ESR Graphic Fuzz [schematic]

[rocket88]

I haven't commented on this for awhile. I built peeps' layout and it works exactly like the original, spacey cocked wah, whistling and all. I even compared it to a real one, not mine nor could I pull it apart. I never got around to trying inductions suggestions about modifying the circuit to allow for daisy chaining and an led.

I have Tom free thst the self oscillating whistle and cocked wah thing gives this effect it's personality. I usually leave it full blast just to have the fuzz, which is massive, but messing with the filter knob and not playing you get that 50's sci fi space sounds, which is awesomely cool. The only change thst I would make is getting it to be daisy chainable, and have an led to let you know its on. I'm thinking about emailing jimmy to see if he would help give some idea about what he did, but I think that if it's this hard for us to figure it out I don't know if he will be willing to saw......

[tabbycat]

Hey Tabbycat! Any luck? Thanks!

fuck yeah. or as close as i can call it with a mini amp in the small hours of the morning. 'verofied'. my first working vero layout.

earth sound research graphic fuzz vero layout verified 21.1.15.gif

esr graphic fuzz vero.JPG

esr graphic fuzz breadboard.JPG

i used 2x 1m resistors stuck together in series because i'm cheap and lazy and impatient and didn't have any and didn't want to order any and have to wait until i did. had a 2m2, maybe that would have done, but induction specified 2m and so 2m it is.

have included a breadboard shot which should help you with offboard wiring if you get stuck with that.
i made a curious discovery while playing with it trying to make it work ('research and development'), the thing didn't really work until out i linked the wiper of the level pot to ground. don't think this was in the schematic. but it seemed to make the magic happen so i'm not going to question it. will try putting that level 2 to ground it on the vero instead of offboard, but seeing as it works like this may as well go with it. it's hardly a heartbreaking addition. a dealbreaker though...

will look into adding the 'offset null control' larsjm mentions tomorrow. if it works well i will add it onto the vero layout and post it here this weekend.

but as it stands it's like a very metallic gritty fuzz (like putting tin foil over a speaker) with the filter control acting like a wah on a pot to give you dialble sweet cocked-wah tones.
it's pretty interactive with guitar controls too. lots to play with for such a tiny layout.

as i say, because it's late i've only played with it with a mini-amp but it seems to do all it should. so get in.

tabbycat.

[Nocentelli]

...didn't really work until out i linked the wiper of the level pot to ground. don't think this was in the schematic. but it seemed to make the magic happen so i'm not going to question it. will try putting that level 2 to ground it on the vero instead of offboard, but seeing as it works like this may as well go with it. it's hardly a heartbreaking addition. I typed in

The induction schematic has lug1 grounded, lug2 from the output cap, and lug 3 to output. This is how many DBA and some catalinbread volume pots are wired, and it works ok. I swapped lug 2 and 3 for a standard volume pot. The vero doesn't match either of these arrangements, even allowing for the pot working backwards.

[tabbycat]

...didn't really work until out i linked the wiper of the level pot to ground. don't think this was in the schematic. but it seemed to make the magic happen so i'm not going to question it. will try putting that level 2 to ground it on the vero instead of offboard, but seeing as it works like this may as well go with it. it's hardly a heartbreaking addition. I typed in

The induction schematic has lug1 grounded, lug2 from the output cap, and lug 3 to output. This is how many DBA and some catalinbread volume pots are wired, and it works ok. I swapped lug 2 and 3 for a standard volume pot. The vero doesn't match either of these arrangements, even allowing for the pot working backwards.

hey nocentelli, many thanks indeed for having a look.

basically i got my build to the (much protracted) point last night where it made gritty fuzz and cocked wah noises through the little mp3 amp i was using and decided that must be it.
but checking the schematic on here today while looking for it to post ( download/file.php?id=21595&mode=view ) reveals something that the black and white print out of the schematic i've been working from doesn't. i.e. that the line from pin 4 of the 741 to ground is red (a mod propsed by induction? rest of his proposals have question marks next to them). on my black and white print out red is black, so i read that whole line from pin 4 to ground as a single uninterrupted ground, hopping over the line it crosses instead of joining to it. that's why i didn't ground that line going across, as i thought that was some separate feedback loop or something.

well i've made a verified schematic of something, but not an esr graphic fuzz it appears. 'the esr monochrome special edition' maybe. have a feeling it shouldn't be a big deal to put right. will get to it later this evening.

meanwhile, do you think i should buzz a mod to pull the schematic asap before anyone builds it and hates me for it? not sure of the protocol for that. crap layouts should be pulled really, i think. lest they get posted out of context (i.e. this thread). i don't really care about embarrassing myself, i'm beyond that, but i'd feel bad if a newbie made it up for their first build and then found it isn't what it is supposed to be. there's no edit option on this thread.

oh well, my rep is dirt until i get this reposted. he who dares, etc.

thanks nocentelli,

tabbycat.

