Some simple improvements to the Green Ringer

[Mark Hammer]

I was putting together a little thank-you for a fellow who sent me a lovely retirement present, and went with a Green Ringer, since I had the sense he did not have any octaver.  On my own unit, I installed a FET booster as the front end, but didn't have room for it in the 1590A enclosure I decided to go with.

Looking at the schematic, it occurred to me that more gain could be achieved from the simple front end it used.  I added a path in parallel with R5, shown here, to do just that.  I put in a 10k resistor in series with a 10uf cap to ground, and used a 3-position toggle to place an 8k2 or 1k5 resistor in parallel with the 10k, yielding moderate and higher gain from Q1.

I used Schottky (1N5817 in this instance) diodes for D1-D2 since they yield a more dependable octave-up in my experience. and also a 100pf cap in parallel with R1, as Earthquaker des with their Tentacle clone of the Ringer.

Finally, as I've noted in other posts, I stuck a back-to-back pair of 1N914 diodes in place of R13.  This additional "compression" makes the octave stick out more as one sustains the note to the point where fundamental predominates.

So, not much in the way of bells or whistles.  Just a lowly volume pot and toggle, for better matching to different circumstances.  As always, and as others have often pointed out, feeding the RInger to some sort of overdrive or distortion can make the octave stand out better.

Enjoy

[thermionix]

Mine started as a straight clone, but since I only use it in front of fuzz (usually a Fuzz Face) I thought it made the tone too trebly/thin because of the buffering.  So I took inspiration from PJP and stuck a series resistor at the output.  I tried 22k first and that was way too much.  So I tried 3k3 and that was much better.  I've left it like that so far but I think really 4k7 is probably ideal.  I guess a switch would be best, to bypass the series resistance, but I kinda like having no controls on this particular pedal, and I didn't leave room for a switch in mine anyway.

[BetterOffShred]

I literally made a tentacle board last week,  it's patiently waiting on my desk to be boxed with a Harmonic Percolator and Majikbox venom boost, with order switching.   I am going to test these mods before I box it up, thanks for sharing Mark!

Edit:  Is there a suggested Vf  (VF?) forward voltage drop on the diodes for maximum effect?  Or is it more important that they are matched?   I built a green ringer clone as my second project ever a couple years ago and I didn't match the diodes, but it still works, Maybe I just got lucky.  When I built the tentacle I matched 4148's and I think it's more noticeable, but this could be just because of the mods. 

And by "Back to back" 914's instead of R13 do you mean an Anti-parallel pair?  I am going to try all these mods!

-Brett

[radio]

I wonder if you put an electra distortion kind of circuit in front of it

what that would do. Just waiting for some parts to test it.

Just found this link,the  Atmos,I don't see it at work but 2Leds are involwed in the modification.

http://johannburkard.de/blog/music/effects/Building-pedals.html

[Mark Hammer]

1) I opt for simply the lowest forward voltage from D1/D2.  Matching is good.  Given how low Schottkys tend to be, it won't take much sorting to find a pair that are within a few millivolts of each other.  Consider that the Vf poses a kind of gating effect. So if you try to extend the note with finger vibrato, a higher Vf will cause the circuit to cut the note off, rather than let it sustain a little longer.  And I can't emphasize enough that the octaving is strongest as the harmonics in the original note disappear and leave the fundamental.

There is no need to match the silicon diodes I suggested for compressing the output.  As noted in an older post on the topic, dino/digi2t suggests sticking a smaller-value resistor (e.g., 1k) in series with that additional diode pair to soften any clipping they provide, while still retaining the compression effect.

2) Any "thin" quality is simply because with everything doubled, even bass notes are an octave up.  Unless one designs in some sort of parallel path for blending in unaffected low end (which the Prescription Electronics Clean Octave Blend tries to do), there won't be much bass in the effected output.

