Event Horizon, or a Rotobone puts on a space suit

[kamadude1216]

UPDATE guys!! This was typed on my phone so the structure may be a bit messy, but here ya go!


Trombetta Rotobone Y-input rotary caps:


0.0082 uF/8.2 nF (yellow Cornell dubilier cylinder style) OR 10 nF since label is MKT .01

0.012 uF/12 nF (poly)

0.022 uF/22 nF (poly)

0.047 uF/47 nF (poly)

0.068 uF/68 nF (grey film)

0.082 uF/82 nF (poly)




Trombetta Rotobone X-output rotary caps:



0.0082 uF/8.2 nF (yellow Cornell dubilier cylinder style) OR 10 nF since label is MKT .01

0.022 uF/22 nF (poly)

0.047 uF/47 nF (poly)

0.068 uF/68 nF (grey film)

0.082 uF/82 nF (poly)

0.22 uF (poly)



Tested with Peak Atlas DCA55

Q1 transistors (NPN)


Si 2N5088: Hfe - 424; 0 mA leakage

Ge AC130 Hfe -109; 0.010 mA leakage



Q2 transistors (PNP)


Si 2N2907A: Hfe -225; 0 mA leakage

Ge 2N404: Hfe - 174; 0.137 mA leakage


Other important circuit notes (may be all over the place but I think I covered everything/most)

Diodes: 1N4148 symmetrical but opposite directional arrangement after Q2E but before Vol.


Input-Y and Output-X rotaries in parallel w/input and output caps


*I believe in the V1 Rotobone, the Presence switch is an SPDT On/Off/On with a 1n cap wired in Up position, and a 2n2 cap in Down position, just like the Minibone. Location is after the Volume pot but before the Output .


In either Rotobone version, 1n Output cap wired after Vol. Pot Lug 2/Middle Lug


22n cap wired in series off Q1B 560k ohm resistor


560k/150k ohm voltage divider on Q1B


Q1E to SPDT switch(Q1E to middle common lug and 1.2k ohm resistor wired to terminal ends for Hi/Lo power switch)


Second 560k ohm resistor in series with 560k ohm resistor off Q1B voltage divider section AND Q1C 3.3k ohm resistor


Input signal into 2k ohm resistor and other end of resistor goes into Gain pot "In" Lug (left most lug if viewing pot from the underside)


2k ohm resistor between ground and ground lug (right most lug of Octave pot when viewed from underside). Middle lug connected to Q2C? (I may be getting slightly confused since this portion is PNP transistors. It's either off Emitter or Collector, whichever is electrically correct). The experts can help clarify this one for me lol.



Here's a recap list of all the resistors used in Rotobone V2:


x2 560k ohm 1% tol.


x1 150k ohm 5% tol.


x1 3.3k ohm 5% tol.


x1 1.2M ohm 1% tol. (I believe this one is wired near input to prevent pops when switching the pedal on/off)


x1 1.2k ohm 5% tol. (for Q1E)


x2 2k ohm 5% tol. (one for the Gain pot and the other for the Octave pot)



Apologies on any confusing details. I'm no expert on schematics but I think I know enough to be able to decipher most of a somewhat simple circuit. The existing Trumpeter/Minibone PCB should be able to do the Rotobone V2 circuit. Some resistors may be ever so slightly different, but the overall wiring and configuration is the same.


You'll just have to find a way to do the input and output rotaries and the transistor switching. Vol and Gain are wired the same (more or less). On Rotobone, there's no Bone pot (which looks just like a simple tone panning switch to me like in the MK3 Tonebender/Buzzaround circuits). Octave is essentially an external Bias pot off Q2 and the Hi/Lo is just a switch off Q1, which I believe coltonius has implemented a very similar function in his modded Minibone/Rotoclone builds.


The only thing I haven't done is take voltage readings off Q1 and Q2 with the circuit activated, but I'd be glad to take those if needed. I just need some guidance on what ranges to look for.



Hope this gives the experts enough to work with!

*Anyone willing to return the favor by offering up a Tornita/Tornita KS2 circuit next? Pretty please? Lol

@coltonius @PedalBuilder @Robert

 

[PedalBuilder]

Awesome work! Thank you so much for the detailed measurements.

