Ibanez MOStortion MT10

[jrc4558]

Recently received one for fixing.
It souns so good, I want to make one for myself. I found a schem:
http://www.diystompboxes.com/pedals/MT10.gif
Now I want to ask you gentlemen, what do you think could be among main factors contributing to its good sound? Is it the bi-MOS chip it uses?
http://pdf1.alldatasheet.co.kr/datasheet-pdf/view/66349/INTERSIL/CA3260E.html
Or is it the tone stack?

Would anyone be interested in a project for it?

[dano12]

The GIF you posted appears to blank.

[NoFi]

I have one and i love it too. With humbuckers, just turn down the mids and it becomes "straty". With a single coil just pump up the mids for more honk. The circuit bares ressemblance with tubescreamers. But the diodes are special... 0.9V forward voltage if i recall correctly.
Not sure they are still made (the diodes).
I modded the input cap because i was under the impression it was removing my lower frequencies.
I dont think it's the chip because i swapped it out for a burr brown/tl072 etc. and the character remained.
The pedal is all about clean/slightly overdriven sounds that responds to the playing. IMO it is awfull for anything else.
It's the pedal i use the most.
++
Greg

[MartyMart]

You could get that forward voltage from three 1N4148's in series, or a 1N4001 would be
a little higher ( 1.1v ? )
Those "edge of breakup" sounds are VERY useful and quite hard to get right

MM.

[stm]

IR (infra red) LEDs may have such voltage (0.9-1.2 V).I remember someone in the forum over a year ago reported its use, perhaps for a bass OD, IIRC, with apparently good results.

[RDV]

Yeah, there's nothing there. I've always wanted to have a look at it, and it's always been "not there". 

Edit: Right click and 'save as' and there it is! 

RDV

[Mark Hammer]

Not sure where I got it from, but *somebody* posted this one a while back.  Don't have it here with me or else I'd repost.

I first heard about this pedal from an interview with country/blues slide player Lee Roy Parnell in an old interview in Vintage Guitar magazine.  He swore by the thing.  Looking over the schem, the only real distinctive thing I remember is the presence of the 3260 dual op-amp.  This is NOT the same as a CA3240 (which is a dual 3140) nor the same as a pair of CA3130 op-amps, as far as I know.  Just exactly WHAT it does in that application is beyond me, since I have never knowingly heard one and have not tried one either.  I do know that I caught Parnell on Austin Cty Limits and liked his tone very much (the reason for me reading the interview, actually), but whether what I liked about his tone was due chiefly to THAT pedal, I can't say.

Lest this thread start some sort of rampant JRC4558-like mania and frenzy of chip-swapping, let's wait until a posted schem allows people to identify a specific set of reasons for why the pedal might sound different in certain ways, and a set of mechanisms for why it might do so.

[kusi]

click right to the schematic-link and save it. works by me

mfg kusi

[stm]

Gentlemen, let's do some dissection of the circuit:

1) The schematic IS there.  It is a 276k GIF.  My browser didn't show it, so I just copied the link into a download manager and let it download it to my machine before viewing it.  It took quite a while (several minutes) but after downloading it I could see it.

2) The buffer stage looked odd to me initially due to the assymmetric biasing resistive divider (9.1k and 22k).  My first thought was, aha! that will introduce assymmetric clipping.  No.  The 10k resistor at the emitter of the transistor is seen as a 1.5 to 2.5 Meg ohm resistor at the base, which in conjunction with the 510k base resistor and the 0.6 or so Vbe voltage drop leaves you with roughly 4.5V at the emitter output.

3) The clipping stage is of the TS kind, however the (-) input cap and resistor (220nF & 2k7) have a corner frequency of 267 Hz instead of the typical 718 Hz of the classic TS / Boss SD-1.  More bass are let through.

