Steam Trains, Cranks and the Strength (or not) of Glue

I mentioned in the first post of this series the importance of thinking carefully about what you're doing and not just going ahead and building something because it'll look right. If you know how to calculate something, like the expected strength of a joint, do so. At least estimate it. Please.

Turns out that I'm terrible at listening to my own advice, as this tale of woe will explain.

The Luggage needed to have "hundreds of little legs" to properly emulate the original Sapient Pearwood version. I had long thought that this would be too complex to have The Luggage actually use these legs to move around with, instead they were going to serve as elaborate decoration. But they still had to move. Obviously.

The basic crank mechanism turns rotary motion into reciprocating linear motion.

The basic crank mechanism turns rotary motion into reciprocating linear motion.

Cranks were an obvious choice as they are very simple and can both apply a push and a pull force (like on a steam train). "All" you need to make one is a rotating shaft with something freely rotating about a point that's offset from the centre of that shaft. Well, there were plenty of rotating shafts in The Luggage so far, just need to glue something onto the edge of one of those and I'll have my crank!

Spot the deliberate mistake there?1

In my enthusiasm to get a mechanism assembled, I didn't think at all about how I was constructing it. Glue is very good at doing a very specific job, and not very good at doing most other things, so when I asked some glue to do something that wasn't its specific job, it almost immediately failed, much to my dismay.


You may be forgiven for thinking that the thing glue does well is sticking things togther. It is not this.

Well, not quite. Glue's good at bonding two surfaces together, but it's great at resisting shear forces once the joint's set. If you've got two surfaces which have been roughed-up well for the glue to grip to, they'll be able to support a large shear force. However, it's fairly likely that your glued joint will fail if you apply a force that tries to simply pull the two surfaces apart (a tensile force).

You'll notice in the first of the photos above that I'd used a small amount of glue to bond together two very smooth surfaces in such a way that they would be under tension for half of the shaft rotation2.

Shear + Glue = Good
Tension + Glue = Bad

The above is the take home message from this post. If I'd even stopped for half a minute to think about what I was trying to make, I'd have realised the folly of my ways (or at least my 20-20 hindsight tells me so). Once that truth had hit home, it was the work of moments to drill some bits of scrap wood and insert the shafts. Bingo: working cranks.

Brilliant working cranks. Easier to make, stronger and generally more awesome than my first stupid attempt.

Brilliant working cranks. Easier to make, stronger and generally more awesome than my first stupid attempt.

With that bit of learning done, how could I possibly fall into the same trap and fail to do some basic calculations that would risk the success of this project? Only time (or the next post) will tell....

1 2 Internet PointsTM available if you get it.

2 Internet PointsTM now unavailable.