I have a few questions, and a draft schematic.

Input-Y and Output-X rotaries in parallel w/input and output caps

Do you mean that these switch the input and output caps? I didn't see any capacitors in parallel with the capacitors on the switches in the pictures.

In either Rotobone version, 1n Output cap wired after Vol. Pot Lug 2/Middle Lug

I'm assuming that this 1n cap goes from volume pot 2 to ground?

2k ohm resistor between ground and ground lug (right most lug of Octave pot when viewed from underside). Middle lug connected to Q2C? (I may be getting slightly confused since this portion is PNP transistors. It's either off Emitter or Collector, whichever is electrically correct). The experts can help clarify this one for me lol.

I think that this should go from the positive power rail, and the middle and remaining lugs should be connected to Q2E.

The only thing I haven't done is take voltage readings off Q1 and Q2 with the circuit activated, but I'd be glad to take those if needed. I just need some guidance on what ranges to look for.

That would be awesome. I'd recommend taking voltage readings at Q1B and Q1C and Q2E and Q2B. Q1E and Q2C should both be at ground (unless the high power switch is turned to low power, in which case Q1E will be above ground), so no need to measure those.

Here's my draft schematic:

 

[PedalBuilder]

Comparing this schematic with the Trumpeter/Mini-Bone, I think that it's interesting that he changed the value of R7 (it was 18k in the Mini Bone/Bone Machine) and removed the 47pf capacitor that was in parallel with R7. I'm guessing that the reduced gain from a lower collector resistor eliminated the need for the cap. Adding R2 is an interesting decision, too.

 

[kamadude1216]

Awesome work! Thank you so much for the detailed measurements.

I have a few questions, and a draft schematic.

Do you mean that these switch the input and output caps? I didn't see any capacitors in parallel with the capacitors on the switches in the pictures.

I'm assuming that this 1n cap goes from volume pot 2 to ground?

I think that this should go from the positive power rail, and the middle and remaining lugs should be connected to Q2E.

That would be awesome. I'd recommend taking voltage readings at Q1B and Q1C and Q2E and Q2B. Q1E and Q2C should both be at ground (unless the high power switch is turned to low power, in which case Q1E will be above ground), so no need to measure those.

Here's my draft schematic:
View attachment 80719

Just looked at things again. So the 47pf in parallel with R7 still exists. I snipped out and measured that mystery axial cap. It reads between 49-51 pF. The black diode is just a reverse biased polarity protection diode . Now here's the interesting parts, R7 is the 1.2M ohm resistor. R4 is the 150k resistor, and R5 is the 3.3k. I'll update again with voltages once I get the circuit back together since it's still in dissected form. Haha.

Also, there's no 1n cap to ground near the output. That's what I originally thought the 47pf cap was. Also, you're right, the input and output rotary cap switches don't connect to any in/out caps in parallel. I mistakenly assumed they were since I thought they'd sum capacitances with an already existing default cap for the ins/outs.

Also, for Q4/the second transistor pairs, I think Emitter and Collector labels should be swapped in the diagram since they're PNP.

I also think that the R3 1k resistor between Gain pot Lug 1 and ground doesn't exist in this one.

 

[PedalBuilder]

Just looked at things again. So the 47pf in parallel with R7 still exists. I snipped out and measured that mystery axial cap. It reads between 49-51 pF. The black diode is just a reverse biased polarity protection diode.

Both of those make sense to me and match the circuit for the Bone Machine/Mini Bone.

Now here's the interesting parts, R7 is the 1.2M ohm resistor. R4 is the 150k resistor, and R5 is the 3.3k.

Hmm, this seems off to me. The Bone Machine and Mini Bone have R4 = 560k; R5 = 150k; and R7 = 18k (using my schematic's designations). So does the Jordan Bosstone, which is the starting point for the Bone family of fuzzes (the Bosstone schematic is at the bottom of this post). So I'm pretty confident that R4 should be 560k and R5 should be 150k. I think you might have read the 1M2 resistor backwards, and if it's one of the Tayda resistors, the pale yellow might actually be gray, which would make it an 18k resistor, and a good fit for R7. Can you try measuring the 1M2/18k resistor to see if that's the case? I also spent some time tracing out the picture, and it looks to me like the 3k3 resistor is connected from the 150k resistor to ground, like this:

I wonder if the 3k3 resistor was added to make up for the 5% tolerance of the 150k resistor; if the 150k resistor's actual value was 2% below its nominal value of 150k, adding a 3k3 resistor would bump it back up to 150k.

