Equal gain but different amount of clipping in bjfe honey bee and dyna red

[hans h]

Hi all,

Something has me puzzled. The honey bee from bjfe is a low gain overdrive (see procyon legacy schematic online).
The bjfe dyna red is a medium to high gain overdrive / distortion. The schematic can be found here:
http://4.bp.blogspot.com/-UbPB-VgJTEE/TxCpOqhAmxI/AAAAAAAAAe4/e25XRy5BOOM/s1600/Bearfoot+DRD.jpg

So listening to them online and based on impressions I read, there is supposed to be a huge difference in the maximum amount of clipping, yet the q1 gain is equal (500 k pot with 1k to ground), both have leds in noninverting q1, both have passive clipping diodes of CA. 0.6 forward voltage after q1. Only difference I see is green leds in the honey bee and red in the dyna red, but that difference is really small. Can anyone enlighten me on what is going on?

BTW the freq response is quite different, but one would expect that the larger amount of pre-clipping bass in the honey bee would actually lead to more apparent clipping.

Thanks in advance, hans

[ElectricDruid]

I don't have an original Honey Bee schematic, but checking against MadBean's Yellow Shark clone, I agree the two circuits seem very similar.

Could it just be that one person's "low gain overdrive" is another person's "heavy crunch"? People do vary *a lot* in how much distortion they like in their signal, and perhaps their descriptions reflect that?

[hans h]

I added some YouTube clips for reference. Would say that the dyna red definitely sounds gainer. I think this type of circuit is a nice starting point for exploring different Amp like tones so that's the reason I'd like to understand. Might be nice to take the dyna red in a mesa'esque  direction with a third clipping stage based on an inverting opamp and some scooping post-clipping. I'll play around

https://m.youtube.com/watch?v=axYZmqvvGiU&t=12s

https://m.youtube.com/watch?v=cZmpn1Ijiys&t=104s

[ElectricDruid]

Perhaps it *is* just down to the frequency response. If you put some fizz in, you'll get lots of fizz out. If you roll off the tone going in, you'll get something that adds a bit of dirt, but doesn't sound like it's "turned up" as much.

Maybe. Try it and see.

[idy]

Difference between 2 and 3 volts is small for clipping? Lower clipping = more distortion.

[hans h]

Good suggestion Tom, had not thought about the way we experience clipping. It may indeed be the case that higher frequencies sound gainier somehow. In regards to idy: I borrowed this from another diystompboxes thread: "red is generally under 2V, orange and yellow about 1.8-2V, green about 2-2.4V and blue 3-5V". I measured some greens this morning that came in at 1.8V. So there doesnt seem to be a whole lot of difference. And the higher clipping threshold of the greens will subsequently cause more clipping in the passive 1n400x pair that follows. At any rate, I'll play around with the circuit and let you guys know if I come up with a useful Mesa makeover

[m4268588]

These three have different sounds. Are there many differences?
What is the opinion that your ears claim?

[ElectricDruid]

These three have different sounds. Are there many differences?
What is the opinion that your ears claim?

There aren't *many* differences, but I'd say there are *significant* differences. Red and Blue LEDs typically have quite different forward voltages, so swapping between those two is going to provide either more clipping or more headroom and a more dynamic sound. They're also wired in the feedback loop, so once the diodes conduct, the gain drops to unity, whereas with the silicon clipping diodes to ground, you've got both a very low clipping threshold (0.6V or so) and no gain at all once the diodes conduct.

So yeah, they sound different, a bit.

[hans h]

Hi, sorry for my absence from the thread.

I've been running some ltspice simulations on the honey bee up till the fet and found the answer. The reduced clipping I and others seem to hear in the honey bee compared to the dyna red is easily explainable by the additional feedback loop filters which I somehow missed at first. These reduce the gain from approx 46 db to approx 36 db. This is all visible in the images I uploaded below. Note that the nature pot is turned all the way anti-clockwise, so the bassiest response. Honey bee 1 only has the filters activated which are directly coupled to the opamp out. Honey bee 2 has one additional feedback filter which comes after a 1 k resistor. Honey bee 3 has yet another feedback filter that comes after a 10 k resistor. At this last position there is also a passive 22 n lowpass filter.

For reference: the blue line is the freq response directly at the opamp out, the green line is after the 1 k resistor and the red line after the 10 k resistor.

So the gain and amount of clipping now make sense to me. However, the additional filters do not. How should I calculate the cutoff frequency for these filters? I can conceive two ways:

1 is like a passive filter. 1k with 22n gives a cutoff at 7.2 kHz and 10k with 4n7 gives 3.3 kHz. However, for a cap in the feedback loop one would normally take the parallel resistor which is approx 250 k when gain is halfway up. This would give a cutoff of 29 hz for the 22n and 135 Hz for the 4n7. Then there is the fact that the filter action is reversed due to being in a negative feedback loop, which results in me being a bit lost about what is going on exactly 

My best guess at the moment is that the frequencies below 7.2 k and 3.3 K are attenuated. At low frequency this is compensated by the increased gain due to the 2u2 cap from the feedback loop (note that I turned the "nature" pot all the way to the bassiest response) to ground resulting in a somewhat humped frequency response and overall attenuation compared to a schematic without the additional filters. BTW, the end result is a mild marshallesque tonestack.

[Elektrojänis]

Difference between 2 and 3 volts is small for clipping?

It is... It's only about 3.5 dB which is not insignifignat, but definitely not huge. If you compare it to gain control on most OD/distortion/fuzz stompboxes it would be like turn the gain knob just a bit.