I’m taking some liberties with today’s topic. Although it was meant to be a conversation between workers and their tools, I have much more to say about the interactions between my knitting (designing) work and the work I do in my day job.
You see, whilst I’m a knitter by night, by day I’m actually an incredibly glamorous Shielding Analyst in the nuclear industry. Yes, it’s about as complicated as it sounds but maybe not for the reasons you’d think! As much as I would like to wax lyrical about the basics of particle physics (really!), that’s not the part of my job that has the most impact on my knitting.
Instead it’s the practical aspects of building a 3D computer model of the shielding that needs analysing. The main thing you need to know about radiological analysis in civil nuclear is that tried-and-tested wins out over new and potentially bug-ridden software every time. I have no problem with this cautious attitude, but it does mean I’m using codes rooted in Fortran that were first released around the year I was born. The input is…clunky. I’m manually defining polygons in absolute co-ordinates whilst the mechanical engineers are using CAD packages that look like Minority Report in comparison. I’m not bitter I swear! It keeps my trigonometry razor-sharp, and we all know trigonometry has a very real use in designing knitting patterns. So, your maths teacher was right about all this stuff being useful on two counts.
The alternative title for this post could be “How Radioactive Waste Drums are like armpit shaping”. This is nerdy. This is how thinking about maths leads you in random directions if you’re not careful. I don’t know if this will provide anyone with insight but it’s definitely something.
Below is a screenshot of the software I use. You can see some sample code, and a nice plan view of waste drums in a concrete overpack, closepacked in a project specific 8-drum arrangement.
Did I mention I had to work out the co-ordinates of those drums manually? (fun aside: the software also has very little clash detection until you try to fully load the model and then it’s all like “OH HELL NO” so getting it right first time is a good idea). That’s a perfect example of a problem that can be solved with a bit of trig! It looks like this:
Now you can either recognise that this is a classic 30-60-90 triangle and that x=√3r, or do what I did and work out the much longer way that x = 2rSin60, proving that numeracy is not the same thing as common sense. But it started me thinking about the exact point the circles ‘kiss’ on the x-axis, which here is exactly half the distance between origins.
At the same time I was in the middle of designing a cardigan for the Knitter, and I wanted the armpit shaping to be dead simple – one straight bindoff row then a bunch of alternate row decreases until the correct armpit depth was achieved. But what width should the bindoff be compared to the decrease section? Perhaps it should be just before the point where the curve of the armpit becomes closer to vertical than horizontal? And as this is meant to be a close fit maybe we can simplify the armpit shape to a circle with a radius of required armpit depth? And that would make that point the exact point where a 45° line intersects the circles circumference? Which would be rSin45! Or…roughly 70% of the total armpit depth! Yes! Another successful day at t’mill!
Although the ideal proportions for this kind of armpit may be up for debate, I found this ratio to work very well for my purposes.
I understand this is a very quick run through more maths than most people have to worry about but nonetheless, I think it makes for a nice story of how you can get from here
through the magic of Maths!
All of my posts for this week are collected here.
You can see what everyone else is posting today here.