next topic »

[Reply #15 on:]

As of yet, a few things I have discovered.

1. The power supply was the culprit in disabling all audio signal. (It also powers my processor/preamp)
Not sure theoretically how an AC audio signal would cancel the digital signal through the processor's power(maybe inverted phase cancellation?)
I had to run a separate power supply to get it to work.

2. I have only been able to get any sound with ceramic disc caps. I've tried electrolytic caps back to back(+ - +), metal film caps, and polyester film caps.

3. The more resistance, the stronger the signal is. (Because its a bigger load?)

4. The capacitance didn't seem to change sound level. I used as small as a 1nf, and as much as 0.5uf with no noticable difference.

5. Multiple caps in series seemed to increase the sound level just barely. While multiple in parallel greatly decreased it.

As for the variable resistor, it seemed to not make a difference. And the RLC method is for a low pass filter isn't it? I could try an LC method, but not sure how I would incorporate the resistance. Also, I've read places that a guitar's pickup is basically an inductor in series with a cap and resistor in parallel(RLC)

The op-amp circuit you describe.. by which you mean as a buffer to keep a specific resistance? That may be a good idea to try out, but would be sensitive to different pedal layouts(different IO impedance).

There also seems to not be any 20hz high pass filters available in a small package. I found this schematic however:
http://www.eeweb.com/blog/circuit_projects/20hz-to-200hz-variable-high-pass-filter
May be useable, but the schematic states 15v input voltage, and the tl072 is rated ±15v.
Could I just swap for a lower voltage chip, or just run this with 9v?

Thanks for the help Far-seeker, you have really helped me out so far.

[Reply #16 on:]

1. The power supply was the culprit in disabling all audio signal. (It also powers my processor/preamp)
Not sure theoretically how an AC audio signal would cancel the digital signal through the processor's power(maybe inverted phase cancellation?)
I had to run a separate power supply to get it to work.

Hmm, I didn't think of that but in retrospect perhaps I should have thought to check the power supply.   However, I’m not sure about your proposed theory of inverted phase cancellation, instead my initial suspicion would that preamp was too much for the power supply to handle along with everything else.  In any case this is progress.

2. I have only been able to get any sound with ceramic disc caps. I've tried electrolytic caps back to back(+ - +), metal film caps, and polyester film caps.

Welcome to the world of real, rather than theoretical, components.    I know it hasn’t been discussed in this thread, but material and construction can cause a lot of subtle and not-so-subtle differences between two parts with the same basic ratings.  In this case the difference is probably due to the ripple current limitations and higher ESR for  film and electrolytic capacitors. 
I probably should have brought this up on my own.   However, I presumed that if you were using electrolytic capacitors they would be intended specifically for audio applications, not the general purpose ones.  However, it seems you don’t mind learning through experimentation, which is a good thing.

3. The more resistance, the stronger the signal is. (Because its a bigger load?)

 Yes, since we are mainly concerned with voltage levels and not power levels in this application.  If you want to know more about this, read up on the concept of  impedance bridging. 

4. The capacitance didn't seem to change sound level. I used as small as a 1nf, and as much as 0.5uf with no noticable difference.

Remember the intended purpose of the capacitor(s) in this application is to separate the variable signal from the constant DC and that will determine the target value.   In this case the goal is to have capacitors that block the DC without attenuating any of the audio frequencies when combined with a given load.   There should be a range of capacitances that will do  this, especially with a load that can vary by thousands of Ohms.

5. Multiple caps in series seemed to increase the sound level just barely. While multiple in parallel greatly decreased it.

Given the differences capacitors combine in series versus parallel that’s to be expected.
Furthermore, even if you are using multiple capacitors configured in different ways to achieve the same total capacitance there could be some small but detectable differences in circuit behavior.   Although capacitors can be combined to form an arbitrary effective value in the filter, there will be some differences between using one physical capacitor and multiple ones.  This is because in reality they don’t all charge and discharge at exactly the same rate.  In series they will tend to charge and discharge one after the other, in parallel each individual capacitor can charge and discharge (mostly) independently of one another.