3) Any sort of boost ahead of the Ringer tends to improve its performance.  That could be a clean boost, op-amp or transistor based, or use of the internal front end as I depicted in the initial post.  Ideally, the top end of whatever you feed the Ringer should be rounded off with some form of lowpass filtering.  Realistically, one can do that by turning the tone down on one's guitar.  But it is far more convenient to be able to leave the guitar tone up full and have the pedal do all the compensating for you.  The additional 100pf cap that Earthquaker added to the input for the Tentacle does some of that, but if you can do a little more, so much the better.

[amptramp]

It might be interesting to try a variable resistor to ground from the base of the output transistor.  The bias of the output transistor is set by current through the diodes and if you can increase that, you have the effect of changing the crossover distortion during the time the diodes are at the same anode voltage.

[thermionix]

2) Any "thin" quality is simply because with everything doubled, even bass notes are an octave up.

I meant the Fuzz Face sounds thin like it always will with a buffer in front of it.  I know the GR isn't just a buffer, but acts as one to some extent, and the last stage is an emitter follower.

[bluebunny]

. . . since I had the sense he did not have any octaver.

Wow!  How do you do that, Mark?  Did you detect a disturbance in the Force??   

[Juicy_scooby]

Hey guys!

Love this thread and the others like it...I've found some seriously awesome mods for the Ringer I can't wait to try.

Mark, I wanted to ask a few clarifying questions for these two modifications you outlined here:

I used Schottky (1N5817 in this instance) diodes for D1-D2 since they yield a more dependable octave-up in my experience. and also a 100pf cap in parallel with R1, as Earthquaker des with their Tentacle clone of the Ringer.

Looking at a schematic of the Tentacle clone of the ringer:



I noticed that cap in parallel with R1, is this to reduce "pop" like in a fuzz face or does it do something else for Q1?

I also notice another Diode (D3) in the upper right hand corner going to ground after C6 (which I notice is conserved in all the GR schematics)
What is this for?

Mark, you mention replacing R13 in this schematic with diodes as well to give it "compression".

Finally, as I've noted in other posts, I stuck a back-to-back pair of 1N914 diodes in place of R13.  This additional "compression" makes the octave stick out more as one sustains the note to the point where fundamental predominates.

Can anyone elaborate on why these diodes add something a resistor can't, and maybe why EQD also put a diode to ground in a similar part of the circuit? Are these unrelated or speaking to a more fundamental use of diodes?

Thanks guys for another awesome thread! I wish I knew more people in real life to show this site to because I'm so excited about it!!!

[Rob Strand]

Another mod is adding a resistor in series with C2, like the circuit in the other thread.  This acts as an impedance the balancer.   I'm sure the mod has for and against factions.

I noticed that cap in parallel with R1, is this to reduce "pop" like in a fuzz face or does it do something else for Q1?

The resistor R1 prevents pop.  It works because it make sure the input side of the cap C1 is always at 0V DC.
The cap helps prevent RF interference.

I also notice another Diode (D3) in the upper right hand corner going to ground after C6 (which I notice is conserved in all the GR schematics)
What is this for?

It is a simple form of reverse power supply protection.

[Mark Hammer]

Mark, you mention replacing R13 in this schematic with diodes as well to give it "compression".

Finally, as I've noted in other posts, I stuck a back-to-back pair of 1N914 diodes in place of R13.  This additional "compression" makes the octave stick out more as one sustains the note to the point where fundamental predominates.

Can anyone elaborate on why these diodes add something a resistor can't, and maybe why EQD also put a diode to ground in a similar part of the circuit? Are these unrelated or speaking to a more fundamental use of diodes?

Thanks guys for another awesome thread! I wish I knew more people in real life to show this site to because I'm so excited about it!!!

With any analog octave-up design, making an octave is the easy part.  The hard part is making it obvious and easy to hear. 