[deltafred]

oh well, my rep is dirt until i get this reposted.

Far from it, as we used to say where I worked "if you never make mistakes it's because you never do anything" (and there were plenty who never made mistakes!)

[Nocentelli]

oh well, my rep is dirt until i get this reposted.

Far from it, as we used to say where I worked "if you never make mistakes it's because you never do anything" (and there were plenty who never made mistakes!)

+1

If you want to get rid, you can pull attachments yourself from user control panel -> manage attachments -> delete attachment.

[Nocentelli]

Looking again, i think if you just swap volume 1 and 2 wire connections and ground strip C it's pretty much the graphic fuzz.

[tabbycat]

Looking again, i think if you just swap volume 1 and 2 wire connections and ground strip C it's pretty much the graphic fuzz.

thanks for your help and encouragement nocentelli and deltafred.

grounded c and swapped level one and two (thanks nocentelli) and it absolutely roared into life. this thing is totally dirty now.

really verified.

earth sound research graphic fuzz vero layout final.gif

didn't expect it to sound this good. secretly thought the one in the video had been treated and massaged to get such a big sound from so little, but they really do sound that good.

so now have a paint stripper and black hole in one. fuzz blisters and spits, proper nasty, and the filter does everything from a wuthering breeze on a midnight moor to genteel whistling in a tiled underground station to spaceship teleporter beams.

so totally verified. it's a keeper.

tabbycat.

[deltafred]

Just looked at the schematic. The voltage divider for the half supply voltage bias is all wrong.

It doesn't want a cap across the upper resistor. This makes an AC potential divider* instead of a low pass filter.

The value of the cap needs to be higher, and/or the resistor values need to be higher as the cut off frequency is 160Hz, way too high to filter out mains hum fundamental. I would go for 100k resistors and 1uF electrolytic which would bring the cut off freq down to 1.6Hz (only because I have about 4000 100k resistors and a huge bag of 1uF caps).

*If it is run on a power supply and there is any hum on the 9v it will be halved by the potential divider of the 2 x 100nF caps (and 2 x 10k resistors) and fed into the opamp. Removing the top cap creates a low pass filter with the 10k and the lower 100nF with a cutoff frequency of 160Hz,

[tabbycat]

hey deltafred, thanks for the insight. am just making sense of, according to my limited understanding.

Just looked at the schematic. The voltage divider for the half supply voltage bias is all wrong.

It doesn't want a cap across the upper resistor. This makes an AC potential divider* instead of a low pass filter.

between the 9v and the 10k you mean?

download/file.php?id=21595&mode=view
induction's post to accompany this schematic with his proposed mods with explanation of circuit viewtopic.php?f=19&t=22197&start=20#p221373

i think induction took his lead from this drawing at d*a*m http://www.stompboxes.co.uk/forum/downl ... hp?id=1835
found here http://www.stompboxes.co.uk/forum/viewt ... 6&start=40

same drawing is disected in this thread at diysb by john lyons among others (who also comments in earlier pages of this thread about it) http://www.diystompboxes.com/smfforum/i ... ic=83300.0

The value of the cap needs to be higher, and/or the resistor values need to be higher as the cut off frequency is 160Hz, way too high to filter out mains hum fundamental. I would go for 100k resistors and 1uF electrolytic which would bring the cut off freq down to 1.6Hz (only because I have about 4000 100k resistors and a huge bag of 1uF caps).

it would seem that the resistor values are authentic going by this post from a former esr employee.

Hey guys, I worked at ESR as one of the two technicians there in the customer service repair. It was back in 1978 and my first electronics job. Sometimes we would just play guitar all day.
What I remember about the Graphic Fuzz box is they oscillated like crazy. The other Tech, Tony, made some improvements.
As far as the schematic, I remember there were two 10K resistors as a divider with the center going to ground to make the +/- supply and whatever caps on that.
The IC was a 741.
A lot of amp designs were ripped of from Peavey and they got in trouble eventually for that..

viewtopic.php?f=19&t=22197&start=20#p221556

so if it's to be upping the cap what do you recommend and where? or could it be that the original was just badly designed from a technical perspective, but sonically makes good of that flaw?

*If it is run on a power supply and there is any hum on the 9v it will be halved by the potential divider of the 2 x 100nF caps (and 2 x 10k resistors) and fed into the opamp. Removing the top cap creates a low pass filter with the 10k and the lower 100nF with a cutoff frequency of 160Hz,

when you say "remove the top cap" do you mean top in the schematic (100n) or top in my vero layout (1uf)?

sorry if this is a bit of an involved post, but i am on a mission with this one. have to say the pedal i have roars and does all the original seems to, but if you think it is merely a lucky work-a-like i am ready to persist to nail the original.

many thanks,

tabbycat.

