Solar Music Box

Inventor · Last edited by Inventor on Mon Apr 02, 2012 11:55 pm; edited 1 time in total

Now that i am nearly done with the first Thermaview prototype, I am looking forward to the next project. I have practically fallen in love with this little company at www.sparkfun.com, which has the coolest parts at the best prices, so most of the parts will be from them.

I'm using these excellent little clear plastic cases now for my projects (gasp! yes, I finally learned to put stuff in boxes!), see attached pdf for engineering drawings of the box. I will run the power from a solar cell they sell that fits almost precisely within the box and use CDS cells (LDRs) for all controls. That way I accomplish my goal of having no batteries and no wallwarts to worry about. Plus the solar cell can be covered by the hands to do suply starving of the circuit. The solar cell is specified at 4.5Voc and 100mAcc, so the circuit will have plenty of juice even in low lighting conditions but the voltage will be on the low side.

The top of the box will hold all the CDS cells and the solar cell, and will connect by ribbon cable to the bottom of the box where the circuit board (designed to fit the box) and two audio jacks exist. The audio jacks take line-in and deliver line-out on 3.5mm stereo lines. The line-in is for any audio input, but specifically to come from the line-out of my computer so that my fancy Blue Yet microphone and ChucK pre/post processing can get in on the action if desired. However teh box will also act stand-alone.

I like that stand-alone idea such that I can bring this box and nothing else to an electro-music festival and do an full hour or half hour performance with it.

Oh, and the design will be BSKS2, which stands for Boolean Sequenced Karplus Strong Duo. Note that this is different from a BSKSKS, which has a BS feeding a KS feeding a KS. I am not entirely sure what a KS2 will be exactly like, but i believe it will be a KS with two feedback paths and only one summing amp.

That should make for a most interesting project, and I have just ordered the first $77.78 worth of parts for it! (well i already have the case.) I'll keep you posted.

Les

SFE_Box-1.pdf

Description:

The sexiest little project box in town!

Download (listen)

Filename:

SFE_Box-1.pdf

Filesize:

128.25 KB

Downloaded:

1288 Time(s)

Inventor · Posted: Fri Mar 30, 2012 10:43 pm Post subject:

Here is the BS section that I just dreamed up with 25 CDS cells. It's a take on the linear cmos logic but with a 12 bit parity generator chip as the logic and CDS cells at both feedback and input positions.

This circuit is cool because it has that groovy XOR based sequencer logic, plus it's all analog with feedback and input signals controlled by the CDS cells, plus it's only two chips (well, need a tempo oscillator too). I think I may change it a bit because it's still too much drive current for the TLO74, so i might use the full quad with three outputs each - wild huh?

Should sound really freaky!

Les

BSKS2 BS Section with parity analogic.jpg

Description:

analog logic (analogic?) Boolean Sequencer with parity core

Filesize:

369.55 KB

Viewed:

1279 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Inventor · Posted: Sat Mar 31, 2012 12:32 am Post subject:

See this page:

http://www.sparkfun.com/products/10917

for a fascinating concept in speaker technology, again available at www.sparkfun.com. It is a transducer that acts like a speaker with no cone. You just mount it on an object and press the other side against something and the resulting vibrations will cause a sound. In other words it turns anything you put it on into a speaker.

I plan to use it on this BSKS2 project by attaching it to the circuit board and the other side to the enclosure top. That way the case itself will be the speaker and i'll pick up the audio with a microphone.

This removes the 3.5: jack and associated wire, which also does something amazing: it makes the entire project a sealed-in-the-case device. The power will come from the solar cell which is mounted to the bottom of the circuit board, and the CDS cells (LDRs) will be mounted inside the case also. So the thing will be fully self-contained and will operate non-stop (put it in a dark box to shut it up).

Wow, how cool is that? This is turning out to be a very fun design!

Les

Inventor · Posted: Sat Mar 31, 2012 3:11 am Post subject:

Well here i have been up all night fueled by energy drinks and I have come up with another wacky design, shown below. I'm not sure if it will work but I can breadboard it and see once I gather up all the parts. I'm not sure if there is a correct inversion structure or if the combination of all inputs to one output in the parity generator is going to work, but the idea is as follows.

