Bliptronic 5000 LED Synthesizer

[electri-fire]

I got two delivered yesterday. First impressions:

Eight cheesy sounds, most of them with a piano like attack.
1: synthy sound
2: thin piano
3: Rhodes piano (my favorite)
4: thin clavi
5: harp/guitar like
6: reed like
7: bell/metallic
8: sax?

The lenght of the notes increases with slower BPMs. All sounds have a similar (fast) release. I suppose note lenght is shortened at faster BPMs to retain the polyphony for each step.
The bliptronic retains the last pattern when turned off and on again.

I sent it through Reaktor effects and was able to get some soundscaping stuff going. Ah well, even a Tammy Wynette or Willy Nelson record can be turned into a screaming noisefest with effects. Many 80's toy keyboards will give you more versatility and better sound for the buck. There's a serious lack of lows. I wish the designers had opted for a lower octave.

Line out is fairly hot, the volume knob has just sufficient response at low volume. Good to drive distortion devices (even a headphone, albeit at slightly lower volume).

All but one button feel responsive and reliable. Out of 128 (2x64) one I have to hit slightly oblique. Maybe it was somewhat displaced during assembly, there's some sideways slack in them. As the board needs to be pressed sideways to get the in-out jacks into their holes, I'll file some bevels at the inside of those holes first before reassembling.
I hope it works then. Even one slightly less functional button is distracting.

In-out sockets are soldered to the circuitboard, mine dropped 60 cm (nearly two feet) with the headphone attached, and damaged the headphone jack.

So today I opened it up. Four screws underneath the rubber pads: easy. Then remove the knobs. Now careful, the in-out jacks coming through the holes in the case fit rather snugly. I decided to press the jacks in from the outside with the tip of a ballpoint to prevent further damage.

[electri-fire]

[michaeluna]

Greetings all. I was able to bend the Bliptronic's clock using a Getlofi precision timer kit, and I found some bend points on the audio IC that create a rich harmonic drone.

The sync in/out seems to be 5v pulses, and using the HighlyLiquid MD24 kit to translate, I was able to sync the Bliptronic up with Ableton live by sending it a MIDI note at the start of every measure. I need to take some photos and video, but I'll post here when I've got that up.

[RF]

[creekree]

yesterday i found out that the bliptronic keys are glow-in-the-dark! nice touch...
anyways, there is this video on youtube where a guy called gjyamada shows off his midi interface for the bliptronic:
http://www.youtube.com/watch?v=Jd-iiyE1lvQ
- and he has the schematics and code for download! i have not tried it myself though, as i do not have a PIC burner (anyone from germany willing to program the PIC for me?)

the midi interface obviously is "only" for syncing.

[slacker]

[electri-fire]

[urbanspaceman]

here
http://blog-imgs-38.fc2.com/y/a/m/yamadanohito/midigate1.png
http://blog-imgs-38.fc2.com/y/a/m/yamadanohito/midigate2eng_asm.txt

ciao!!

[creekree]

[electri-fire]

[electri-fire]

[slacker]

I had a play around with it this morning and still haven't found any interesting bends.
Like RF already discovered the EF pins change the voice and the speed. You don't seem to be able to get any voices or tempos that aren't available from the switches though.
Connecting some of the pins to OUTL or SIGINL through a resistor or cap gives some nasty squarewave buzz on top of the audio. I don't think this really counts as a bend though, You're probably just mixing some logic pulses from the chip with the audio.
Poking around the AB pins and CD0 and CD1, sometimes changes the pattern. So maybe these pins are the ones that talk to the button board, I haven't traced them. Some of the AB pins also affect the speed and voice as well.
Interestingly, the synth board will work without being plugged into the logic board, it just plays whatever the last pattern was. If I'm feeling brave later I might try hooking the Lunetta up to some of the pins and see if I can make it play itself that way.

[electri-fire]

As can be seen BPM and Tone rotary switches are connected, so interaction is bound to occur.

---

bliptronic_CMOS_ logic_trace_colored.jpg
Description:
Filesize:	130.14 KB
Viewed:	562 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

---

CPU pots marked.jpg
Description:
Filesize:	93.89 KB
Viewed:	340 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[electri-fire]

[slacker]

Thanks for posting those diagrams, that's useful stuff.
I tried just firing random logic signals into the "Column" pins of the synth board and you can make it play itself.
Like electri-fire pointed out though, as the signals were not what it's expecting as well as changing the patterns it also changes the tempo and voice.

I also tried bending the clock using the Lunetta's oscillators with some success. My fastest oscillator only goes up to a couple of hundred KHz though so it just resulted in sonic mayhem, seems to confirm that it can be bent this way though.

[electri-fire]

Memory for bliptronics 5000 switch/LED matrix
I devised some idea of using the switch/LED matrix with external memory, polling the switches, and storing their state. Reading the memory and driving the LEDs has not been addressed yet.

Disclaimer: untested theory, also made to get my head around what's happening. Using parts that may not be the most appropriate, but that I happen to have available. posted for your enjoyment and inspiration. Errors bound to be present..


