Bob and Mike,
The "can opener effect" from a lug is easily avoided by doing two things. Never put a true point on the bottom of the lug profile. All it takes to eliminate this is a radius of .030" or greater and the stress riser is gone, in addition it is reccommended that the thickness of the lug be under 1 mm (.040"). In other words, round the point a bit and file the lug to a reasonable thickness and all of you worries are gone. If one uses pressed steel lugs this effect seems to not be a factor almost regardless of the presence of a point or how thick the lug is.
Also, regarding prepping lugs for brazing, wheather for brass or silver, cleanliness is always the most important thing along with proper joint tolerances. On the rare occassions I use IC lugs, it is neccessary to remove the cast "skin" from inside the lug. One need not get carried away and effect the critical dimensions, and this is more important when useing silver than with brass. Brase ons are the same thing; always remove the skin before attaching to a frame, even the machined surface on the underside of a shift lever boss or a 621 must be lightly filed or sanded for best results.
I also sand my brazing rod (brass or silver) and wipe it clean before using along with sanding (after a preliminary degreaseing and sandblasting) to clean everything. Tubes and lugs are all cleaned the same way. You can't be too careful or too clean in prep for brazing anything. It is a foundational principle of brazing.
So all the talk about where the heat effected zone is on the tube goes out the window as far as I'm concerned. Still don't consider the testing in question anywhere near valid in my world. There is a place for brass brazing. It's for "production" type work relative to bike frames in my opinion. I don't do production work; so for me I have no use for it. I am however a fan of brass fillet brazing, which BTW takes place in the range of about 1400 degrees-ish whereas a brass brazed lugged joint is in the 1750 degree range. Silver brazing takes place around 1150 degrees. I've had the same opinion for over 30 years and I haven't seen nor heard any valid reasons for that to change.
Brian Baylis
La Mesa, CA
Thanks for your comments Mike. I may not have made that clear. The point was that the margin of the heat affected zone, which represents a discontinuity in the strength of the tubing (stronger on the farside than the nearside), when it coincides with the lug point, multiplies the chance of fatigue failure at that point. That discontinuity, by itself, will not likely cause a fatigue failure, but when you put it right where the point of the beer can opener is, it can. Putting it out beyond the lug point (but still in the butted part of the tube) would seem to be a good thing. There is no way to avoid having a discontinuity or you would have a cold joint (no bonding of the materials). And remember, in the test, the strongest joint, requiring the most force to buckle, was the one done with the rosebud torch to evenly preheat the material, and brazed quickly with brass. Silver and brass brazed only with a small torch and no preheating required about the same force to buckle.
And the point of shining the lugs was not to clean them. They are pretty darn clean right out of the box, and are degreased in solvent prior to use. It was to break the surface layer which may be weak. Remember, Bill observed a thin layer of metal attached to the brazing material, meaning it is not a dirty braze that is coming apart but a shear layer in the lug material. Could be only a few molecules thick. And that observation was not a result of our test but of seeing bikes that had been crashed and trying to figure out why the lug sometimes separated. Some lugs have a tight fit out of the box. Too tight in some cases to allow full penetration with even silver. The amount of abrasion needed to shine them will not increase the gap enough to make any kind of brazing a problem. I am sure every good framebuilder shines up his tubing and lugs before assembly. I just looked at a Henry James seat lug that we have already shined and it is still so tight you can barely get it on the tube. That tube will be dressed a bit more to provide the proper gap.
I will bow out of this discussion now as I don't have much more to add. We don't want to bore these poor guys silly. They want to talk about nice vintage bikes (Wait til you see pics of the Paris I am restoring!). But thanks for the opportunity to share our experiences. Anyone is welcome to ask questions off-list.
Bob Freeman Elliott Bay Bicycles 2116 Western Ave Seattle, WA 98121 206-441-8144 Home of Davidson Handbuilt Bicycles
In a message dated 9/27/2005 1:57:44 P.M. Pacific Standard Time, hersefan@comcast.net writes:
Ouch....I think there be a logic problem here...
The claim that the stress riser being lower down improves fatique life does not follow. Where the weakest point of the joint is may well become irrelevant if all parts of the joint are now much less prone to fatique failure than they would be with brass brazing.
Again, it comes down what others have said - it is not what methods or materials are used, but rather how appropriately the material of choice is used in the application. And the mention that after testing it was decided to "shine" the lugs seems to confirm that the lugs may not have been really "clean" on the failed joints. Of course, too much "shining" (by use of a reamer or grinding) and the tolerances of the lugs could be shot with respect to use of silver braze.
Mike Kone in Boulder CO