Go Down

Topic: EL Wire (Read 9 times) previous topic - next topic

tazma

hello, is
may work the wire
without investing?
or power
to inverter
recycled materials?

Paul Stoffregen

#6
Dec 17, 2011, 10:45 am Last Edit: Dec 17, 2011, 10:57 am by Paul Stoffregen Reason: 1
I would advise you NOT to use Sparkfun's board for 50 feet of wire!

Several months ago, I worked on a big project with 114 pieces of el wire, each 8 feet.  Here's my blog about it:

http://dorkbotpdx.org/blog/paul/hand_eye_supplys_starlight_parade_float_wins_award

Unfortunately, we used those Sparkfun controllers, and they were nothing but trouble!  Here is a blog post I wrote about all the technical problems.

http://dorkbotpdx.org/blog/paul/sparkfun_el_sequencer_trouble

The Sparkfun board probably isn't so bad for a tiny project, where you have 8 pieces of wire ranging from several inches to maybe a couple feet.  We used 8 feet on each channel, and it was a new higher brightness wire which has more capacitance.  So 8 feet of this wire is like using perhaps 40-some feet of "ordinary" wire (though the trend seems to be newer wires are brighter and heavier loads).

The really terrible problem with those Sparkfun designs is you can't switch at the AC waveform zero crossings.  With el wire (or any purely capacitive load), switching near the zero cross point is critically important.  If you trigger the triac near the crest of the waveform, you're suddenly applying a massive voltage to a capacitor.  The current flow is incredibly high for a very brief moment.  For short, older-style wires, it's not enough to damage the triac.  But 50 feet is quite a substantial load.

If you search the web for info about triacs, everything is regarding resistive and inductive loads, at 50 or 60 Hz.  The rules are entirely different for a capacitive load.  With an inductive or resistive load, you can switch at any point in the waveform, and dimmers do exactly that to control effective brightness.  The problems with inductance come in regarding excessive rates of voltage change due to the inductance.  But with capacitance, the issues are completely opposite.  Switching capacitors causes huge surges in current rather than incredibly fast changes in voltage.  Small triacs simply can't handle those huge peak currents reliably, so you must switch at the zero crossings.

The simplest solution, and what I believe Sparkfun is eventually planning, involves using an optocoupler which has a zero cross switching detector built in.  If you're going to control 50 feet of wire, do yourself a big favor and look for a solution with zero cross switching built in.  You could also build a zero cross detector circuit and interface it to an interrupt pin.  But that solution is a bit risky, because a bug in your code could stress or destroy the triacs.  With the Sparkfun design, you don't even have any way to know when the zero cross is near, so your only option is to just trigger the triac at some random moment (relative to the AC waveform) and hope you don't stress the triac too much.  It's a very poor design.

If you go with Sparkfun's board, you'll have nothing but trouble using such long wires.


tazma

hello
thanks for the response
what I want is to ride my
own inverter
therefore
wonder if it is possible
and if anyone has any tutorial
  I have to mount a tron costume
the wire

Go Up