Santana sells similar numbers of tandems with both wheel sizes and has no ax to grind.
Four years ago a respected bike designer confidently predicted the impending demise of 700c tandems. While this was not particularly frightening to Santana (we had designed and promoted 26" road tandems as far back as 1983), it also didn't ring true. I've watched as hundreds of customers tested both sizes before making a decision --- was it possible the majority who chose 700c tandems were mistaken?
Many people who advocate one size over the other insist on comparing fat 26-inch tires and skinny 700c tires. Some make recommendations based on the availability of a particular tread pattern. Others confuse the issue by comparing 700c tandems designed for pavement with 26" tandems designed for dirt. Santana's question was simple: if you were to eliminate the differences in tread width, tread pattern, inflation pressure and frame geometry, is a 26" wheel superior to a 700c wheel strictly on the basis of its diameter? If yes, why? In an attempt to discover the truth, we prepared some test tandems and asked a number of teams to evaluate them.
To reduce extraneous perceptions our test bikes used identical tubing and direct-lateral frame style, 26" and 700c rims produced from the same extrusion, and tires with the same width and tread pattern inflated to the same pressure. The honest attempt was to discover the best size --- after all, life here at Santana would be a whole lot simpler if we could standardize on 26" wheels.
But first, some background. The argument over wheel size did not start with tandem riders. Alex Moulton of England produced pro racing bikes with 14-inch sew-up wheels in the late-'60s. In the mid-'70s, Tarn Cycles of Chicago built a series of Campy-equipped full-race singles (and at least one tandem) with 20-inch wheels. In the early-'80s California's first production mountain bikes, built by Victor Vincente, were equipped with 20" wheels. All of these builders argued that bikes with smaller wheels would be superior due to lower weight, stronger wheels, quicker acceleration, and less wind resistance.
Critics of these designs claimed bikes with smaller wheels were slower and less stable. While slower was difficult to prove, some organizers banned small-wheeled bikes from racing (where they might have disproved the "slower" argument) fearing "diminished gyroscopic effect" would inevitably lead to crashes in pack racing events.
Fred de Long, Technical Editor of Bicycling, disproved the "gyro" theory in the late '60s when he assembled a unique bike with side by side front wheels --- a normal front wheel plus an identical counter-rotating wheel slightly above and to one side. The second wheel (which rotated at the same speed but never touched the ground) offset the gyro effect of the first. His finding: a bicycle's gyro-stability is a myth. He postulated (and I agree) that all us cyclists remain upright by continually steering through/across the path of our imminent fall. (You can quickly prove this to yourself by riding a bike with an over-tightened headset --- the results are extremely convincing).
Three years ago there was yet another resurgence of interest in road-racing bikes with smaller-than-700c wheels. For a time you could buy road racing bikes with 26-inch wheels from many serious builders including Serotta and Paramount.
While a few large-frame time trial and triathlete bikes are still built around a pair of 26" wheels, the designers of these bikes are admittedly chasing tiny aerodynamic and weight advantages that will be lost on a tandem (where doubled power reduces the significance of these advantages by 50%).
So what did we learn during Santana's testing? Our panel of testers uniformly found 700c tandems were more stable at higher speeds. Most testers also believed the tandems with 700c wheels were faster. The difference in "feel" was substantial enough so that an envisioned follow-up "blind" test with carefully shielded-from-view wheels was deemed unnecessary.
Why were 700c tandems clearly more stable? At the time of the testing, none of us had a clue. I later developed a theory, first published three years ago, that the answer was due to the shape of the tires' contact patch (footprint). If the same mass is supported on tires inflated to the same pr essure, the area of the contact patch must also be the same --- this is, after all, the meaning of p.s.i. or "pounds per square inch." The difference in wheel diameter causes the footprint of the bigger wheel's tire to be more elongated than the footprint of the tire on the smaller wheel. I reasoned a longer footprint would provide greater directional stability at high speeds (as is the case with longer skis, surfboards, and skates). Until someone comes up with an alternative explanation, this theory not only explains the increase in high speed stability, it also explains why off-road riders might reasonably prefer 26-inch wheels --- the rounder footprint provides less steering resistance and easier maneuvering at low speeds.
