All the bits for the final version of the bevel gear project are cut out and ready to go. I'll complete the photography in the morning.

Each of the two bevels is based on an arc of paper. The one pictured here is the sixteen tooth bevel. The teeth are also based on an arc, they curve round fairly tightly so I have divided them into blocks, each block being used to make four teeth. The left hand arrows show the grey areas where the teeth glue down.

I've had no internet for the past few days (grrr!) so I've being availing myself of the free wifi at the Coffee Kitchen in Cockermouth. I'm the JK Rowling of the paper engineering world. Without the money obviously, and fewer wizards. Less writing as well. In fact - forget that. I'm the Rob Ives of the paper engineering world and I like a good coffee.

Keep up to date: Receive the latest blog post by email

I do most of my design in Adobe Illustrator. When I'm designing a part I usually make up the parts with boxes, these are quick to draw and easy to edit. The piece here is the upper arm. It is a little rough and ready but is the layout I'm using for the final model. Now to convert it into a usable piece of artwork.

The first thing I added are the guidelines to construct the tabs on the top of the arm piece. I made a V shaped line on either side of a central line...

...then copied it six times rotating it 30° each time.

With the guidelines faded into the background I drew on the tabs. The guidelines ensured that they were neat and evenly spaced.

I then copied the tabs and modified them to fit above the right hand side of the arm. I then completed the outline of the part so that it is a single line incorporating both sets of tabs.

To finish off the part I added the dashed fold lines and grey areas where glue is to be applied.

I then printed out the part, scored the crease lines and cut it out.

And here's the finished thing. Once I'm happy that the rest model goes together properly I'll add colour to the Illustrator file.

The rest of the parts with our new part in place.

Keep up to date: Receive the latest blog post by email

One final tweak to the arms, I've added an extended cover to the elbow so that even as the arm is fully folded there is no gap in the side view of the arm.

I fitted a slip ring round the shoulder tube to act as a mounting point for the neck. So long as it is a tight fit it should hold the head into whatever position it is placed.

I'll add proper artwork onto the foot along the lines of this pencil version.

And here is the completed model with the head fitted into place. Last step, colour. I'll leave that until tomorrow, time for a quick strum on the old guitar.

Keep up to date: Receive the latest blog post by email

Well I must have made a dozen (or metric equivalent) different size arms trying to get the proportions right on the push up model. Frustrating but sometimes that's just the way it goes. This is one of the failed efforts.

And another - this time using an I beam rather than a box section.

Too long.

Arm grave yard.

I'm finally sorted. Just the head to go. I think I'll add a semi-pose-able head onto the circular shoulder tube. I hope that goes together a bit quicker!

Keep up to date: Receive the latest blog post by email

I've put together a box with sides as cut-away as possible so that you can see the mechanism at work. Triangular paper tubes help keep the structure surprisingly strong.

Rather than a circular axle  running through cut out circular holes (left) I've opted for a rolled tube fitted to the top of the body with a square pin running through it. (Right) This has the double advantage of reducing the number of circular holes that need cutting out and square pin fitting nicely to the arm top.

End view of the crank.

Keep up to date: Receive the latest blog post by email

Smelter and Coolio both drew my attention to this paper model of a push up character on a Japanese karakuri website.

Unfortunately it is impossible to tell from the picture just how it is driven. One thing is for certain though, the drive is through the feet, not the arms just as in my proposed model.

This plastic, wind up model which I found on the Hawkin's website has powered arms.

Master automata maker Paul Spooner produced this delightful model some years ago, again driven via the feet.

This time a circular cam with a slot round its circumference pulls on a string as it is turned, in turn this pulls the feet and lifts the body.

Pneumatics power...

...and hydraulics power and both beyond the scope of paper projects. So far.

Cam power, however, is not as per this model that I produced for a company some years ago which appears to be no longer in print.

So we come back to the design I'm working on now. It is fascinating to see how your approach to a similar problem changes over the years.

