Friday, November 28, 2014

What's Wrong with The Education System

When I was a grad student twenty-some years ago, my profs knew me as a guy who liked to come up with his own way of analyzing things. My thesis advisor was in Electrical Power, and everything there is done by "modelling". You have a motor or a transformer, and there is a "model" which represents its internal parameters like "core losses" or "magnetizing inductance". You do some external measurements from which you calculate those parameters, and then you analyze the machine as a simple circuit using the parameters you just calculated.

I never did this. I always worked from physical logic and analyzed things from the ground up. Once my advisor asked me, "Marty, why don't you like models?" (He was Rob Menzies, actually a very capable engineer and a pretty good prof.) I told him I didn't like them because they encouraged you to work by numbers without actually understanding what you were doing. I don't know if he got my point, but the other day I had a flashback to that moment.

I was working with some Engineering students the other day and one of them asked me a question from his Power Systems course. It was about transformers. You do some measurements on the transformer and calculate its equivalent circuit parameters. I asked him to show me the question, and he did. It started off something like this:

"You have a single-phase transformer rated 20 kVA, 2200V primary and 220 secondary. You do an open circuit test and measure 220V, 2.5 A and 100 Watts. Then you do the short circuit test and measure 150V, 4.5A and 250 Watts. Determine the equivalent circuit parameters."
 I asked him to show me how those measurements were done. So he started to draw out the circuit model, which looks something like this:

Okay, I said, where do you measure the 220 volts? He started to point to one of the components, I don't remember which one, it might have been Xm, and I said: "No, you can't measure the internal parameters, those are only theoretical constructs. You can only measure the actual transformer".

He didn't exactly get it. So I drew this picture:

"THIS is what a transformer looks like", I told him. "There are only four wires. You have a voltmeter, an ammeter, and a wattmeter. Where do you hook them up?"

He was stumped. He had to admit that he had no idea.  They never talked about that in class. The prof told them that there was something called an "open circuit test" and something called a "short circuit test", that you get these measurements, and then you put them into these formulas, and the result is the equivalent circuit parameters (the ones you see all over the transformer model in the first diagram). No one ever talked about what it actually means.


And that's why I don't like models. Because they fool you into thinking that you know what you're doing, when you really don't. Actually, this particular student was pretty smart. He recognized right away that he'd been strung along, but he'd gone along with it because he had no choice. You follow directions or you fail. There's no time to second-guess the system and question what you're learning. And that makes him an exception.

The problem with education system is that by the time they've gotten this far, most students are no longer capable of recognizing what's wrong with the whole scenario. If I'd have confronted the typical engineering student with the fact that he was doing the calculation blindly even though he didn't even know where the voltmeter was supposed to be hooked up, he would have simply replied that it didn't matter, that you didn't need  to know that stuff because the right way to do it was just to follow the steps that the professor had laid out. And he would get the right answer on the test.

But that's not the biggest problem. The real tragedy of the education system is that it is doing exactly what society demands of it: churning out obedient workers for the government/industrial bureaucracy who will do what they are told without questioning or even trying to understand the reason behind it.