4) Looking at the clipping diodes, there are two in series of a less known type: MA165.  First, having two of them produces more apparent clipping than just one since the contribution of the clean signal component at the output is smaller in comparison to the two-diode voltage drop of the distorted output.  Second, someone suggested those diodes have a 0.9 V forward drop and may not be made anymore.  There is no particular mojo in these diodes.  I downloaded the MA165 datasheet by Panasonic and verified a 0.95 V drop at 100 mA, which is far more current than the microamps you'll actually have in the feedback loop of the opamp.  Then, I downloaded the 1N4148 datasheet by Hitachi, and found these diodes are essentially the same in terms of average diode current capacity, recovery time (both are fast), and also verified that at 100 mA the 1N4148 has a forward drop slightly above 0.9V.  Bottom line: you could use two 1N4148 in series.

5) The feedback cap is 51 pF which is pretty stock, so nothing new here.

6) The opamp is of the rail-to-rail output kind.  Looking at its datasheet you see the output configuration is nearly identical to a CD4049/69 inverter.  Also, the numerical values indicate the rail-to-rail capability is degraded with increased loading, just as in the CD4049/69.  This might add some additional compression to the sounds, particularly during the initial note pick or when strumming chords hard.  Whether this effect is noticeable or not is open to discussion.  The TLC2262 seems like a sensible substitute for this opamp.

7) Then, there is a three-band tonestack.  I haven't had the time yet to redraw the circuit in a more traditional way, but apparently it is similar but not exactly a Fender or Marshall device, since I see at least one too many capacitor and resistor in there.  Certainly a design to be further studied!

Next, there is a gain recovery stage formed by the second opamp. The thing to keep in mind here is the 47k and 1nF in the feedback loop that filter out the high frequencies starting from 3377 Hz.  This certainly tames the srhill.

9) Finally, there is a transistor buffer stage pretty much standard.


To summarize, the particular aspects that might make this circuit different are:

a. Lower bass cutoff frequency (267 Hz v/s 718 Hz)
b. Use of two diodes in series in the clipping stage (varies the clean to distorted mix--this is inherent to this particular type of clipper and has been discussed previously)
c. Rail to rail MOSFET opamp--perhaps
d. Three band customized passive tonestack

Regards.

[Mark Hammer]

Thanks for the tip, Kusi.  A few things jump out about this particular pedal:

1) Uses a 2+2 diode complement in the gain/clipping stage.  As you might expect, it aims for more gain than a TS-9 does (x186 vs x118) to get closer to a higher clipping threshold produced by use of a 2+2 diode arrangement.

2) Uses a somewhat different passive 3-band EQ than one often sees on Ibanez and other pedals.  Bass control appears to be cut only, though the treble seems to be boost/cut type.

3) Second op-amp is uses as gain-recovery after tone controls rather than as part of active tone control as in other Ibanez pedals.

4) The clipping stage differs from the TS-9 in terms of rolloffs.  C5 and R8 give low-end rolloff around 268hz, compared to 720hz for TS-9.  Retention of more bottom is also why gain doesn't have to be substantially greater than TS-9 even though clipping threshold is effectively doubled by diodes.

A quick glance at the schem, and consideration of what we know from other pedals, suggests a bit more dynamics than a TS-9 and SD-1, not as much clipping as an SD-1, and a fuller less strident sound than a TS-9 or SD-1.  My guess is that the reason Parnell spoke so highly of it is because of these features.  Whether the CA3260 is a big player in all of this....who knows.  ut the general design approach seems interesting for those who like to fiddle with TS-9/SD-1 variations.


SELF-EDIT:  Sebastian posted his analysis while I was working on mine.  I'll post anyways and we'll see what comes out in the wash.

[WGTP]

Thanks for the analysis.  You could change those diodes to "Real" Mosfet clippers.   

[jrc4558]

nice analyses, gentlemen. I will try a PCBfor it over the weekend (still too far)

[NoFi]

Well from what i hear, yep it's dynamic, it's transparent in the sense that's there's not much difference between the "buffered bypassed" sound and when the effect is engaged if the gain is low and the eq properly setup. That's really good because the pedal allows to just dial in some light clipping to the usual clean sound.
It's smooth and not fizzy sounding in the top end that's a fact. Is has some gain but really to me if you push the knob past noon it sounds strange in a bad way. Like one is increasing the clipping but there is no oomph happening and it all remains in weak flabby sounding distodrive territory. It is good to drive other pedals or amp though.
And if there's no mojo in the diodes, that's cool i might tinker lol. I didn't know about the current thing.
I have an OPA2604 in the pedal right now. Tried 4558, tl072. I don't have the original op amp anymore since i thought the usual ones sounded at least as good. But it was well before i frequented this forum so i was a really stupid young fellow, even worse than now lol.
Thanks for explaining how the thing works, it'll help bring out the most of this pedal i really like.
I think it has quite a few ceramic caps, i'll take pics of the board tomorrow. Er and i'm not saying there is mojo in ceramic caps lol.

