Adding Bias Mod To Marigold (Yellow Hybrid Tone Bender)

debrad

Active member
Hi everyone,

Slowly getting back to my build projects after our move last fall and one of the first things I'm going to tackle is the Marigold.

I love having an external bias control to adjust how smooth or spitty the fuzz sounds so I'm wondering which resistor would be the one to replace...R12 (1k2) or R14 (3k3). In either case, would you recommend a 5K pot, a 10k pot, or something even bigger (I used a 10k pot in place of the 5k6 resistor on TagboardFX's silicon bender and thought it worked quite well...)?

Thanks in advance!

- brad -
 
I don’t have direct experience with the
Marigold but if we were to help we’d need to know the following:

1.) Q3 Collector Resistor you have installed.

2.) Q3 Collector Voltage.

I can tell you from experience with 2-3 transistor English fuzzers that things do t get glitchy until you get below 2 vDC and harder gating doesn’t come until above 8 vDC so you need a substation swing in resistance to get that kind of voltage swing.
 
Thanks @Big Monk . The schematic shows R12 (1k2) and R14 (3k3) tied to Q3 (along with D2 (4V7)) and from what I have read on the PedalPCB threads where they traced this circuit, folks are recommending a Q3 voltage of about 2.5V. My IMPRESSION is that R14 would be the likely candidate to replace with an external potentiometer but seeing two resistors in that path leaves me a little uncertain (plus, I've read discussions where people said adjusting the Q2 bias was more effective).

Marigold Schematic.JPG
 
R12 and D2 form a 4.7V regulator. R12 is a good value for making that regulator work properly. If you change R12 enough to make the voltage on D2 change, then you've broken the regulator.

R15 forms a feedback loop around Q3. That feedback loop will oppose any change in Q3's collector voltage. That means that you would need to make a very large change in R14 to get a big enough change in Q3's collector voltage to cause or cure gating. That makes R15 the most logical place to adjust Q3's collector voltage because R15 has the greatest influence.
 
The bias section of this circuit is new to me so thanks so much for the explanation, I really appreciate it.

In terms of more details on the values you've recommended, I assume that the R15 swap to 100k ensures there is always be SOME resistance in that feedback loop but, with that in place, I would have thought something like a B250k or B500k pot would put the stock 220k "closer" to the centre of the rotation (i.e. 100k from the fixed resistor + 125k at the mid sweep) but perhaps they don't add enough resistance to really starve Q3? Does the B1M pot just give more range to "unbias" or is there another reason why you'd suggest B1M?
 
The bias section of this circuit is new to me so thanks so much for the explanation, I really appreciate it.

In terms of more details on the values you've recommended, I assume that the R15 swap to 100k ensures there is always be SOME resistance in that feedback loop but, with that in place, I would have thought something like a B250k or B500k pot would put the stock 220k "closer" to the centre of the rotation (i.e. 100k from the fixed resistor + 125k at the mid sweep) but perhaps they don't add enough resistance to really starve Q3? Does the B1M pot just give more range to "unbias" or is there another reason why you'd suggest B1M?
your goal is to create a way to add significantly more resistance in that feedback loop as well as the ability to decrease the resistance to 100k. Try that value and see if you like the range of adjustment it provides. And if your useful range is only a part of the range of the pot you will still have more information about what other pot value you might prefer.
 
So, after tackling a couple of other builds, I finally managed to get my Marigold put together using the B1M and 100k resistor mod that @Chuck D. Bones recommended. At first, the external bias knob was having ZERO impact on the Collector voltage so I double checked the pin configuration on my MP38 transistor, flipped it around, and now the Collector voltage ranges from 1.7v - 3.5v (~3.0v at mid-rotation) with a 9.42v input.

Given those numbers, does it make sense that I am not hearing any difference in the sound of the fuzz over the full sweep of my external bias knob? Do you think that I would get more of a spitty fuzz at one end of the spectrum if I changed the 1N4732A-6 I'm using in D2 to something that would bring my max. Collector voltage closer to 2.5 or 3.0 volts?
 
Are you aiming for the sputtering/dying battery effect? Do you hear any difference in the output when you change the bias control from the higher to lower voltage? How high of a signal level are you putting into the pedal?
 
