envelops on the sk-1

[sensor]

is there a way to change the envelop-parameters via pots on the sk-1?

[gmeredith]

Yes!!

I did a mod for my SK-8 which I call the "Slo-A/D mod", which changes the attack and decay rates of the sounds. The only thing is that it uses capacitors to change the rates, not resistors, so you can't use a simple pot setup to change it. I've tried it and it doesn't work. Instead, you can use sets of different value capacitor banks, which you can switch between using a rotary switch.  Basically, the SK keyboards generate a set of 4 control voltage envelope signals (one for each of the 4 voices of the 4-note polyphony of the SK) that vary between 0 and 2.5V DC. This gets sent to a quad VCA unit that opens and closes in response to the shape of the control voltage signal.

If you put an electrolytic capacitor in parallel with the ground line and the control voltage (CV) line of each envelope signal line, the CV will charge up the capacitor. In doing so, it reaches its proper voltage more slowly. When the CV stops, the capacitor discharges into the CV line and so extends the time that there is a voltage in the CV line. The VCA sees all this as a slower attack and extended decay.

If you use a resistor instead, all it does is short out the CV voltage to ground, and reduces its maximum voltage that goes to the VCA. The result is a quieter volume of output for that voice, depending on the resistor value. Which is an interesting bend in itself, but it doesn't change the attack/decay rate.

You can find my Slo-A/D mod here:

http://www.circuitbenders.co.uk/forum/index.php?topic=485.msg2533

Download my illustrated Slo-A/D guide, and also the SK-1 service manual to find the CV points on the SK-1 to tap into.

This thread here will help you to find the envelope lines, with the service manual:

http://launch.groups.yahoo.com/group/CasioSK/message/373

Cheers, graham

[Gordonjcp]

Makes me wonder, is there somewhere that you could get a trigger output from?  There are plenty of real ADSR envelope generator circuits out there.  You'd just need to trigger it and use its output in place of the one from the chip...

[gmeredith]

Yes, there is a form of trigger out. I had also thought about that possibility. You can use the "default" envelope setting on the SK - the one that has the sharp attack and minimal decay, mimicking a gate output. Then you could tap out the envelope CV lines, which output +2.5V DC on each voice and run it to a quad ADSR.

The problem I came across when exploring this option was the lack of single rail +9V DC powered ADSR chips. They're all dual rail +/- 12V to 15V DC. They were also quite large and complicated - certainly nothing that would fit easily inside the SK casing.

If someone knows of a small, simple ADSR unit for 9V DC, post it up here.

Cheers, Graham

[sensor]

has somebody an idea where to find the envpoints on the board to connect the capacitors with? i just have a really blurred servise manual pdf.

thanks
matthias

[Gordonjcp]

Graham, you probably won't get an ADSR chip these days, except on eBay where you might score a mindboggling expensive Curtis or SEM chip.

You can build one using discrete components and opamps, and try running it off 4.5/0/4.5 - it should still work.

[gmeredith]

Hi Sensor,

If you could take a good sharp photo of your SK-1 circuit board (both sides) and post them up here or email them to me I could probably pinpoint them for you.

One dead giveaway as to the CV envelope lines is the 4 little diodes on the board, near the processing chip. See the photo below. This is for the SK-8, so it won't be the same on your board. But look at the diodes - these WILL be the same on your board, and these are what you are looking for. Hook up your CV tap wires to the ends of these diodes somewhere convenient. You can verify them to be the CV points by putting a + multimeter probe on the diode leg, and the - probe on the - battery terminal spring. Set the multimeter to DC volts. Switch your SK on and play a 4-note chord and hold the notes. The meter should read about +2.5V DC. Release the notes and it should drop down to zero or close to it. These confirm it is a CV point. Do it for the other 3 diodes to confirm that they also show this, and then they are your 4 CV points, one for each voice of the 4-note polyphony.

You'll notice in my pic that my CV wires are off wire jumpers or resistors - this is because they were directly after the diodes, and were an easy soldering point. But you might want to solder them from the circuit track side of the board instead, and so can solder them directly to the track legs of the diodes if that suits you.

Hope that helps,

Cheers, graham

[sensor]

thanks graham. it helped a lot. i found the diodes.
is there a way to solder all capacitors to one spdt? i thought about one diode after each of the capacitors and then all together to the spdt. please tell me if this is bullshit. i just don't have 4pdt's.

thanks
matthias

[gmeredith]

Unfortuntely they have to be individual caps on their own switches. I tried connecting them to 1 switch originally but it didn't work - they discharged into each other when the switch was disconnected and wouldn't bypass the effect. So you'll need a 4 pole single (or double) throw switch (4pst), or 4 individual SPST switches. You also need to put the switches on the + side of the cap, not the ground side,  because if the + leg of the cap is not the one switched, it will keep its charge on the CV line and will not bypass the effect either.

You can also use a 12-pole rotary switch - you can configure these to be a 4-pole 3-way switch. This will enable you to have 3 banks of different sets of capacitors for different A/D rates

Cheers, Graham

[wax+wire]

any thoughts on good suppliers (australia) for 4PDT switches.  i'd love an 'off' position not On-On.

there is a number on ebay, but you know, china/hong kong or $20 from USA