Casio SA-20 crystal problem

[peter]

Hi there!

I'm just started to experiment with circuit bending and i chose the SA-20 (awesome keyboard). And recently I found a cool thing on http://casperelectronics.com/finished-pieces/casio-sa-keyboards/ adding an extra crystal to change the pitch. But here it comes...i tried it on my SA-20 but it didn't work. Somebody knows why?? Or maybe someone knows another cool way to do it??

Tricky thing circuit bending but damn it I like it!!
Peter

[Gordonjcp]

I'm surprised that works at all, with those big long dangly wires.  They probably have an unbelievable amount of capacitance at the sort of frequencies the clock runs at.

Basically what you're trying to do there is replace the crystal with one of another frequency.  The crystal resonates at a very steady frequency, only drifting a little with temperature.  You could replace it with a tuned circuit and have a variable oscillator if you had the space, or a different crystal for different pitches.  Some people have built external oscillators and driven one of the crystal pins (with the crystal removed) to get a variable oscillator frequency.

[PolyPhuckin]

Hi

I trie the casper way but came a little unstuck when it came to wiring up 3 pins to a DPDT switch, in the end, i just bought a 3PDT switch.
Works wonders 

[Gordonjcp]

The middle pin is earth on the ceramic resonators.  As the linked article a couple of posts back show, if you're using a crystal you need to add a couple of capacitors.  Bear in mind that the length of the wires has an effect too, because of their capacitance.

[peter]

Thanks for the responses!

How do I know witch capacitors to use, is there some way to calculate the amounts of farad I need with each capacitor?

[Gordonjcp]

Look up "crystal load capacitance" on Google.  Basically most crystals need to be loaded with a certain amount of capacitance, and will have the amount on the datasheet.  As a rough guess, the two capacitors should be a bit more that the load capacitance value for the crystal.  Somewhere between about 15 and 33pF shouldn't be far wrong for most crystals.

[peter]

Thanks a lot for clearing things up for me!!

Gordonjcp you're right. I used first the 33pF caps, but didn't work. But then with the 15pF caps there's such a great low pitch..amazing. By solving this i fall into another problem. When i change the switch from the old crystal to the new it kinda tweaks and stops immediately. Some idea's??

[Gordonjcp]

There's probably not a lot you can do about that - there will be a point where neither crystal is connected to the circuit, and the oscillator will just stop.  Maybe you could try using one pair of caps on the middle terminal of the switch, and two crystals (or leave the centre leg of the ceramic resonator disconnected)?

[Circuitbenders]

Would a rotary 'make before break' switch sort that out? Or would it just crash when both are connected

[PolyPhuckin]

When i made this mod i used two 22pf caps and lived with the fact you have to turn it off and on again to change the pitch.

[peter]

I think that's how it is. When I turn it off and switch then it works wondrous.

On GetLofi.com I saw the LTC1799 mentioned. Would this be a good alternative?? I read from a lot of people that it's an intressting thingy. And how about the 555, I read and read and tried, but could'nt work out how this could work for this as oscillator?

What it comes to is that I'm teaching myself how all off it works, but sometimes I just get stuck in it. So I really appreciate the help!!

[Gordonjcp]

LTC1799 would be good - you could have either a knob for variable frequency or have some switched "preset" frequencies.

I'm not sure the 555 would go fast enough.

[peter]

yeah, that's what i thought. I ordered the LTC1799 chip so i can experiment. Updates later....thanks!