Too much horror business, drivin' late at night. Bendix '76.

Fixed up an old Bendix '76 coaster brake hub a while back. Here's my sketch diagram that kept me straight. It might help you out when you do that kind of thing. Click on image for larger image.

Optimum

This is a dull and geeky update to my last post about pistons. I posted some diagrams for different ways to connect a rotor to a piston, like this:


The part labelled c1 is the connection from the rotor to the piston. I knew that the length of c1 had to be longer than the radius of the rotor, otherwise the machine would completely fail. To be safe, I noted on the diagrams to make c1 twice the radius at least. This is a good value, but I wanted to figure out the optimal length for c1.


The optimal length of c1 is about 1.5333r. This makes the optimal angle about 49.29°, not 45°. This is the point where you get the most angle increase for c1's length increase. This has nothing to do with the amount of force transferred, and if it does it's a coincidence. But I do think it's more efficient the greater the angle. Ideally the angle would be 90° at all times but obviously that's impossible. So, the longer c1 is the greater the angle, but 49.29° is the optimum trade between c1's length and the angle.

Here's a more accurate sketch:


Now you can sleep at night, eh?

Here's the working, if you're interested.

Pulleys and Pistons

I have a fetish for pulleys. I find pulleys and rope very interesting. Anything involving rope usually interests me. No idea why. I know about 40 different types of knot, and I know how to bind and suspend a woman securely from the ceiling.

Last week I started to sketch out some ideas for good-looking configurations of pulleys and rope. No real use for it, just to play.
(click on any image for larger image)




This is one way you can rig pulleys to allow a small helium balloon to lift a 1 lb weight (assuming the ropes and pulleys weighed nothing). The second sketch is two equivalent pulley systems joined together. The left side uses fixed pulleys to give a 1:4 advantage; the right side uses a movable pulley to give a 1:2² advantage. So they balance.

So I started reading about pulleys, and then I read about simple machines. I decided I wanted to design a wind-powered water pump that fits onto a standard 55 gallon food-grade barrel. This barrel is a rain barrel. The idea is that the water will be pumped to a filter, and then drain back into the barrel. With the wind powering it, the barrel will be self-cleaning. The filter will be easily removed, cleaned and replaced to continue filtering. Not only will this filter the water with no effort, it should also look awesome.

So that's where I started. I watched some videos, read about simple machines and sketched a few ideas. Then I really got into it. I figured four ways to get a fan to pump a piston.

DIRECT ROTOR TO PISTON CONNECTION

ROTOR TO PISTON VIA LEVER

ROTOR COG TO PISTON COG TO PISTON

ROTOR COG TO PISTON COG (VIA BIKE CHAIN) TO PISTON

I also figured out something that took me a while to understand. A yaw. A yaw on a wind turbine is the big wing at the back of a wind turbine that turns the turbine around so that the blades always face upwind. I couldn't figure out how to connect the turbine to a piston pump while at the same time allowing the turbine to rotate 360°. I figured it out today and did a sketch.

MECHANICAL YAW MECHANISM

The yaw wind turbine sketch isn't to scale but everything is there. I don't know the proper symbols used on professional technical drawings but hopefully it's clear. The thing in the centre that looks like a box of marbles is a type of joint that allows the piston connection to rotate above the 'box of marbles', but not below it. The other 'box of marbles' above it is to allow the entire top part of the machine to rotate. Hope that makes sense.

All these wind turbine sketches are a simple one turbine, one pump setup. I had other ideas for water pumps that pump in both directions, a couple of different ways to connect 2 pumps to one wheel, and combining those ideas together to get two double pumps connected to one wheel. I haven't sketched them yet though.

I haven't included in the sketches the valves you require for the piston to become a water pump. There should be two one-way valves (or check valves) attached to the piston at the base, one that only allows water in from the water supply, and one that only allows water out to where it's to be pumped. I was going to make these but I think bought ones would be more reliable, and their reliability is crucial.

I think this level of mechanics was probably taught in many schools to young children, but I never did this at school, so it's new. Learning a new mechanical technique is like finding a different type of Lego block, and it introduces an asston of new permutations.