My dad was a design and technology teacher. I was exposed to the arts of woodwork, construction, design and so on as a child, and – while my brother went off and became an engineer – I have to say I’ve not particularly made use of those skills since I was doing my GCSEs. That said, I always enjoyed technical drawing, the precision and 3D spatial planning (and indeed, that same 3D spatial awareness that is so important in many aspects of geology).
While I was doing my PhD and had to design some flume equipment I had a flash of interaction with things again, but it was an absurdly simple three-part flume that a 7 year old could have designed and assembled. For this current project, however, somewhat more time and effort has had to go into flume design. In fact, the full first month was spent doing little but designing and refining the flume, to ensure it was capable of producing the results I want to achieve without being blown apart by high pressure gases or collapsing under its own weight.
So I found myself with a ruler and protractor drafting out isometric sketches and trying to make sure everything fitted together. Then a stroke of genius hit and I booted up Google Sketchup. I’d used it a few times before, but I reasoned that while I could sketch by hand as much as I wanted, by producing a 3D model I would ensure that all the separate parts would fit nicely, and I could read the numbers for part ordering straight out of the model.
And I was right – in fact, it was even quicker than drafting by hand. A couple of people in the lab were quite impressed by these models I was producing and got interested. I took images of the models down to our outstanding workshop engineer, and he was very pleased with the detail – illustrations of where cutouts needed to be made, clear ideas of where we would need PVC, where we would need aluminium, where perspex, and even where particular joins should be made (and how). It enabled me to design the somewhat complicated hopper shape (do a Google search for ‘hopper geometry’ and you’ll quickly realise that simply whacking a great big box on top of a release mechanism is not the done thing), and was even able to ensure it would be of a certain volume and mass.
So I was ever so pleased with myself when the purchase order went off with all the outputted measurements for the 75 or so separately cut pieces of material, in a variety of materials and thicknesses.
I was less pleased this morning after an hour in the workshops ensuring that all the pieces fitted together nicely to discover that the 3000 mm x 100 mm x 10 mm perspex flume base had been delivered at 3000 mm x 130 mm x 10 mm. Sounds trivial, but Perspex is really horrible to cut, even in a well equipped workshop. A 3 m length of it is unwieldy and impossible to deal with without some very specialist equipment.
I checked the purchase order. 3000 x 130 x 10. Bugger. How did they make that mistake. Why did I not notice it? I spent some time cursing myself for not checking the purchase order more carefully. I went and double checked again. Then I went and looked at the original paperwork. The spreadsheet I had specced everything on was correct. Then I checked the order email. I’d sent. Hmm. That was wrong.
So, what had gone wrong? Idiot cut and pasting. I had copied the details from one part, and pasted them, then pasted the same values for the next part on the list.
In summary, after a morning of ferrying a 3m length of perspex to and fro between the suppliers and their enormous cutting machine to remove 30 mm of excess plastic, I propose a 21st century modification to the age-old adage ‘Measure twice, cut once':
“Measure twice, cut once, paste once, check”.
At least it was cut too big rather than too small – that would have been both annoying and expensive.