MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Circle Machine
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author Circle Machine
davebr
Has anyone ever thought of making a Raymond Scott Circle Machine? I've done a Hammond Vibrato Scanner module and I always thought a Circle Machine would be another fun gadget. I guess it's the thought of a whirling arm and lights that I think would be interesting.

I'm going to try and put one together and I thought I would use a stepper motor so I can rapidly step to each lamp. I haven't decided if I want to use lamps or LEDs yet. I'm leaning towards the lamps to be closer to the original design. I think lamps may help the crossfade between steps, although with a stepper motor that may not matter. It will also be much less work to use lamps although it will take much higher power.

I'm also thinking about whether I want to build this on a panel for my modular or as a stand-alone design. I'm leaning towards a stand-alone design because of the size, heat, and power.

I've got the stepper motor drive electronics built and I've written code for my ComputerVoltageSource to generate the appropriate stepper motor clocks. I have CV control of the rate and number of steps. Later, I can always port this development code to an AVR processor for the final design.

I usually don't post my projects until complete but this one is going to take some time to develop. I started a web page for my progress on my circle machine project.

I've got a schematic of the stepper motor drive, some oscilloscope timing images, and a video of an eight step sequence for a 16 position circle machine (e.g. the last step is 180 degrees). Unlike Raymond's original design, I will be able to set the number of steps in the sequence.

Dave
fluxmonkey
i've often thought abt it, tho in a lower-tech way than your approach... will be watching your progress with interest!

b
SynthBaron
Awesome DIY vibrato scanner.

A circle machine would be awesome. Maybe the oscillator, filter, and shaper modules from Scott's synthesizer too. That thing sounded great. I wonder if there's any documentation on it left at all, though.
wetterberg
I've often considered a modern variation of this... but my brain asplodes every time... here's why, perhaps you can help me solve the conundrum.

Say you place an LED or an LDR on a servo motor and control it to spin in a circle; how would you then connect this LED to the rest of the electronics? Anything with a wire would spin and get fucked up, basically.

Anyone?
rafe127
wetterberg wrote:
I've often considered a modern variation of this... but my brain asplodes every time... here's why, perhaps you can help me solve the conundrum.

Say you place an LED or an LDR on a servo motor and control it to spin in a circle; how would you then connect this LED to the rest of the electronics? Anything with a wire would spin and get fucked up, basically.

Anyone?



http://en.wikipedia.org/wiki/Slip_ring
wetterberg
sure, but that typically takes us well beyond what's easily DIY-able... unless there are simple implementations of this that I've missed.
neandrewthal
Cool stuff!

I don't think I could tackle such a crazy project, but I'm interested in your motor stepper circuit. I've always wondered if it's possible to use such a circuit to create some sort of master sequencer controller that can give the coveted 960 "skip" feature to any old sequencer with only a clock input.

For example, if you set it to skip step 7, when it reaches that step it puts puts out 2 pulses microseconds apart and skips to step 8 so fast you don't even notice it was on 7.

Does this sound like something that might work?
rafe127
wetterberg wrote:
sure, but that typically takes us well beyond what's easily DIY-able... unless there are simple implementations of this that I've missed.


I didn't know you wanted EASY diy... razz

I certainly don't know of any solutions that do not involve something like a slip ring...

However, I think it might be within the realm of possibility to use something like this:

http://www.hurricanewindpower.com/servlet/the-145/wind-turbine-wind-ge nerator/Detail
daverj
There's a variation of this out there with a printed disc on a turntable and a photocell over the disc getting the reflected light. It's not the same in that the discs have to be made and printed with your pattern rathing that turning knobs to change the pattern in real time. But the discs contain many patterns and you can move the sensor to play different ones or slide between them.

As for slip rings, there should be some available in surplus outlets. New ones start at about $100. I've used them in a number of devices that I've built for artists to pass power or video and audio signals.

They can be built DIY style with a couple of washers and nylon spacers held together with a nylon bolt. Then use some bristles from a wire brush as the contacts. Probably easier to find them surplus.

A quick check on ebay turns up this one:

http://cgi.ebay.com/Wind-Turbine-Generator-Slip-Ring-No-Twist-BEST-PRI CE-/250460995182?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item3a50a37e6 e

I know nothing about this specific device or that seller, but just saw it with a quick search. Might work.

Another method would be to mount a mirror on the end of an arm at 45 degrees and another one over the motor shaft at 45 degrees to point the light at a stationary photo cell above the motor shaft.
Just me
wetterberg wrote:
I've often considered a modern variation of this... but my brain asplodes every time... here's why, perhaps you can help me solve the conundrum.

