Improvements to the Slacktave/Truth Table

[egasimus]

Hey guys! I've successfully breadboarded the Truth Table, save for the octave up part (yet), and it's been working out really great so far; so good, in fact, that I'm gonna use it as one of the main modules in my bass synth project. I'm gonna have some fun with a 4051 sequencer at the output...

I have a couple of questions:

1. Does opamp choice influence tracking much? I've built it with two 072s, but maybe there's another chip that'd be an improvement?
2. Any other mods I could do to fine-tune the circuit for bass?

[~arph]

No the opamps don't do much, perhaps with a rail to rail opamp it sustains a little bit longer, but not much and nothing you can't achieve by fine tuning the gain of the first stage either. For bass it might be good to adjust the sallen key low pass filter around the second opamp to filter even more highs. (try increasing the 10n to say 47n)

[egasimus]

Great - I'll have a look at that filter!
I don't get the purpose of the last opamp stage though... Are the clipping diodes used to provide a safer output level? I mean, it'd otherwise be 9Vpp, right...
So can I safely omit the last stage and just have four separate outputs, or would they be loaded down if I don't buffer them?

Also, reading up on Sallen-Key filters and this comes up:

If you're using LM324s because they're cheap and because you have 1/4 LM324 just sitting around, consider that the LM324, because of its class-B output stage, is one of the worst op amps for a Sallen-Key filter. That's because its output impedance is potentially soft. (source)

Good I didn't use a 324, I guess? I'm still pretty much working blindly, there's a lot I don't understand.

[slacker]

Nice to see someone playing with this again. Rick has just taken the last opamp stage from the original Slacktave, in that it was used as part of the blend control between the straight fuzz and the octave down, here it's being used as a mixer for the 4 outputs. Yeah the diodes are there to limit the output to a sensible level, they might also add a bit of extra fuzz.
You should be fine omitting the last stage and  having separate outputs.

[~arph]

Also, reading up on Sallen-Key filters and this comes up:

If you're using LM324s because they're cheap and because you have 1/4 LM324 just sitting around, consider that the LM324, because of its class-B output stage, is one of the worst op amps for a Sallen-Key filter. That's because its output impedance is potentially soft. (source)

Good I didn't use a 324, I guess? I'm still pretty much working blindly, there's a lot I don't understand.

Good to know. Though I assume that with soft, they mean low. I've heard of high and low impedances.. Not soft and hard. 

[kingswayguitar]

could someone be kind nough to explain what the 1st op-amp is doing with both sides referenced to 4.5V?  wouldn't that chop your signal?

[~arph]

Yes, that's a standard non inverting gain stage with a gain of (15k / 1.2k) + 1 = 13.5

http://www.radio-electronics.com/info/circuits/opamp_non_inverting/op_amp_non-inverting.php

The opamp that feeds the 4024 clk pin is doing the chopping here. That is comparing the guitar signal with the Vref.  If the input is above Vref it outputs the supply voltage, if it's below it output 0V so that is converting your signal into a squarewave.

[egasimus]

Nice to see someone playing with this again. Rick has just taken the last opamp stage from the original Slacktave, in that it was used as part of the blend control between the straight fuzz and the octave down, here it's being used as a mixer for the 4 outputs. Yeah the diodes are there to limit the output to a sensible level, they might also add a bit of extra fuzz.
You should be fine omitting the last stage and  having separate outputs.

Thanks Ian - you've given us a great design!

On my first try, I actually attempted to put together my proposed CMOS version from a year ago, but that didn't work for some reason so I decided to play it safe and did the original circuit word for word, step by step. I might have a try at running it from a dual supply later on, though. But most importantly I'm really looking forward to finalizing this on vero (my preferred medium after my laser printer died) and moving on to the sequencer; that's gonna be the bomb y'all!

