Zipp’s Sub 9 is the perfect storm of geek. First of all, it’s TT equipment. TT = nerd. Secondly, Zipp claims it produces thrust in a 15 degree crosswind. Physics = pocket protector. Finally, it gives me another chance to build a homemade wind tunnel to verify Zipp’s claims (I love debunking dishonest marketing). Science experiments = atomic wedgies! But enough about my high school experiences. On to the geek.

For a history of the development of the Sub 9, and a quick explanation of the sail effect, go here. I came up with a very simple low friction rig to measure the relatively tiny forces. A steel rod goes through the test wheel’s skewer hole. One end of the rod rests and pivots on a monofilament loop, while the other end is supported by another wheel, which rolls on a tabletop. To test the concept, I put together the basic rig and held a Vornado fan in front of the test wheel. With the fan pointed straight on, the wheel got pushed back. With the fan at an angle, the wheel moved towards the fan like a horror movie zombie. Amazing. The claims of thrust are absolutely legit.

Then I remembered something I heard. Paul Lew said that pretty much any disk wheel will produce thrust in a crosswind. Maybe Zipp just happens to be the first (or most cynical) to market it. I put my flat Zipp disk in the rig and…the same thing happened! This is not totally surprising, considering that any ol’ piece of cardboard will fly like a frisbee. So is the Sub 9 a clever piece of marketing that touts something a sheet of plywood can do, or is it an honest innovation that produces so much more thrust that it warrants Zipp’s claims? It was time to build a tunnel to isolate the wheel, and rent a high power fan.

I enclosed the wheel and fan in set paper. A string and pulley system pulled at the end of the wheel support rod to measure the thrust force. Since the forces in question were so small, the setup was very sensitive to tiny errors. The table had to be perfectly level, the support wheel had to be perfectly balanced. Even then, we’re talking about a wheel suspended on a bowed rod, held by string at one end, supported on the other by a wheel listing from the weight of the test wheel. The whole contraption rattles and shakes when the fan is on. Let’s just say that this test is far from perfect, and if you want to discount my results entirely I won’t argue with you.

Here’s the numbers. Measurements were taken at approximately 15 degrees and 22 mph:

Sub 9 disk with Zipp Tangente tubular: 62 grams

Clincher disk tubeless with Hutchinson Fusion: 29 grams

Zipp Tubular flat disk with Schwalbe Stelvio: 28.3 grams

808 front with Schwalbe Stelvio: 18.3 grams

Duchamp inspired pulley system records the forces and doubles as a cat toy.

Zipp’s site claims 80g at 15 degrees, but doesn’t specify wind speed. If Zipp’s measurements were taken at 25 mph, then my numbers are dead on. If it was taken at the more typical 30 mph, then my numbers are way off. The front 808 number is there just for fun. A front hub should have much less drag than a rear hub, so you can’t compare it to the rear disks.

So it seems the Sub 9 does generate enough extra lift to be considered a whole new class of wheel. But here’s where the marketing talk gets slightly squirrelly. I don’t doubt that at some wind speed you can get 80g of thrust from this wheel, but that’s only a 43 gram improvement compared to a flat disk. Now, let’s assume my numbers are reliable. At 28 mph the Sub 9 would generate 54 grams more thrust than a flat disk. For a rider with 2400 grams of drag at 28 mph, that would translate to 24 seconds saved over 40k. And that’s the best case scenario, where the wind stays at an optimal 15 degree angle over the whole ride. Also, it’s pretty safe to say that the wheel would perform best in isolation rather than at the back of a bike, right behind two churning legs.

So far I’ve only had the chance to ride the clincher outside, and it seems to be much less skittish than the flat disk. Maybe that bulge is absorbing some shock, maybe I’m imagining things. I’ll be grabbing some people to help me do side by side tests for ride quality.

So in the end we’re talking about an $1875 wheel with undeniable geek appeal that saves you 24 seconds over 40k. We cyclists spend so much money on gear I won’t comment on whether that’s money well spent. The wheel should be faster, it’s up to you to decide if it’s worth it.