Post by thetragichero on May 1, 2018 15:09:32 GMT -5
been decluttering my life by selling stuff on ebay (recently flat round pieces of black or funny colored vinyl) so i had a paypal balance (DANGER) and i got to thinking about how i can march to the beat of my own drummer (which is good because i can never find one to play with) and how i'd like to have a space for the consensus worst pedal ever on my board... will definitely bust out the soldering iron to remove all of the bees that live inside that metal box, but this partly is commentary on gear-snobism so the plan is to keep it as stock-looking as possible i have a wordpad file of some ideas from another forum but i was curious if anyone here had anything special they like to do to these (besides toss `em) will be using for punk/metal/fast loud stuff in front of a peavey xxx head that already has more gain on board than i could ever need. i'm not a mids pooper-scooper guy, though. that's where all the good stuff is
to sum up: i bought a pedal most guitarists that i know hate and i'm gonna mess with it until i can get something useful out of it
I havent tried that pedal, but Ive had a lot of fun twisting the tone of a couple of other pedals until I liked them better. These were a Boss Sd1 and also Marshall Guv'nor2.
My approach was to play them a while to figure out what I liked and didnt like, read up on how they work, then build a Spice sim to experiment with ideas for circuit mods ahead of diving in to hack the circuit.
Id suggest to resolve your opinion on a few characteristics of the stock pedal eg: The general amount of distortion available; just right or too much?(or not enough!!?) Tone control range; more mids needed? less high-end fizz? Output level; enough?
Post by thetragichero on May 3, 2018 18:38:28 GMT -5
pedal should be here tomorrow i have ordered some SIP sockets for diode and ic swaps. since these ICs are not the DIP configuration that i have/ordered, i will be making DIP-to-SIP adapters on perfboard need to look at some datasheets, but I'm gonna try (if possible) to make a 14pin DIP to 2x 8pin SIP adapter to try a tl074 sockets should help make this a few minutes here and there fun project
I think you're barking up the wrong tree thinking about swapping opamps. Especially if you're going to such a plain vanilla part as a TL074. An LM324 with its crossover distortion and assymetrical limits (will go all the way to the bottom rail, but clips a diode drop from the top) might make it a little more interesting, but really this quote from the link JohnH posted sums up my thoughts pretty well:
"Whereas on many other distortion or overdrive pedals, the op-amp used or the clipping diodes have a major influence on the final output, in the MT-2 it’s really all about frequency response. It’s heavily shaped every step of the way, and those frequency responses determine the sound of the pedal, much more than diode choice ever would."
Edit - Not that I really have any experience with this pedal or modding it.
Last Edit: May 8, 2018 11:55:35 GMT -5 by ashcatlt
Post by thetragichero on May 8, 2018 12:36:44 GMT -5
i'm actually thinking tlc2262 might do the trick. i've read that's a big part of the tech 21/sansamp sound, which i dig for solid state i've got a whole slew of op amps either here or in the mail, and sockets for all four in the circuit (plus for the clipping diodes) my goal will be less gain and more mids, but if i get something usable then i'll be happy and certainly will have learned some stuff through the process stock it doesn't sound all that bad in front of the crunch/ultra channels of my peavey xxx. in front of the clean channel is a different story....
The tone shaping looks to be very extensive and powerful, and would appear from the graphs to be capable of anything from a mid-hump to a mid-scoop. But is that how it sounds?
If so, then I reckon the best place to focus is the source of the bees, which is the stage described as 'Gain stage and clipping diodes', and try to soften the clipping.
Things that might help with that include:
Increasing r33, maybe adding one more diode in series with one of the pair for a bit more volume an asymetrical harmonics. Increase in r33 should let the start of diode conduction compress the signal sooner and more gradually.
Maybe a small resistor in series with the diodes.
Maybe try a diode group in tbe opamp feedback loop like an SD1 or a TS. This is usually a softer place to clip and stops the opamp from slamming into power rails.
This circuit area is quite simple and could easily be modelled to take a look at distorted waveforms and play with virtual values to see what might be worth trying.
Different opamps must make some difference but as Ash points out, maybe not so much. An opamp of any kind is a very pure almost perfect gain device, particularly at audio frequencies, with its performance shaped by the components around it. You can understand 99% of most opamp circuits just by knowing that it is an opamp, even if of unkown type. That's different to other active devices such as transistors, diodes and valves which have more individual quirks and characteristics.