Also, for Q4/the second transistor pairs, I think Emitter and Collector labels should be swapped in the diagram since they're PNP.

That's one of the weird things about the Jordan Bosstone (PedalPCB Executive Fuzz) family of fuzzes. It uses NPN and PNP transistors, and the PNP transistor is wired as a buffer, with the collector connected to ground and the emitter connected to VCC through a resistor. Compounding the weirdness, Q1 and Q2 are DC coupled (there aren't any caps blocking DC between the collector and base of Q1 and the base and emitter of Q2). So weird as it is, the emitter/collector labeling should be correct.

I also think that the R3 1k resistor between Gain pot Lug 1 and ground doesn't exist in this one.

Yeah, that one is an addition of my own. I added that because otherwise the pedal mutes when you turn the gain all the way down. I'm a firm believer that a pedal should only have one knob that mutes it, especially if the solution is adding a single resistor.

 

[kamadude1216]

Both of those make sense to me and match the circuit for the Bone Machine/Mini Bone.

Hmm, this seems off to me. The Bone Machine and Mini Bone have R4 = 560k; R5 = 150k; and R7 = 18k (using my schematic's designations). So does the Jordan Bosstone, which is the starting point for the Bone family of fuzzes (the Bosstone schematic is at the bottom of this post). So I'm pretty confident that R4 should be 560k and R5 should be 150k. I think you might have read the 1M2 resistor backwards, and if it's one of the Tayda resistors, the pale yellow might actually be gray, which would make it an 18k resistor, and a good fit for R7. Can you try measuring the 1M2/18k resistor to see if that's the case? I also spent some time tracing out the picture, and it looks to me like the 3k3 resistor is connected from the 150k resistor to ground, like this:
View attachment 80727
I wonder if the 3k3 resistor was added to make up for the 5% tolerance of the 150k resistor; if the 150k resistor's actual value was 2% below its nominal value of 150k, adding a 3k3 resistor would bump it back up to 150k.

That's one of the weird things about the Jordan Bosstone (PedalPCB Executive Fuzz) family of fuzzes. It uses NPN and PNP transistors, and the PNP transistor is wired as a buffer, with the collector connected to ground and the emitter connected to VCC through a resistor. Compounding the weirdness, Q1 and Q2 are DC coupled (there aren't any caps blocking DC between the collector and base of Q1 and the base and emitter of Q2). So weird as it is, the emitter/collector labeling should be correct.

Yeah, that one is an addition of my own. I added that because otherwise the pedal mutes when you turn the gain all the way down. I'm a firm believer that a pedal should only have one knob that mutes it, especially if the solution is adding a single resistor.

You're right, R7 is 18k. Same with R4 560k and R5 150k. That grey looked yellow to me in certain lighting. That makes way more sense now! Haha! Also, yes, the 3.3k is connected from the 150k to ground like you stated. Used my DCA55 again to read the second (PNP) transistor pair pin outs and it keeps saying that the side of the 2n404 that's connected to ground is the emitter?

Gotcha on the 1k resistor preventing the full mute, quite clever!

Attached are pics of the PNP transistor pair readings I get though. Please help me understand and explain that to me in caveman terms you said collector is the one that should be connected to ground, right?

 

[PedalBuilder]

cYou're right, R7 is 18k. Same with R4 560k and R5 150k. That grey looked yellow to me in certain lighting. That makes way more sense now! Haha! Also, yes, the 3.3k is connected from the 150k to ground like you stated.

I've had the same experience with the gray on the Royal Ohm brand of resistors. Spent way too much time trying to figure out where the 140k resistor on my breadboard came from before realizing that it was a gray stripe and a 180k.

Used my DCA55 again to read the second (PNP) transistor pair pin outs and it keeps saying that the side of the 2n404 that's connected to ground is the emitter?
. . .
Attached are pics of the PNP transistor pair readings I get though. Please help me understand and explain that to me in caveman terms you said collector is the one that should be connected to ground, right?

