Boss OD-3 Gain & diodes Mod

[eh la bas ma]

Hello,

I opened my OD-3 unit for the first time and realized it was a through-hole build, no smd parts. There are also some spaces to add toggles or 9mm pots...

I am perfectly happy with the EQ on this overdrive, I like it the way it is, and i am not looking for a bass cut.

I wonder if it would be possible to get more gain from this circuit ? Is there a simple way to push it into distortion territory ?

Also interested by a diode clipping mod, but i am not sure if the best location is D10, D11, or D9, D12, or both ?

How would you do it ?


I searched the forum and could only find this thread from 2004, with a couple of unanswered questions :

https://www.diystompboxes.com/smfforum/index.php?topic=21170.msg131084;topicseen#msg131084

Every suggestions and observations are welcome,

Thanks for your help !

[Steben]

Add a booster or low gain overdrive pedal in front. It keeps the original and you have versatility.
I have always found the most useful mods on Boss pedals to be cutting gain rather than adding.
Best example is DS-1 where adding some treshold voltage to the clipping diodes (for example LED's instead of silicon) gets you more output with less gain/compression.

[bartimaeus]

the magic of the OD3 is the cascading gain stages. mess with one stage too much, and it'll change the feel and response of the pedal.

imho the diode clip mods are silly. they replace the soft clipping with a bunch of opamp hard clipping. totally changes the character of the pedal, making it harsher and lower gain.

better to double up the diodes or replace them with LEDs. that will increase volume and decrease gain without extreme changes to the pedal. if you do, change D10+D11 AND D6+D7. don't mess with D9+D12.

you could increase the value of the drive pot and you'll increase the gain range of that first gain. a 500k or 1M pot would give a lot more clipping for example. alternatively you can lower the value of R51, but at 470R it's already quite low.

[eh la bas ma]

Thanks for your replies !

So far I added an extra toggle allowing to choose between two D9E in series instead of anti-parallel at D9-D12 and D10-D11 (4 ge diodes total), and original boss diodes.

Ge diodes sounds slighly brighter, loosing a bit of sharpness at max gain, without noticeable changes in saturation level. I remember both diodes setting reads around 0.6V on diode test.

I tried with D9E in series with 1N4001, around 0.9V on diode test, and it decreased the saturation along with a volume drop.
I didn't touch D6-D7 yet.

Another extra switch at R51, toggles between 470R and 330R. I'd say it cuts a slight bit of low-ends, doesn't increase gain but acts more like a subtle "bright" switch. 10R gives a more drastic change, cutting even more low ends. R51 action reminds me of some"presence" controls on other gain circuits. I am sure I didn't confuse the resistor location, because the one I desoldered reads 470R.

I will try 250K, 500k and 1M pots on the gain control, 3mm red leds at D6, D7, D10, D11, and report.

Thanks for the suggestions !

[Rob Strand]

Another extra switch at R51, toggles between 470R and 330R. I'd say it cuts a slight bit of low-ends, doesn't increase gain but acts more like a subtle "bright" switch. 10R gives a more drastic change, cutting even more low ends. R51 action reminds me of some"presence" controls on other gain circuits. I am sure I didn't confuse the resistor location, because the one I desoldered reads 470R.

The open loop gain of the amplifier is low so when you change "gain" resistors for more gain you reach a point where you don't get any more gain.

There's a difference between gain and signal level as well.  Once you stick in clipper diodes the level becomes fixed.

In this thread I make a BJT only amplifier have the same properties as the Boss JFET design.   You could go the other way and replace the JFETs with BJTs and get more gain.   You have to watch out for DC through large gain pot resistors as it will shift the bias point and also create pot crackle.  Best to leave the gain pot and reduce the gain resistor to ground.  FYI, Ibanez use all BJTs in some of their designs.

https://www.diystompboxes.com/smfforum/index.php?topic=129424.0

[eh la bas ma]

I resoldered R51, because lowering it sounds more like an EQ variation, and used the same switch to toggle between stock 1N4001 and 3mm red leds at D6-D7. Diode test on the switch reads 0.6V and 1.6V.

Both switches set downward I have standard OD-3 diodes and sounds, upward I have ge diodes and red leds which gives something interesting, hard to describe, but i like it. It's closer from what I am looking for. The saturation feels somewhat less processed, less sophisticated, unleashed at max gain.

I will try this configuration for a few days, see how it behaves with a volume boost and other effects.

I do have some BJT transistors in my stash, like 2N2222. I might give it a try, thanks for the suggestion !

[Rob Strand]

I do have some BJT transistors in my stash, like 2N2222. I might give it a try, thanks for the suggestion !

You can try playing with the 2.2k resistor on my circuit, 2k7 on the OD3.   The actual value depends on the source resistor 4k7 on my circuit and 6k8 on the OD3 schematic.   (Then IIRC on top of that Boss chose the value to be a bit skewed in that the two transistors/JFETs don't bias at the same current.)

FWIW, that OD3 schematic has an error.  The "Discrete Op Amp" is operating at a lower voltage, like some of the the Boss pedals with similar circuits.  If you have an original then it will be right.

[bartimaeus]

rob makes a very good point. i was thinking of the fulltone ocd, which switched from between a 500k and a 1M pot. that pedal has a 2.2k resistor, so the switch was from 227x gain to 454x gain. i do wonder if that extra gain range with the 1M pot is totally usable, or if the pedal maxes out after a point.

on the od3, the resistor is already a low 470R. so with a 100k pot you get 212x gain at max. so even a 250k pot (531x gain at max) would be pushing the limits of that stage, but probably worth hearing if you have one to hand.

edit: just tried it in simulation, and the amount of clipping only slightly changes as max with a 250k pot. so maybe not worth trying after all haha!