Where do you have the FUZZ control set? You may only hear a difference at very low FUZZ settings. Q2 can drive Q3 hard enough that it won't matter what the bias is. Try making the 100K resistor smaller.

I'm not convinced that a bias control is appropriate for this type of fuzz.
 
I was hoping to get a sputtering/dying battery type of effect but, as Chuck is saying, maybe this isn't the circuit to do that with? For the record, the sound of this build (once I got the transistor oriented properly) is VERY good to me and I have other pedals that do the velcro/sputter quite well so, if this mod isn't really going to work, I don't have a problem with it. At the end of the day, I really just wanted a ToneBender type pedal that sounded as good as the vero-based silicon version I built before but with the option to adjust the treble/bass content and this one does that in spades AS IS!

So far, my tests have been done with a Strat with vintage output pickups straight into the Marigold. I've tried sweeping the bias control with a wide mix of minimum and maximum level, fuzz, and tone control and...so far...the only things that seem to create any kind of sputter are turning down the level and turning down the guitar's volume knob.

If I opt to play around more with R15...and that may depend on how eager I am to pull the circuit out of the enclosure...how low do you think I can go with a replacement @Chuck D. Bones? What happens if I replace D2 with something that puts my max Collector voltage closer to the recommended 2.5 volts (i.e. something that lowers the minimum voltage even further below the 1.7v I'm getting now)? I'm only asking the latter because I have D2 in a socket so I can play around with that easily enough.
 
Try 10K. If you find a useful range on the BIAS control, measure the resistance of the pot at both ends of that range. That will guide you on selecting the best values for the pot and R15.

What you want to do is drive the collector voltage down to around 100mV. When Q3 is saturated, almost no sound gets thru. When you hit Q3 with enough signal, it overcomes the bias and drives Q3 out of saturation. That's how gating is done.

Fiddling D2 is not going to get you there.

Are you sure you have D1 pointed the right way? Which diode are you using for D1? Russian diodes are mostly marked with the band on the anode.
 
Thanks @Chuck D. Bones. I will continue to play around with this one but will focus my attention on that R15 resistance rather than worrying about the other aspects of the circuit around Q3.

I am using a D9G diode in D1 and, as you describe above, I believe it has a band on the anode so my band is pointing to ground rather than Q3 (i.e. completely opposite to build document).
 
OK...just swapped out the 100k I had in series with the B1M pot to replace R15 with a 10k and the voltages now range from 0.38 - 3.75v. I still don't get the gating I had hoped for but I do hear the slightest variation and, as Chuck stated, maybe there might be a little more once I start fiddling with the interaction of bias, level, and fuzz.

One question I have is whether it matters which lug of the pot I tie the resistor to? I switched them when I swapped from 100k to 10k (moved resistor to middle lug instead of outside) and I feel like there is now an abrupt change at the end of travel that wasn't there when they were wired to opposite way...

Regardless of all of that, I'm VERY happy with the sound of this one...especially since I just plugged in the highest hfe MP38 of the five I got from ebay (~75hfe) and used a random 4.7v zener as per the build documents! Hats off to PedalPCB and all the folks to worked on figuring this circuit out!
 
I'll close the loop on this one by saying I have called this one "complete" (for now ;)) with:
  • 9.4v input
  • "mod" = 10k resistor and B1M potentiometer replacing R15
  • max bias on Q3 collector = 3.27v
  • min bias on Q3 collector = 0.367v
  • 2.5v bias achieved at approximately 11:30 on bias knob
  • 2.7v bias achieved at Noon on bias knob

No real "velcro" or "gated" fuzz with starved bias but the mod does seem to alter the "saturation" of the fuzz.

So, not exactly what I was aiming for but, all in all, a very satisfying build nonetheless since this is the best Tone Bender style fuzz I have ever had the chance to play with.

I'll leave you all with a couple of shots I took of the final product:

Marigold01.JPG Marigold02.JPG
 
Last edited:
Thanks @mdc. Just scanning the interwebs for tips on dying battery simulators (schematics, potentiometer values, etc.) as I'm now thinking a separate box might be useful for test driving that kind of mod on other pedals too. In my case, I don't really need it to mimic ALL of the parameters that change as a real battery dies so simply lowering the voltage is probably good enough. If I really like what it does, I can then implement it directly on the circuit.
 
Back
Top