Uh, I broke it I had a hunch that the dynamics would change once pressure was added to the system... I made that verticle dowel a little half-assed and it snapped after about the third gallon. I emptied about 3 gallons in 1.5 minutes. It looks like a little over one cup leaked out the front of the pump onto the floor. I'm not discouraged.

Besides the other enhancements I already mentioned, I would try to take as much pressure off the system as I could by reducing the size of the cylinder or increasing the size of the valves and hoses. Rev 2 will (eventually) be two small cylinders. Two makes sense for the following reason: currently only 50% of the hose resources are being utilized; one hose always stands idle while the other is working. Two pumps would utilize more of the resource since neither hose would stand idle... it follows that the system would be less strained under that arrangement. My intuition tells me to avoid using valves and hoses much larger than .75" dia - could be wrong.

The teeter moved really slowly sucking all that water into the 3" cylinder; I think I would prefer a more aggressive teeter. I'll make that failure point different as well. My single kayak will eventually have a homemade footpump in it but don't hold your breath.