It's obvious that I've been out of touch these past few decades. For instance, I didn't know that Peavey had started making pornographic amplifiers... I really have to get out more often.
As close as I'll ever get to being serious, at least here in The NutzHouse:
The truth is not propaganda.... unless you believe the truth can hurt you.
It's obvious that I've been out of touch these past few decades. For instance, I didn't know that Peavey had started making pornographic amplifiers... I really have to get out more often.
the original grill has a mudflap girl on it like so:
the guy i bought mine off of swapped it out for diamond plate steel, so it looks more like a mesa booger amp
Last Edit: May 8, 2018 17:10:15 GMT -5 by thetragichero
Post by thetragichero on May 16, 2018 21:35:59 GMT -5
rereading the thread, i feel as if i came off only wanting to change the op amps. that is only part of it. i've socketed them along with the clipping diodes so that swapping them will be easier, but i've also got a word pad document with various mods from around the web, plus the schematic and the excellent analysis. i'm gonna do a lot of tweaking
Maybe try a diode group in tbe opamp feedback loop like an SD1 or a TS. This is usually a softer place to clip and stops the opamp from slamming into power rails.
This circuit area is quite simple and could easily be modelled to take a look at distorted waveforms and play with virtual values to see what might be worth trying.
In my opinion that's one of the most interesting opportunities for tinkering. The ratio of the feedback resistor and input resistor (the one between the inverting input and reference point) sets the 'early gain'. The back-to-back diodes in parallel with the feedback resistor limit the 'late gain'. As the output signal just begins to go above or below the reference point the diodes are barely conducting and don't play much of a factor at all. But as the output signal moves significantly above or below the reference point, one of the two diodes comes into the equation, shifting the gain from high to extremely low and essentially clamping the signal.
Personally, I think adding a pot (connected as a rheostat) in series with the back-to-back diodes might bear some additional fruit. In any case, clamping in the feedback loop is a very slick method indeed.
Post by thetragichero on May 17, 2018 16:48:21 GMT -5
so far i have done some of wampler's mods: removed c35 changed c34 to.047uF removed c25 changed c42 to .1uF changed c29 to .033uF changed c21, c11, c4 to 1uF non-electrolytic changed c32 to 120pF (with r44 value of 220k that lowers the corner frequency of the rc filter from 7.2kHz to 6kHz.. according to an online calculator. wampler listed c32 as 100pF and r44 as 100kOhm which have me like 15kHz... either calculator is wrong or this is some misprint) r33 changed to 10kOhm and .047uF cap in parallel, with 2 1n60p diodes in opposite directions connecting to c27 negative lead instead of through the trace on the board (said to reduce noise) socketed clipping diodes and op amps, working on dip-8 tho sip-8 adapters
managed to disconnect white wire that is on the footswitch (white and purple on switch, red and black are power)... can't find where it is supposed to go!
Just to say that the TS style diodes in the feedback loop don't really stop it from hitting the rails. They make it less likely in some situations, but a big enough input voltage will still hit the rails. In fact, it is very much like a diode-to ground clipping arrangement with the clean signal mixed in. The Output will (try to) be the input signal plus the diode drop. They actually tend to clean up with larger inputs...right up until they hit the rails...
I kind of disapprove of trying to make this sound better. Most of the fun of these pedals is exactly that they sound horrible. At least that's why they're so popular with the real black metal folks. There are plenty of other easier ways to make "good" distorted sounds. Let this one be ugly. Make it uglier if you can.
Post by thetragichero on May 18, 2018 9:33:49 GMT -5
hey man we have different philosophies on this one and i'm cool with that
had to put it down last night to get to rehearsal. was pondering (while watching tv) about this wire that came off the switch, did a bunch of fruitless googling. looked at schematic. looked at pedal. dangling wire from the switch just goes to ground lol. much ado about nothing
will finish up the socket adapters this afternoon and probably will fiddle with op amps tomorrow (figure i'll be a good team player and clean the kitchen?start dinner)
my work on this has been slow but it's a good learning experience... and then i will turn to the dd-3 i clipped the direct out jack from (so that i could remove it and have a hole for an analog delay mode toggle switch) and hasn't worked since... seems like these jacks are shorting jacks so something isn't connected to what it wants to be...
...and then i will turn to the dd-3 i clipped the direct out jack from (so that i could remove it and have a hole for an analog delay mode toggle switch) and hasn't worked since... seems like these jacks are shorting jacks so something isn't connected to what it wants to be...