Keep up to date: Receive the latest blog post by email

Exercise is good, so I'm told. But if you don't have time for it you can always create a paper avatar to do it for you. Hence the push-ups model I'm working on now. Soon I'll have the vicarious physique of a seven stone weakling. 

So on to the mechanics! Sometimes with mechanisms the best way to create a movement is not to mirror reality. When I'm doing pushups (ha.) The muscles in my arms, chest and shoulders contract to lift my body from the ground. In this project I'm actually lifting the body via the feet.

I see two main competing ways to achieve this, via a cam or using a crank. Cam first. I fitted a cam to a drive shaft and threaded it into the box. The cam rests against a cam follower linked directly to the feet. Turn the handle and the body lifts up and down.

It works reasonably well, the plus side of the cam is that you can create any movement profile you want within the limits of the cam follower. In this case I was able to add a little dwell at the top and bottom of the travel, a slight pause at each extreme of movement. The downside - there is always a downside - is that the movement is only driven up and relies on gravity for the return. Not a huge problem but it does give a slightly floaty feel to the movement.

The red writing on the side is where I measured up for fitting a crank. 

Onto the crank. I used the crank template I'd created in the recent crank slider project with the throw set to 9mm. Everything fitted together nicely, I simply needed to turn the box top round so that the feet were at the other end away from the drive handle.

Looking good! I must say I was surprised just how much more satisfying the movement is using a crank. Very positive in both direction and the lack of dwell doesn't seem to be a problem. Yep. I'm going with that.

It'll look better with two arms and a head. I'm still not sure about the character though. Robot? Pirate? Gentleman in a top hat and tails? Hopefully inspiration will strike soon.

Keep up to date: Receive the latest blog post by email

I'm almost done with this co-axial crank-slider, just the assembly photographs to process.  A crank slider mechanism uses a crank, the rotating bit at the bottom, and a slider tube hinged onto the top of the box to convert rotary motion into an up and down, back and forth motion.
The co-axial part means that two sliders share the same axis to make a more interesting motion.  

In this project there are three tubes nested one inside the other. The longest tube, 6mm on a side, is fixed to the right crank, the middle tube, 7mm, is fixed to the left hand crank and moves 90° ahead of  the innermost tube.. The final tube is 8mm on a side and is hinged onto the top of the box.

This is a promising looking mechanism which I'm hoping will be useful for all sorts of paper projects. The 90° lag in the motion makes for an interesting, organic looking movement.

Keep up to date: Receive the latest blog post by email

I've been creating parts for my forth coming gear zine. If you've been following along you'll know I'm planning to use photographs of paper gears to illustrate how different types of gear mechanism work. I'm also planning to provide links so that readers can make their own gears to try them out first hand. 

It turns out that the design process for bevel gears is an interesting process and will be rewarded with its own blog post shortly. I settled on an 18/24 teeth gear set.

Shown here are the inner supports holding up the cone of the 24 tooth gear. 

...and here are the actual teeth.

The two parts fit together to make this rather satisfying gear.

Both gears meshing together. I'll be putting together a downloadable project of these parts shortly.

Keep up to date: Receive the latest blog post by email

Okay, back to the gear zine. I'm not sure whether this will work but I've been experimenting with cycloid gears. They are looking promising so far. The cycloid is the shape traced by a point on a circle as the circle rolls.

In this case, I've wrapped seven cycloids round the outside of a circle generating this hypocycloid internal gear.

The six lobed gear uses lobes the same size as the circle used to create the cycloid. These fit neatly into the seven segment hypocycloid.

The inner piece rotates in a sort of eccentric way within the outer advancing one tooth at a time as it does so.

I'll see if I can put together a stop motion or a flash animation of the mechanism shortly to more clearly explain what is going on.

It is certainly looking like a promising mechanism. If I can get it working smoothly it will make an interesting project with some interesting uses. A single gear like this reduces speed by a ratio of 1:7 adding a second gear back to back with the first should reduce the ratio a further 1:7 making a total reduction of 1:49 with only three moving parts!

Fingers crossed.

Keep up to date: Receive the latest blog post by email