[stm]

Well from what i hear, yep it's dynamic, it's transparent in the sense that's there's not much difference between the "buffered bypassed" sound and when the effect is engaged if the gain is low and the eq properly setup. That's really good because the pedal allows to just dial in some light clipping to the usual clean sound.
It's smooth and not fizzy sounding in the top end that's a fact. Is has some gain but really to me if you push the knob past noon it sounds strange in a bad way. Like one is increasing the clipping but there is no oomph happening and it all remains in weak flabby sounding distodrive territory. It is good to drive other pedals or amp though.
And if there's no mojo in the diodes, that's cool i might tinker lol. I didn't know about the current thing.
I have an OPA2604 in the pedal right now. Tried 4558, tl072. I don't have the original op amp anymore since i thought the usual ones sounded at least as good. But it was well before i frequented this forum so i was a really stupid young fellow, even worse than now lol.
Thanks for explaining how the thing works, it'll help bring out the most of this pedal i really like.
I think it has quite a few ceramic caps, i'll take pics of the board tomorrow. Er and i'm not saying there is mojo in ceramic caps lol.

OK, so let's see some additional points:

1) Based on your experience, you haven't performed a *conclusive* A/B test against the original opamp, so this leaves the possibility of being even better than a standard unit. These guys (Ibanez) named it MOStortion for some reason; maybe it was just marketing, or maybe they found this MOS opamp had some (despite small) sonic difference with the rest.

2) The flabbyness at high gains might have to do with excessive bass at the input.  It would be just a matter to change the the 220nF cap (C5) to a lower value, like 100n, 82n or even 68n (82n produces a response similar to a TS).  It might be great to be able to improve the high gain territory of the pedal.

3) One way to force additional soft saturation at the opamp's output would be to connect a 10 uF cap at the output in series with a 1k to 4k7 resistor to ground just to add extra loading.

4) The other day I was puzzled by the Golden Drive schem.  It is a TS variant, but with much much smaller resistors around the feedback path of the opamp.  The gain pot in this case is 25k, while the resistor from the (-) input to ground is 620 ohms.  This indeed increases the current through the feedback diodes, thus slightly increasing their forward voltage drop.  This not only loads the opamp more heavily, but also makes the clipping curve look a bit more round and soft in the SPICE sims. I don't know for sure if there is an audible effect here, but something to take into account nevertheless.  At least the diode resistance is more relevant at higher currents, which is particularly noticeable with germaniums, then 1N4148's, and lastly for the silicon rectifier types (1N400x).

Cheers.

[Mark Hammer]

I have to say that I've never really understood the use of the term "flabby" around here.  That's not a criticism, it's just that I think there may be a bunch of us whose amps and speakers just don't permit such properties to be noticeable.  At the very least, it is worth noting that every large speaker has distinctive resonances which may or may not be compatible with rolloffs in a given pedal.

[tcobretti]

It seems to me that "flabby" is often how many of the fuzzier pedals are described.  I have learned that if someone calls a pedal flabby that I'll probly like it.  They never sound flabby to me, just fuzzy. 

[$uperpuma]

I say lets start a discussion about "flabby" in a brand new thread... I've never understood the term myself... How bout it Mr. Hammer?

[Mark Hammer]

You're on.

[RDV]

  "FLABBY"

[stm]

OK, here I redrew and simulated the characteristic curves of the Mostortion tonestack.
Looking for a similarity, it is not Fender, not Marshall, but it is the same topology as the tonestack used in the Umble, apart for some different values and an extra capacitor in parallel with the Middle pot.  The Treble control has a very wide action, spanning near 30 dB at 10 kHz.  The Middle control even has some limited boost capabiliy.  Interesting.