[kingswayguitar]

Yes, that's a standard non inverting gain stage with a gain of (15k / 1.2k) + 1 = 13.5

http://www.radio-electronics.com/info/circuits/opamp_non_inverting/op_amp_non-inverting.php

The opamp that feeds the 4024 clk pin is doing the chopping here. That is comparing the guitar signal with the Vref.  If the input is above Vref it outputs the supply voltage, if it's below it output 0V so that is converting your signal into a squarewave.

thanks
so it's a non-inverting amp
i'm just more accustomed to this (Figure 8 - only 1 reference voltage)

http://www.ti.com/lit/an/sloa030a/sloa030a.pdf

[~arph]

It's the same actually, only difference is that Rf is tied to GND. (and Vref is 0 ??)

That schematic will clip the negative halve of the input signal (as stated in the text Vin <= 0 Vout = 0) .

When the + input of the opamp is biased around Vref (1/2 Vcc) (connect Rf to Vref instead of GND), the output will swing between 0 and Vcc, centered around Vref.

[egasimus]

Just built the opamp part on vero, and it seems to work pretty nicely. Laying it out was a bitch (even though using Eagle helps me a whole lot!) so I decided to put two of my TL074's gates in parallel - the comparator stage, to be exact. The notes now end quite quickly and abruptly, but I don't know if it's because of this or because of me not having my bass and playing bass samples from my laptop's soundcard instead. As soon as I get some more IC sockets, I'll continue my exploration of this circuit's marvelous properties

[slacker]

This any use  

[egasimus]

I know that one, man, but I wanted to fit the octave up on the board, too (I briefly entertained building that on a daughter board, but there's really no point in that unless I use SMD parts), besides my vero was 19 holes wide, and cutting it for this layout would leave me with a 5-hole x 19-track piece of vero... not too useful, alright maybe a fuzz would fit though. It's 19x18 now, not too bad; I wanted to cut tracks at holes, not between holes, otherwise it'd be even smaller but I probably wouldn't have gotten it to work at all due to some invisible solder bridge somewhere. I use a 10mm drill bit which I turn by hand, could use a better tool (abrasive bit in mini drill) but this one does the job fine enough.

[egasimus]

Hey all

Finally got around to testing the board with a real bass, and I get the same result - the sustain is annoyingly short, and it's not because of the doubled comparator stage. Would increasing the gain of the first stage help, or should I look elsewhere?

[kingswayguitar]

also, could i get some reference voltages from some kind soul...just the first 3 opamps would be great

thanks!

[slacker]

The inputs and outputs of the first two opamp stages should be whatever your vref is, 4.5 volts if using a 9 volt supply. The inputs of the comparator stage should be vref and the output will probably be near the supply voltage or near ground.
How short is the sustain?

[egasimus]

I can't really record a demo right now, but it goes like this - light playing usually goes undetected aside from the usual pop caused by the string attack; if I mash the strings hard, I get sound, but  it's a bit unreliable, with notes tending to cut off earlier than they should. The breadboard didn't exhibit that issue; it hummed like crazy due to some longish unshielded wires that I had used, but otherwise was really responsive.

[egasimus]

I was right - I soldered a 33k pot in place of the 15k resistor, and things improved significantly. I still don't know if it's due to some error on my part somewhere later in the circuit, but it works great. Gonna put a 100k pot there as a sensitivity control and leave it at that; I've still got a divider and a doubler to solder in.

[slacker]

It shouldn't behave like that, something is wrong somewhere. If you can post a picture of your layout, I'll have a look and see if I can see a problem.
Or if you have an audio probe you can use that to try and find where the problem is. The output of the first stage should be a louder version of the the input, the second stage should be similar but with less treble and the third stage should be a very loud square wave with plenty of sustain.

EDIT: posted the same time as your last post

[egasimus]

Soldered everything together, and the 4024 and 4030 stages sadly don't work. I couldn't get them right on the breadboard, either... I'll swap in some new chips when I can (next week maybe) but just wanted to share how frustrated I am; also the fact that I'm using cheap vero and I'm not sure how reliable it is; just wanted to say, fffffffuuuuuuuuuu............