Naughty Fish Help

[mgwhit]

Well, Mycyk, if no one else chimes in here anytime soon, don't despair.  I ordered the components to build mine today.  Between this thread and the fact that my band is rehearsing "Sugar Magnolia" I decided it was time to get on it.  In a week I should either have a nice reference model to compare yours to or my own Tech Help thread going.

[RobA]

I'll try to pull mine back out and take some measurements with different settings tomorrow. Although, I can't guarantee the accuracy -- I've got a nasty head cold right now and it's leaving me kinda goofy.

A couple of notes about my build that might be relevant; It seems to be pretty sensitive to the Gain pot setting, Mine is really distorted in some setting when I first turn it on.

One the first point, in some toggle settings it just doesn't respond much at all until the gain setting is just right. When the Gain is dialed to the right point, it pops into being very touch sensitive and a ton of fun to play with.

The weird thing about the second point is that it goes away after about a minute or so and become completely clean and it doesn't seem to do it with my Marshall amp. The Marshall thing could be because of the difference in the tone settings and responses of that particular amp, or maybe it's the input impedance or ... I don't know. At this point, I don't have a grasp on why it's noisy at startup or why the distortion stops after a bit of warm up. I really need to do some more testing on why this happens. But, it doesn't bother me in practice because it goes away pretty quickly and the thing sounds great.

Edit: Don't know why I didn't think of this before, but it just hit me that the most obvious guess for the initial distortion is the high dark resistance of the Vactrols at startup. I'll investigate that.

[mgwhit]

Thanks, Rob!

[Mike]

I would be happy to take some voltages if they are still needed. What settings do you want?

In the mean time, I have recorded a (crappy) demo of mine on bass. It's pretty bad in terms of sound quality, playing (I play guitar), and as a bass effect generally. But maybe the video will help.

[mgwhit]

Mike, if you wanted to post voltages with the Up - Down switch in the "Up" position, that would be great.  That's the setting we're trying to debug.  RobA was kind enough to post his voltages, but his switch was in Down mode and that totally changes the readings on IC2_B.

[RobA]

Sorry about not getting the voltages in yesterday. The cold got the better of me. Warning: these voltages should be thought of as tentative, I've been drinking tea with a bit of whisky, taking decongestants and antihistamines, and haven't slept in more than 1 and 1/2 hour stretches for two days. So my thought processes may be slightly altered .

With that warning in place,

Knobs at noon,  LP, Up, Hi

Vin = 9.23V

IC1:   
   1   0V
   2   0V
   3   0V
   4   -8.21V
   5   0V
   6   0V
   7   0V
   8   8.65V
IC2:    
   1   ~330mV drifting
   2   22mV
   3   22mV
   4   -8.15V
   5   ~40mV drifting slightly
   6   41.1mV
   7   -450mV
   8   8.64V

IC3:
   1   0V
   2   0V
   3   0V
   4   -8.20V
   5   0V
   6   0V
   7   0V
   8   8.64

IC4:
   1   1.37V   
   2   4.55V
   3   0V
   4   -3.83V
   5   -8.19V
   6   2.55V
   7   1.41V
   8   8.59V

Vactrol 1:
   1   -620mV drifting
   2   36mV
   Resistor side both pins at 0V

Vactrol 2:
   1   -520mV drifting
   2   36mV
   Resistor side   both pins at 0V

Leaving everything else the same but moving Gain to 9:00

I only put in a voltage if it changed significantly from the noon voltages.
IC2:    
   1   ~300mV drifting
   2   22mV
   3   22mV
   4   -8.15V
   5   ~20mV drifting slightly
   6   ~25mV drifting slightly
   7   ~ -1.18V drifting slightly 
   8   8.64V

Vactrol 1:
   1   -1.16V drifting
   2   22mV
   Resistor side both pins at 0V

Vactrol 2:
   1   -1.19V drifting
   2   22mV
   Resistor side   both pins at 0V

[Mike]

I took voltages with every switch combination.

XLS isn't an allowed filetype, so here you go: Naughty Fish Voltages [XLS]

Mike

[Mycyk]

Thanks for all the input guys. I'm still not getting anywhere new with mine. I replaced the 2 tants (C1 and C9) with bipolar electrolytics, and I don't think it sounds as good, but maybe that's just frustration clouding my hearing. 

Rob I hope your feeling better.
Mike your's sounds great and looks good too!
Let us know how your build goes Matt.

Thanks again all.

[mgwhit]

Thanks everyone for uploading voltages!  I've made a table of what I think are the relevant test points with as similar settings as I could get (RobA LP/Up/Hi Gain=9:00, RobA LP/Up/Hi Gain=12:00, Mike LP/Up/High Gain=12:00, Mycyk Range=Up w/ 9V supply):

Test Point	RobA (Gain 9:00)	RobA (Gain 12:00)	Mike (Gain 12:00)	Mycyk (Gain ?)
IC2_B 4	-8.15	-8.15	-7.58	-7.8
IC2_B 5	0.020~	0.040~	0.16	3.3
IC2_B 6	0.025	0.041	0.17	3.4
IC2_B 7	-1.18	-0.450	1.85	2.5
IC2_B 8	8.64	8.64	8	8.4
VACT_1A (a)	-1.16~	-0.620~	1.85	4.2
VACT_1A (c)	0.022	0.036	0.17	2.4
VACT_2A (a)	-1.19V~	-0.520~	1.85	4.2
VACT_2A (c)	0.022	0.036	0.17	2.4

I expected to see lower voltages on the anodes of the vactrols than on the cathodes: RobA has that, but Mike and Mycyk don't.  I expected to see the same voltage on IC2_B pin 7 as on the vactrol anodes, which Mike has exactly, RobA has approximately and Mycyk doesn't have at all (but could possibly be backwards).

