Hello all - I am working on a project where I want to have EL wire light up when I strum the strings on my guitar, with strum intensity / volume correlating with brightness of the EL wires. I need a few tips from those more experienced with electronics than I (really, I am trying to use this project as a way to teach myself more about electronics).
I have purchased an EL sequencer (which contains an ATmega328, so it's bascially an arduino with triacs built-in) from Sparkfun, and plenty of EL wire to mount on the guitar. I have assembled the unit using some of the tips posted online (for example, see http://bildr.org/2011/06/el-sequencer/ and http://blog.makezine.com/archive/2010/04/programming-el-wire-fashion.html ), and the wires will light up as the sketch from Make specifies. So, it works prior to being connected to the guitar.
Here are my two main problems at the moment:
How should I connect it to the guitar, so that the guitar output (typically 0-200mVAC) will be properly read by the analog inputs in the E: Sequencer? Assume that I know nothing about this. It would seem that I would want to isolate the guitar signal (I know how and where to get that in the guitar, which I have also just refinished and rewired) and convert it to a max ~5VDC signal. But is the EL sequencer sensitive enough to measure the voltage if it is simply changed to DC, without being amplified? Long story short, I'm looking for a circuit diagram and a list of components here. You know, the whole thing, basically.
The EL wire and sequencer puts out a nice ~1kHz signal that can be detected by not only the guitar's pickups, but also the cable! That's right - I can hear the whine without even connecting the guitar, and once I plug it in, it gets a little louder. So I need to build a filter (which I assume I would connect between the guitar cable and guitar amp) that will selectively knock out this tone, but leave others unaffected (like a one band EQ). What type of filter is best, and how to I build it? I assume that this is simple enough, but as to what resistors / caps / etc I would use, I'm just not knowledgeable enough to figure out at this point.
Thanks for any help you can give me folks - I'll post pics / vids when it's completed. And, fyi, the guitar is a 1999 Platinum series BC Rich Warlock. A fun guitar, hopefully soon to be more fun.
What you are asking for is a notch filter sometimes called a band stop filter.
However it is important as to where you place this filter as it can't affect any interference pickup after the filter. It is not as easy as you may thing doing this. The exact values of R&C depends on the exact frequency you want to blot out. You might have to make two or three of the resistors from pots so you can tweek the exact cut off frequency.
I'm working on an amplifier circuit (gain of about 82 dB or 12,000 in my case) that maybe you could adapt for your purposes:
You could change the 1M feedback resistors to a lower value if you need less gain which I guess you do if you're really getting 200 mV in.
The circuit uses the LM324 quad op-amp chip in a single-ended supply mode, with a virtual ground set at 2.5 volts by the 100K/100K voltage divider. Gain is in two stages to keep the overall bandwidth high enough--should be about 10 KHz in this case, more if the gain is lowered.
The voltage sent to the Arduino is conveniently between about 0-5V with 2.5 being the virtual ground level.
As for the noise issue, maybe this device would help which I'm also looking at but haven't actually tried yet:
Tiny audio optocoupler is about the size of a pencil eraser and has 4 leads. These are used extensively in audio compressors, audio level controls, audio limiters, expanders/noise gates, guitar tremolo effects, guitar amplifiers and music effect boxes. Inside the black epoxy case is a 2 lead photocell (not a photodiode or photo-transistor) and a 2 lead LED. This is why the device finds extensive use in audio applications (photocell instead of photodiode). The LED requires about 10-16mA current and the resistance of the photocell varies from 150K up to megohms when the LED is off to under 400ohms when the LED is on.
Joe
PS pardon the typo of two caps designated C4 in the diagram!
Thanks for your help - some additional notes from my end:
First off, as I learn more about this process, I'll post the information here. Hopefully, that will lead to more precise answers.
I think I included too many zeroes in my pickup AC voltage. I'm going to spend some time with the guitar this afternoon so that I can get a more accurate read on the voltage levels. I think the pickups are BC Rich's version of passive EMG's, at least in styling. Not really sure on the voltage output yet, as I haven't measured it, and all pickups are different.
I was hoping to use the analogRead function to read the guitar's output, and then set levels for turning the wires on or off. If the signal exceeded that level, the wires would turn on.
After I figure out what frequency(ies) are put out by the arduino/sequencer and the wires, I'll post them here. I was not sure exactly which components to add into the band stop filter. That obviously is needed to pick the proper components. I have found that different sequencers (I have a few) put out different frequencies. They are all around 1 kHz though, I believe. The filter will be between the amp and the guitar cable, hopefully minimizing any chance to get the whine into the signal chain, unfiltered.
Would that amp circuit be appropriate here? I was thinking just trying a rectifier and seeing if the arduino was sensitive enough to be able to tell a strum from silence. If not, then a simple DC amplifier to bring the max level into the AC range.
I am well aware that the 1 kHz is in the audio spectrum - It's around a C6 (two octaves above middle C - high E string, 8th fret). Personally, I don't mind losing that one note - this guitar is supposed to be fun.
I need the arduino for its programming capabilities, and just because I want to learn how to use it, or at least this iteration of it. Also, the EL wires give off this frequency independent of whether or not an arduino is used. Lastly, and I'm sorry if this wasn't clear, the guitar cable itself is enough to pick up the hum. Even if the guitar is not hooked up, you can still hear the hum thought the amp. I would argue that this makes the selection of pickup type somewhat irrelevant - hence the need for a filter.
I would love it if there was a more elegant solution. Maybe my cable does not have enough shielding - combine a better shielded-cable with a different type of pick up? But for this project, that isn't an option.
Also, remember - to make the EL wire light, you need to convert the battery's low voltage DC up to high voltage AC. Hence the triacs, etc.
Hopefully more testing will help narrow things down some.
I would argue that this makes the selection of pickup type somewhat irrelevant
Not true. With nothing attached to the amplifier input your amplifier is a high impedance input. This is very susceptible to picking up interference. With a low impedance pickup the interference is reduced considerably.
I think you have a bit of an idealised view of a filter. While notches are very sharp it will not affect just one note. As I said before you will have to locate any filter inside the amplifier to try an minimise post filtering pickup.
