My lad loves to draw, so I recently bought him a 1950s A0 drawing board and drafting machine on eBay. The whole thing cost me less than thirty quid all in, because there’s a lot of old drafting equipment about. Those who claim that CAD has revolutionised engineering are right, and the biggest change is computer aided drafting (rather than computer aided design). Roomfuls of drawing boards – and the skilled drafters who used them – became redundant back in the 1990s.
Computer Aided Design, in the sense of computers supplanting engineering judgement, is however still at the stage of artificial stupidity (as opposed to artificial intelligence). Not only are “design” programs stupid, they make those who use them stupid as well.
Rather than being required to exercise judgment about which calculations to do, (because every hand calculation can takes ages), and using methods known to be rough and ready, (because rigorous calculation takes too long), people – I hesitate to call them engineers – operating modelling and simulation programs can do a million pointless calculations in a second.
They don’t need, and have had no opportunity, to develop judgement of what simply isn’t going to work, and of the always approximate nature of engineering calculations. Unfortunately, these two things are quite possibly the two most important aspects of engineering judgement.
Where did they get the idea that being a Hysys jockey was engineering? At university, of course. Almost all of today’s chemical engineering departments neither possess nor recognise the validity of engineering judgement. They get free access to programs like Hysys, (which is as far as I know a great application when used for the purpose it is intended for, i.e. modelling oil and gas plants), and they use it for a purpose it is ill-suited for, the process design of all kinds of plants.
Consequently, graduates are often under the impression that once the key recycles converge in a steady state model in Hysys, (even if you cheated a bit to make that happen) you have designed a plant which will work. There really isn’t room in this article to explain just how stupid an idea that is – I recently completed a new article on Process Scaleup for the Kirk Othmer encyclopaedia which covers it in about 30 pages – but if you work as an engineer, and get a new graduate on your team, you may well already know what I am talking about.
But to return to drawing, chemical engineering graduates are rarely taught to draw, whether by hand or using CAD. They will often have to teach themselves CAD for their capstone design project, based not on professional engineering drawings, but on the dodgy approximations of these which feature in a lot of academic textbooks.
These drawings are a kind of mix of PFD and P+ID, because academic authors are all too often unaware of the role chemical engineers play in layout, and proper layout drawings don’t feature in academic textbooks, because academics have never seen one.
Layout is the part of design most incompatible with the idea (commonplace in academia) that engineering is applied maths. However, maths tends to be the one thing that academics are good at, and as a result layout has gradually been edged out of university curricula. Layout, though, is the kind of thing that people are great at, and computers (and mathematicians) are not.
A lot of piping engineers work in 3D CAD nowadays, and the artificial stupidity of such programs is good at spotting any clashes which humans did not. There is however no program on the market anything like as good as an experienced piper at laying equipment out in space.
Academics and big industry alike are hoping that graduate engineers can be churned out as almost interchangeable parts, at a volume which means that the only people making money out of the process are those who produce and use us. Being able to draw is actually a way to avoid that. Being able to use CAD reasonably competently is part of how I can work successfully as a sole practitioner.
I have no need for a big organisation. I can receive, read, and modify drawings electronically myself, then send back the revised version. I don’t need a draffie, or a CAD station. When I was stuck in Panama last year, I was designing and drawing plant from my laptop whilst living in a shipping container in the jungle.
Attempting to deskill engineering benefits only those who want “cheap” engineers, and nothing could be stupider than that, because there’s no such thing as a cheap engineer. In the long term, spending as little as possible to consider the allocation of millions of pounds to build or operate something really dangerous really doesn’t add up.