The four twin-tee filters will have component values selected to produce not a bandpass or notch filter, but a wideband filter result. The idea there is that they can produce tones within a range, and four separate bands can be selected by component value choice, covering the audio range of the output transducer. These filters combine with the parity chip to form an XNOR based linear CMOS amplifier.

If it is stable and works like i think it will, it should certainly make some unique sounds! Design critiques anyone?

Les

BSKS2 twinn tee design.jpg

Description:

BS with analog logic section including four twin-tee filters

Filesize:

325.02 KB

Viewed:

1220 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

drapdap · Posted: Sat Mar 31, 2012 6:16 am Post subject:

Inventor · Posted: Sat Mar 31, 2012 9:52 am Post subject:

drapdap · Posted: Sat Mar 31, 2012 10:30 am Post subject:

Ah, ok, i see it now. Please post a bigger if you get on with this, i like to read schematics. Very Happy

Inventor · Posted: Mon Apr 02, 2012 11:58 pm Post subject:

ask and ye shall receive, drapdap! Here are close-ups of the current, simplified design. I found some problems with my previous design so I scrapped it and created this simpler one. The design works as follows:

1. Relaxation Oscillator
This is a classic square wave oscillator whose frequency is controlled by the light incident on the CDS cell. The tempo will increase with decreasing light.

2. Boolean Sequencer
I was very excited to discover the CD4531 12 bit parity generator IC as it is the perfect companion to the CD4040 12 bit binary counter, generating an XOR sequence on a 12 bit count with only two chips. I have tested the XOR of all bits in ChucK and it makes a pleasantly musical sequence. Nice!

3. Analog Instrument
The concept here is that it's a twin-T network extended into a ladder network. I don't know what the frequency response will be but it should be periodic in some sense, yet controlled by the CDS cells, it will change with the incident light. I will mount the CDS cells close to each other so they have similar values for proper periiodic response.

4. The Little Gem
This is the simplest of the LM386 speaker drivers. The speaker it drives is a special cone-less device that turns whatever you mount it to into a speaker, in this case - the case!

As mentioned previously due to the solar power cell, CDS cells for control, and special speaker, the entire music box can be mounted in a clear Sparkfun case and selaed to become a stand-alone fully solar music box.

Critiques and suggestions are very welcome!

Les

Relaxation Oscillator.jpg

Description:

Filesize:

103.75 KB

Viewed:

1170 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Boolean Sequencer.jpg

Description:

Filesize:

179.67 KB

Viewed:

1129 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Analog Instrument.jpg

Description:

Filesize:

193.09 KB

Viewed:

1141 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

The Little Gem.jpg

Description:

Filesize:

111.26 KB

Viewed:

1174 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Inventor · Posted: Wed Apr 18, 2012 8:56 pm Post subject:

OK folks, here is the constructed prototype minus the output stage. I have made a few design changes, but first let's talk about the physical construction.

The box is a Sparkfun clear project case, which i truly enjoy working with (on two projects now). The red board is a printed through-hole prototyping board which mounts in the case with perfect alignment, dividing it into a top and bottom section. I am not using the board to solder parts onto, but rather as a structural element. Adhering to the top of the board is a small breadboard that fits nicely with chips, resistors, and wires in it (no caps in inventory yet). On the flip side (not shown) is just barely enough room for a 4.5V solar cell designed for Sparkfun.

So in the physical situation we have a nice tightly packed arrangement showcasing a breadboard with chips running down the middle and the noticeable white wires on the sides. The use of the breadboard instead of soldering into the protoboard allows me to easily make design changes, which I have done a few times.

Electrically we have power from the solar cell with intentionally zero power conditioning circuitry such as bypass caps (might add those later) for maximum weirdness in the sound. The 8 pin opamp chip at the top has a relaxation oscillator with LDR (CdS cell) frequency control on the right side and an incompleted output amp on the left (more later on that). Then we have my old friend the CD4040 counter with a 12 bit output + clock in is 13 bits of counter bus, or 8192 step sequence. The third chip (on the bottom of the photo) is a CD4531 parity encoder chip which is pretty much just a giant 12 bit XOR dealie with a separate XOR input for odd/even parity.

Rather than wire the parity chip as shown above in the schematic, with bit 0 goint to bit 0 and so on, i put bit 12 (the MSB, or most significant bit) on the even/odd parity control and assigned all the other bits including clock to the remaining 12 inputs according to physical proximity on the board. This makes the board neat and clean instead of the tangled wire mess it would have been, and also makes for a different sound than the pure parity would. This is because there are upper bits in lower places and lower bits in upper places, resulting in a scramble of the even/odd parity function which affects only the upper 1/3 of the bits as near as I can tell from the schematic.