I'm thinking you clipped the wrong jack. ()
There's no switching or shorting at all on the normal mono output jack. You can run all day long with nothing plugged into the other and it should be fine. That's exactly the same as if you removed the extra jack entirely. That other jack, though, I think only ever gets the dry signal. Plugging a TS plug into that TRS jack does short the portion of the dry signal which is feeding the mixer going to the other jack so that it would be 100% wet, but you're not listening to that one anyway.
This does seems strange to me since the Direct Out is pretty clearly labeled, and I'd have thought that would be a hint... I suppose the other hole is a more convenient spot for the switch? Cut off and tape up the wires going to the jack that's left and connect the wires that you cut off before. I'm thinking it might just be the tip wire since the sleeves should find ground through the chassis. (). You neither need nor want the wire going to the ring of that jack.
Edit to say - The MZ is actually a pretty powerful and sophisticated pedal capable of quite a range of different tones. It really is a good platform for this kind of experimentation. There's no good reason not to mess around with it if that's your thing.
Last Edit: May 18, 2018 11:43:41 GMT -5 by ashcatlt
Post by thetragichero on May 18, 2018 21:32:03 GMT -5
replaced the input and output jacks with switchcraft (i try to replace everything with them and i had to fix a bunch of the connections to the jacks anyway) she's all ready for op amps and clipping diodes tonight i tested every single led, diode, and fet voltage drop across a 2k2 resistor
still have some math to do my assumption: there is no magic sound juju that would make me use, say, a 2n5457 jfet over, say, a 3mm red led, or vice versa. 1.65v is 1.65v, right?
Well, that's a can of worms I'm not sure I wanna pop. It IS conceivable that one part might have more self-capacitance or static resistance or leakage that might make a minuscule difference. I don't honestly believe those things make enough difference in actual practice to even notice. Like might be measurable, but any audible difference is probably from other things that get changed (on purpose or not) in the process. Or confirmation bias. Or just the fact that we don't remember the minute details of sounds as well as we might like to believe.
I play with digital non-linearities pretty regularly and in my experience it requires a pretty major difference in the curve shape to make a noticeable difference in the sound or even "feel" of the distortion. I think I can hear the difference between a hyperbolic tangent and a sine-based clipper, and I can definitely tell the difference between either one and an actual discontinuous hard ceiling. These curves are all a lot more different from one another than any two diodes once you normalize them, and just like diodes the difference between the curves is minor compared to differences in clipping threshold vs input voltage. (tanh, btw, is really darn close to a diode-for-ground arrangement)
If you pump 1V into a silicon clipper, it clips off about 30% of the signal. A "typical" red LED needs 2V to clip off 30%. If you normalize the outputs they look and sounds as much alike as any two of either might (or within tolerance of the part) with the same input. All bets are off if the signal going into the LEDs clips before it hits 2V, and of course without the normalization, it's going to hit whatever it follows twice as hard...
Now, if you put even a relatively small resistance in series or even a very large resistance in parallel with the diodes it will start to make a real difference. Germanium diodes I think act like they've got both, so the curve will be a bit different. There are also parallel capacitances here, but they're generally too small for us to worry about here in audio. If we were designing X-ray equipment...
But that's kind of why you're doing this, no? To try and see what you think makes a real difference. Have at it. Draw your own conclusions. Record your results and share them with us please.
Last Edit: May 18, 2018 23:20:16 GMT -5 by ashcatlt
Post by thetragichero on May 19, 2018 12:44:47 GMT -5
not going to test ALL of the permutations because there is an incredibly large number of them, but so far an interesting finding is that using a jfet quiets down the expected amount of noise (open enclosure in front of monitor, speakers, equipment, ceiling fan, etc) another interesting side effect is that most of the jfets clamp down the output on the clean channel quite a bit, but the j201 (which i thought would be useless because i measured a voltage drop of like 6.something last night) does not still working on recording, will post later
Post by thetragichero on May 19, 2018 15:05:01 GMT -5
here we go...
i'm not going to copy all that down... but i go through each on the clean channel and then on the crunch channel. first it's with the pedal off, then with no diodes, etc. i end with no diodes and then will pedal off again, for reference. i think it gets interesting when i start throwing in the FETs and they started blocking the low frequency noise/oscillations (not sure which, post above mentions that i was expecting noise even though i was using new active guitar) i did not change any pedal or amp settings throughout, so the only changes were the clipping diodes D3 and D4. i messed around with some parallel capacitors at the end as well
Last Edit: May 19, 2018 15:08:45 GMT -5 by thetragichero
Post by thetragichero on May 19, 2018 21:40:47 GMT -5
will mess with op amps on Monday i like the j201 1n4738 (with or without the 220pF cap), but will also try clipping diode Laura i didn't try this afternoon (thinking mainly combinations of 1n34a 1n4738 1n4148 1n4001) and then roll op amps with whatever i like best, record and post to be honest, the fact that the friggin thing worked shocked me the most!
