designing a keyboard that can be patched into a toy in place of a clock resistor

[madstringbean]

I've been imagining how I would build a keyboard where each key passes the signal through at a different speed.

So instead of using a potentiometer to replace the clock resistor on a device, you would have a keyboard, with each key representing a different resistance.

This would involve a big array of switches to route the signal through the correct resistor whenever that resistor's key is pressed, which i am not sure how i would design.

Also, do you think it would be possible to tune it?  Are there any rules governing the relationship between resistance and pitch?

then i would give each end of the keyboard's "resistor array" an alligator clip and it could then be clipped into any circuit with a clock resistor.

Has anyone seen any documentation of something like this being done?  I would really like to learn more about it if so, so I can avoid making idiot mistakes.

thanks

alison
http://www.moonpieinthesky.com

[mikebike]

im working on somthing similar to this.

i was gonna try having a chain of trim pots for each "switch" (read key on a keyboard)
that i would tune by ear the first time and hope fully would corispond to other instruments,

each lower "note" would be in series with the key above it. these would all have a master pot in front of them to adjust the overall tuning

there is also ken stones mini keyboard pcb, but i dont know if volt/octive is gonna translate to what your wanting to acomplish

id like to hear somemore ideas, i think the trim chain is complicated and has its faults

[madstringbean]

Mine is probably going to be way simpler than that.  I think I'm going to make a quick ultra simple one first and then think about making it better after that.

I just drew this quick schematic:



That's really the minimum, but it should work.  Each key on the keyboard will be a momentary switch.  The other thing I realized about this is that pressing more than one key at once will have the effect of turning on those resistors in parallel, which is interesting.  I've given up the idea of tuning it altogether though I may add a master tuning pot to change the range of the whole thing.  I'll have to see what seems to be the best when I'm experimenting with it later.

let me know if you have any other cool ideas!  after i do this i might try mike's version.  (the way i designed it just isn't quite as awesome as I thought it was when the idea first drifted into my skull - this is never going to rival an sk5! )

alison
http://www.moonpieinthesky.com

[expanoncolin]

Mine is probably going to be way simpler than that.  I think I'm going to make a quick ultra simple one first and then think about making it better after that.

I just drew this quick schematic:

That's really the minimum, but it should work.  Each key on the keyboard will be a momentary switch.  The other thing I realized about this is that pressing more than one key at once will have the effect of turning on those resistors in parallel, which is interesting.  I've given up the idea of tuning it altogether though I may add a master tuning pot to change the range of the whole thing.  I'll have to see what seems to be the best when I'm experimenting with it later.

let me know if you have any other cool ideas!  after i do this i might try mike's version.  (the way i designed it just isn't quite as awesome as I thought it was when the idea first drifted into my skull - this is never going to rival an sk5! )

alison
http://www.moonpieinthesky.com

Just a note that any instant where you remove the clock resistor completely and then put one back in (eg, no switches down vs a switch down in your schematic) requires the circuit to start back up, and could yield some big pops, clicks, and start-up delays... it would be better to put a relatively large resistor always wired in and just connect things in parallel to it.

-Colin

[theshame]

I've built something similar to this, but I don't think mine has anything to do with actual clock signals. I think the important thing is to find bend points in the circuit that will dramatically alter the pitch when connected with little resistance. If you are there already, then my circuit should work for you.

Like mikebike was saying, I made mine with a pot connected to each switch to allow for tuning, rather than hard-wiring the resistances. I think you will find tuning with resistors difficult to do, especially if you are planning on using this across multiple toys. In my case, the pitch of the signal the keyboard effects is not affected in a linear way. Being able to tune on the fly is quick and easy.

The only thing I don't like about my circuit is that it plays a tone when no key is pressed. I think that is what Colin was getting at, as completely cutting the connection would crash the circuit. However, if you can get the pitch above audible frequency, you could make your circuit default to that.

The schematic is essentially the same as what you drew, but with pots in place of resistors. I also added a LDR with a switch for alternate control.

As for connecting with alligator clips, if you are looking for a cleaner alternative, a simple audio cable with jacks attached to bend points works great.

Check it out here:
http://www.vimeo.com/4230827

[madstringbean]

thanks, you all made some very good points. 

it seems the biggest remaining question is the best way to set up what happens when no key is pressed, because i don't want a signal coming through.  But having the "default" signal passing through maintained at above-audible range (therefore, passing through @ very low resistance) will mean that any key pressed will add resistance in parallel to the always-on one, which will just mean i will have to use higher value resistors or pots for those.

this is coming together nicely! I can't wait to get home from work and work on this

alison
http://www.moonpieinthesky.com

[mikebike]

does a signal allways come through normally?

maybe you could switch in the resistor at the same time as hitting the note?