Snow White speed mod

[Kipper4]

Has anyone ever heard of a speed mod for the Snow White auto wah?

Here's the schemo if it helps

http://revolutiondeux.blogspot.co.uk/2012/07/mad-professor-snow-white-auto-wah.html

Thanks
Rich

[Keppy]

Increase R13 (or add a pot inline with it) to increase the attack time. There's already a variable decay time. Is that what you mean?

More info: http://www.geofex.com/article_folders/ecftech/ecftech.htm

[Mark Hammer]

R13 will indeed determine attack time, and reducing it will make a faster attack.

I'll also note that the circuit shares much with the DOD FX25, including the ability to provide both low pass and bandpass outputs.

[Keppy]

R13 will indeed determine attack time, and reducing it will make a faster attack.

From the man who wrote the article! The reason I didn't suggest reducing it is that the load on the opamp looks pretty low to me already. I read once that opamps don't like driving shunt capacitors straight to ground, and <330R seems pretty close to that to me. Is this not a concern?

[Mark Hammer]

Well I don't think I'd reduce it a LOT, although fast attacks on up sweeps can sound like a reverse sweep.  But I imagine values as low as 100r should be fine.

[Kipper4]

I was thinking maybe adjusting the sweep speed more?
if i have it right, re; faster or slower bwow.

[Keppy]

I was thinking maybe adjusting the sweep speed more?
if i have it right, re; faster or slower bwow.

That sounds like what happens when you turn the Decay knob down. The Decay pot slows the rate at which C8 discharges, which in turn slows the change of the filter once the input signal drops. Have you actually tried this effect out yet? Do you need something other than the controls that are in the schem? It might already have the control you're looking for.

[Mark Hammer]

Time constants (attack AND decay) on simple half-wave rectifiers like that used on the Snow White and so many others, are a tricky beast.  Half wave rectification can be more susceptible to "envelope ripple".  That is, the removal of half the signal results in something that is sort...of....DC, but not smooth-like-battery DC.  Fullwave rectification doubles the rate at which those little ripples/variations from pure DC occur, and can often make the ripple more tolerable.

But it is important to keep in mind that ripple in rectifying AC is a function of the AC signal to be rectified.  And in the case of guitars, it is generally the case that the sorts of momentary changes in input signal that would result in ripple occur during the "decay" phase of the note/chord.  That is, after the initial post-pluck peak, when the strings have relaxed a bit and are starting to die out (albeit slowly).  That ripple is often reported by listeners as a kind of "distortion", a sort of trill (not thrill ).

IF the filter "gets in and out" - i.e., completes its sweep cycle - before you get to the wonky parts, then you don't hear the ripple.  If the filter is completing its sweep as the smoothness starts to disappear and is replaced by instability, the ripple becomes more obvious and annoying.

The capacitor that provides the "averaging" (essentially gathering all those little micro-peaks and storing them in a basket, a little like a tachometer averages out each motor turn over time) sets the rate at which transitions/variations in the DC are summed.  If the cap value is larger it makes for slower transitions, much like an elephant doesn't turn around quickly.  But, like an elephant, it has a hard time responding quickly.  The optimum is for a "horse-sized" cap value, that allows for fast-enough attack, but can be persuaded into a slow enough decay.  You don't want the decay to be too long, since that ends up making the ripple more audible.  But if the decay time is really long, that can smooth out the ripple nicely.  You do that by making the bleed resistance higher, but also by using a bigger averaging cap.  The tradeoff is clearly inability to get fast attack with long decay.

The bottom line is that simple half-wave rectifiers require one to select component values to suit the style you think you're going to need.  None is better or worse, just better suited to the properties of the input signal and the intended sweep you're trying to get.

[Kipper4]

I built it from the vero above with the 2x tlo72s
Its nearly all clean signal. I'm just not getting the qauck. I mean i can hear it underneath but its way too subtle.
I have another etched board on the way but meantime i'd like to sort it out.
subbed the 2SK170 with a 2n5458 (twisted the legs to match board pinout)

[Keppy]

I built it from the vero above with the 2x tlo72s
Its nearly all clean signal. I'm just not getting the qauck. I mean i can hear it underneath but its way too subtle.
I have another etched board on the way but meantime i'd like to sort it out.
subbed the 2SK170 with a 2n5458 (twisted the legs to match board pinout)

The 5458 should be fine there. We'll need more info to help you. Check the pinouts of the other two transistors and the regulator. If they're correct, check your voltages.

[Kipper4]

Ok will do thanks for the offer as soon as I've finished work I'll post 'em up thanks Keppy.

[Kipper4]

somethings not right here
bc550c
Q1
8.71
8.33
8.27

bc550c
Q2
8.71
8.73
8.33

regulator
Q3
8.33
8.73
8.71

Q4
2n5458
d8.83
g4.32
s8.82

IC1
lm13700
1   1.29
2  0.14
3  4.08
4  4.08
5  5.27
6   0
7  5.27
8  4.08
9  4.08
10  8.23
11  8.83
12  8.24
13  8.72
14   8.71
15   8.27
16   8.32

Ic2
tl072
1  4.06
2  4.07
3  3.99
4  0
5  4.07
6  4.05
7  8.43
8  8.43

Ic3
tl072
1  3.75
2  4.07
3  4.08
4  0
5  4.08
6  4.08
7   4.08
8  8.44


referance voltage source 9.19v


looks like the regulator is a gonner

[Kipper4]

I replaced the regulator it now works and the voltages are right.
I tried some bc109 with almost identical hfe (which i've left in for now)
but im still not getting any bwow wow
frustrated villager.
Rich
updated voltages

13700

1.25
0
4.06
4.06
5.25
0
5.25
4.06
4.07
5.26
8.5
5.26
4.06
4.06
0
1.125


ic2
4.06
4.06
3.96
0
4.06
4.06
3.6
8.5

ic3

3.8
4.06
4.06
0
4.07
4.07
4.07
8.5


q1

5.07
3
2.47


q2

5.07
3.6
3


jfet

8.5
4.3
3.9

regulator is now as expected

[Kipper4]

I found a mistake on c2 but it's too late to test it geez
I'll follow up tomorrow
Rich

[Keppy]

According to your voltages D5 is backwards.

[roseblood11]

Is it possible to reverse the sweep in the SWAW?

[nocentelli]

Apparently. It's hinted at within the thread over at eff ess bee. Also, HP/BP/LP switching is supposedly possible but I could not work out how. Perhaps comparing the schematic to another state variable filter schem might yield the answer.

[Mark Hammer]

I wouldn't consider implementing HP mode unless you can also implement downward sweep as well.  LP and BP work fine in either direction, but HP sweeping upwards is like an automatic make-my-guitar-sound-wimpy effect.

[Kipper4]

its sorted now
im finally getting some bwow
i had subbed d3 with a 1n34 , i read it in a forum somewhere dont ask!!!!!!!!!
once i put the 2u2 in the correct holes and undid the D3 sub and matched (as close as possible the hfe on the bjt's Higher the gain the better from what i've seen too )
I've finally got some qauck.
I still want more but I'm thinking of building a mutron 3 but thats another story. I might need some help with that Mark and the rest of the guys.
Thanks for all the help.
Btw the parting question is
Can i use differant bjts to closer match for the HFE or do they both have to be the same transistor?


eg
a bc550c with a 2n3904?

[roseblood11]

unless you can also implement downward sweep as well.

Do you think that's possible?