I see that in the Alpha Dog there are two clipping sections, "Hard" and "Soft". I understand that the normal Rat clipping is the Soft section. What is the "Hard" clipping section... the output stage? It looks like the options are the BS170 (Mosfet) transistor, no clipping, and the 1n34a(Germanium) diode.
I use to own a Deucetone Rat and it had lots of clipping options but I remember having the option of Germanium diodes in what would be the "Soft" section in the Alpha Dog. I guess I don't understand why the 1n34a is where it is?
The Deucetone clipping options were:
Clean Rat = no clipping
Vintage Rat = silicone diodes
Turbo Rat = LED
Dirty Rat = Germanium diodes
Also, I see the four 1n4148 diodes on one side of the "Soft" clipping section and the center position is no clipping. Is the other side the 2n5458? Isn't that a silicon transistor? Why use two silicon options instead of LEDs for one side?
Thanks folks!
You've actually got the hard and soft backwards. Hard clipping is the default Rat style which are the two 1n4148 that go straight to ground. It's a bit more complicated than the traditional setup since it has two more 1n4148 in parallel with the soft pot to blend additional clipping diodes in series. Soft clipping is in the feedback path of the op-amp. The Soft pot mixes between the two back to back clipping diodes in the op-amp which are the 1n34A and the "body diode" of the BS170. At center it has equal resistance on both sides meaning the least amount of soft clipping from either diode.
The 2n5458 has nothing to do with clipping. It's an output buffer.
Wow, Thanks. The Soft knob makes total sense now
Sorry, I guess I remembered the labels on the wrong sides.
So... Is the Hard pot blending between asymmetrical and symmetrical silicone clipping?
Also, you said: "it has two more 1n4148 in parallel with the soft pot to blend additional clipping diodes in series".
Did you mean?... in parallel with the Hard pot
Thanks Again!
The 'hard' section is always symmetrical. In the middle of the pot rotation, there will be basically no hard clipping at all. To one side, you blend in standard RAT clipping (two silicons). To the other side, you blend in four silicons plus a cap (which will roll off some highs). The configuration of the four silicons is non-standard as they are connected in the middle (like the Timmy does, but it has it as soft clipping). Rather clever tricks that VFE used to get many clipping options with a single pot.
Quote from: Aentons on August 28, 2017, 07:54:20 PM
Did you mean?... in parallel with the Hard pot
Oops yes I did.
I finally finished my Alpha Dog and it sounds pretty good... but boy does the Soft knob unleash a big bucket of farts when counter-clockwise. I used the 1n34a "workalike" from Small Bear and put it in a socket. Does anybody have any suggestions for alternatives or mods for that particular clipping option?
I'm going to order another round of parts for the Old School, White Horse, and Dark Horse once the BOM is released for the Dark Horse.
Btw, my vote for the next release is the Bumblebee
Pics
Is the BBoFs at the extreme of the pot? I seem to recall that at the very end of rotation it got kinda mushy.
It's not just at full rotation, it's the whole left side if the dial. It gets a little less noticable the closer you get to the middle due to the clean being mixed in.
I actually like gatey, voltage starved sort of sounds, but I don't care for this particular flavor.
I was already wondering what such extreme asymmetric clipping would sound like in this circuit. It might sound better when dialling in the hard clipping after it? Extreme asymmetric clipping does not sound very pleasing, to my ears at least. You can always replace the single diode with a pair (one in each direction, same or different type), and play around with different diodes until you get a sound you like better (very much like people do in other soft clipping circuits like the TS).
I built mine with on/off/on switches like the first versions, instead of pots. Maybe I'm missing out on some interesting "in between" sounds, but I wanted easier combinations. Would noon be the equivalent of diode bypass, or would the diodes with highest forward voltage take over? I love the versatility, from hard rock to odd sputtery type fuzz.
Noon would effectively take the diodes out of the circuit, they are never really really entirely out.
I've been playing around with a replacement for the 1n34a side of the soft pot and have tried various sym and asym combinations using ge, si, led, zener. I can't really find anything that sounds better or different. It seems any sym combination I try sounds essentially the same, and any asym combination is still essentially a BBoF. I'm curious what the parallel cap on the BS170 is doing to smooth the asym out? So, I guess my question is... why is there a cap on the BS170 side and not the 1n34a?
Also, I broke my sockets off for the 1n34a so have to go back in and do some desoldering so while I'm in there I'm going to replace a couple of the MLCC pf size caps with silver mica caps. I only have 68pf and 100pf values. Are these too high of values for replacements to the 10pf and 33pf caps?
This is somewhat based on the rat isn't it? The 33 pf is probably the the slew rate compensation for the 308. No idea how much changing the value will change the sound.
What exactly do you hope to gain by putting mica caps in there? You could probably replace the mlcc with mica or ceramic an not notice a difference.
Quote from: bsoncini on March 08, 2018, 03:01:27 PM
This is somewhat based on the rat isn't it? The 33 pf is probably the the slew rate compensation for the 308. No idea how much changing the value will change the sound.
What exactly do you hope to gain by putting mica caps in there? You could probably replace the mlcc with mica or ceramic an not notice a difference.
Yeah the 33pF is compensation for the IC. Don't waste a silver mica cap on it, its not in the audio path.
Thanks, I wasn't sure about the 33pf cap. My main target for the mica swap is the 10pf low pass cap in the feedback loop which is supposed to affect the clipped waveform corners.
One of my first and still favorite pedals is a 1998 Maxon made Ibanez TS9DX that was modded (not by me) with a silver mica cap for the 51pf cap in the feedback loop. I have bought several other TubeScreamers that didn't have that particular mod and none can hold a candle to the smooth sweetness that comes out of the one with it. I have a BYOC Overdrive 2 that at first had a ceramic and then was modded to a mica and it now has a very similar smoothness as well. Btw, I believe the AM "Silver" mod is essentially the TS9DX in + mode(4 diodes), the 808 mod, and the 51pf cap changed to a silver mica.
I am hoping to achieve a similar result with the Alpha Dog
I finished my Alpha Dog build and I am very happy with the range of adjustments in the clipping options. I get noticeable volume drop dialing away from the center position on the hard clipping know, but no BBOF anywhere in the settings that I have played with. I am using the OP07 chip in my build, so perhaps that makes a difference.
Good idea. I have a couple of OP07 chips that I pulled from the aforementioned Duecetone that I didn't even think to try. I also have a metal can LM308 from Small Bear on the way. The chip I'm using now is a National Semiconductor version so maybe it is different enough to make a difference.
My burning question is still... What does that cap across the BS170 do?
Quote from: Aentons on March 09, 2018, 07:40:05 AM
My burning question is still... What does that cap across the BS170 do?
I imagine its rather like the cap across the LEDs in a Tube Screamer, in that it "softens" the clipping edge. To quote master R.G. Keen:
"When a diode (any kind) clips, it goes from effectively non-conducting to conducting sufficiently that the voltage across it doesn't change much any more.
Before the diode turns on, it has essentially no effect on the signal impressed across it. Once it turns on, it acts like a very low resistance.
A capacitor across a diode acts like... a capacitor. Capacitors eat current to keep their voltages from changing quickly. So with the diode not conducting, the capacitor is the only operational device. It slows the rise and fall of the signal in concert with the signal source impedance.
Once the diode starts conducting, it limits the voltage change of the signal as well, so the capacitor has less work to do, and participates less. When the signal declines below the diode conduction voltage, the capacitor sources current to keep the voltage from changing as fast.
The net effect is that the capacitor keeps the "corners" of the diode's clipping from being as sharp. This reduces the higher order harmonic production of the clipper. "