Original effects with schematics, layouts and instructions, freely contributed by members or found in publications. Cannot be used for commercial purposes without the consent of the owners of the copyright.
Alright, I got it finally figured out... an SHO-style MOSFET boost with NO CRACKLE!!![/b]
I came up with a new gain control for it that allows for the same range of gain adjustment without having to adjust the Source. Since the Source stays fixed, there is no crackling.
While thinking about how to solve this problem, I realized I could treat the basic gain stage as an inverting opamp. So, I set a fixed max gain and tried a variable resistance in series with the input and voila, works perfectly.
Check it out and give it a try; it's a great improvement over the SHO and it's easy to tweak the sound to taste.
Here is the schematic:
2009 is looking good so far!
vero layout:
"Analog electronics in music is dead. Analog effects pedal design is a dead art." - Fran
way to go, now...now if zvex comes out with an improved SHO, we know where he got it from. im sure he won't be able to resist trying incorporate your improvements.
Not trying to look down upon this excelent design by soulsonic (to whom I give my kudos), I don't think (and I'm sure you're all aware of this) that Zachary ever thought that the crackle was a problem in the SHO design. Contrarily, it was a clever marketing idea. I'm sure that the "crackle ok" thing broght the SHO to attention. If it wasn't for that faulty feature, probably the SHO wouldn't have been that popular an effect.
So I don't think Zachary will ever try to replace the current design of the SHO with a crackle-less SHO.
Just a thought, nothing to see here
first the Klon-buster, now the Krackle-killer? what's next?
"You've converted me to Cubic thinking. Where do I sign up for the newsletter? I need to learn more about how I can break free from ONEism Death Math." - Soulsonic
I doubt you'll see a Zvex gain pedal sans-crackle anytime soon. People have come to expect it and might even think it's broken if that particular problem was fixed.
While this looks like a fine circuit, you can't really say it does what a SHO does, for the simple reason that the bias is not adjustable. I never noticed it when I had the SHO on the breadboard, but there are supposed to be subtle changes in tone (distortion) with different bias points. But love the simplicity (same total number of parts). A much more elegant solution than the one I was thinking of doing for a pedal of the same name (Crackle-Not-Okay with gain/bias adjustable with MOS current mirror).
Anyway, thanks Soulsonic!
rocklander wrote:hairsplitting and semantics aren't exactly the same thing though.. we may need two contests for that.
I used to think that bias changed as the gain adjusted in an SHO; then I measured it on a breadboard and realized I was mistaken. I believe the Gate Bias divider keeps the bias constant no matter how the Source resistance is set; that's the really clever thing about the original design. As the Source resistance is changed, the bias "wiggles" for a moment and causes the transistor to turn on and off, but after a second, it stabilizes and returns to proper bias... that bias "wiggling" is what causes the crackle. The distortion that happens at high gain is simply because the gain of the circuit is so high that it clips the signal and distorts. The character of the distortion doesn't have that "smashed" quality that you'd hear if the bias was hot.
I swear, I hear almost no difference between my new gain control and the original other than the lack of crackle. In any case, even if the bias were different at say, halfway up the gain control, it's still identical to the bias of the original SHO at full gain, and that's were the magic sound is (to my ears).
The version in the schematic has a lower input impedance than the original SHO, so it won't have exactly the same response (it's actually smoother with these values), but you could precede it with an ultra-high impedance buffer and get that same kind of unreal clarity that's typical of the SHO.
"Analog electronics in music is dead. Analog effects pedal design is a dead art." - Fran
It's a standard cheap bipolar cap; I like Nichicon ones. They're nothing fancy, and I prefer them in most cases to standard polarized electros for signal coupling. You can substitute a standard polarized electro or even a tantalum; just be sure and connect the positive leg to the side where the 5k1 resistor connects to the Drain of the transistor.
stephenxvp wrote:Very elegant circuit; thank you.
Can you explain why the 10u bipolar cap and is there a work around ?
"Analog electronics in music is dead. Analog effects pedal design is a dead art." - Fran
soulsonic wrote:It's a standard cheap bipolar cap; I like Nichicon ones. They're nothing fancy, and I prefer them in most cases to standard polarized electros for signal coupling. You can substitute a standard polarized electro or even a tantalum; just be sure and connect the positive leg to the side where the 5k1 resistor connects to the Drain of the transistor.
stephenxvp wrote:Very elegant circuit; thank you.
Can you explain why the 10u bipolar cap and is there a work around ?
Wouldn't 1µF suffice?
I admit, it's hearsay....
analogguru wrote:Look at the Zvex SHO: it is the same there..... and everybody is copying it without doubting the sense...
10µF would be necessary with a following (input) impedance of 2 kOhm only and 22µF if the following impedance is 1 kOhm.
So take a 1µF polyester-film-capacitor instead of it and be happy....
soulsonic wrote:I used to think that bias changed as the gain adjusted in an SHO; then I measured it on a breadboard and realized I was mistaken. I believe the Gate Bias divider keeps the bias constant no matter how the Source resistance is set; that's the really clever thing about the original design. As the Source resistance is changed, the bias "wiggles" for a moment and causes the transistor to turn on and off, but after a second, it stabilizes and returns to proper bias... that bias "wiggling" is what causes the crackle. The distortion that happens at high gain is simply because the gain of the circuit is so high that it clips the signal and distorts. The character of the distortion doesn't have that "smashed" quality that you'd hear if the bias was hot.
I swear, I hear almost no difference between my new gain control and the original other than the lack of crackle. In any case, even if the bias were different at say, halfway up the gain control, it's still identical to the bias of the original SHO at full gain, and that's were the magic sound is (to my ears).
