This Week on the Breadboard: The FireDrake

Chuck D. Bones

Circuit Wizard
Cooder had asked me if there was a way to build the VFE Dragon without having to use the C10K/C100K dual pots. I have been tossing that same thought around in the back of my mind for some time and I knew it was possible. The only tricky bit might be doing it with a reasonable parts count. Here is the Dragon schematic. Think of it as a Timmy with steeper high-pass and low-pass filters. Right after the input buffer is a 2-stage passive high-pass filter. The trick Peter Rutter uses is he scales the impedances of two stages 10:1 so that the filter has a sharp corner at the -3dB freq. He does the same thing with the low-pass filter after the distortion stage. There's no free lunch and the price we pay to do it this way is we need a dual pot with two different resistances. It's a custom part and difficult to come by. One of the alternatives is to build an active 2nd-order filter. I went that route. It only took one more opamp and a few more R's and C's to do it.

VFE Dragon.gif

If we want filters that have HUGE range and are very well-behaved like the ones in the Dragon, we still need dual pots, but at least the two halves can be matched. I did a little experimenting to see if I could make some acceptable compromises in the filter design so that I could use single-gang pots. This was the result. It works well. I'm still tweaking a few component values, but it's very close to done. I kept the input buffer (U1B), but upped the gain from 1x to 5x. If you're running hot pickups you might want to reduce R5. U1B, C2-C4, R6-R8 and VR1 form an active 2nd-order high-pass filter. It's tunable from 40Hz to 1.2KHz and has 12dB/octave slope for most of the range. The midrange frequencies are attenuated at more like 6 to 8 dB/octave. I wanted to have different clipping options, so I added a 6-position rotary switch. So far I've only run the breadboard with the blue LED / 1N4003 combination and with no clipping diodes. The COMP control is anything but subtle. I has a major influence on the compression and harmonic content. U3A, C8-C10, R11-R13 and VR4 form an active 2nd-order low-pass filter. It's tunable from 600Hz to 7.3KHz and has 12dB/octave slope across the entire range. I tried an NE5534 and a CA3130 for U2. Both work well. The NE5534 clips asymmetrically when driven to the rails. The CA3130 is much more symmetric. Many different single opamps will work, just pay attention to whether a compensation capacitor (C6) is needed and which pins it uses. With the clippers out of the circuit and the GAIN kept low, this pedal will play clean. By adjusting the controls you can get just the right amount of dirt, from a wee-bit to a metric shit-ton. Added bonus: this circuit can be run on 18V for additional headroom and volume.

FireDrake OD v1.01a sch.png
 
Last edited:
This is the high-pass filter's freq response for various BASS pot settings. Green is BASS at zero; dark red is BASS at 10.
FireDrake v1.0 bass response.png

And the low-pass filter's freq response for various TREBLE pot settings. Dark red is TREBLE at zero; green is TREBLE at 10.
FireDrake v1.0 treble response.png

Here is the gain stage freq response. Max gain in this stage is about 56dB. Not huge, but enough. The input buffer and HP filter add another 20dB. Green is GAIN at zero. At higher gain settings, C7 shaves off some of the higher frequencies.
FireDrake v1.0 gain response.png
 
I agree 100% @cooder .

I'm starting this today. I'm going to put all of the clipping options and switch on a separate board, so I can use it again with future builds. I like the choices listed in the schematic.
 
I'm "rounding the bend" on the schematic and have a question; Why not use a quad opamp?

Edit: I'm trying it with a TL074.
 
Last edited:
Great stuff indeed! How useful is it to have an empty spot #6 on clipping rotary if we also have the comp pot, wouldn't it a fully CCW be pretty much like no clipping or is there enough discernible difference? Or is it just because we can ease of use?
 
Any more opinions on the Comp control? I'm suspicious I'd be inclined to make it a set value.
 
  • Like
Reactions: fig
I'm "rounding the bend" on the schematic and have a question; Why not use a quad opamp?

Edit: I'm trying it with a TL074.
You can do that. There are two potential downsides.
1) Two duals vs. one quad doesn't save any real estate on the circuit board but it does complicate the layout. This could actually increase the required board area.
2) There is 56dB of gain between U1A and U3A on my schematic. We want to have as little stray coupling as possible between the input and output of the U2 gain stage. It is much easier to ensure isolation between input circuitry & output circuitry with two duals.

With low-gain circuits like phase shifters and uni-vibes, quad opamps are fine.
 
Any more opinions on the Comp control? I'm suspicious I'd be inclined to make it a set value.
You should try it and decide for yourself. Breadboard it and see if you like varying the COMP control.
It's equivalent to a clean bleed. Makes a big diff with the Ge + Schottky clippers. It also lets you vary the asymmetry when the clippers are asymmetric (blue LED & 1N4003). The other thing it does is allow both kinds of clipping at once: opamp saturation and diode soft clipping.
My personal opinion is COMP belongs on the front panel.

One more thing... Different clipping diode selections may very well require different COMP settings.
 
Last edited:
Great stuff indeed! How useful is it to have an empty spot #6 on clipping rotary if we also have the comp pot, wouldn't it a fully CCW be pretty much like no clipping or is there enough discernible difference? Or is it just because we can ease of use?
The COMP control never completely removes the diodes from the feedback path. If you want only opamp clipping, or a Clean mode, then you need the "no diodes" option on the rotary switch. 6 is not a magic number. The rotary switch is a single pole, so it could be up to 12 positions. Some rotary switches come with an adjustable stop so we can get from 1 position (not very useful) to 12 positions, or anything in between.

There might be 1 or 2 more asymmetric clipping options we'd like to have.
 
If you want a 6/12 dB/oct switch on the TREBLE control, install a SPST toggle in series with C8. Put a 1MΩ anti-pop resistor across the switch.

I don't think it's necessary to add a switch on the BASS side.
 
I'd read that quads aren't as good at handling high-gain, but I've still always wondered about quads vs 2x duals.

Hang out in the Boneyard and you can't help but learn somethin'.

Dojze sai, Sifu Bones.
 
  • Like
Reactions: fig
I find it a tad funny thinking about a VFE with more options. I'm watching how this unfolds because I really like the Dragon and am always up for a fun CB build.

The Dragon was a mainstay on my board for a few years till it got bumped by Bio77's superb enclosure for a Seymour Duncan pickup booster clone. I used it as more of a boost with a tilt in the top end.
 
I find it a tad funny thinking about a VFE with more options.
I hear ya. If I could only have one clipper, it would be a toss-up between the blue LED/1N4003 and the back-to-back MOSFETs. They both sound good to me, but are quite different.

As I'm putting together a BOM, I've decided to go with an 8-position rotary switch so I can play with ALL of the clipper options on my prototype.

I'm still tweaking the component values in the filters, but it's damn close.
 
Back
Top