Hall effect sensors (by alaric)

At the Bristol Hackspace this evening, I powered up a Hall sensor.

The Allegro A1321 takes a 5v supply and outputs 2.5v - plus or minus 5mV per gauss of magnetic field in the sensor.

It turned out that small magnets don't produce many gauss, so I added a 741 op-amp with a gain of about 100 to get more useful output voltage deflection!

Here's the experimental setup:

WIth no nearby magnets, we read about 3v:

But with a very nearby magnet, we get a large deflection, which drops as we get the magnet many cm away, but is still significant:

Clearly we still get significant readings a good distance away, and the nonlinearity isn't too bad; it doesn't max out with the magnet very close but still varies many cm away.

For my next trick I'll make a few of this circuit on some veroboard and hook the outputs up to an AVR with several ADCs that I have lying around, and try comparing the results to measure position (and maybe orientation?) of a magnetic probe...

Wearable computers (by alaric)

One of my too many projects is to make a wearable computer.

Lots of people are interested in making wearables, but nobody's yet come up with one that hits a "sweet spot" of decent functionality along with it being unobtrusive enough to not be a pain.

Well, I'm a nerd, so I'm far happier to put up with obtrusiveness to get my pervasive cognitive-assistance fix... I've been fascinated by pervasive computers since I was a kid; I read about Steve Roberts' recumbent bicycle as a youngster, as well as plenty of fiction about brain implants and the like.

Read more »

High power LEDs (by alaric)

I've got a few 3W RGB LEDs that I've been meaning to play with, so over the Christmas break, I decided to hook 'em up to the bench PSU and have a play.

As I have but one variable bench PSU with current limiting, I could only easily light one LED at a time. I didn't have big enough resistors to build individual LED current regulation circuits - I just set the current limit on my PSU to 0.35A and cranked the voltage up until it maxed out, hooked up to one LED in turn.

They are certainly dazzlingly bright:

Since the green and blue LEDs both have the same forward voltage, I figured I might be able to drive them together by using a pair of resistors as a current splitter, and setting the PSU for 700mA, thus ensuring that 350mA went to each LED.

However, my 0.25W resistors started to smoke when I got to about 400mA, so I shut it off - if one of the resistors burnt out then the entire 400mA would go into the surviving LED, overloading it (until its resistor also burnt out), and possibly making the thing explode. I ended up with a nice pair of burnt-out resistors:

Which is a shame, because I'd love to see how bright the thing is at maximum, with all three LEDs going!

My lab partner was most impressed, and asked me lots of questions about current and voltage; I had to resist her demands to keep making things, so I could go inside and write this blog post:

Bench PSU made from an old PC power supply (by alaric)

My first PC compatible was an actual PC. As in, the original IBM PC - not even an XT. It died a death, but I salvaged the power supply from it, with the nice red switch, to be a bench PSU for my electronics experiments:

It supplies -12v, -5v, 0v, +5v, and +12v. Which served me well for some years, but these days, everything's 3.3v or 1.8v. And my termination setups - first wiring directly into the row of screw terminals on a breadboard, then later a bit of wood with metal strips nailed to it so I could attach croc-clips easily - all left something to be desired.

So I got busy and built this:

Complete with a power LED so I can see at a glance if it's live:

Inside it's quite simple. Most of the terminals are fed directly from the PSU via the black cable, but I added an extra wire for an earth connection to the chassis (which I checked was really earthed) for ESD wrist straps. I mounted 3.3v and 1.8v linear regulators on a bit of stripboard along with the LED and its series resistor, and hooked it into the 0v and +5v lines.

Job done. Now the last bit of major infrastructure I need set up is to get Ethernet into the lab so the PC I have in there for driving the dev boards can access the Internet and my version control system...

The workshop’s getting there (by alaric)

My good friend and colleague Andy came over to stay for a while, which meant I had 'entertaining a guest' as an excuse to do some of the fun stuff I've wanted to do for ages.

So we finished making the welding bench! I'd made the top and gotten part way through cutting the legs to length, so we finished cutting them and welded them in place, then welded extra flat strips around the bottom to make it reasonably rigid. It's all just tack welds, and it'll almost certainly need some diagonal struts added, but it stands upright and is surprisingly sturdy; I'm going to experiment with it a bit to see just how many more struts it needs. I'd also like to drill holes so I can mount my vice on it, too.

Then we cleared a space and moved my electronics workbench down into the workshop! This is great news, as it clears up space in the office (albeit revealing the piles of junk that were lurking beneath the bench), means all my tools are in one place (which is most convenient, as things were always in the wrong places), and creates more storage space in the little garage, meaning less stuff on the floor.

I've since reincarnated my power distribution rail, which I had when we lived in Ealing, but haven't used since; the idea being that it'd be good to do my electronics experiments on the end of a dedicated RCD so I don't trip the one in the house. It also splits the output into four circuits, each with a six amp circuit breaker (the smallest I could obtain easily).

I still need to get rid of a lot of waste cardboard that's sitting around, and we're still looking after Seth's motorbike, and there's still junk to be sorted - many things need to be elsewhere; the little garage isn't a place to store things we only use a few times a year, as we have the Big Yellow for that!

Although the lighting's not really good enough when I'm explaining the power rail, here's a quick video tour:

For my next trick, I'll stop procrastinating by building infrastructure, and get on with actually making myself a digital watch with an embedded ARM processor and colour dot-matrix LCD. Watch this space.