Guitar strings generate a lot of harmonic content when you first pluck/pick/strum them.  That harmonic content dies down as the string sustains for a while.    When all that harmonic content gets doubled, the octave-up gets lost in a haze of harmonic hash that's a bit like listening for the telephone while you're in the shower; you catch a hint of it now and then, but can't be certain that's what you're hearing.  Just about anything you might read about how to get the best octaving will make recommendations that reduce that harmonic content, like using the neck pickup, turning down the guitar Tone control, adding a cap like C1 (or even increasing the value of C1), and picking above the 7th fret (shorter strings are stiffer, and stiffer strings exhibit less harmonic content).  All of that is intended to make the octave stand out and be easily heard.

Keep in mind, however, that once plucked, the volume of a string also dies down fairly quickly.   So the challenge is "How do I make the octave stand out when it is initially overshadowed by all this other hash, and the string amplitude dies down quickly?".  The answer is: force the string volume to remain as steady as possible.  There are many ways one could do this.  You could stick a compressor in front of the octaver, or use a sustainer pickup or E-Bow.  Some folks also like the sound of an octaver in front of an overdrive - which has the property of putting a sort of clamp on the volume. 

I have suggested adding a pair of diodes to ground at the output of the GR as a quick and dirty hard limit, so that the octave appears to "bloom" as the harmonic haze fades away but the volume remains steady, while the octave comes out of hiding.  Indeed, if one studies the circuits of most cherished octave-up units, one of the things you'll regularly see IS a diode pair to ground after the octave-doubling part.  Many people think that these are necessary for providing the "fuzz", but actually they aren't.  If you lift that diode pair in any of hose designs, you'll find the rest of the circuit is plenty fuzzy.  The diodes are there to limit the volume in a way that make the octave more perceptually obvious.

And, lest there be any misunderstanding, that diode pair is not suggested as a replacement for the resistor to ground on the output, but is suggested as a supplement, to be placed in parallel.  What I think may have confused you was my additional suggestion to vary the gain of Q1.  If one does that, there may be an issue with effect/bypass volume-balance, in which case replacing R13 (47k) with a 47k volume pot (log taper) may be helpful.

All of that said, you may rightly wonder "But how does adding a pair of what look like clipping diodes NOT add harmonic content of its own?  I mean, isn't that what we usually do when we want to produce distortion?"  KEEE-wrecked.  So think of their addition as 3 steps forward and one step back.  One way of making that 5 steps forward and only 1 step back is a suggestion from Hen's Tooth Cafe proprietor and valued forum member Dino/digi2t.  Add those diodes, but stick a small-value resistor between them and ground, like 470R.  That will retain the limiting function, but make any clipping they provide softer.

There, I hope this makes sense of it all for you.

[Juicy_scooby]

Damn Mark, Damn Rob, very well put. Thank you.

Understanding the signal from when the string is struck to it's decay and release is really helpful to think about; I feel like as someone with a synthesizer background when it comes to audio signals this makes particular sense.
And thanks for clarifying about the parallel wiring with the resistor, I did think it was "instead of".

I have suggested adding a pair of diodes to ground at the output of the GR as a quick and dirty hard limit, so that the octave appears to "bloom" as the harmonic haze fades away but the volume remains steady, while the octave comes out of hiding.  Indeed, if one studies the circuits of most cherished octave-up units, one of the things you'll regularly see IS a diode pair to ground after the octave-doubling part.

Would "back to back" in the schematic appear as  --|<l--l>|--ground?

I also really like the idea you mentioned courtesy of Dino/digi2t to add a resistor after the diodes, makes me want to read up on limiting and diodes 


And Rob thank you for clarifying about the other diode and C1. I realized I have actually read about decoupling capacitors before and forgot what they do!

Can you (generally; in theory) put this kind of capacitor here in any circuit that could "buzz" so to speak?


Thanks again guys!!

[Mark Hammer]

Would "back to back" in the schematic appear as  --|<l--l>|--ground?

Not quite.  When people talk about a back-to-back pair, they generally mean two diodes in parallel, "facing" the opposite way.  But as your query illustrates, it is perhaps not the very best of technical terms! 

[Rob Strand]

Can you (generally; in theory) put this kind of capacitor here in any circuit that could "buzz" so to speak?

The buzz the actually harmonics of the mains.  Almost impossible to removed (other than preventing it getting in in the first place.)

