Circuitbenders Forum

Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: diagnose a broken circuit - seeing the flow - help?  (Read 11929 times)

volg4

  • Newbie
  • *
  • Karma: 1
  • Offline Offline
  • Posts: 27
diagnose a broken circuit - seeing the flow - help?
« on: April 05, 2009, 12:02:22 AM »

im still pretty new to all the circuits and schematics

 i dont know what im looking at when i see a diagram, i managed to build an lfo using a picture of how its put together, i dont know how or why this lfo works but if i follow the picture, like a join the dots, it works... lol

ive got an amdek pck-100 percussion synth and it sounds nothing like the you tube video of the same thing...
it works but makes a high pitched noise all the time, no sub or bass or mids at all...
i have the schematics and building instructions and some mods (which i did thinking they might make it sound better, ...they didnt)

i dont know where to start looking, has the oscillator gone? any suggestions?

one of these..  http://filters.muziq.be/model/amdek/pck100

video of how it should sound

http://www.youtube.com/watch?v=td8EqQU80R0






Logged

Gordonjcp

  • This person is dangerously insane.
  • *******
  • Karma: 78
  • Offline Offline
  • Posts: 1005
    • http://www.nekosynth.co.uk
Re: diagnose a broken circuit - seeing the flow - help?
« Reply #1 on: April 06, 2009, 12:59:35 AM »

Okay, the circuit is a little odd, so we'll go through it bit by bit.  You very quickly learn to recognise "blocks" of circuits, so when you see something you can kind of guess what it's supposed to do.  It's a bit like, even if you don't entirely understand a word you haven't seen before, you can figure it out from the context.

I don't know how much you know about electronics, but there are masses of sites out there that will tell you all about it.  So I'll assume you know what resistors and capacitors are, explain opamps briefly, and crack on.

The circuit uses a bunch of opamps (Operational Amplifiers) which are chips that can be wired up to do various clever things (not just amplifying, as we shall see!).  Look at the top left corner, and you'll see IC3 used as an amplifier (with a twist which we'll come to) and a buffer - the buffer has no gain but "isolates" the input from the load on the output.  In the datasheet for the opamp it describes pins 1, 2, and 3 as being opamp A, and 5, 6, and 7 as opamp B - dual opamp, two in one package.

Okay, IC3, then.  You tap the piezo beeper - <BONK> - and you get a big wobbly pulse.  The sensitivity pot controls how much of this is fed to the amplifier formed by one half of IC3, through a simple highpass filter formed by the capacitor and 22k resistor.  The signal goes in the non-inverting input, where it's amplified by 1000 (that would be 4M7/4k7 = 1000).  The clever twist is the diode, which ensures the output pulse only ever is positive, and the output of the opamp doesn't discharge the envelope cap.  C8 charges up from the pulse, and discharges through R17 and the decay pot - R17 ensures that there's a minimum decay time and we don't just short the cap to ground.  NOTE: It's a very good idea to use a resistor in series with a pot if you're adding a pot to ground to fuzz up a circuit  that you're bending.  It might save delicate parts from getting fried.  That buffer we mentioned earlier consists of the other half of IC3, with the inverting input directly connected back to the output, and the signal fed to the non-inverting input.  The output follows the input exactly, but if you load the output with something it won't affect the current at the input.  Think of it as power steering for C18 ;-)

I'm not totally sure what the Attack circuit does, so I'll worry about that tomorrow.  I *think* it adds a bit of a "click" to the output, but I don't know.  Anyway, IC2 forms a VCA, with the control voltage from the decay control being fed in through R19.  It's a special kind of opamp called a "transconductance opamp" - what this means isn't very important but basically you can control how much gain it has with an external source going into pin 1.

The funny bowtie and semicircle symbols indicate solder pads that can be cut or bridged to change the sound.  You could break these out to switches.  Anyway, we'll move into the good bit - the oscillator.

The oscillator is formed from IC4A and IC5B in the middle of the circuit.  It works like this.  C15 charges up through R24, and the voltage on pin 1 increases as it does.  IC5B is wired as a comparator, and when the output of IC4A reaches a certain point the output of IC5B turns on, which turns on the transistor, which starts to discharge C15 again through R29.  Eventually the voltage drops far enough that the transistor turns off again and C15 starts to charge again, and we're off round again.

The rate at which C15 charges is determined by the voltage on the output of IC4B.  It has various inputs from the LFO, envelope and pitch control.  The voltage on pin 7 of IC4 should vary when you turn the pitch knob, trigger the drum, or increase the LFO mod.  If it doesn't, it might be that one of the pitch or mod controls is miswired.  The build instructions are out there on the 'net, along with the circuit diagram.  It's probably worth checking through the build instructions to ensure that everything is wired up as it should be, before you start.
Logged
If at first you don't succeed, stick it through a fuzzbox.

volg4

  • Newbie
  • *
  • Karma: 1
  • Offline Offline
  • Posts: 27
Re: diagnose a broken circuit - seeing the flow - help?
« Reply #2 on: April 06, 2009, 06:37:29 PM »

 ;D what an awesome reply!! thanks... lots to get my teeth into

there was a slight whoosh as it went over my head but i reckon with a few nights of reading i might be able to get my head round this...  i understand mostly what the bits do in isolation (resistors and capacitors mainly :)) but when they are all joined together it starts to get a bit hard to fathom

definitely worth the effort to read up on this tho...

thanks again for your time, most appreciated
Logged

volg4

  • Newbie
  • *
  • Karma: 1
  • Offline Offline
  • Posts: 27
Re: diagnose a broken circuit - seeing the flow - help?
« Reply #3 on: April 09, 2009, 12:19:33 PM »

so i spent a couple of hours (maybe more  :-\ ) reading over your descriptions and managed to get a clearer picture of whats going on with the synth...

still not totally comfortable with the ins and outs but much more clued up than i was

i checked pin7 on ic4 and it seemed to stay at 8.5v regardless of moving the pitch or mod pot...
gonna rewire the pots and see where that takes me
Logged

Gordonjcp

  • This person is dangerously insane.
  • *******
  • Karma: 78
  • Offline Offline
  • Posts: 1005
    • http://www.nekosynth.co.uk
Re: diagnose a broken circuit - seeing the flow - help?
« Reply #4 on: April 09, 2009, 02:47:02 PM »

Is there a good reason to rewire the pots?  Are they wrong, or is the wiring messy and badly done?
Logged
If at first you don't succeed, stick it through a fuzzbox.

davethedoobie

  • Newbie
  • *
  • Karma: 3
  • Offline Offline
  • Posts: 36
Re: diagnose a broken circuit - seeing the flow - help?
« Reply #5 on: April 24, 2009, 08:09:25 PM »

Gordon jcp - I don't own this perc synth but I've learnt alot about bending just from reading your instructions.  Why don't you explain how the Casio vl-1 circuit works?  I am a newb - I can do a-b bends, add the odd cap etc to make them give different sounds but when it comes to theory it like sitting in a physics lesson again!  (I wish I'd listened now). 

Tell me this does putting a cap with a resistor work at changing the sound?  I won't try this incase I fry it and I'm a terrible cook.
Logged

Gordonjcp

  • This person is dangerously insane.
  • *******
  • Karma: 78
  • Offline Offline
  • Posts: 1005
    • http://www.nekosynth.co.uk
Re: diagnose a broken circuit - seeing the flow - help?
« Reply #6 on: April 24, 2009, 09:17:06 PM »

I haven't got a circuit diagram for a VL-1.  If you can find one, then let me know and I'll take a look at it ;-)
Logged
If at first you don't succeed, stick it through a fuzzbox.