MM2102A Static RAM (16pin)

Fixed pins:
Pin9= Ground
Pin10 = +5.25V max
Pin13 = LOW: enable

11 = DATA IN (serial in)
12 = DATA OUT (serial out)

pin3 High = read
pin3 LOW= write

5 + 5 address pins ( row x collum):
pins 1 2 4 5 6 7 8 14 15 16

3x3 bits is 8 x 8 bits of memory (64 values)
For convenience: pin 4, 5, 6 for Row and pin 14 15 16 for Collum

(This leaves 4 bits to select 16 additional banks on pins 1, 2, 7, 8 )

xxxxxxxxxxxxxxx
additional parts:

some master clock oscillator for counter
Binary counter like 4040 (6 outputs used to sequentially select memory locations)

4051 8-1 Multiplexer/Demultiplexer to select collum to be "scanned".

a X-OR gate

4 SPDT switches to select one of 16 banks.

xxxxxxxxxx

fixed connections:

4040 binary counter:
pin16 V+
pin 7 Ground
pin 9 (reset): Ground
pin 10 clock in

pin 9 > 4051 pin 11 , 2102 SRAM pin 4
pin 7 > 4051 pin 10 , 2102 SRAM pin 5
pin 6 > 4051 pin 9 , 2102 SRAM pin 6
pin 5 > 2102 SRAM pin 14
pin 3 > 2102 SRAM pin 15
pin 2 > 2102 SRAM pin 16

........................
4051:
pin16 V+
pin 6, 7, 8 Ground
pin 11, 10, 9 is A, B, C binary value from 4040 pin 9 , 7, 6
Inputs: 13,14,15,12,1, 5, 2, 4 to collum pins 1-8 at the Bliptronic header.
pin 3 (output) > NAND gate input a (to toggle current memory state into the opposite value when a switch at the matix is pressed)
....................
NAND gate
in a: from 4051 pin 3
in b: Latched "memory state" bit
NAND output > 2102 SRAM pin11 (input)

...................
Bliptronic 5000 read/write:

remark: U7 is clocked to send power to the switches one row at a time (clock in at U7 pin8 = Bliptronic Header pin 10, as counted at http://www.straytechnologies.com/bliptronic-5000-hacking-full-pin-out-to-
the-64-button-pcb/ . This clock needs to be 1/8 of the master clockspeed into the 4040, so at each Row driven, the 4051 goes through 8 clock pulses to select each collum sequentially.

Therefor the U7 clock is taken from (ehrm..i think ) 4040 output pin 5 QD.
After 8 master clock pulses U7 needs to power the next row of switches, but the serial inputs at pin 1-2 only want a new Hi

step1

2102 SRAM pin12 (serial out (= state)) is the counter clock (to retain (latch) the state value)

state and collum go into a XOR gate, XOR out is write value, so:

state = Hi + collum = Hi : write value = Low
state = Hi + collum = Low: write value = Hi
state = Low + collum = Hi : write value = Hi
state = Low + collum = Low : write value = Low

2102 SRAM pin 3 is Low: write

Ok, done. Next clock.

step2 QB pin 4 is Hi */ Row2 is powered, now read collum 1-8 )

etc., until

step8 QH pin 13 is Hi (now read collum 1-8 )

next U7 selects Row1 again....


EDIT1: memory state and switch condition needed XOR gate (not NAND)

.

[michaeluna]

I was able to find some good bends on the Bliptronic and sync it up to a MIDI signal. Details at getlofi: http://www.getlofi.com/?p=2244

[slacker]

Cool, thanks for sharing those

[johnnymad]

could the sync in/out be used to clock a modular? it seems like it could if it was sending out 5v pulses. has anyone tried this?

[RF]

It can clock my Klee and my Baby 10 sequencers (and presumably MFOS) - kind of...

Only one pulse out for every eight steps - and it must be in "Loop" mode, with a Y adapter into the "Link In" and "Link Out" jacks of the Bliptronic, and the other end to the clock input on the sequencer.

Without The "Y" you get one pulse out and the loop continues - with no more pulses.

bruce

[johnnymad]

thanks for letting me know. i don't know whether i want it or not after hearing that.

[RF]

"Bliptronic 5000-to-Monome Clone conversion kit:

This kit includes Everything you need to convert your ThinkGeek Bliptronic 5000 synthesizer to a fully working “Arduinome” style open-source Monome Clone controller."

http://www.straytechnologies.com/products-page/bliptronome/

http://www.straytechnologies.com/bliptronome-v2-tests-kits-and-code-released/

Not what I was after - but some of you may be interested.

bruce

[slacker]

Saw that the other day, looks quite interesting. I wouldn't have a clue what to do with it though.

[Houseman]

Hi forum.
I am going to buy a bliptronic 5000 to do a Ableton controller but before spending 99 bucks I would liker to know the external dimensions of the box.
Having think big fingers (hee9 is it useful to type on the pad?
How many room I mean height is available in there? all out except the key pad?

Best regards and thanks to all.

Steve