Hi

Ive recently built a synth based on CMOS chips. I am using a 40106 to generate 2 seperate VCO's, one with Tri and Saw waveforms and another with just the square wave, I am also modulating the 1st osc with an LFO. The square wave osc is being run through a CD4040 so that I can stack 3 different octaves and then being run out to a 100k pot for overall volume. The Tri/Saw is being run through a CD4069UBE to buffer the signal and provide some overdrive, although it sounds more like a basic filter to me. This is then run through another 100k pot as the overall volume for that osc. These two signals are then put through 2 100mF caps before going to a stereo socket.

Everything is working fine except I'm getting a periodic click coming through the speakers, when that happens I can also hear the LFO driving the 1st osc but very quietly, when the click happens again it goes silent and then the process repeats. This doesn't happen when both master volumes are turned right down though. Does anybody have an idea how this signal is leaking through and what are the click sounds and of course how can I fix it.

Any help would be really appreciated.

Thanks

Gface83

 

Hello,

Can you post a schematic of the current setup?

How is the decoupling done?
Decoupling or Bypass Capacitors, Why?

Bertus

 

Hi Bertus

How things? I have been following along with a series called Logic Noise on hackaday.com so I haven't really drawn up a schematic, I have just been learning the separate sections and then trying to combine them all. Anyway, here are a few schematics from hackaday that I have been using.

This is the section i used for the second (stacked octave) osc. The only thing that I changed was that I am using a resistor ladder keyboard instead of the 100k pot (RVAR1), and I have switches for the octaves so I can turn them on and off. As mentioned above, where it says output on the schematic I have used a 100k pot as a volume control which then goes to a 100mF cap and out top stereo socket.

This is the first osc schematic: I have copied this exactly except I've added an LFO to the input of the osc with a switch so that I can turn it on and off and the output also goes to a 100K pot (volume) and onto another 100mF cap and onto the stereo socket.

TBH, I don't know much about decoupling or bypass caps. I have just read that post you posted and it sounds very good so I will try and get to grips with decoupling. When it says to put a decoupling cap across the power lines would that be positive and negative/ground? and how big of a cap should I use?

 

Hello,

Usualy the decoupling capacitors are 0.1 μF and placed directly on the supply pins of the chips.
They will prevent pulses wandering from one chip to the other, wich will often lead to distortion in analog circuits or false triggering in digital circuits.

bertus

 

Ok brilliant ill try that, thanks. So just so I'm certain the cap will go on the + power pin and then to ground? Do I have to use one on the positive pin of every chip? And just out of curiosity, decoupling means to decouple from the power supply or decoupling the chips from each other?

 

Hello,

The decoupling capacitors are used on the powerpins of each chip (from + to -).
The decoupling capacitors will buffer the powersupply to prevent dips created by the transients in the digital chips.

Bertus

 

That's great, thanks for the help

 

I'm getting a periodic click coming through the speakers

I'm guessing that could be due to a brief burst of spurious oscillation due to circuit instability. The unbuffered 4069UBE would be my prime suspect. The addition of decoupling caps would hopefully cure that.