Battery-----Driver----ELWire  in order to make it work.  Now all of the
drivers I show use a 9 V DC batter  You can do others things, and use
other types of batteries, but this is the simplest. 

Depending on the amount of wire you want to run, will determine how many, and what kinds of drivers you will want to get.  There is a guidline on how many feet each driver will power next to the item number in the order form. 

 The current draw is miniscule.  I think it's something like 1 amp/800 feet or on that order.  Very low draw.

A few notes:
The intensity is based on voltage and the color is based on frequency.
I've used their inverters, and they work with 9V, 12V  and 18V battery
sources.  The last time I checked, they had 4 different inverters.  They
were low power/low frequency, low power high frequency, high power low frequency, high power high frequency.  The low power inverters are used for strands up to 15' and the high power ones can be used up to 50'.  We always pushed those limits and was able to run a 200' strand with a
12VDC to 110VAC RV inverter.  The base phosphor glows with a blue/aqua and the other colors are derived by colored coatings.  The colors I thought were best were the blue/aqua and the violet.  I didn't like the orange/gold very much, it seemed somewhat off colored.  Be careful, If left in the sun, the colored coatings will bleach out with time.  One very nice thing about this EL wire is it's lack of latency.  When the electricity is off, it does not continue to glow. 

I had the chance to speak to one of their engineers once for about a half
hour.  I ~wish~ I could remember his name.  Anyway, he explained to me
that they have a blue-green wire which will change colors as the
frequency varies.  He further explained that the blue-green was the only
color which would do this.  All of the other colors will simply light
brighter or dimmer.

I just dug out my data sheets from a while back.

The efficiency of the wire peaks at 100VAC but is still acceptable as low
as 60VAC and starts tapering off above 120VAC.

Power consumption is higher at higher frequencies.  It seems to average
200mw/meter.

Brightness at 100VAC is nearly 3 times greater at 1000hz compared to
200hz.

Current consumption is averaging around .002 Amps per meter.

This is a clue only to how they decided to do it.  SMT SOT-8 is a package type not a part #.  See if you can get a part #.  When I get a chance, I'll try and look at a Sipex web site.  It's probably a micro power low voltage oscillator that drives some sort of step up device (cheap switchmode or IF transformer probably).  Maybe the step-up's already on the chip...

Ok, it's later now - I've been to the web page.  Sipex makes Electroluminscent Lamp drivers.  I'm printing out some app. notes right now.  I'll look it all over.  Yes, this seems like the way to go if this stuff applies well to EL Wire (I'm sure it does).  I know some of this stuff is for watch dials but it looks like theres some others for more power.  I'll have to look over this data.

I have the feeling we'll be seeing a lot of this stuff in '99, should be fun to see what people do with it. Hint: Moto HC05's are cheap and have alot of pins that can be pulsed.

Hmmm, I'm pretty sure he/she's referring to a *series* of Motorola CMOS low voltage digital parts but I'm not sure.  HC05 alone is not a good # I'm pretty sure....

As far as sequencing goes, simple logic gate based circuits are probably the easiest thing unless there's a Gigsvillian who's really quick with something like the Basic Stamp.  I'm guessing tho that for each "switched segment" of wire, there should be a separate little power supply (which probably isn't as bad as it sounds specially after seeing the Sipex stuff).  I say this simply because it *may* be a pain to switch 200V even when there's almost no current involved (I've never tried it). I'd love to see a schematic for some of the switched stuff from last year just

Another tip:  The best way I found to strip this stuff is to use needle
nose pliers.  Grip the wire 1/4" from the end with the tip of the pliers
and pull.  This removed both the inner and outer plastic coatings,
leaving the two fine outer wires intact.  The easiest way to connect to
the finer outer wires is to use a small strip of copper foil wrapped
around the outer case near the stripped section.  Fold back the outer
wires over the copper foil and solder them to it.  This gives you a
solid base to which you can attach the power leads to the very fine
outer wires.  Take note, this stuff can bend, but not too much.  The
problems that we ran into most were shorts caused by excessive bending
of the wire.

That's an interesting comment on the color.  I found that there is a
notable color change with all of the wires.  It was more pronounced with
the blue/aqua though.  I think that it's like that because it's the base
color.  All of the other colors are derived by coating the wire with
another color.

It's not too difficult to switch those voltages.

If you want to take a HV DC power supply and produce switched HV AC from it, you can do it with 2-3 high-voltage transistors per circuit if you use a double-ended HV power supply (which is easy: two diodes and two capacitors on the output of an inverter) or 4-6 HV transistors on the output of a single-ended power supply.  This approach has the advantage that you can control the frequency and duty cycle of each segment individually, for color and brightness variation.

If you want to take the output of a HV AC power supply and switch it to various segments, use reed relays (cheap and good for millions of cycles), solid-state relays, or just a triac.  Triacs and some solid-state relays will only work if there is a brief off period in the inverter output while it changes polarity. If you're designing your own inverters, that's easy to arrange. If not, you need either a rectangle wave or sine wave inverter, not a square wave inverter.

Another approach is to use a single high-power low-voltage double-ended inverter driver connected to a bunch of inverter transformers, with each segment enabled by selectively grounding the other side of a transformer's primary winding with a triac or high-power MOS transistor.  This avoids any high-voltage switching, at the expense of one transformer/segment.  It also lets you select the voltage for each segment individually.

If you have lots of segments and don't need varying voltages/frequencies for them, I think the triac-switched HV AC output will end up being the cheapest, simplest way to go.  If you're tricky and add a delayed turnon synchronized with the AC drive, you should be able to dynamically control the brightness too.
    Here's a question that hasn't been asked before:  How does lifetime
and brightness vary with the type of driving waveform?  I'm thinking of
the 'modified' sine waves put out by most inverters nowadays.
 

We're still using 12V fixed-rated inverters designed to operate within a given range of wire length.  My next goal is to redesign the inverter and integrate it into my switching circuits.  If there are any magnetics people out there interested in tackling this project I'd like to talk to them.

I think it never actually burns out but it does burn down.

- The fibre's brightness is a function of the voltage applied.
- The colour (to a degree) varies with the frequency, and changing the frequency can yield some interesting effects.
 

 .... No such inverters that I'm aware of, but it should be pretty straightforward to build one.  If you want something highly portable, it'll take a dedicated-hardware design.
     If it's for a base camp, though - the easy way would be to use a programmable frequency generator to drive an inverter.  Some frequency generators have e.g.  serial ports.  You could also use the programmable squarewave generator that's built into all PCs - the speaker driver.  It's clocked at 1.193182 MHz.  You write a value to the divisor register and it produces a square wave by dividing the clock by the divisor.   This lets you produce a frequency from 1.193182 MHz down to 18.2 Hz, which I suspect encompasses the entire useful range :)
     If the lowest frequency of use is 50Hz or so, you could probably get away with modifying a generic 60Hz 12VDC->120VAC inverter: replace its clock circuitry with a feed from your programmable frequency generator.  Then put a step-up transformer on the output to bring the 120VAC up to whatever you want. Below 50Hz, the cores of the inverter transformer and step-up transformer are liable to saturate.