h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Having pretty much finalised my new kick-ass cave-light design, (high efficiency, built-in redundancy, variable power, twin Luxeon LEDs, wide and spot beam, simple and advanced operating modes, etc.), streamlined the production process, and sold versions relatively cheaply to most of my closer caving mates (hoping to move into more profitable wider distribution soon), I'm trying to work out what to make next.

I started toying with a smart charger for cavelight battery packs, but got distracted into making a rough-and-ready amp-hour meter, since I have multiple uses for one, and it should only take a couple of days to knock one together. However, I can feel feature creep starting already on the amphour meter design, which is almost unavoidable given the possibilities opened up by circuits built round microcontrollers.
The cave light started out rather simpler than it ended up, though something I'm expecting other people (especially people I like) to rely on as their main source of underground light is probably better for the inclusion of the extra redundancy, and for some of the user-interface simplifications.

It *would* be great to have a battery simulator that I could use to check out the lamps - being able to fast-forward and rewind in time to the voltage/internal resistance performance of multiple battery packs could be even more useful than the amphour meter, but a bit more complicated to make, and it would really require the existence of a simple datalogger to use to record voltages and currents over time.

A datalogger would make testing virtually all the other things easier, but would probably be trickier to make than all the other things I want to build at the moment, and is probably something best worked up to gradually.

Datalogger...
Your pc has a soundcard? Can't you use that. I've seen various things use the soundcard for other than making sound, eg turn it into a basic oscilloscope ( http://home.conceptsfa.nl/~rona/e-guide/main/downloads/oscilloscope.html or http://www.muzique.com/schem/tips.htm ) or use it for all sorts of amateur radio applications.

I'll have a look at that possibility - cheers.
In the long term, I'd like to build a self-contained unit that can connect to my PC later, but it could be useful to have a (free) realtime version for the meantime.

Rather more mundane stuff today, fixing my car's headlight switching.
A while ago, some little bit of camming plastic must have broken off inside the indicator/headlight switch, which meant that though I could still flash main beam if the headlights were off or on sidelight, when they were twisted up to dip beam, I couldn't flip them onto main beam.

I worked out a temporary fix - twisting back to sidelight, puling to flash main beam and then twisting back to dip beam while still pulling resulted in the switch locking on main beam instead. To switch back required twisting to sidelight than back to dip again. A little fiddly, but I got quite quick at it. Still not really ideal, though.

Anyway, with an MOT imminent, I called the Renault dealer to enquire about a new switch, and I thought the guy was reading out a long part number until I realised it was actually the price he was quoting. £85+ for a switch really is taking the &!$$.
A quick poke with a meter around the defective switch resulted in a decision to go for a homebrew repair. One trip to the local electronic store to buy a 12V/30A relay (£1-49) and a few crimp terminals, and a scrabble round my parts boxes for a bit of wire, a toggle switch and some scrap aluminium to mount it on provided all the bits I needed. 10 minutes thought, and 15 mins crimping and soldering resulted in a fully working (and perfectly reversible) repair for about the price of a pint.

Fingers crossed for the MOT next week.

Yesterday I was for the second time defeated in trying to change one of the dipped headlamp bulbs on MC's car.
So much for simple roadside repair. The two options were remove the grill and headlamp assembly (fine until I realised that there must be a bolt holding it in from underneath that would need the car up on a ramp to get at!) or remove the battery which after faffing with the other method I didn't fancy.
The user guide for the car shows a nice simple way. Shame it cannot be done with the particular engine size that MC has got.

Carsdon't talk to me about cars

Hey, if you're coming to the meet Potholer, you might be able to flog a lamp to Bruce .

I'd hoped to attend, but only found out last night I wasn't needed elsewhere for family transport duties, so hadn't been able to plan either accommodation or cheap transport.

After my unexpected New Year success (I'm a good engineer, but a crap salesman, and only expected to make one pre-arranged sale), I need to build a few more.
Got to decide whether to just build a couple for now (which I have all the components for), or more (which will involve buying £200 worth of LEDs to get a reasonable discount). I'll probably end up taking the plunge and hoping I can sell another dozen, since I could really do with having a few units out as samples in various places, and a few for me to be able to sell.

Hmmm. I've just realised a cave lamp redesign may be a good idea.
A user with bad battery connections that are typical of old H******e packs has suggested a couple of changes. One seemed at first like it would have required a completely differently operating unit, but I've realised that I may have worked out how to add very useful extra functionality on the current design without having to modify my circuit boards layout, which would have been a *real* pain.

In fact, the changes would virtually all be in software, and the extra hardware needed would be simple and extremely cheap, but would result in a device hugely better when powered through unreliable battery connections, and no more complicated in operation.