Mycyk, this should give you some things to think about.  The first question I'd have for you is, Are you labeling the anodes and cathodes of your vactrols correctly?  The second question is, How are you getting over 3V on IC2_B pins 5 and 6 with the Range switch in the Up position?  That's probably where your problem is.  If you can remeasure with your knobs at 12:00 and your switches at LP/Up/High that might help, too.

[RobA]

Cold's still here -- but no antihistamines yet today so maybe I can reason a bit better now. The square pad of the Vactrol should be the more positive pad. So my readings confused me. And then I thought to put it on the scope. My DMM sucks at averaging AC voltages and has a pretty slow response time. So, what I got when I put it on the DC coupled setting on the oscilloscope with no guitar signal coming in is an AC signal of about 400kHz that just barely pokes positive but with the majority of the sinusoidal-ish waveform on the negative side. That isn't really going to do too much with the filter as the LED is never really going to turn on. So, I plugged in the guitar and hit a string. The oscillations stop instantly and the now flat voltage pops up to about 9V (gain setting at 12:00) and proceeds to track the fading input signal down to 0V at which point the oscillations come back in.

The gain setting sets how high the voltage will go up and to a degree how fast it fades.

I think getting the readings for IC2 in the Up setting would be more useful with an input signal. It should start high and track down as the signal level decreases. 

[Mycyk]

Thanks Matt for pulling all this together.

Okay, here's measurements with switches in LP/Up/Hi and pots at about 12 noon:
IC2_B 4   -7.76
IC2_B 5   2.65
IC2_B 6   2.65
IC2_B 7   4.48
IC2_B 8   8.36
VACT_1A (a)   4.48
VACT_1A (c)   2.71
VACT_2A (a)   4.49
VACT_2A (c)   2.70

When I pluck a bass string the voltage at IC2_B 7 and both VACT anodes (square pads marked + on the schematic)  goes up to about 6.3 then decays. So the main differences are that I have significant voltages at my IC2 B 5, 6 which the other builds don't and mine is not producing the same voltage change with an input signal.

This still makes me think that there's a problem with the circuit supplying (or controlling) IC2. Will look at this more today or tomorrow.
Thanks again guys.

[RobA]

I think your analysis is right. Check the connections around R17, R16, and C10. Maybe there is a problem with the switch. Check for continuity across the switch between the outside pins. Check the resistance across them too. Check the value of R20 because it works differently in the Up mode from the Down mode.

I'm also wondering what would happen if there were either low or high frequency inaudible noise hitting the envelope generator. If there were noise getting in there, it might cause the envelope to be offset and not get the full sweep. I guess that C10 should be filtering out most of the high frequency content. Maybe it's not working right?

[Mycyk]

Next steps:
I replaced C10 with one I tested to ensure it was working = no change
I removed the Range switch and tested it = it's fine.
I replaced the Range switch with jumpers and got the same behavior I've had all along in both positions.

I then opened the jumpers and inserted a 1M pot between R5 and Pin 5 of IC2B (Pins 2 & 3 of Range A switch on the schematic). By adjusting this pot I can get the unit to give a mediocre Up (autowah) sound. However it does not seem as sensitive as the typical Down position sounds.

I also have significant high hash noise when the Mode switch is in the HP position. This may be the issue you mentioned Rob, but using an audio probe I couldn't figure out where it was coming from. Basically the input pad is good, C2, C3, IC1_A1, R2, R3, all are noisy. Do you think it could be the Gain pot? It sounds fine when I touch each leg so I don't don't understand what's going on with this. I also don't understand why I can't hear audio when touching either leg of C1 but get audio after it......  Am I not using the audio probe correctly? Can one of you that have a working unit try this and let me know what you find?

Any help is appreciated......   Thanks again!

[RobA]

IC1_A is in an inverting op amp amplifier configuration. What this means for what you are seeing at C1 is that the op amp tries to move the output voltage so that the feedback through R3 and C2 will keep the voltage at pin 2 of the op amp zero. How much it needs to amplify the signal will also depend on the attenuation of the input signal by the Gain pot and R1 (and a bit by C1 in really low frequencies). So, the audio at pin 2 will be tiny. The inverting input of an inverting op amp amplifier is considered a "virtual" ground point. So, what you are seeing there makes sense.

The Gain pot will add a lot of resistor thermal noise to the circuit when it is set to a low gain value (high resistor value). If the pot is noisy in addition to this, it could be adding a bunch of noise. One thing you could try is clipping a 33pF cap or so across C2. That'll reduce the gain through that amplifier for higher frequencies.

[mgwhit]

My switches finally arrived today.  I'm taking the day off tomorrow (very important soccer match to watch!), so hopefully I can get mine populated pretty quickly and start looking at this again.