gswce.net

Just about to plunge into the marking season here, and this can be good and bad… if the marking goes well I usually reward myself by doing a bit of programming between every ten scripts or so. On the other hand, if I don’t get myself into the zone, and the marking is going really slowly, I’ll probably not get the time to do another demo in the next month. Not really a big issue… I don’t actually have one in mind to do at the moment.

I did attend a lecture by a colleague recently in which he was demonstrating crosstalk… and that struck me as a fairly quick demo to get up and running. Might be based on the existing transmission line demo, which itself might get a bit of polishing up… it’s not exactly a model of user-friendliness at the moment.

All I have to show for this week is a slight mod to the analogue ac simulator. The frequency response curve used to be rather unintelligent; it just sampled the outputs over a set series of frequencies equally spaced between the smallest and largest frequencies currently displayed. This resulted in some rather unhelpful and inaccurate graphs when there were some high Q-factors being used. The new version concentrates the readings where the output is changing the most rapidly, which helps give much more accurate-looking readings in these cases. So I’ll be able to stop apologising about these misleading graphs when I use this myself.

Right – that’s all for a little while now. Tomorrow a huge pile of scripts hits my desk, and life will get a bit strange for a while…

In theory it’s a bank holiday here today, which means I don’t have to work. In practice, of course, there’s no way I could do this job if I didn’t work evenings, weekends and bank holidays, so today I’ve been moderating some exam questions and preparing some mini-lectures for the labs later this week. However, I’ve allowed myself some time off, just enough to get the new transmission line demo into some sort of working state.

It’s not very polished this one, and I’ve no doubt I’ll be returning to it later, but I thought I’d put it up here so I could, in some sense, call it “finished for now” and get on with something else. I won’t be needing it myself until next year.

It’s my first attempt at a transient circuit simulator. I’d never done one of these before, and after getting myself rather tied in knots trying to think of a good way to do this, and searching the Internet rather fruitlessly for some ideas, a student lent me a copy of “Electronic Circuit and System Simulation Methods” by Lawrence Pillage, Ronald Rohrer and Chandramouli Visweswariah. It seems to be a difficult book to get hold of for a sensible price, which is a shame, since it was very useful. To the authors: thanks, just reading the first chapter was enough to solve my problem.

Two chapters of the book now available… progress. Slow, admittedly, but at least it is progress.

With a few hours to go, I’ve finally managed to persuade myself to put the first chapter of the proposed book up here. It’s a brief introduction to the decibel, part of the “What You Need To Know About…” section. You can find it here.

Phew.

Another milestone passed, another deadline (just about) hit. I’ll see what I can do about getting one chapter per month up from now on.

I wonder if anyone will find these and start commenting…

It’s been a while since the last post – and I had such hopes for making progress with this site over Easter. Sadly the usual story wrapped itself around my hopes, and I find myself here with four days left before term starts to try and do something. It’s been a while since the last demo as well; but I’ve not been entirely idle on this front – the Analogue AC Simulator has been upgraded, a few bugs removed, a couple of new features (including a greater frequency range) some more preset circuits added (the ones I need for the labs here next term: in particular a voltage-controlled voltage source filter and Baxendall tone control circuits). The one in the works is a transmission line demo showing the effect of terminations of reflections and pulse shapes, but I won’t need that one myself until next year, so I might start and finish another one before that one’s ready to release.

The thing I’m more disappointed about is the lack of progress on getting chapters of the book up here. I guess I could just stick a couple of chapters up in the state that they are in now – maybe with “draft” written all over them. After all the entire idea was to get feedback from the readers to improve things and correct the mistakes. As long as I put on a suitable disclaimer… I’ll see what I can come up with.

I’ve now lost count of the number of times I’ve wanted to know how to disable comments on pages, but leave comments enabled on blog posts. So I thought I’d start to keep a record here.

The method seems to vary with which theme is being used. Currently I’m using the twenty-ten themes, and here unchecking the “Allow people to post comments on new articles” box in the Settings/Discussion page turns off comments on all new items.

To do it on for pages but not posts, what seems to work is editing /public_html/wp-content/themes/twentyten/loop-page.php and commenting out the comments_template line like this:

<?php //comments_template( ”, true ); ?>

I’ve no doubt I’ll return to this post and refine it when I change themes, or upgrade WordPress to the next release.

I’ve just installed a captcha. I was getting more than a little irritated by the large volume of spam I was getting, and I’ve finally given in. Akismet was one possibility, and while that won’t put anyone off commenting because it’s transparent to the commenter, it costs money. A captcha is free, but might put some people off – on the other hand, since 100% of all comments on this site have been spam so far, there isn’t really much to lose.

What else? Well, I’ve decided that by the time of the first anniversary of this site (which will be in April), I’ll make a serious attempt to publish it. That means making some videos to illustrate the demos. This might work (to my astonishment, one little YouTube video I did a couple of years ago has had 5000 hits, and I only put that up to show my nephew how to solve a puzzle). I also aim to put the first few chapters of the book up, register the demos on some OER (open educational resource) indexing sites like Jorum and OER Commons, and investigate Wikieducator.

One issue I’ll have to sort out is the licensing… currently the demos have a rather odd licensing agreement, due to the fact that they are live web applications, rather than anything you can download. So it doesn’t really make sense to release them under any of the usual licensing agreements, there’s nothing really to release. I’ll have to investigate this one a bit more.

