What is causing the square drop in my guitar effect's sine wave?

[danieluyter]

TL;DR

I built a guitar (fuzz) effect, and analysed the sine wave through an oscilloscope. Wondering what the wave means and how it got to be the way it is.

I built a guitar (fuzz) effect, and analysed the sine wave from a tone generator through an oscilloscope. Wondering what the wave means and how it got to be the way it is. I will attach a picture of the oscilloscope screen + the circuit I used for the Fuzz effect. My question is what the cause of the square drop in the sine at the top is. Any information/help would be immensely appreciated!

 

[berkeman]

Paging @DennisN

 

[DennisN]

I built a guitar (fuzz) effect

That's cool! I was actually taking a pause from testing my new Zoom guitar multieffect box when I noticed an alert in this thread from @berkeman.

Wondering what the wave means

Hmm... my first impression that it looks like a voltage drop due to a part in the circuit going over a certain "limit" voltage.

Wondering what the wave means and how it got to be the way it is. I will attach a picture of the oscilloscope screen + the circuit I used for the Fuzz effect. My question is what the cause of the square drop in the sine at the top is. Any information/help would be immensely appreciated!

Since the voltage drop only occurs in the positive region of the sine wave, I think it is the diode (1N914) that is the origin; it is connected to the base of the transistor so it probably alters the transistor base-to-emitter voltage (V_BE). But it was a while since I last did electronics myself, so I can't say for sure, but that is what I think.

If I were you I would try to use SPICE (software that does electronics simulations and analysis) or some other alternatives to it. With such software you can quite easily modify and experiment with circuits. I have used some free SPICE-like software for electronics, but I don't remember the name of it at the moment.

In short, since the sine wave is significantly altered by the circuit, you have succeeded in building some sort of distorsion effect. But I don't know how it sounds, of course.

 

[DennisN]

I have used some free SPICE-like software for electronics, but I don't remember the name of it at the moment.

It was AnaSoft SuperSpice I used (though it's long ago now), and it is now freeware. Download page is here.

 

[DennisN]

I think it is the diode (1N914) that is the origin

Oh, by the way, diodes are very often used in distorsion effects, and often in pairs, like in the famous Tube Screamer. See the first diagram on this page, section "Clipping amp"; it is the two diodes D1 and D2 that do the actual clipping of the signal.

 

[danieluyter]

Thank you! Could it be because of a single diode that only the top of the sine wave is clipped?

 

[DennisN]

Could it be because of a single diode that only the top of the sine wave is clipped?

That is what I think. Though the diode in the circuit also works together with the transistor. I've built a tubescreamer once, and when I was experimenting with different diodes I noticed that using just one diode in a small circuit can introduce subtle distorsion. https://www.electronics-tutorials.ws/diode/diode-clipping-circuits.html describing the basic clipping in simple diode circuits (without transistors).

 

[danieluyter]

That is what I think. Though the diode in the circuit also works together with the transistor. I've built a tubescreamer once, and when I was experimenting with different diodes I noticed that using just one diode can introduce subtle distorsion. https://www.electronics-tutorials.ws/diode/diode-clipping-circuits.html describing the basic clipping in simple diode circuits (without transistors).

Thank you so much, doing this for a physics project, but it got me into building pedals! Planning to make a more advanced fuzz pedal!

 

[danieluyter]

Just one more thing, would you have an explanation for the sort of dug out flat part in the middle? From what i understand, signal clipping just limits the amplitude of the sine wave, but in my case the clipped part is lower than the rest of the peak.

 

[DennisN]

Thank you so much, doing this for a physics project, but it got me into building pedals! Planning to make a more advanced fuzz pedal!

That sounds cool, I wish you luck! There are a lot of audio effect circuit diagrams available on internet.

Just one more thing, would you have an explanation for the sort of dug out flat part in the middle?

As I said, I think it is the diode working together with the transistor via the base, altering the base-emitter voltage of the transistor V_BE. I can't do an exact analysis at the moment, one thing that is needed for an analysis is the data sheets for the particular components, e.g. diode voltage data, transistor base voltage data etc. That's why software like SPICE or AnaSoft SuperSpice can be very useful for things like this. You drop components into a virtual circuit and then you can do various analyses, in this case small signal AC analysis.

 

[tech99]

Summary:: I built a guitar (fuzz) effect, and analysed the sine wave through an oscilloscope. Wondering what the wave means and how it got to be the way it is.

The transistor might be going into high frequency oscillation when the collector current exceeds a certain value. This will build up negative bias on the capacitor and block the amplifier until it discharges.

 

[gneill]

Interesting little gadget. Looks like the only source of base bias for the transistor is either the diode or the source signal. I doubt that the transistor is acting as an audio amplifier, but rather as a switch. When the source signal drives the base above the junction threshold the transistor will quickly turn on and slam the collector voltage down to nearly ground potential. That would explain the sharp-edged "bite" in the signal.

So essentially the real signal path from the input to the output is via the diode! As the input signal varies, the current through the diode and 100 k resistor will vary, and no signal inversion will occur as it would if the transistor were behaving as an amplifier.

I took a few minutes and put the circuit into LTSpice.:

I had to use "best judgement" for the battery and capacitor values as none were supplied on the original circuit. Note that I disconnected the load to rule out any possible effects by the output capacitor. Turns out the load didn't affect the results anyways, but I thought it prudent to check.

Here's some voltage plots for the circuit. The input signal is in red. The voltage on the collector is in green.

The input signal is a 1 V 600 Hz sinewave.

Notice how the output follows the input, rather than being an amplified inversion of it.

 

[DennisN]

@danieluyter,

the two capacitors (green in your schematic and C1 and C2 in the schematic by @gneill ) are common in such transistor circuits, and they are usually simply input (C1) and output (C2) filters, more specifically two highpass filters (see e.g here and https://www.allaboutcircuits.com/textbook/alternating-current/chpt-8/high-pass-filters/), which block DC (direct current).

And nice analysis, by the way, @gneill !

 

[Klystron]

My study group in electronics designed a bass fuzz tone for a musician member almost 50 yrs ago based on a ring modulator IMS. Wild sound. This diagram shows one possible diode config but the trick was to match LC components (not shown) to a Gibson bass guitar. Diagram from here.

 

[DennisN]

And here is the core of the classic Tube Screamer, the clipping amplifier part, which consists of an operational amplifier with a pair of diodes in the feedback loop. It is the diodes that do the actual clipping; one diode clips the positive part of the signal while the other clips the negative part. There is also a logarithmic potentiometer (500K) which controls the amount of distorsion (drive). C2 is an input filtering capacitor and C3 and C4 are distorsion filtering capacitors which control which frequencies that are distorted.

Source: http://www.electrosmash.com/tube-screamer-analysis#clip-stage

 

[danieluyter]

Thanks everybody!