[induction]

Just looked at the schematic. The voltage divider for the half supply voltage bias is all wrong.

It doesn't want a cap across the upper resistor. This makes an AC potential divider* instead of a low pass filter.

That depends what you mean by "all wrong". I didn't redesign the sonic characteristics of the circuit, I just converted it to standard negative ground so it could be daisy-chained. People like the original design, so I left it alone as much as possible, "all wrong" or not. The caps and resistors in the bias network are true to the original, based on the schematic I was working from.

i think induction took his lead from this drawing at d*a*m http://www.stompboxes.co.uk/forum/downl ... hp?id=1835
found here http://www.stompboxes.co.uk/forum/viewt ... 6&start=40

Nope. I did it on my own as an exercise in understanding how the circuit works. If it came out similar to someone else's attempt, that's not too surprising because it's a very simple task to change the biasing arrangement for unipolar power, once you understand how the original circuit creates square waves.

Could it be improved? Probably. If you like the original circuit, this one should sound the same, at least with a battery. If you 'improve' it in a way that changes how it sounds, some people will like it and others won't. That's how opinions are. That being said, it may not be optimized for psu hum reduction, which sounds like what deltafred was getting at. In my experience, deltafred tends to know what he's talking about. So if he suggests a modification, it's probably worth trying out.

Often, the design 'defects' are what gives a pedal it's characteristic sound, so 'improving' these defects away kills the charm. (Incidentally, that's how blue led's were invented: an engineer tried a bunch of random stuff that introduced flaws and lattice defects into a semiconductor until he got it to emit blue light in response to applied voltage. He didn't know he had damaged the lattice, or even how the thing worked until he gave it to a physicist who analyzed it and explained it to him. The physicist said that he would never have come up with that technique in a million years because the goal with semiconductors is always to produce a defect-free lattice with predictable properties.) That being said, few people equate psu hum with 'charm'.

And nobody should ever make the mistake of confusing me for an EE. I'm just a physicist (aka an electronics hack). I learned more about electronics from playing with these little boxes than from those limited electronics courses I took so many years ago. There is every possibility that I overlooked something in the schematic that a real EE would have learned about in second grade. I haven't piped up before this because I haven't played with this circuit since I made up that schematic, and I didn't really remember what was going on in my head at the time. I breadboarded it back then, decided not to build it, and forgot all about it. I'm very happy that somebody found my work on this thing useful, but I shouldn't get any credit for the sound of the circuit, or even for the conversion to unipolar power, since apparently other people figured it out and published it way before I did.

[deltafred]

Just looked at the schematic. The voltage divider for the half supply voltage bias is all wrong.

It doesn't want a cap across the upper resistor. This makes an AC potential divider* instead of a low pass filter.

That depends what you mean by "all wrong". I didn't redesign the sonic characteristics of the circuit, I just converted it to standard negative ground so it could be daisy-chained. People like the original design, so I left it alone as much as possible, "all wrong" or not. The caps and resistors in the bias network are true to the original, based on the schematic I was working from.

Sorry I didn't make myself clear, I was not criticising anyone's work here just the original designer's. This is what amazes me about stomp box "design", the way good design practice is ignored for the sake of "mojo". (e.g.. Devi Ever - )

As has already been stated in this thread the originals were prone to oscillation, I wonder why? Infective power supply decoupling is a well known cause of that.* If that is your design goal then fine but to me it just looks like sloppy design, or more likely not knowing what the fuck you are doing.

*In a properly designed circuit the power supply rail and the bias rail should both be decoupled (have a largish value cap to the ground rail).

Again I will reiterate I am in no way criticising the work done by the guys here on FSB, reverse engineering circuits is a skill in itself, as is doing layouts.

tabbycat - Yes the cap between the +ve supply and the bias voltage need removing and the lower cap needs to be larger. Like I said I would go for 100k resistors and 1uF/2.2uF/10uF, whatever you have and will fit in the layout. I would also put a 100uF across the 0v and 9v. (unless you are always going to run on batteries and don't mind if it oscillates a bit).

[induction]

Sorry, I guess my post came out snarkier than I intended it.

[tabbycat]

many thanks deltafred and induction for your posts.