What else is new? Well, I’m just finishing off a demo about the token bucket algorithm for a module I’m teaching on Internet Protocols, and I’m finding that it wouldn’t be difficult to expand this to cover a range of other QoS routeing and queueing issues as well, so I might keep adding things to this for a week or so before releasing it. After that, one on multicast routeing would be useful as well, but I doubt I’ll get time to finish that one before I have to give the lectures this year.

The Karnaugh map demo is done now, finally. Another one that took much longer than I thought it would, as it turns out that designing and solving Karnaugh maps isn’t as straightforward as it looked at first sight. That’s why when you find a solution as good as the built-in solver can find, the demo just says “Optimum?” with a question mark – I cannot prove that the solver always finds the optimum solution in every case.

It’s been an interesting first year. Most of the demos are translations of some of my old VB6 demos, but the two most advanced ones here, the digital and analogue simulators, are new. So are the Oscilloscope demo, the Fourier demo, the Veroboard demo, the Log/Lin Graph demo and the Resistor Colour Code demo. So my initial target of one demo per month seems to be sustainable, even after the old ones have been translated.

What hasn’t happened at all is any progress on the chapters of the book. I haven’t posted a single chapter yet, despite several of them being very close to a releasable state. Why not? Well, I suspect some of the reason is that since I no longer teach communications, I don’t think many people would be interested; certainly the levels of feedback from this website suggest that the numbers of people who might find them useful would be very low. However, I do have a lot of students who I know appreciate the demos. Another reason is that I had a plan to integrate the demos with the notes chapters; but that’s going to take a lot of time that I can’t really justify.

This might be about to change: my bosses here seem to be trying to get me to go back to my old job of teaching communication theory, which would mean more time developing notes on communications theory. Just when I thought I’d left that behind and was free to start a new life. Oh well…

If all goes according to plan, I’ll put the analogue simulator up here tomorrow. It’s just about got op-amps sorted out now, although it took a little longer than expected to deal with the saturating states with positive feedback. (One of those things that’s obvious once you see it, but I’d never thought of before: in many cases there is an unstable equilibrium point in an op-amp with positive feedback; and the program had to be bright enough to detect that, and kick the solver off to find a stable solution instead.)

The simulator still doesn’t know about diodes or transistors, that’s the next step. However, since I don’t teach transistors until next November, there’s no hurry for this from my point of view. I am keen to get on with some other demos – the Karnaugh map one for a start, and a breadboard layout tool. (One of the most common mistakes in the lab is students just not building on the breadboard the circuit in the lab scripts, and I think an automated and fun way for them to plan their layouts before they build them would help a lot.)

There are just a few sample circuits to program as built-in demos, and that’s it for now. Brand new and not much testing done, so any comments particularly welcome: there are very likely to be some bugs.

Update… damn, a load of **** just landed on my desk that I have to deal with urgently. So it’ll take a bit longer. Oh well… maybe I can get this one finished and on-line before Christmas. I’ll certainly try.

Update… right, that’s it. Well, it’s not perfect, but it’ll have to do for now. I think it’s good enough for what I need for my first-year course, anyway. I hope someone out there finds it useful as well.

There haven’t been any new demos for a little while. There are a few reasons for this (it’s term time, so there’s a lot of work to do; I was given to couple of PhD theses to examine and that takes a huge amount of my spare time), but one major reason is that the one I’m working on is rather large, and is likely to take a bit longer before I’m happy enough to release it. However, today, it sat up, compiled, accepted its first task, and got the answer right! The analogue simulator is born! It’s got a lot to learn (it only knows about resistors and voltage and current sources so far), but the basic engine that solves the analogue circuits does seem to be working. I’m delighted, and had to tell someone.

There’s a lot left to do: it keeps writing currents as “2m4 A” rather than “2.4 mA” at the moment as it’s using the same print-formatting routines that the passive components use; it has a habit of writing negative currents when a positive current with the arrow in the other direction might be more friendly; and it needs to be able to deal with op-amps in saturation. Hopefully all that will be done by next week, and I might be able to release an initial version then.

After that, I’m going to try and get diodes and transistors into it. Non-linear components present rather a greater challenge, so that might take a while… but it’s a good start.

I’ve been tearing my hair out recently trying to figure out how to make some video tutorials for the demos without spending money I don’t have. This should both be more useful, and faster, than writing some more conventional help pages. At least that’s what I thought. This post is a record of my current thoughts and attempts so far…

Firstly, I’m trying CamStudio to capture the screen while I demonstrate what the programs can do. I found several problems with this. The most worrying being that when I first tried it, it generated .wav files over 2 Gbyte long (I’m not joking), and refused to record any video more than about 20 seconds long. The program needs careful setting up to work well – the default settings did not work at all for me.

For the record, these are the settings I’m currently having some success with:

Region: Select “Window”. This restricts recording to just one window, and seems to work well.

Options -> Video Options: Select “Indio video 5.10 codec”, quality about 80%, capture frames every 40 milliseconds (25 frames/second). Don’t use the Xvid MPEG4 codec (it didn’t work at all for me), and don’t use the Microsoft Video 1 codec (see http://screencasttutorial.org/18/best-settings-for-camstudio-to-sync-audio-and-video-28)

Options -> Audio Options: Select “AK5370” as the capture device (this is my logitech desk microphone). Set volume to maximum. Leave recording format as 22.05 kHz stereo 16-bit (mono doesn’t seem to work for me). Check the “Use MCI Recording” box, this means that CamStudio isn’t compressing the audio as well, which makes its job a bit easier. (Can always compress the audio later.)

Options -> Program Options -> Directory for recording: Set away from my C: drive that’s rapidly running out of space.

Don’t use any effects.

This is for an XP machine with CamStudio recorder v2.6.