Basically, a PNP transistor needs to be flipped relative to an NPN because of the reversed polarity. In a PNP transistor, current flows from the emitter to the collector, unlike NPN transistors, where current flows from the collector to the emitter. The direction of the arrows on the symbols is indicative of this. As a result, the emitter should be more positive than the collector. So if you connected the emitter to ground and the collector to VCC (through a resistor), the transistor probably wouldn't work. Like the Bosstone, the Mini Bone/Bone Machine have the PNP transistor wired this way. My best guess for what you are seeing is that the transistors are wired differently from my schematic, with the emitters wired together and that wire is connected to the octave pot, and the base and collector are wired to the switch?

 

[kamadude1216]

So from what I can see in my V2 Rotobone, Q1 switch common wire/terminal that has NPN collectors is linked to Q2 PNP base, Q2 PNP emitters wired together and tied to ground. Q2 PNP collector common terminal off Q2 selector switch wired to X/Output rotary "in" terminal.

*NPN Q1 emitters soldered together and goes to common middle terminal on Hi/Lo switch, Q1 bases link to common DPDT terminal that's wired to the "out" lug of the Y/Input rotary. Hopefully that's not too confusing.

 

[PedalBuilder]

So from what I can see in my V2 Rotobone, Q1 switch common wire/terminal that has NPN collectors is linked to Q2 PNP base, Q2 PNP emitters wired together and tied to ground. Q2 PNP collector common terminal off Q2 selector switch wired to X/Output rotary "in" terminal.

*NPN Q1 emitters soldered together and goes to common middle terminal on Hi/Lo switch, Q1 bases link to common DPDT terminal that's wired to the "out" lug of the Y/Input rotary. Hopefully that's not too confusing.

Nope, quite clear and easy to follow. I'll try it out on the breadboard tomorrow and see what happens if the PNP transistor is run with the emitter connected to ground.

 

[PedalBuilder]

So I breadboarded the Rotobone, and you're 100% correct. Reversing the PNP transistor eliminates a lot of the unpleasant low end oscillation and gives much more intense trumpet tones. So with that, I think we've got a final schematic!

 

[kamadude1216]

Yay!! I'm glad it worked out! So the 3.3k resistor off the 150k on mine is most likely there to compensate for the tolerances then?

 

[PedalBuilder]

Either that or to confuse people like us. The 150k resistor has a ±5% tolerance, so anything less than 7.5k is within the margin of error. The version on my breadboard sounds exactly like the demos, so I’m pretty sure we’ve got this one right. Thank you again for the pics and measurements!

 

[kamadude1216]

Either that or to confuse people like us. The 150k resistor has a ±5% tolerance, so anything less than 7.5k is within the margin of error. The version on my breadboard sounds exactly like the demos, so I’m pretty sure we’ve got this one right. Thank you again for the pics and measurements!

Absolutely! Glad I could contribute! So what's the next step/timeline on a board being made for this? Also, curious what transistors you used on your breadboard tests

 

[PedalBuilder]

Absolutely! Glad I could contribute! So what's the next step/timeline on a board being made for this?

I don’t know if @Robert has any plans, but I’m in the process of laying out a PCB and should have something ready for fabrication in the next week or so. I’ll keep you posted.

Also, curious what transistors you used on your breadboard tests

I verified it using transistors with very similar specs to the ones in your unit:
NPN Silicon - BC550B, hFE 425
NPN Germanium - Motorola house number (equivalent to 2N1306) - hFE 93, leakage 10µA
PNP Silicon - 2N3906, hFE 225
PNP Germanium - 2N1309, hFE 172, leakage 125

I tried quite a few others to test how sensitive the circuit is to gain and leakage. Here's what I found.

NPN Germanium - You need really low leakage here. Higher leakage quickly leads to more aggressive gating. All of my NPN germaniums except the one 2N1306 have more than 70µA leakage, so I couldn't experiment with the impact of hFE on the tone.

