Parasit Studio 8-Bitar clone motorboating problem

[mrmet5]

Hi guys.
I built a 8-Bitar clone on the breadboard, and everything seemed to work well except for this motorboating problem. I researched for a bit, finding possible solutions by increasing the value of R4 and trying all 5 of CD4069UBEs I have but they didn't solve this issue. I also found a perfboard layout on Perf and PCB.com during my research, and since 5 or so layouts I tried using his layouts worked without any issues before so I gave it a shot. The exact same problem occurred. I also tried different electrolytic caps to no avail. In addition, it happens when the guitar volume is all the way down.

It seems like quite a few others have had the same problem, but there are also those who succeeded. I just wonder if there's something in the circuit design or I'm just missing something. I'd appreciate any advice/help you can give, as I'm still learning pedal building/electronics. Thanks!

Video: https://www.reddit.com/r/diypedals/comments/wncfsk/parasit_studio_8bitar_clone_help/

Schematic: https://www.parasitstudio.se/uploads/2/4/4/9/2449159/8bitar_doc.pdf

[anotherjim]

Looks like the last part is an oscillator being modulated by the squared-up guitar signal with the transistor. Hmmm... I'm not sure if it can always be certain to stop the oscillator when you stop playing which needs Q1 to turn off. U1.6 pin12 needs to drop to 0v when playing stops. See if a resistor from U1.1 pin1 to 0v can bias it towards 0v enough to stop Q1 - try 1M for that.
Also, I know the input of this type of thing can be too sensitive to noise and a breadboard build is very open to noise. Try changing R1 from 1M to 100k.

[mrmet5]

Hey It'll have to be tomorrow but I'll definitely try them! Before that though, does '0V' mean ground here? Forgive me if it's a dumb question. Thanks for your advice!!

[anotherjim]

Hey It'll have to be tomorrow but I'll definitely try them! Before that though, does '0V' mean ground here? Forgive me if it's a dumb question. Thanks for your advice!!

Powers supply 0v which in this case is ground or power supply negative.

[mrmet5]

Hey Jim,
So unfortunately the motorboating persisted. A few observations though: 1) the motorboating noise now sounded lower; 2) I was able to make it go away, however, as I turned the volume (not the guitar volume) down all the way, as opposed to neither guitar nor effect volume would make it go away before. Do these give you any clue? Thanks again for your help!

[anotherjim]

You know what, I didn't look at the video before. That's not breadboard but perfboard. Breadboard is what we call the solderless plugin things.
If you can power it without anything plugged in, measure the chip pin voltages.

[mrmet5]

I know it's not a breadboard in the video but I did test it on a breadboard before soldering but the results were exactly the same.

Now the voltage readings on 1 to 6 are all between 1.8V and 2V, and 8 is 1.94V, 9 is 0.09V, 10 is 1.99V, 11 is 1.95V, 12 is 0.02V, 13 is 2.5V and 14 is 4.7V.

[anotherjim]

Shouldn't pin14 be nearer 9v? What are you powering it with? A bad battery can measure good when not supplying anything but drop the voltage a lot when it's got work to do.
That said, the chip can work at 4.7v and the voltages on some of the pins look sensible but 8,9,10 & 11 look wrong but let's have 9v power first!

[amptramp]

CMOS inverters take about 8 mA in their linear range at the middle of the swing from low to high or high to low.  You might need more than 10 µF for a bypass capacitor.  The current drain could couple into other stages and since all six stages pass through the Vcc/2 point at the same time, you have a 48 mA drain to contend with.  This is near the limit for a 9 volt battery and definitely requires more filter capacitance.  Back in the days of germanium transistor radios, motorboating was often the sign of a failing battery or a bad filter cap.

[mrmet5]

Jim,
I used a brand-new battery, and the readings changed to: 8-2.32V; 9-0.07V; 10-3.61V; 11-5.08V; 12-0.10V; 13-4.60V; and 14-8.78V. However, the motorboating did not go away...

amptramp,
I tried 22uF, 47uF and 100uF, but they didn't really change anything.

I appreciate your continuing help!

[anotherjim]

The only thing I can think of now is to try increasing R8 above 10k. It should make it less sensitive.

[mrmet5]

Hi Jim,

Tried 22k and 47k but did not make the motorboating go away.

I'll continue learning electronics and I'll come back to this again some time in the future. Thanks for helping, I really appreciate it!

[antonis]

Maybe an RC supply filter could eliminate your issue..

Just interpose a 10R resistor (or bigger if you can afford the voltage drop across it..) between +9V and C8 (which should be made 470μF..)

[mrmet5]

Hey antonis,

Thanks for your suggestion, but before I try that I just wanted to make sure '10R' means 10 ohm. That's correct?

[Rob Strand]

Thanks for your suggestion, but before I try that I just wanted to make sure '10R' means 10 ohm. That's correct?