Link

as the sound of what i have built from the vero i posted sounds immense and pretty much there compared with the youtube clips i’ve heard of it (though i haven’t had the chance to get it into anything big and loud yet, still a tabletop amp setup) i will keep that one as the benchmark (thanks again for the schematic induction) and will build another with the ‘corrections’ (in the pure technical sense) deltafred is suggesting so i can compare them head to head.
it will be an interesting experiment to me. it’s a relatively small breadboard-able circuit and i’m curious about the whole artist/artisan/technician pay-off/trade-off slant to the thing.

do you know the francis bacon quote “there is no excellent beauty that hath not some strangeness in the proportion”. that’s the history of human subjectivity in a single sentence.
i always keep that quote on a post it stuck to my desk because it also sums up why i always end up with mad women. because the sane ones bore me. it rationalises my irrationality to me in a way that stops me losing myself in angsty retrospection and regret. i choose mad over sane because that’s who i am and what i do. ‘a man’s character is his fate’ (heraclitus).

maybe the art of stompbox designing is to design-in enough sanity to ensure that you don’t get electrocuted every time you hit a chord, and enough madness to ensure that sounds you haven’t got bored of yet keep coming out of your speaker.

@deltafred many thanks for applying your wisdom to this one deltafred. much appreciated and respected input.

going straight for the practical, just to double check my understanding;
i’m dropping the 100n cap between 9v and bias and replacing it with a big one (100uf) between 9v and ground.
i’m trading up those two 10k resistors for 100k.
and i’m replacing the lower cap, between bias and ground, with 1uf or bigger.
that is my mission as i understand it.

@induction

thanks for the schematic induction, the pedal i built from my vero of it sounds superb.
i formally click-thanked your post on that a while ago, and your detailed explanation that went with it, but worth saying it here too.
apologies if when i said “i think you took your lead from...” you felt i was detracting from your efforts. that was not intended by me. you made it clear in your post that this was something you had worked out for yourself by investigating comparators. i was just trying to place disparate elements into some kind of loose ‘possible’ narrative so as not to overwhelm deltafred with a pile of disparate links that seemed to come from nowhere. historian’s vice.

re the snarky thing, your comment was unapologetic and forthright, which i suppose to some (read very early in the morning) could seem ‘bold’. but not entirely snarky i don’t think.
in any case, i read it (twice) and still managed to go about my business all day feeling thoroughly unsnarked.
so that must mean something...

tabbycat.

[storyboardist]

Here's a perf and PCB layout based on Induction's schematic. Should fit in a 1590a.

[deltafred]

Sorry, I guess my post came out snarkier than I intended it.

No worries (as Dr Nomis would probably say), I was probably a bit abrupt in my first post, it was near midnight here and I'd had a few scotches. Never a good idea to post (or go on Ebay) when you've had a few.

(I saw your post yesterday but was on my tablet and can't be doing with typing on those poxy little touch keyboards.)

[storyboardist]

Selfdestoryer verified my layout last night. The above image is wrong though. Here's a link to the verified layout:

http://effectslayouts.blogspot.com/2015/01/esr-graphic-fuzz.html

[EBRAddict]

It was mentioned earlier in one of the ESR Graphic Fuzz threads here or at DIYSB that this circuit is sensitive to the source impedance feeding pin 2. It doesn't like it. In fact when I had my bass roll-off knob engaged on the guitar all the pedal would do is squeal.

I put a simple JFET buffer feeding pin 2 of the 741 op amp and the squealing/oscillation went away. The circuit I used is straight from muzique, I subbed 1.5M resistors for the 1M because I have a bunch and used a 2N5457 JFET:



In that configuration it has a high input impedance, low output impedance, and near unity gain. From the 10uF I went straight to pin 2 of the op amp and pin 2 of the Filter pot.

Also earlier, larsj contributed that adding a potentiometer between pins 1 and 5 (the offset null setting) produced some interesting results. The datasheet for the LM741CP says to use a 10k linear pot, I grabbed a 50k linear by mistake, but playing around with the pot you can generate a square wave or a sawtooth wave at the output, and several interesting permutations in between and at the extremes. The sawtooth and square wave generate very different harmonic profiles.

[greenskull]

Thanks for revisiting this circuit! I had a go at it on the breadboard some years back, but for some reason I didn't commit to a build. I think I'll give it another shot and try some of those tricks.
I wish I knew what John over at Basic Audio added to his version of this pedal...... it sounds extra nasty!

[EBRAddict]

Thanks for revisiting this circuit! I had a go at it on the breadboard some years back, but for some reason I didn't commit to a build. I think I'll give it another shot and try some of those tricks.
I wish I knew what John over at Basic Audio added to his version of this pedal...... it sounds extra nasty!

I wish I could go back and edit my previous post. That JFET buffer makes the "FREQ" pot practically useless. You still get the crazy fuzz and it kills the out of control oscillation, but it kills the frequency resonance effect. I put mine on a DPDT bypass switch for when I want the fuzz but none of the squeal. The pedal needs direct interaction with the guitar's controls without a buffer to get many of the sounds out of it.

I also added a squeal tamer as others have shown in the posts above, along with a switch to go back to stock. It keeps some of the functionality of the FREQ pot like the weird sub-clipping resonance and just a tiny bit of squealing, but not the crazy oscillation every time I stop playing.