• Motorola house number (equivalent to 2N1306) - hFE 93, leakage 10µA. This sounded the best to me. It didn't get very gated until high levels of the octave knob. The gating had a low threshold, making it very musical and easy to play with.
• Texas Instruments house number (equivalent to 2N1308) - hFE 150, leakage 70µA. More distorted than the 2N1306, but it got gated very early in the octave knob, and not in a pleasant or useable way. The threshold for the gating effect was very high, so only the peaks of notes were audible.
• Soviet MP35 - hFE 60, leakage 120µA. Very sputtery and weak, very gated.

NPN Silicon - Pretty much anything works here. I tried a wide range of gains, from 2N2222A (hFE 140) to BC108 (hFE 710), and lots of others with gains somewhere in between (2N3904, 2N4401, 2N5088, 2N5089, BC109C, BC549, BC549C, BC550B, etc.). All of them worked well and sounded good. The lower gain transistors were less compressed and slightly darker, but the differences were subtle.

PNP Germanium - I tried a bunch of different types here—2N404, 2N1307, 2N1309, CV8252, GT309B/E, and GT310B. Anything with leakage under ~200µA and hFE over ~90 works, but the best results are with an hFE above 150, with higher gains having more trumpet-like tones. Interestingly, it's not a linear progression to my ear. Transistors with gains below ~90 don't get any horn sounds. Anything between 90 and ~150 sounded roughly the same, with a slight increase in trumpety tones as the gain increase. Somewhere between 150 and 165, there's a sharp inflection point, and the horn sounds suddenly become a lot more pronounced, with increasing trumpet tones as the gain increases. Leakage needs to be below 200µA or the circuit gates as soon as the octave knob is past 9:00.

PNP Silicon - Like the PNP germanium, the trumpet sounds become much more pronounced once the hFE is above ~150, and become more pronounced as gain increases. I didn't test anything with an hFE above 250, but everything between 150 and 250 worked really well.

 

[Ginsly]

I was going to chime in and say "it'll take some time for PedalPCB to get to this (if inclined), but I bet @PedalBuilder already has an order placed for boards! Looks like I was right! Ha... I'm behind on builds, but I'm verrrrrrrry interested in this, as a Bosstone fanatic. Incredible work, @kamadude1216 and @PedalBuilder.

In the meantime I have a Trumpeter pcb, but I'm not quite sure what I'd have to do to it... My guess is that it wouldn't be terribly difficult, but I'm still pretty green.

Sorry for the aside, but regarding the Rotobone and other rotary-switched pedals - can I get a recommendation for A) a decent pcb-mounted rotary switch and B) a decent solder-lug version from Tayda et al? Seems like 1P8T and 1P12T are the two major ones to look at. I haven't delved into that world yet, and I can already see that the shafts are typically knurled (and oddly short?) or D-shaped, making it possibly difficult to have a matching set of knobs on a pedal that presumably would have mostly 6.35 smooth shafts.

 

[PedalBuilder]

In the meantime I have a Trumpeter pcb, but I'm not quite sure what I'd have to do to it... My guess is that it wouldn't be terribly difficult

No idea if PedalPCB plans on making a board, but in the meantime you’re right that it’s not very difficult to build one of these on a Trumpeter board. All the information that you need has already been posted in this thread, so feel free to get building if you’re so inclined!

 

[Ginsly]

No idea if PedalPCB plans on making a board, but in the meantime you’re right that it’s not very difficult to build one of these on a Trumpeter board. All the information that you need has already been posted in this thread, so feel free to get building if you’re so inclined!

I’ll try my best! Need to get rotary switches sorted out, though. I’ll make a separate post.

 

[kamadude1216]

I’ll try my best! Need to get rotary switches sorted out, though. I’ll make a separate post.

If it helps any, I read Alpha on both 6T rotaries Paul put in my unit

 

[PedalBuilder]

So what's the next step/timeline on a board being made for this?

Initial draft of the board is done. I'm still working on the artwork, so there might be some changes to the board depending on how that goes, but here's what I have at the moment:

 

[kamadude1216]

Initial draft of the board is done. I'm still working on the artwork, so there might be some changes to the board depending on how that goes, but here's what I have at the moment:

View attachment 81053

Oooh looks exciting! I'm interested in a couple pieces too. I'll message you privately