The RF thing would be like when you can hear a radio when you touch a part of the circuit.   Sometimes the radio signal get in even when you don't touch a circuit and that's when that cap helps.   An old day's example was you could hear people speaking on a CB radio through your stereo system when they drove past.  Todays example might be the buzz/beep you hear when your mobile phone is close to audio electronics.    The RF is much higher frequency than audio but it still gets in.  The simple explanation is the circuit behaves like a crystal radio set all by itself.

The input cap can also stop oscillation problems.   This is completely different to the RF scenario.  It's more to do with the output signal getting back to the input through the air because the wire are close together, or at the switch where there are wires close together.

These thing scan be tricky to understand but one easy recipe is to stick a cap on the input.

[Juicy_scooby]

The input cap can also stop oscillation problems.   This is completely different to the RF scenario.  It's more to do with the output signal getting back to the input through the air because the wire are close together, or at the switch where there are wires close together.

Ah! Is this induction? That's crazy!

Just in general for these mods...I know PCB's for this circuit exist, a few even for these modifications or the "21st Century" GR...I'd kind of like to pick and choose a few of these modifications (the cap, diodes, and null trimpot).

Looks like the best way to do that is using veroboard and basing it on a layout. I'm having some trouble translating the schematic I've drawn to a layout that works and I think it's because I'm having trouble twisting the circuit around in my mind to fit on a perfboard. I've also heard breadboarding the circuit first (which I plan to do to test it) is a great way to wrap your head around these layouts.

I know this is kind of basic stuff, but do you guys have any tips for translating layouts to veroboard? Is there a way to modify a PCB to have an extra resistor or op-amp in the circuit?

[amptramp]

This thread (which is the top one in the shaded section of Building Your Own Stompbox) has a layout creator that works for Veroboard and other purposes:

https://www.diystompboxes.com/smfforum/index.php?topic=44838.0

[trevor78]

Looking at the schematic, it occurred to me that more gain could be achieved from the simple front end it used.  I added a path in parallel with R5, shown here, to do just that.  I put in a 10k resistor in series with a 10uf cap to ground, and used a 3-position toggle to place an 8k2 or 1k5 resistor in parallel with the 10k, yielding moderate and higher gain from Q1.

Quick Question, unless I am blind I am not seeing this on the schematic? Any way to show us visually how this is done? I am pretty greaan and still navigating the awesome world of building pedals!
Thank you!

[trevor78]

Mark, you mention replacing R13 in this schematic with diodes as well to give it "compression".

And, lest there be any misunderstanding, that diode pair is not suggested as a replacement for the resistor to ground on the output, but is suggested as a supplement, to be placed in parallel.  What I think may have confused you was my additional suggestion to vary the gain of Q1.  If one does that, there may be an issue with effect/bypass volume-balance, in which case replacing R13 (47k) with a 47k volume pot (log taper) may be helpful.

All of that said, you may rightly wonder "But how does adding a pair of what look like clipping diodes NOT add harmonic content of its own?  I mean, isn't that what we usually do when we want to produce distortion?"  KEEE-wrecked.  So think of their addition as 3 steps forward and one step back.  One way of making that 5 steps forward and only 1 step back is a suggestion from Hen's Tooth Cafe proprietor and valued forum member Dino/digi2t.  Add those diodes, but stick a small-value resistor between them and ground, like 470R.  That will retain the limiting function, but make any clipping they provide softer.

There, I hope this makes sense of it all for you.

One more question, so we are not actually removing the 47k resistor but adding a pair of diodes with a (470R resistor in between the two diodes) in parallel with the existing 47K resistor? 
Thank you!

[PRR]

Welcome!

...I am not seeing this on the schematic?

[Mark Hammer]

I suspect one would likely be better off simply sticking a 10k linear pot in place of the original 6k2, one outside lug going to ground, the other going to the emitter of Q1.  A 10uf cap goes from the wiper of the pot to ground (+ side tied to the wiper).

But Paul gets a point for the assist.