Anyway so we have a boolean sequencer with light-dependet frequency control (brighter slower, dimmer faster) and a funky XOR output of one bit from all 13 input bits. This will go into the modified twin-tee bass drum analog section powered by the left half of the opamp chip. More on the twin-tee modification in the next post since this post is getting long.

Finally, I have decided not to use the speaker transducer thingie to vibrate the case for sound, and instead to have a 3.5mm (1/8") jack for headphone, computer, or mixer output. That's it so far, read on for info about the analog section.

Les

Solar Music Box 2012-04-18 at 22.32.jpg

Description:

Gasp! It's in an enclosure!

Filesize:

73.81 KB

Viewed:

1138 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Inventor · Posted: Wed Apr 18, 2012 9:05 pm Post subject:

You know you've got a pretty good stronghold of experience in a particular category of knowledge when you can imagineer something in that realm and then find it in the collection of existing knowledge in the world according to Google. That's why I was pleasantly surprised to find the exact circuit that I thought up in the below image in Google Images. It's got a comment about not being the best of it's kind for the Q control practical issue, but I'm willing to take that chance and we'll see what happens. Anyway patting my own back there, sorry.

Anyway, I'm modifying that circuit to include three tees squooshed together to create a multi-resonant structure that will create multiple voices of drums depending on the frequency of the stimulation at the time. When the music changes a lot it will trigger the upper drum harmonics and slower, lower ones if you get my meaning.

My friend just walked in the door and we are going to record some Gospel Rap, so I'll have to get back to a more thorough discussion of the circuit later, but suffice to say it should be a drum cacophany of music, I hope!

More to follow,

Les

active twin tee bandpass filter.gif

Description:

Active Twin Tee Filter image from Google Images

Filesize:

14.11 KB

Viewed:

1123 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

elmegil · Posted: Wed Apr 18, 2012 9:23 pm Post subject:

pretty cool looking Smile

Inventor · Posted: Wed Apr 18, 2012 10:23 pm Post subject:

Ty elmegil, I'm having fun to boot! The circuit is quite interesting as well, to me anyway.

In the illustration below we have the analog section detailed. I have made a generalization of the twin-tee circuit shown two posts ago, making it into a ladder network or more accurately a pair of ladder networks combined together. in a twin-tee fashion.

The idea here is that the ladder networks will form a notch filter with multiple notches due to the multiple resonant points that are formed by the network. For example, we should have resonances at RC, RRC, and RRRC at least, and probably other cross-resonances if you will permit that term.

In this way the circuit is not just a drum, but multiple drums of different sizes related by harmonic intervals, all being struck at the same time. It should sound rich, full, and unique.

That is, if the thing doesn't self-oscillate because of runaway Q, we shall see.

Les

Multi-Drum Amplifier.jpg

Description:

The weirdness continues with this Multi-Drum Amplifier
Based on the Twin-Tee amplifier shown above

Filesize:

172.18 KB

Viewed:

968 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Inventor · Posted: Wed Apr 18, 2012 10:28 pm Post subject:

Here in the illustration below we have the whole circuit including extension to three counter chips with three logic elements. This way the circuit does not repeat it's sequence for well over 100 years of music playback, wow!

In the prototype, however we have only room for just the one counter and logic chip pair. The rest of the circuit is previously described.

I am ordering some capacitors tonight and an audio jack for completing this project, and will do so when they get here.

Les

Solar Music Box Schematic.jpg

Description:

The whole shebang, showing three BS logic elements instead of one

Filesize:

136.21 KB

Viewed:

1033 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

wmonk · Posted: Thu Apr 19, 2012 8:52 am Post subject:

Way cool Les!

Bbbuuut where did you leave the beard? Shocked

RingMad · Posted: Thu Apr 19, 2012 3:46 pm Post subject:

Inventor · Posted: Thu Apr 19, 2012 4:36 pm Post subject:

Inventor · Posted: Thu Apr 19, 2012 4:37 pm Post subject:

Inventor · Posted: Sun Jun 10, 2012 2:27 am Post subject:

I finally finished wiring up the circuit and tested it at night by holding it up to a light bulb (incandescent), and guess what? It works! well, I just heard some thumping bass-drum-like sounds but that's enough to verify that at least something is working right.