The version in the schematic has a lower input impedance than the original SHO, so it won't have exactly the same response (it's actually smoother with these values), but you could precede it with an ultra-high impedance buffer and get that same kind of unreal clarity that's typical of the SHO.
Ah well that's what I get for just taking Zach's word for it... Thanks for actually doing the measurement, the DIY world is the better for it.
One thing I might add on would be a small-ish cap in parallel with the 1M gain pot, so it stays bright at minimum gain. Well, that and to do something about the Zener. I never liked the fact that the Zener diode will clip on -0.7V and would prefer two 5V zeners connected cathode-cathode.
rocklander wrote:hairsplitting and semantics aren't exactly the same thing though.. we may need two contests for that.
The SHO is a deceivingly simple circuit. There is a lot going on when the source/bias pot is adjusted.
First, as the bias pot is adjusted from max (5k) down to min (0 ohms), the DC drain voltage changes roughly from 7.2V down to 4.0V. I have measured this in the breadboard and verified it in the sim (using a BS170 Spice model by Zetex).
Second, the gain varies from 0 dB to almost 40 dB as the bias pot is adjusted. Notice how the variation of the DC drain voltage tends to keep the dynamic range nearly centered. At min gain, drain and source resistors are 5k and 5k, respectively, thus the output can vary from about 4.5V up to 9V, and the 7.2V bias voltage is nearly centered. At max gain the output can vary from about 0V up to 9V, so the 4.0V bias voltage is again close to the theoretical center of the dynamic range.
Third, the input impedance of the SHO is not 5 Mohm as the popular belief dictates. No, it varies big time with the bias pot. At min gain, the end of the upper 10Meg resistor is tied to an inverted copy of the input voltage, not ground, so overall input impedance actually becomes 3.3Mohm when you do the proper math to find out the equivalent thevenin resistor (this has been verified in the sim as well). As the bias pot is reduced and gain increases, something very convenient happens: the input impedance starts going down, down to 140 kohms at max gain. This certainly helps taming the high frequency resonance of any guitar pickup (around 2-5kHz), thus taming highs and helping to the musicality of the clipping. Again, this is because the upper 10Meg resistor is effectively tied to a voltage source that is an amplified version of the input voltage, and not a simple ground point.
estragon wrote:Again, this is because the upper 10Meg resistor is effectively tied to a voltage source that is an amplified version of the input voltage, and not a simple ground point.
I recently did a layout from Analogguru's schematic of the SHO, and in the notes it says the 10M resistors can be reduced to 1M. (Also, he ditched the zener and used two 1N4148s, but that's another discussion, I suppose.) I built mine with the 1M resistors, and I'll be damned if I can hear a difference either way. Is there some special significance to the 10M resistors? Why would those have been chosen over a lower value, such as 1M?
I gotta measure this stuff again... I might be completely mistaken.
But my EARS are not mistaken, my circuit gets the sound. Even though the dynamic headroom may not change as estragon pointed out, the input impedance still changes as the gain is adjusted - that is probably why it sounds so close to the same to me.
"Analog electronics in music is dead. Analog effects pedal design is a dead art." - Fran
estragon wrote:Again, this is because the upper 10Meg resistor is effectively tied to a voltage source that is an amplified version of the input voltage, and not a simple ground point.
I recently did a layout from Analogguru's schematic of the SHO, and in the notes it says the 10M resistors can be reduced to 1M. (Also, he ditched the zener and used two 1N4148s, but that's another discussion, I suppose.) I built mine with the 1M resistors, and I'll be damned if I can hear a difference either way. Is there some special significance to the 10M resistors? Why would those have been chosen over a lower value, such as 1M?
Thanks.
Consider a humbucker guitar with 500k volume and 500k tone pot. With both pots dimed, the resistive loading for the pickup is 250k (notice that the tone control has a very large cap in series, thus it is effectively loading the pickup as well as the volume pot). So, as the load is around 250k, any extra load higher than 500k will have very little impact. On the other hand, when the additional loading approaches 250k or goes below this value, then you should start noticing how the resonant peak of the pickup is slightly reduced--a reduction in brilliance, rather than in volume. This should be more noticeable when playing full chords. The effect is subtle to some people (including me), but others will claim it is day and night regarding guitar tone.
You can do the following simple experiment and see (hear) for yourself: add a 100k resistor in parallel with your guitar connected to the clean channel of the amp (no pedal in the line) and listen how it affects the tone. Then you can repeat this with the overdrive channel of the amp. Some guitar pedals take advantage of overloading the guitar with a low impedance in order to reduce excess brightness, such as LPB booster, BMP, and others.
soulsonic wrote:I gotta measure this stuff again... I might be completely mistaken.
But my EARS are not mistaken, my circuit gets the sound. Even though the dynamic headroom may not change as estragon pointed out, the input impedance still changes as the gain is adjusted - that is probably why it sounds so close to the same to me.
The drain should be biased close to 4V in this condition.
I agree that even though the two circuits may have a difference in the dynamic range at different gain settings, overall sound could be very similar, most likely because a 9V peak to peak output will clip any amp's input stage, so if dynamic range is a little more or less it won't really make such an audible difference.
This is a clever solution, indeed. You don't sacrifice maximum gain and retain the varying impedance effect, which by the way I believe it was a lucky side-effect, rather than a design objective.
![[smilie=a_goodjobson.gif]](./images/smilies/a_goodjobson.gif "a_goodjobson"){kind=small}
Thanks for actually doing the measurement, the DIY world is the better for it.