<VV> how to break a crankshaft, and your heart

[John Beck]

Great post! --J.B.

djtcz at comcast.net wrote:
> 
> 
> The reduction in a shaft's bending strength and stiffness going from 1.700 to 1.680 is about 3.5% . No big deal. It's related to shaft diameter, cubed. (1.700^3 vs 1.680^3)
> 
> Around 1967 Chevy increased the 327/350's rod journals from 2.0 to 2.1, and the main journals from 2.3 to 2.45 inches.  Drag racers promptly turned the rod journal back down 0.100" (with such generous larger than stock journal radii that te rod bearings had to be narrowed) to reduce friction and allow smaller, lighter rod big-ends
> http://www.hotrod.com/thehistoryof/hrdp_0706_10_historic_hot_rods_performance_engine/index.html
> 
> Of far, FAR greater importance is the 200%, 300%, or even higher stress concentration that results if an undersized radius, step or (yikes!) a corner is created during the crank re-grinding process.
> http://www.pirate4x4.com/tech/billavista/PR-BV60/Materials/8.jpg
> 
> Smoothing the crank's exterior surfaces is nice. http://images.hotrod.com/techarticles/p116312_image_large.jpg
> 
> But, that is like looking for my lost keys under the street light just because the lighting is better there, because cracks 'most invariably start at the rod journal radius nearest the main web or overlap area.
> http://www.bmbikes.co.uk/photos/museumphotos/Broken%205%20Crank.JPG
> http://home.hiwaay.net/~langford/corvair/flexplate/05062401m.jpg
> 
> An example of the importance of geometry on part "strength" (resistance to fatigue cracking under the repetitive loading which is a crankshaft's life) is shown here.
> http://www.ellweinengines.com/jezus2/BBCnose3.jpg
> Curiously, even though vast amounts of material is removed from the rod journal's core, the resulting stress profile can be MUCH better, reducing the high stress at the radius in the overlap area, forcing other areas to "share" the work. That stress reduction via geometry and, surprisingly,  sometimes material removal is one reason why (expensive) aircraft cranks often are designed with heavily cored journals. http://www.sacskyranch.com/lyc_crank_gear.jpg
> 
> After the geometry is good, hardening and surface compression processes can boost the fatigue strength even higher.
> 
> All this assumes the crankshaft first passed a thorough magnaflux inspection.
>  http://www.thurstonengineering.co.uk/images/enginereman/magna.jpg
> http://www.aston.co.uk/restoration_services/images/01-007.jpg
> A visual inspection, or the famous "ring" test is no where near good enough, although may identify total junk early.
> Sloppy crank grinding can induce grinding burns and cracks, and a post-grind Magnaflux is the only way to know.
> 
> Sometimes when one or more journals get badly torn up, the crank will be welded before re-grinding to restore it. While welding >>can<< be an acceptable repair, the best welded material is really an uncontrolled casting, and also highly vulnerable to a host of deadly metallurgical problems like inclusions (built in stress concentrations)  and cracking.  Plus, welding even one journal kicks the whole crank out of alignment that requires some straightening (some straightening methods are forbidden on TuffTrided cranks) and probably some journal grinding just to get the mains running true, which can leave the run out of other non-trivial crank features like flywheel pilot and flange, snout, cam gear, seal surface, and balance compromised (messed up).
> 
> This does not begin to address any of the truly bad bearing journal geometries that a handsomely polished reground (or new, or used) crank may have.
> 
> So, a well done 20-under crank >>can<< be better than new.
> A welded non-magnafluxed crank with bad radius geometry and less than 100% geometric inspection is a door stop, or, along with the busted engine, will be soon.
> 
> Dan Timberlake