I mean, it seems like the ones marked "quiets noise" also quiet everything else.
I have to say that while it was kind of interesting, it didn't really give me a whole lot of useful information. I'd like to see/hear these direct in to take all of the variables of the amp itself out of it. If I could just have the file to mess with, that would be cool too. I don't do sequential A/Bs very well. If I could cut it up and do L/R comparisons, it might help me.
Last Edit: May 20, 2018 14:06:23 GMT -5 by ashcatlt
listening with fresh ears, i lose a lot of that mid/high freq dirt/CHARACTER with the J201 but that reduction in bass rumble... maybe i need to revisit the schematic and some of the filtering? or could it be caused by environmental stuff? guess tomorrow i investigate
edit: my usual one spot supply was with my bass pedal board at church and i was using a 9v supply i found and switched to negative ground. a brief *bingle* search brought up bad supply filtering as a possible source for lfo
Last Edit: May 21, 2018 23:28:57 GMT -5 by reTrEaD: Added local playing links.
I'd like to see/hear these direct in to take all of the variables of the amp itself out of it.
i had thought of doing this, but my end game is to play it through this amp (and maybe others), so even if it sounded like a cranked plexi (or whatever your dream amp is) DI, if it still sounded like doo doo through my setup then i'd be unhappy
i set the amp levels so that the clean and crunch channels peaked at approximately the same level. i was surprised (until i thought about it for 3 seconds) that most of the clipping diodes made the clean channel louder than the crunch (until i realized i was feeding that channel a compressed and boosted signal, whereas the crunch channel was already compressed)
I get it, of course. My problem with it is that it is impossible to tell what changes in tone/character that we hear might actually just be due to changes in level affecting how you're hitting the amp. Even the "clean" channel probably isn't completely linear, and the crunch channel is definitely going to add different amounts of its own harmonics to the result. In a real world scenario, you'd probably set the amp itself differently for each of these different inputs to compensate, but here we have none of that. The fact that they are all different levels make it an unfair comparison to begin with because of the way our ear/brains work, and the amp complicates that even more.
There's no switching or shorting at all on the normal mono output jack. You can run all day long with nothing plugged into the other and it should be fine. That's exactly the same as if you removed the extra jack entirely. That other jack, though, I think only ever gets the dry signal. Plugging a TS plug into that TRS jack does short the portion of the dry signal which is feeding the mixer going to the other jack so that it would be 100% wet, but you're not listening to that one anyway.
This does seems strange to me since the Direct Out is pretty clearly labeled, and I'd have thought that would be a hint... I suppose the other hole is a more convenient spot for the switch? Cut off and tape up the wires going to the jack that's left and connect the wires that you cut off before. I'm thinking it might just be the tip wire since the sleeves should find ground through the chassis. (). You neither need nor want the wire going to the ring of that jack.
the ground wire from the switch was wired to the direct out jack. fortunately on the metal zone the wire broke off so as soon as i saw it on the dd-3, i knew why the durned thing wouldn't turn on. time to fire up the gun!
Post by thetragichero on Jun 9, 2018 21:45:23 GMT -5
so i went with a red led and an infrared led i salvaged from an old tv remote for the clipping diodes
that means i spent the evening testing op amps
if you'd like to save yourself some time: i can not really hear much of a difference between the combinations. a couple (i think with tlc2262 in ic3 and ic4) squashed/neutered the sound. tl072 in ic1 SQUEALED and gave probably the only drastic tonal difference, but not necessarily something i'd want/that squeal would need to be filtered out somehow. i like the ne5532 in ic3 at one point but now i don't think i can really tell a difference.
i have seen something about stacking two dual op amps in place of one, but with so much gain already on tap i don't even know if that would be useful
part of me wants to put the ne5532 in ic3 just because, but the practical guy in me says to put back the stock ics and save the good stuff for something else
)
