PSS-31

[SineHacker]

I opened up a PSS-31 today, another of yamaha's AWM instruments in hope of finding similar glitchy bends to the PSS790. I didn't find anything for ages except for some short but sweet lockdown crashes. I started using a pot between two crocs and then a capacitor, still with no new results.

- I was thinking of cutting a hole in the top of the keyboard so that I could easily access one of the chips and just sample some of the crashes with ease, but I then stuck a small signal diode between the crocs and got some amazing results!

a number of the points that were previously causing breakdown crashes now started causing more stable... er... crashes - where the keyboard kept on trying to do whatever it was doing while it was crashing, sometimes the beats and tones start sounding like they're in jellyland and other times you get aggressive washy loops of glitch and crashing sound that often have trouble stopping when you break the bend FUN

I haven't laid down any solder yet and I HAVE taken photos *for once*  to log the finds - I can post them up should anyone be interested


Aid

[Gnome]

Yeah post them sounds cool.  And post samples of it too i'd love to hear it

[SineHacker]

Ok, I only added six switches in the end - I thought I had found more but a number of the glitches were very similar if not identical - there probably are more in there though, I was only focusing on one of the chips really:





a quick breakdown of my jobby is

1 (+diode-,switch) 3
1 (+diode-,switch) 4
1 (+diode-,switch) 5
1 (+diode-,switch) 6
1 (+diode-,switch) 7
2 (switch) 8

Point 1 is the sweet spot, using a diode (positive side facing point 1) this can be connected to points 3, 4, 5, 6 and 7 (and others) for different glitches on the tones, drums and backing tracks... I used some vero board, soldered one diode and a hookup to point 1, then linked up a switch in line with the other points - this kinda prevented me from throwing more than one switch at a time as it makes connections between points 3 4 5 6 and 7 usually resulting in a crash, but turned out not to be too bad as sometimes you can control weird harmonic drones by flicking between them. If you want to avoid this you will need a diode before each switch I think (i'm just being really cheap at the moment because my bank account is kinda dry)

Point 2 can be shorted directly to point 8 for a kinda sqwelchy sounding glitch on all tones and beats

the other points I noted (9 to 14) are 'points of interest' I didn't end up using but generally produce cool crashing glitches - I didn't use them because they are pretty unreliable, sometimes just crashing the keyboard with no resulting sound. but may do some cool stuff if approached the right way!

I also marked on the top piccy (with the crude x) what I'm guessing are a pair of amplifier chips (this keyboard isn't stereo though?) which may be usuable for some distortion bends or the likes. I didn't jot down the codes foolishly, but I want to install a break switch so i will be opening up again. I also forgot to take pics of the inside after I did the bending. d'oh! so I can do this if it helps visualise my madness

[SineHacker]

at long last spring has arrived:

Circuit bent Yamaha PSS-31 fun and fail

[MAGTIG]

Nice work! I have some of the more modern PSS's lying around so this could potentially really help me.

I have a question though: I know diode's prevent the current from traveling backwards, but I don't know WHY that would make these glitches work for you, or why there seem to be diodes on a lot of triggers/switches. Can someone please explain the theory to me a bit?

[SineHacker]

I'm no electronics expert so I couldn't go into theory but basically, making simple shorts between most of the pins on the chip will just cause the keyboard to crash, however using a diode means that the short is only going from one pin to another and not bouncing back, one pin of this chip was really useful for creating controlled crashes, but without the diode in place this pin would also kill crash the keyboard.

But you already have the right idea, the diode basically allows the current to flow in one direction, I can't say for sure that this is true for all diodes, but this is my understanding of them, when in doubt, wiki it: http://en.wikipedia.org/wiki/Diode

as for triggers/switches, I guess It would depend on the application really


If you play with oscillators at all, they can do some really cool things to the sound / oscillation, I couldn't tell you what they do again though as I don't have any decent equipment (such as an oscilloscope) other than my ears!

[sk-1]

Having a bank of samples to bend turns out much nicer results, even though getting there is not as easy.  This video has convinced me to grow my pile in the direction of AWM and not just FM and PCM.  And these bends sound nice.  Even nicer that you've left yourself a lot of options if you ever want to dig deeper and go the full hack.  So I can't wait to see what you discover next!

I wonder if the AWM-based PSR keyboards would be as bendable with similar results.  After all, they are just full-size PSS boards with similar architecture inside, aren't they?

Excellent job... keep it up 

[SineHacker]

I wonder if the AWM-based PSR keyboards would be as bendable with similar results.  After all, they are just full-size PSS boards with similar architecture inside, aren't they?

The architecture can vary greatly, as with the case above I have found with a number of both old and new AWM yamaha's that you need to use components (usually diodes) between shorts in order to get any results - I keep preaching about this, maybe I should give it a rest!