While my original "footprint" theory explained stability, it didn't explain the perceived difference in speed. I originally thought it probable our testers were mistaken about a speed advantage for 700c wheels. If they actually rode faster with 700c wheels, I felt certain it was an ephemeral result of enhanced rider confidence. Put simply, if riders on 700c test tandems felt more confident at higher speeds (because of stability resulting from the shape of the footprint), this confidence might allow a temporary increase in performance. If there was an enduring speed difference, it seemed likely to me the lighter and smaller 26" wheels would have the advantage.
Some of you might think the difference in diameter between 26" and 700c is too small to matter. Actually, even though we all know 700c rim is slightly smaller than 27" rim, a 700c rim is a full 2-1/2 inches larger than 26" rim.
Two-and-one-half inches?! How can difference between 26" and 27" exceed 2-1/2"?! Answer: a ridiculous tradition dictates that sizes of bicycle wheels --- unlike car and motorcycle wheels --- indicate the nominal outside diameter of the TIRE, and not the actual diameter of the rim. While the out side diameter of a traditional 26-inch "balloon" tire is about an inch smaller than the original 27-inch "racing" tire, the rim is nearly 3 inches smaller. The same tradition exists in Europe where there are no fewer than 4 diameters of rims that accept "650" tires (labeled 650-A through 650-D). To compare the "real" size of a rim or tire you must know the "bead seat diameter." Fortunately, this number is found molded into the sidewall of most tires. The real size of a 27" rim is 630mm (about 24.8"), a 700c rim has a bead seat diameter of 622mm, and the "26-inch" rim found on tandems a nd mountain bikes is only 559mm (a mere 22"). If matching width tires are installed, the outside diameter of a 622 (700c) tire is 63mm (2.5") larger than the outside diameter of a 559 (26") tire.
I've since realized the testers who reported faster speeds on a tandem with 700c wheels were correct --- and here's why:
Remember that the area of a tire's contact patch (or footprint), because it is purely a function of weight and inflation, owes nothing to the diameter or width of a tire. It follows that our test tandems with 11% smaller wheels produced footprints that were exactly 11% shorter and, therefore exactly 11% wider. Shorter explains the stability difference and wider explains the speed difference.
Why is wider slower? To apply the extra width against the pavement, the tread and sidewall of the smaller yet equally-wide tire is forced to undergo a great deal of additional contortion --- and tread and sidewall squirm are the primary causes of rolling resistance.
Is the difference in rolling resistance enough to produce a significant difference in speed? Because rolling resistance is a much smaller factor than wind resistance, until a few months ago I would have guessed no. Today I'm convinced otherwise --- whereas aerodynamic and weight differences are probably only half as significant for tandems (because of doubled power), internal tire friction is probably twice as critical (because of doubled mass). Even if 26" someday proves itself the superior size for road racing singles (it hasn't yet), the optimal wheel size for a racing tandem will always be larger.
While determining an exact difference in rolling resistance would be fairly easy, the effect on speed is difficult to ascertain. My best current estimate is a 26" tandem with equivalent rims, tread width, tread pattern and inflation will be 2-4% slower than a 700c tandem. While this will be a small difference for those who want the flexibility of using their tandem off-road, those interested in ultra-fast pavement rides might expect a cruising speed difference of up to one mile per hour (or a century finishing time difference of 5-10 minutes).
A couple of final thoughts about ultra-fast road rides on a 26" tandem. To achieve the same gearing as a 700c road tandem with a 54 tooth chainring, a 26" racing tandem will need a 60 tooth ring --- which is incompatible with the curvature of modern front derailleurs. And when you want to stop, because braking power is a squared function of effective brake radius, a rim brake on a 26" tandem is 19% less effective than the same brake on a 700c tandem.
Does this mean 26" tandems are stupid? Hardly. If you want to conquer the toughest terrain, 700c wheels simply aren't strong enough. And if we built a 700c frame with sufficient clearance for as-yet nonexistent 2.5" knobbies (700x63), captains shorter than six feet would have a hard time straddling the top tube.
If you want a tandem for tackling rugged trails or rutted fire-roads, a 26" tandem with clearance for wide knobbies is the only choice. If you can limit your off-road excursions to graded dirt, a good 700c tandem is adequately strong and will always be faster on pavement.