audger said:
tensile stregnth of silicone hose is extremely low. you want the hose to be squeezed shut widthwise between two surfaces, rather than being forced to kink shut by stretching it legnthwise around an object. you would also have a problem getting that tension on the hose without collapsing at either end.
There are commercial pumps that run the tube that way, but as the silicone tube is the part that sees the wear and is eventually going to need replacing I see no need to stress it out any more than needs be. I am hoping the tube last a fair time, say six months, but have no idea until I start using the pumps. There is no sign of any damage so far and the single pump has run for seventeen hours so far with both hot and cold water, so six months may be a conservative guess I hope.
My Mk3 pump has adjustable rollers so that each roller can be individually adjusted outwards so it squeezes the tube just enough to seal the tube internally but no more. This has made a big difference to how much torque is required to turn the pump. I have run a single Mk3 off a windshield wiper motor and it turned ok, but I would imagine the motor would soon get very hot, but I did not run the test long enough to test this.
At first I ran the pump with the tube fully pushed in, so that it was nicely snugged up against the outer wall. This caused the tube to wander from side to side, and it could easily have flopped out altogether. As soon as I shortened the tube so it sat ever so slightly away from the outer wall it started to track straight, and I have had no further problems.
I have lubricated the running surface to help the tube survive and to stop the cast iron wall from developing a rust film. There is no reason why I could not insert a thin plastic or stainless shim around the drum to get a better running surface than cast iron, but it is ok for now. Thin PTFE sheet would be good perhaps, but it is as pricey as hell. The rollers are polished stainless, so they will stay in good nick longer than me!
Eventually, all three pump shafts will be driven at once, with pumps i wish to use getting individually engaged from the front using the engagement pin like the video shows, or something nicer looking but similar, and to allow for this, every part of the pump is running on ball bearings except for the clamps at each end of the shaft.