able to see the quality of the welds because there was also talk around the same time about powder coating that was so thick that it did not allow a good visual on the welds. I am now thinking it may have to do with this same phenomenon.

 

[DarkMonohue]

Appreciate the knocks on Harley, but that has nothing to do with this bike. The bike is custom. It is only a Harley powerplant.

That brings up a question. Forgive me if I missed this, but is that an OEM hub or aftermarket? I'll admit it has been many years since I worked in a bike shop, but if memory serves, not all of the aftermarket stuff on offer was reputable. Some of it was made in countries notorious for poor QC (being as diplomatic as I can here) and a lot of the old timers would shake their heads at the shiny new bargain parts the younger guys were buying.

Do you have options for replacement in terms of country of origin? Or maybe this is just an oddball failure and not necessarily enough to sway you toward or away from any particular supplier?