As for the variable resistor, it seemed to not make a difference. And the RLC method is for a low pass filter isn't it? I could try an LC method, but not sure how I would incorporate the resistance. Also, I've read places that a guitar's pickup is basically an inductor in series with a cap and resistor in parallel(RLC)

You can configure an RLC circuit as either a high-pass or a low-pass filter.  Look Figure 9 & 10 from the link, the first is low-pass and the second is high-pass.  Did you notice how the positions of the capacitor and inductor are switched?
Also, remember for an AC signal the load is impedance, which is DC resistance plus the  phase varying contributions of the load’s capacitance and inductance, known as reactance.  This matters for the impedance bridging mentioned above.  However, when designing the filter it’s the load’s resistance, not the impedance, that will be used in the formulas.

The op-amp circuit you describe.. by which you mean as a buffer to keep a specific resistance? That may be a good idea to try out, but would be sensitive to different pedal layouts(different IO impedance).
There also seems to not be any 20hz high pass filters available in a small package. I found this schematic however:
http://www.eeweb.com/blog/circuit_projects/20hz-to-200hz-variable-high-pass-filter
May be useable, but the schematic states 15v input voltage, and the tl072 is rated ±15v.
Could I just swap for a lower voltage chip, or just run this with 9v?

This chip can be used at lower voltages and the datasheet shows some testing done at as low as ±5 V.  However, there will be modifications necessary to the values in this circuit, and proper setup of an active filter can be more a bit complicated than a passive one.
 

Thanks for the help Far-seeker, you have really helped me out so far.

No problem, but are you starting to see why I avoid overlaying AC signals on DC when I can?

[Reply #17 on:]

I see what you mean. Better to have tried and failed than to have wondered what if. 

I think I may default to using a stereo cable instead of mono. I'm trying to make something that would be easy to incorporate into guitars without much circuitry. Thought I'd try if it would just be something simple like a cap in line, you know?

But as for the stereo cable, I just need to make a sensing circuit for the pedalboard so it will stop supplying 9v power to the 2nd lead on the cable if a mono is plugged in. In this way, if the 1/4" stereo plug is used with a mono cable, the 2nd lug will connect to ground. I haven't decided on which method I may use: Use arduino to sense current, resistance, or voltage(I think voltage may be the easiest), or to just use a relay that will switch itself given the ground signal returns 9v, and then it will sustain itself there, then when powered off it will reset itself, or along the lines of that..
Can't really think of much else than that.
But if I used the arduino's analog input to determine the voltage(with a voltage divider of course), would the AC signal hurt the input if only for a short moment? I would like to not use an inductor if possible, unwanted distortion and low freq. loss through the arduino to ground.

But onto the communication/wireless. I picked up 4 nrf24l01+'s on ebay for <$8, so my evil idea is to communicate the ardy's with a pair(1 on pedalboard, 1 on guitar) and then transmit that directly on the audio wire rather than air. I just wonder, from the theory you have taught me thus far, I could run the 2.4ghz signal over the audio lines and the +9v line, without issue, right? Or would I be better off running it on the ground line?

Now for the possibly crazy idea. For wireless, run that into another pair of nrf24l01+'s, but with an antenna. I don't know how I would go about amplifying that signal just yet though. Off to google I go!

[Reply #18 on:]

I see what you mean. Better to have tried and failed than to have wondered what if.  

I think I may default to using a stereo cable instead of mono. I'm trying to make something that would be easy to incorporate into guitars without much circuitry. Thought I'd try if it would just be something simple like a cap in line, you know?

Well there are ways to do what want, if the load was a constant or near resistance/impeadance it would be easier.  However, as mentioned the easiest solution would be to keep the power an AC signal on separate wires. 

Edit: Yet, if you still want to overlay AC on the power line, perhaps this filter tutorial will be more helpful.  It's more in depth and much better presented than what I've discussed.

