Conrad Bernier is absolutely correct: for rim brake efficiency the goal is "the greatest force on the rim for the least force on the brake lever." He is also correct that various systems balance "mechanical advantage" (or hand stroke vs. rim pressure) somewhat like a seesaw. While he (and many others) realize that various designs and leverage ratios can be confusing; Conrad stops short of reaching the (wrong) conclusion reached by a few on this list: that all rim brakes are essentially the same. Some brakes are 3-4x more powerful than others because of distortion and flex. Energy-robbing flex between caliper's pivot and the pad is the primary determinant of rim brake performance. The shorter the distance, the less the flex, the better the braking.
Chris Timm, along with Vince and Sue, wonder about mini-Vs. I agree with Rich Shapiro's assessment that these brakes are not the best answer. Most other Burley dealers I've talked to about this have reached the same conclusion. While it is true that you could ditch Travel Agents if a V-brake's arms were 50% shorter, the resulting brake would be too short to clear tires. A mini-V's arms wlll clear 700c tires (but not 26-inch tires) because they are only 20% shorter. Since mini-Vs will cause your brake levers to bottom out unless the pads are adjusted close enough to the rim to prevent quick wheel removal, many Burley dealers installed Travel Agents on mini-Vs in order to keep their customers happy. Put simply, mini-V's (originally designed for BMX) are a poor match for integrated road bar levers. The manufacturer, Tektro, who now combines them with a longer stroke lever and markets them for "comfort" singles, does not recommend them for tandems.
Marten wonders about Magura's hydraulic rim brakes. These brakes, once very popular in Europe, remain relatively unknown in the U.S. At the time of their original early-90s introduction there were a number of small problems. Burley spec'd them on their short-lived Bossa Nova road tandem and had lots of noise and vibration issues (that were solved when Magura supplied some retrofit pieces). Two years later Cannondale spec'd them on a mountain bike tandem, and removed them the following year when V-brakes appeared. Santana has installed a couple of dozen sets of these brakes (on custom frames with special braze-ons), and thinks they can be a fine choice for enthusiasts who do NOT want drop bars and integrated controls.
Mark Livingood provided an interesting link to an independent test of brake pads. While those results are too old to be useful today, in that scientific testing some of the most popular and expensive after- market pads provided woefully inadequate stopping power. How could it be that thousands of bikeshops and customers could have been fooled into buying, installing and/or recommending pricey lower-performance alternatives? Because bikeshop employees and enthusiasts are ill- equipped to test the real power of the brakes they sell, install and use, they are too-easily enamored with the perceived stiffness of brakes and pads that provide less real stopping power. And this explains why many enthusiasts will be very satisfied with long-reach side-pulls. Until the day comes that they need faster deceleration, it will be easy for them to recommend these less-powerful brakes to others.
George and Linda Wells (aka "ronin") wonder about published and peer- reviewed tests. Santana was a supporter of Bicycle Science, the now- gone quarterly magazine that was an outgrowth of Rodale's quarterly newsletter named Bicycle Technology. In both cases there were too few subscribers.
Many manufacturers have large testing facilities. I have visited Campagnolo's testing lab, and have worked with engineers from Shimano's extensive facility. Additionally, Shimano and Campagnolo have both funded research and testing projects at major universities. National and professional bike teams have also funded university- based testing and research. In Taiwan, a large government-funded bicycle test facility was created to boost the quality of that country's exports. Because tests funded by Shimano, Campagnolo, various teams or national governments are all based on securing a competitive advantage, the methodology and the results are closely guarded secrets. While Shimano and Campagnolo have shared results with me (and allowed me to question their test-engineers), they've never provided supporting documentation. Do I believe their results? Certainly.
Finally, the point of my hypothetical suspended-weight brake test was to make it clear that it is possible compare the effectiveness of various designs of calipers without the confusion of levers. It would also be easy to correct for differences in cable stroke by routing the cable via an adjustable length rocker arm. If and when anyone performs these tests, the results will show what every brake manufacturer with a test facility already knows: V-brakes are more effective than traditional canti's, which are more effective than the best side-pull. Among side-pulls, dual pivot designs are more efficient than classic designs. Short reach side-pulls are far more efficient than their longer and flexier cousins. Why? When it comes to efficient caliper design, the primary enemy is the distance between the pivots and the pads. The shorter the distance, the less the flex, the better the braking.
PS: I earlier promised three undeniable advantages of long-arm V- brakes with Travel Agents for tandems. First, as opposed to matched levers-and-calipers this mismatched combination (which is not quite stiff enough to suit single bike riders or their mechanics) is balanced in favor of more stroke and greater power. In short: the resulting tandem-optimized combo feels too spongy---but stops faster with less effort. Second, while they look wrong on a road bike, you can dramatically improve the stiffness and modulation of a V-brake by adding a light and inexpensive brake booster---a performance upgrade without a side-pull equivalent. Third, as opposed to long arm V- brakes that combine power AND clearance, side-pulls instead provide power OR clearance. If want room for fenders or bigger tires, side- pulls will force you to choose a longer and heavier caliper that is less effective---a particularly unfortunate trade-off if you'll only use fenders or wider tires for an occasional tour.
PPS: But aren't side-pulls the best choice for racing tandems? No. If you give an auto-racer less-effective brakes, his lap times will go up. An efficient tandem team will want to delay braking for as long as possible. Unless your braking power overcomes control (a common problem for single bikes, but not tandems), a more powerful brake allows faster finishing times.
PPPS: Because we are extremely busy (tandem sales through May were up by 80%), I've asked Steve Lesse---who monitors T@H and will continue to post private answers---to insulate me until after Interbike. If you have an important question or challenge, simply pick up the phone and give me a call: 800/ 334-6136 ext 11.