An 8-step sequencer

[Attila]

Hi all!

I'm building right now a polyphonic synth from NE555s. It will have one octave (from A1 to A2). I would like to build a step sequencer for it. My solution is simple: i've got running a CD4022 3bit counter. Each output can trigger all of the 555s. The only problem is that it needs 104 switches. Does anybody know an alternative to this? For example I would have 13 buttons (for the tones) and one step inc/dec button. But how to realise it? Can anybody help?

[kokoon]

can you post a few more details about that polysynth?

[Attila]

Why not? ;)

It consists of 13 oscillators, wich generate square-shaped signals. They are astable multivibrators operating at a high frequency. The heart of an oscillator is the NE555 timer chip, wich operates in astable mode. The frequency (and pulse width) is set by two resistors and a cap by the following eqation: f=1,44/( (R1+2R2)*C ). The IC output is only active, when at pin 4 (reset) is a logical TRUE state. The sequencer will drive those inputs. The teory is from http://www.doctronics.co.uk/555.htm but the idea is mine, and it's of coure "open-source" and "public domain", like Throsten's MIDIBoxes. You can build as many oscillators if you want to increase the number of keys. Or you can use an NE556, wich are two 555s in one chip. It's easy, isn't it? :) I will go with this to a school exhibiton of home-made circuits. The first three winners have to do one subject less on the final exams! Also I will go there with a MIDIBox SID (if the PIC arrives - i'm waiting for 3 weeks!!!)

[machinate]

very cool project - I worked with similar stuff a long time ago when I was into circuit-bending - the "light-theremin" style circuitbends are all about the 555 - they often pop up on ebay. Get Lofi have a write-up on the 555 as well:

http://www.uoguelph.ca/%7Eantoon/gadgets/555/555.html

btw, for anyone even slightly interested in lofi gadgetry the GetLofi site is very nice:)

http://www.getlofi.com/

[Jaicen]

you may want to take a look at the Vanishing Point sequencer design which uses a 4017 decade counter. It's possible to expand the design from 10 steps upwards by chaining 4017's. Very simple project which also includes a random mode. I have the schem and layouts here but no idea where the url for them is. Mail me if you're interested.

[machinate]

http://www.geocities.com/thetonegod/vanpoint2/van_point2.html

[Attila]

I have to use 104 switches for an 8 step sequencer! Why would I expand it? Btw: has anybody an idea how to reduce the number of swithces/buttons from 104???

[machinate]

attila, you might want to read the text in the link - as far as I can tell you set the pitches with 10 pots, nothing else - it's a lot simpler, and probably a better solution than all those switches...

[Skunk]

Any experience about (signal & frequency) stability over temperature range? A 555 is not a very stable device... passive components direct the frequency and behaviour over temp. Try at 0°C and 70°C and look for the results and signal behavior.

Cheers,

Skunk

[moebius]

You guys are SOOO lost! (So am I)

Any experience about (signal & frequency) stability over temperature range? A 555 is not a very stable device... passive components direct the frequency and behaviour over temp. Try at 0°C and 70°C and look for the results and signal behavior.

Just see the link above: http://www.uoguelph.ca/~antoon/gadgets/555/555.html

... and the temperature variation is only 50ppm/°C (0.005%/°C).

Jeezz.. this is a GOOD spec for an analogue component! :o and as always, You can choose passive components in timing circuit to improve stability, nothing new here :)

And about "stability over temperature range" - man, it might be an issue in the industrial environments. For a guy playing a synth this means that you have to turn the tuning knobs when you switch locations from the antarctic to near some volcano.

---

First of all - Attila, You have boldly gone where many people have gone before, but all by yourself. Great! But join the SDIY mailing list NOW..

I have to use 104 switches for an 8 step sequencer! Why would I expand it? Btw: has anybody an idea how to reduce the number of swithces/buttons from 104???

I'm giving you an idea:

Use a couple of small 8bit SRAMs. 52256 should be cheap and easily available.

Use a binary counter connected to address lines to step over address locations: 16bit data input/output patterns available to sequence your 555s. Connect switches to SRAM data lines and while SRAMs are write enabled (another switch WRITE) step over adresses and use switches to set bit pattern. While SRAM is in read mode use counter to step thru the adresses and the same pattern plays back. Of course, you'll need reset circuit for the counter, pull ups/downs etc. But if You study hard enough, I'm sure you can do it!!

Moebius

[Attila]

Pfew! It wasn't easy to understand and it won't be easy to do it. Anyway, I found a cheap method to use: jumpers... as in the pc, but not 2 or 3 in a row, but 13 in 8 rows. In the local shop they sale 40x2 pin DIL (am I right???) connectors for cheap. I'm gonna get a couple of them.

First of all - Attila, You have boldly gone where many people have gone before, but all by yourself. Great!

Thanks! I had only one schematic of a 555 in astable mode and 3 equations: (am I right??? are those equations???)

one for frequency: f=1,44/(R1+2R2)*C

one for HIGH time: t=0,69*(R1+R2)*C

and one for LOW time: t=0,69*R2*C

where R1 is a resistor between Vcc and pin 7, R2 is between pins 6 and 7 and C is between pin 6 and ground.

Also I'm only 16 years old... :) and electronics are my life. I started from nothing (ok, I knew a bit of soldering, when I was 13) and the only person who understood electronics was my dad. He doesn't know all the equations (as above) but he knew it only practically, BUT VERY WELL! (I wasn't shouting ;))

I will discuss this SRAM thing with my friend and decide if it worths for it.

And sorry for my English ;)