But as for the stereo cable, I just need to make a sensing circuit for the pedalboard so it will stop supplying 9v power to the 2nd lead on the cable if a mono is plugged in. In this way, if the 1/4" stereo plug is used with a mono cable, the 2nd lug will connect to ground. I haven't decided on which method I may use: Use arduino to sense current, resistance, or voltage(I think voltage may be the easiest), or to just use a relay that will switch itself given the ground signal returns 9v, and then it will sustain itself there, then when powered off it will reset itself, or along the lines of that..
Can't really think of much else than that.

Technically an Arduino can only measure voltage directly, but current and resistance can be determined through Ohm's Law.  So yes if you use an Arduino to detect the signal it's easiest to use voltage.  
However, why not just use a multi-wire cable with discrete pins arranged so it can’t be connected incorrectly, as I’ve discussed previously?  It doesn’t have to be a MIDI cable, I just thought that would be one of the easiest to find a replacement for given your situation.  

But if I used the arduino's analog input to determine the voltage(with a voltage divider of course), would the AC signal hurt the input if only for a short moment? I would like to not use an inductor if possible, unwanted distortion and low freq. loss through the arduino to ground.

First, all Arduino I/O pins are only rated to withstand about -0.5 V, so that is the hard limit for negative input.  Second, I'm not quite sure what the signal is now, the 1 V_AC overlayed on 9 V_DC or just the 1 V_AC.

But onto the communication/wireless. I picked up 4 nrf24l01+'s on ebay for <$8, so my evil idea is to communicate the ardy's with a pair(1 on pedalboard, 1 on guitar) and then transmit that directly on the audio wire rather than air. I just wonder, from the theory you have taught me thus far, I could run the 2.4ghz signal over the audio lines and the +9v line, without issue, right? Or would I be better off running it on the ground line?

OK, I try to be very careful about making absolutist statements, but I want to make this clear it’s NEVER a good idea to put an AC signal on what the rest of the circuit uses as a ground!   Ground is usually the shared reference, so if its voltage level is changing it can affect everything else that goes on in the circuit.  Also, because voltage is an innately relative concept (i.e. it’s the difference in electrical potential of two points), under some circumstances it can damage logic ICs, like microprocessors.  For example, the maximum supply voltage for an ATmega328P used on the Uno is +6 V_DC, so you have a supply that’s giving it a very stable 5.2 _DC, or 5.2 Volts above ground.  Just a 1 V_AC signal on ground will have the resulting power effectively swing between 4.2 V_DC (when the AC is at +1 V) and 6.2 (when the AC is at -1 V), not because the supply’s output change but because the ground changed!  What’s more a voltage regulator won’t help, because the voltage regulator will be using the same ground as the rest of the circuit.
If you want to run the signal on a cable without any issues, the best way is for it to have its own wire.

Now for the possibly crazy idea. For wireless, run that into another pair of nrf24l01+'s, but with an antenna. I don't know how I would go about amplifying that signal just yet though. Off to google I go!

Well there are quite a few options for wireless communication using Arduinos.  Notice that if you scroll down a little, the nRF24L01+  radios have two entries…  

[Reply #19 on:]

Hey Far-seeker,

Almost a year later (correct me if I'm wrong, but I figured it would be better to revive this thread of mine rather then start a new one and have to back-reference to this one) I finally got around to working with stuff again. The issue I was having with the DC blocking out the sound was that I need decoupling capacitors on the ground side too; but I got it to work!

So it would be like the attached schematic: Guitar Power Over Audio.jpg (it's ugly, I know).

Question:
Would I need to get caps rated for at least 9v in this case, or would it not matter as it's being used only to block power, not absorb it?
Still a little confused on the Arduino and power supply side; would/should I use an inductor or diodes to filter out the AC signal, or will it not matter?
Don't want to damage the Arduino, and I may need to filter the 9v anyway otherwise it could bleed into the circuitry that will be processing the guitar's signal.

Thanks,
Matt