Say you place an LED or an LDR on a servo motor and control it to spin in a circle; how would you then connect this LED to the rest of the electronics? Anything with a wire would spin and get fucked up, basically.

Anyone?


Mercoids and/or Hall Effect devices The arm whirls a magnet or reluctor and the hall effect device 'reads' it. Mercoids make good rotating contacts.
Rod Serling Fan Club
on a somewhat related note. I always thought it would be cool to see some sort of dekatron module. The power issues would probably be a problem though.

Luka
i heard that his scott's sequencer was so mechanically loud that he never new what the sequence sounded like until he replayed the recording
davebr
neandrewthal wrote:
Cool stuff!

I don't think I could tackle such a crazy project, but I'm interested in your motor stepper circuit. I've always wondered if it's possible to use such a circuit to create some sort of master sequencer controller that can give the coveted 960 "skip" feature to any old sequencer with only a clock input.

For example, if you set it to skip step 7, when it reaches that step it puts puts out 2 pulses microseconds apart and skips to step 8 so fast you don't even notice it was on 7.

Does this sound like something that might work?

I'm generating the clock pulses with my ComputerVoltageSource. It would be quite easy to generate that type of clock pulse with it. I have been able to write some great programs to connect this-to-that ... like your sequencer example, MIDI to analog clocks and visa versa, drum machine interface, quantizers, analog shift registers, and emulation of other control modules such as the MOTM-730 and the JAG.

I designed the ComputerVoltageSource to be all open source. So far Bridechamber organized one group purchase of about 30 boards and I only know of a couple that have ever been built. I think most people are turn off thinking that programing is quite difficult. It's programmed in Basic which is pretty simple and then downloaded via RS-232. Besides, I've created a program template that has most of the support functions already done - you just add your unique code. The clock generator for the stepper motor took about 30 lines of code and most of that was using two CV inputs to control the rate, number of steps, and displaying the steps in the LCD.

For a sequencer clock the code is simply ...
output high
pause xx uS for the width of the clock
output low

This code sits inside of the code that generates the time delay between clocks. It is very simple to generate a double clock on a particular step.

You might want to add some type of synchronization between the sequencer and the ComputerVoltageSource if there is a reset input or start / stop input to the sequencer.

Then to get fancier, you can use the LCD to display sequence information. This still would be a very simple program. It's just another example of a this-to-that program that I find the ComputerVoltageSource so valuable for. I seem to be the only one writing programs or at least making them public on the web (there were about 45 PSIM modules sold which run compatible programs).

Dave
andrewF
Rod Serling Fan Club wrote:
on a somewhat related note. I always thought it would be cool to see some sort of dekatron module. The power issues would probably be a problem though.



like




more info

the power isn't too hard, needs around 700V, just a voltage tripler did the job.
The hard part with dekatrons is getting them to run reliably on a wide ranging clock frequency, i copped out and used a relay to switch a 120V 50Hz signal on & off. It works fine but is still a finicky module to use.
andrewF
Thanks for posting your work Dave
I have long been keen on the circle machine, did a simple prototype with a record player and a dr-110 drum machine. It works but needs a lot of development.

i will try your stepper motor design for starters thumbs up
Rod Serling Fan Club
andrewF wrote:


like




more info

the power isn't too hard, needs around 700V, just a voltage tripler did the job.
The hard part with dekatrons is getting them to run reliably on a wide ranging clock frequency, i copped out and used a relay to switch a 120V 50Hz signal on & off. It works fine but is still a finicky module to use.


I'll have to check your module out when I get home, that stuff is blocked for whatever reason. I remember seeing a dekatron sequencer floating around but it seemed as though the dekatron might was just running off the voltage and had little to do with the sequencer. I mean, yeah ultimately your using a dekatron for looks more than any kind of usefulness, but they are cool. I'd love to see modules with dekatrons and nixies.
Rod Serling Fan Club
Yeah I think it was your videos I saw. Does the dekatron actually follow the sequencer in a meaningful way or is it just spinning off the mains voltage?
andrewF
yes it follows.
The clock signal closes the relay so the dekatron receives the 120VAC 50Hz (coming from a 240-120 stepdown transformer)
The trick is to feed the relay a clock pulse wide enough for it to close briefly so the dekatron gets just one AC pulse and steps once. If the relay is closed for too long the dekatron is basically in random mode as it receives an indeterminate number of pulses, not necessarily a bad thing.
I use Judd's Bell Gate to condition whatever clock signal is being used so the dekatron sequencer steps properly.