10 ohm

You might also try a higher values like 100 ohm to 470 ohms.   It's possible the 4069 is pulling a lot of current at 9V and promoting motor-boating.   The added resistance will drop the supply voltage which will reduce the current and perhaps prevent the motor-boating from happening.   Good idea to have a big cap like antonis' 470uF.

I'd also try adding 10k to 100k in series with C2.   That will reduce the overall amplifier gain perhaps making it less sensitive to motor-boating.

The last stage two inverters are set-up as a flip-flop.   That's where the divide by 2 comes from.   The way the flip-flop is triggered with a transistor is a little dodgy on one of the switching polarities.   It's worth experimenting with other values of R7,  try values between 4.7k and 100k.   You might find the unit behaves better when that resistor is tweaked.

[anotherjim]

I see a Schmitt trigger oscillator with the transistor forming Rin against R8 positive feedback. Effectively the same oscillator circuit as used with a single Schmitt inverter. It certainly doesn't sound like an octave down to me.

[Rob Strand]

I see a Schmitt trigger oscillator with the transistor forming Rin against R8 positive feedback. Effectively the same oscillator circuit as used with a single Schmitt inverter. It certainly doesn't sound like an octave down to me.

Jim, have a look at this footswitch flip-flop.  It's got the floating switch.
Same circuit used on Dod and Pearl pedals,

http://www.geofex.com/FX_images/ltchrly.gif

(It's possible the 'dodgy' transistor switch is part of the sound.)

There's no hysteresis for a Schmitt osc.

[Rob Strand]

I see a Schmitt trigger oscillator with the transistor forming Rin against R8 positive feedback.

There's no hysteresis for a Schmitt osc.

This morning I see that Schmitt you mentioned.  Funny thing is I don't remember those footswitch circuits oscillating.  The frequency will be quite high if it does (10us time constant => 100kHz). 



I don't have a clear head this ATM but it just occurred to me why the circuit doesn't oscillate.  When the feedback resistor (10k) is small compared to the input resistor (100k), the Schmitt input thresholds are are very high, in fact they exceed the power supply range.  So what happens is the cap only charges to a small voltage and settles.  It never reaches a high (or low) enough voltage enough to cause the first gate to change state.  So oscillation doesn't occur.

[mrmet5]

Hey guys,

So I tried putting in 10 ohm resistor (also 470 ohm) and the motorboating got significantly reduced! I can still hear a tiny bit, but I'd say 90% of it is gone now. As for the 470uF cap, I only have 100uFs (and 1000uFs...) at the moment so I put that in, but do you think a 470uF would be crucial?
That all said, on the other hand, somehow the bit-crushing effect is kind of disappeared... Now it sounds like a booster/fuzz effect (which actually doesn't sound too bad as is haha). Can you think of what might have gone wrong there?

Thanks for all your help!!

[Rob Strand]

That all said, on the other hand, somehow the bit-crushing effect is kind of disappeared... Now it sounds like a booster/fuzz effect (which actually doesn't sound too bad as is haha). Can you think of what might have gone wrong there?

Without building one and going over the design in detail it's hard to know precisely what's going on with these types of pedals.   The circuit around the transistor and the resistors and caps around the base might have been tweaked to make it sound like it does.    The added PSU resistors could be dropping the supply enough to upset the balance.   I'd be tweaking R7 to see if it comes good again, or perhaps even better than before.

IMHO, the best thing to try would be adding the resistor in series with C2 to reduce the gain.   If the amplifiers have too much gain, at some point the circuit is going be twitchy due to just about any source of noise, that includes supply variations.  Tweaking the gain *before* a Schmitt trigger (R4 through U1.6) can be very effective fending off noise.    The fact the issue doesn't improve when you increased R4 is a sign the problem could be coming from the amplifiers.   The caps C3 and C4 help remove high frequency sources of noise but they can't do much for low frequencies, and the motor-boating is a low frequency effect.

To add to this you don't really know how close to the center of the supply the amplifier U1.3 and U1.2 are biasing.   If it were mine I'd be grounding the input side of C1 and measuring the voltages on the outputs of U1.3 and U1.2.  If the DC voltage on U1.2 is close to the thresholds of the Schmitt trigger then the Schmitt trigger might not we operating in an optimal manner.   Normally I'd expect things to bias up OK and the Schmitt to work correctly as all the inverters are from the same chip.   All this is more of a check than a fix.  Reducing the amplifier gain is a known fix.

Other similar debugging things would be to lift C2 then ground the side that when to the U1.3 output.   To some degree that removes U1.3 from contributing to motor-boating but it's not an ultra conclusive test.    The whole idea is to narrow down where the problem is, then find out why, then add a fix.    You could do a lot of debugging only to find adding resistor in series with C2 is the fix  .