The circuit does not have the LM386 driver, instead I chose to send the output to an 1/8" stereo jack directly from an opamp. Since the opamp can drive 25mA, it does a good job delivering the signal to my 64 Ohm Sennheiser Headphones. So at sunrise I'll be walking around the place with my headphones on carrying this little box in my hand and wearing a big smile on my face!

In marginal lighting conditions there is a slight buzz/hum type sound which quickly changes into a soft thumping sound when the light becomes sufficient to power the device. I can't wait to try it in bright sunlight because there will be better control of frequency and offset for varying sounds.

Unfortunately the design has only two output levels, on and off, however I can easily improve that with the addition of more CDS cells from the binary counter output to the summing node. I'll play around with that tomorrow. How exciting!

Les

Inventor · Posted: Sun Jun 10, 2012 3:49 am Post subject:

Here is a photo of the completed Solar Music Box.

Les

Solar Music Box 2012-06-10 at 05.40.jpg

Description:

Solar Music Box

Filesize:

70.94 KB

Viewed:

937 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Inventor · Posted: Sun Jun 10, 2012 11:18 am Post subject:

I played around with the device and modified the circuit somewhat today. It definitely needs the light in the shade of a sunny day to operate. It's kind of like an optical Cracklebox in a way, though not quite as harsh. I can get sounds like motorcycle engine revving in direct sunlight and birds chirping in low light condictions. I will record a jam session with it tomorrow when the light is in my room by the computer through the window.

Les

Inventor · Posted: Mon Jun 11, 2012 8:26 am Post subject:

Here is an audio sample that I recorded this morning in a sunny window. There is a lot more of a range of sounds in full sunlight, but this is what it can do under window conditions.

Les

Solar Music Box Audio Sample.mp3

Description:

In a sunny window

Filesize:

2.22 MB

Duration:

2:25

Viewed:

2705 Time(s)

Inventor · Posted: Tue Jun 12, 2012 2:32 am Post subject:

Since the circuit is on a breadboard, i thought i would consider alternate designs. It would be just a day's work to gut the box and rewire it to another design. The attached audio sample is from another proposed design that actually I prefer. I'll leave it up to the customer to say which one they want.

Customer? yes, it's for sale.

Les

Circuit 2 Audio Sample.mp3

Description:

an alternate circuit's sound

Filesize:

2.06 MB

Duration:

2:15

Viewed:

2675 Time(s)

Inventor · Posted: Wed Jun 20, 2012 1:21 am Post subject:

Well, after 3 tries I now have a circuit that I'm happy with and will stick with as my first completed music box. It's got a nice bass rhythm to it which I like. The attached audio sample has a voiceover of me playing with it for almost 20 minutes. Unfotunately the right audio channel is silent for some reason, I think a loose wire. I'll fix that soon. But I wanted to get this audio sample posted to you now so that you can hear it.

The cost to me including shipping of the box is right around USD$50 including case, solar cell, breadboard, and everything. I'd be willing to sell it for USD$100 + shipping so that I could buy parts for two more boxes in exchange for all my time and effort in developing and assembling the music box.

I'll work up a schematic soon and post it.

Les

Solar Music Box Audio Sample Circuit 3 20120620 0243.mp3

Description:

music box with me talking about it as i control it

Filesize:

16.97 MB

Duration:

18:32

Viewed:

2670 Time(s)

RingMad · Posted: Thu Jun 21, 2012 5:50 am Post subject:

Nice bass rhythms indeed! I could not tell from the photos where the LDRs are located, but a fun thing to do with such things in a darkened room is to have each one looking at a candle of its own, and the random flickering and light changes as it burns down makes changes in the sounds. Distance from the candle to the LDR is also a factor. If one uses the "LED tea candles" found in dollar stores, the flickering is generally faster.

Candles worked well with my Triowaverator (triple oscillator sine/triangle dronebox), and perhaps could produce nice results with this box.

James.

Inventor · Posted: Thu Jun 21, 2012 7:37 pm Post subject:

Candles and other flickering light sources - what a great idea! I am realizing as I play with this new toy that I shall have to build and/or obtain a collection of light sources as I go down this path... Nice ideas!

Les