On the whole, as is, it is okay but not a great module and I am considering to build another dekatron module that will be a trigger source/pulse divider driving a relay based multiplexor/sequencer.
otherunicorn
Slip rings:
You can make them "flat" on a piece of PCB, as long as the tracks are wide enough to allow a brush to run on them without hitting your solder glob! You can turn them up in an electric drill.

In most old VCRs there is a position sensor for the mech. It is a flat PCB with a set of wire brushes resting on it. You can salvage those brushes.

In the old days, people used to fly electric tethered model air craft. They used plastic for the shaft, and ball races as the slip ring. Fixed connection to the center, moving connection to the outer. You can't solder these though. Plenty of small ball races can be found in hard drives etc..

Basically, if you want to desperately enough, it really isn't that hard to achieve.
davebr
I've adapted a Hammond Vibrato Scanner as my base mechanism for the Circle Machine. I now need to find some lamps and bases that are reasonably priced. I'm thinking of using T-1 3/4 lamps with either bi-pin or midget flanges. So far the bases I've found are quite expensive.

Does anyone know of a good source of lamps and bases? The mounting hole is approximately 3/8"

Updated pictures of the mechanism are on my circle machine web page.

Thanks. - Dave
Luka
there are some tube radio part stores on ebay that stock lamp and bases. I got my panel mount lamp base for my krautrock phaser from one
davebr
IT WORKS!! (in concept, anyway).

I chose #2174 lamps which are wire base. The maximum power is 170 mW so I can use 500R 1/2 watt linear potentiometers as rheostats to control brightness (I'd like to keep it true to the original which I'm sure used rheostats).

I mounted 6 lamps and the photoresistor so I could give it a test run. It works reasonably well. The arm has enough mass that I have to use a slower clock for the stepper motor (or I need a more powerful stepper motor). The slower rate works fine except for the longer return-to-home on short sequences. I decided to limit the sequence length to 4, 8, or 16 steps by skipping lamps. A 4 step sequence skips 3 lamps and an 8 step sequence skips 1 lamp. This keeps the sequence timing even and simple.

All my metal surfaces are reflective so there is too much ambient light. I will need to paint the assembly and the bottom vane where the photoresistor mounts flat black. I may have to put a shroud over the photoresistor as well.

I am using a voltage divider with the photoresistor to ground to generate an output voltage. The slip-disc works great and does not generate any noise or variation. There is a slight wobble to the arm when I stop that varies the output voltage. A slight lag cleans up the output voltage nicely. I generate a control voltage for my VCO by scaling the output voltage, adding an offset, and running it through a lag module.

I put a video of a 6 step sequence on my circle machine web page.

This project is fun - lots of mechanics to it.

Dave

[/img]

davebr
I mounted the scanner base on a 10"x10" base with 16 potentiometers around it. I still only have 8 lights installed.

I decided to do software processing on the sensor output since it is easier to prototype and experiment. I simply sample the sensor output with my ComputerVoltageSource. I'm using it for the clock generation so it is just as easy to process the sensor output as well. I inverted the sensor output since I want lamp brightness to increase the output control voltage, and scaled and quantized the output as well. I added a 24 note offset control so I can scale the output control voltage by +/- 12 semitones.

It works pretty good. You certainly have to tune by ear as there is slight variation in brightness from lamp to lamp.

My key issue now is the slowness of the CdS sensor. It slightly detunes at the highest rate. Of course you can simply tweak it to the right setting while operating.

I put another video, pictures, and my learning so far on my circle machine web page.

Dave
davebr
I've more or less finished my Circle Machine project. I finished the 5U panel and built a desktop enclosure for it. It's fun to operate. I've put a video of it in operation on my Circle Machine web page.

I vary the sequence length to 4, 8, or 16 steps by not reading the lamps in-between. This has the effect of changing the tempo by 1/2 or 1/4. Using a control voltage to vary the sequence length in operation has an interesting effect of changing the sequence tempo. In the video I am simply varying the sequence length with a sine LFO.

It was a fun project. - Dave

davebr
I decided to install the Circle Machine vertically in my modular synthesizer. I had made the front panel 5U x 5U so it would fit the standard cabinets. I used one of my spare portable cabinets and installed my PSIM, my PSIM LCD display, and my Circle Machine so I would have a self-contained sequencer. The PSIM is code-compatible with my ComputerVoltageSource modules so I just had to modify the software for the smaller display. Now I have it as part of my modular synthesizer left wing.


Dave
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group