RE: [CR]re: Frame Flex Testing

(Example: Framebuilding:Brazing Technique)

In-Reply-To: <000001c5c4ec$5dd6a080$6401a8c0@maincomputer>
References: <000001c5c4ec$5dd6a080$6401a8c0@maincomputer>
Date: Thu, 29 Sep 2005 05:53:11 -0700
To: "Ken Freeeman" <>,
From: "Jan Heine" <>
Subject: RE: [CR]re: Frame Flex Testing

>The Masters, i.e. Cinelli, Masi, and the other greats had a feedback system.
>The great riders they built for came back to the shop and said "I like this
>aspect" or "I don't like this aspect" or "I can't kick anyone's butt with
>this pos" or "it's wrong, it broke too easily." Quantitative measurement
>wasn't needed.
>My 2 cents,
>Ken Freeman
>Ann Arbor, MI

Indeed, almost all great bicycles I have ridden were based on this sort of experience. Today, this loop has been broken, especially in the U.S. - the big companies don't employ anybody long enough to build experience of the sort required. The guy who designs bikes for Giant has been in his job for 3 years! He was a bike mechanic before... And the small builders, for the most part, no longer have a team of hard riders who can provide useful feedback. Most racers now use mass-produced bikes, unlike the old days where experienced riders rode custom steel bikes.

Of course, the old system was not infallible, either. The racers came up with all kinds of stuff... but the builders seem to have been smart enough to filter this, at least as far as frame design was concerned. That said, I agree rider perception and expectations indeed are influenced easily. For that reason, I only measure the geometry of VBQ test bikes AFTER I have ridden the bike for 2 weeks and am done with the test. And now we have a second rider ride the bike to get a second opinion. That rider does not get to see any of the first tester's evaluation until after they is done with theirs.

Unfortunately, there seems to be little alternative to the "experience is best" approach. Bicycles are such complex machines that according to the book "Bicycle Science," in many areas one has to rely on empirical data (that is, experience of what works and what doesn't), because modeling and calculations cannot express this. Steering geometry is one area - the models assume riderless bikes, but how a bike without a rider goes down a hill and whether it falls over sooner or later, has no bearing on how it rides with a rider.

The book (Bicycle Science) doesn't even go into frame flex. I suspect this is for good reason. Nobody even knows how to measure it - simple deflection at the BB doesn't seem to have any significance. And the current "laterally stiff, vertically compliant" thing has no scientific basis, either. It amuses me that the builders of carbon bikes talk about how they can optimize stiffness in certain planes - which may be true - but this seems to be useless if nobody knows in which plane stiffness should be optimized!

Both flexy Alans and superstiff Columbus MAX (was that the name?) frames have been ridden to victory in great races. So the racers don't seem to mind. Maybe they are so strong that they flex any frame? Or some prefer one type, others the other? In my experience, it isn't a speed difference as much as a difference in how easy it is to keep the bike going. On some, I need to focus totally to keep going at full speed, others seem to make it easy to exert myself. On both, I am tired afterward!

Regarding the harsh ride of a frame - I did not believe it, either, until I got my 1985 Alex Singer camping bike. Despite lightweight parts (including a Duopar Ti rear derailleur), this bike weighs 35 lbs. Some of it comes from the numerous racks, but the frame is STIFF. It even has extra triangulation (like a mixte). It rides great even with 80 lbs. of luggage - even no-hands - but empty, it rattles my bones even with 650B x 32 mm Wolber tires at 60 psi - tires which are known for their smooth ride. I got some wonderful new tires from Japan (test in the next VBQ), which improve it a bit...

Even a truss flexes and springs - stand next to a railroad bridge as a train goes over it. I suspect it all has to do with frequencies - some are absorbed by the tires, others apparently by the frame.

It's a can of worms. But just because you cannot measure it easily (or haven't tried) that doesn't mean that there isn't anything there. -- Jan Heine, Seattle Editor/Publisher Vintage Bicycle Quarterly c/o Il Vecchio Bicycles 140 Lakeside Ave, Ste. C Seattle WA 98122