«

»

Apr 03 2012

MITx 6.002x week 4 – slow going on homework 4, part 2

I’m rather cross with myself – and the course question setters this week. Having found the Lab 4 curve tracer exercise straightforward and also having rattled my way through the first two parts of homework 4 (H4P0 and H4P1 on Zener diodes) without too much difficulty, I got stuck on the amplifier questions in H4P2.

Except, I didn’t really. I’d assumed that a couple of decimal places for reporting the bias current would be sufficient … however, the answer appears to require at least 3 decimal places! I spent ages going round in circles, assuming that I’d got my logic wrong (I hadn’t), only to come back to the original answer, which still wasn’t accepted when I plugged it in.

As I’m currently sitting on a train on my way down to London, I took the opportunity to try it again, having noticed a number of people on the forum say that the question was being fairly picky about what precision you used. Sure enough, going to 3, 4 or 5 decimal places on the answer sorted it for me. At least I should be able to finish the rest of the question fairly quickly now and get myself going on week 5. Thank goodness the Easter holidays are this weekend!

My experience does raise an interesting point however. Most practical electronic circuits operate within fairly relaxed tolerances. For example, typical resistors tend to deliver a value within a few percentage points of their stated resistance. Things becomes even more lax when you realise that you are usually selecting resistance values from either the E12 or E24 range and so can only approximate the ideal resistance you’d like to use. But that’s ok, as there is little difference in practice between a current of (say) 0.17499 A, 0.175A or 0.18A. Yet, in the case of this week’s homework, the first two values are considered correct but the third one is wrong (note – this isn’t my actual answer – just an illustration of the problem).

While I understand that the course designers are looking to ensure that students aren’t simply guessing their answers, I think that as a minimum the questions should state explicitly the number of decimal places or significant figures an answer is required to. That way, I could have spent more time this week enjoying the material, rather than assuming that I’d missed something when I’d worked my answer out originally.

Grrr.

Talk back to me ...