Dual CD4046 Cracklebox

[Repeater]

My tribute to the Cracklebox by Michel Waisvisz and Geert Hamelberg.

Obviously this one's a little different because it uses a pair of PLL chips instead of an opamp, but in spite of different circuitry there are some commonalities:

(1) It abuses cheap and readily available ICs via the simplest and perhaps stupidest means possible: directly poking IC pins. (2) The LM709 has external frequency compensation pins and the CD4046 has its own sort of bandwidth limiting via resistors (or fingers in this case) at pins 11 and 12. (3) Like the original Cracklebox, this circuit is pretty much silent unless prodded. (4) It makes squeals and fart noises that are unique to each player.

There are only FOUR (4) components needed for the barebones version and it costs about one US dollar to make. With some additional circuitry, it can be the heart of a chaotic gesture-controlled synthesizer system. I hope that you enjoy it.

https://soundcloud.com/martin-freem/dua ... cracklebox

[basicbasic]

That's awesome. Seems easy to put into a Euro panel too. I love simple circuits like these!

[ClausF]

Great, as always

[Moos Gummi]

Is there a reason this couldn´t be a single, feeding back to itself? Maybe with a little filter?

[Repeater]

Is there a reason this couldn´t be a single, feeding back to itself? Maybe with a little filter?

Yes.

The additional $.50 in parts buys you stereo operation and opportunities for cross-modulation.

[NS4W]

I'll be adding this to a newbie soldering workshop - should be fun! Thanks for sharing

[mskala]

(1) It abuses cheap and readily available ICs via the simplest and perhaps stupidest means possible: directly poking IC pins.

Is this going to be a static-discharge problem?

[Repeater]

(1) It abuses cheap and readily available ICs via the simplest and perhaps stupidest means possible: directly poking IC pins.

Is this going to be a static-discharge problem?

I've never had problems with modern CMOS and static discharge and I've working with similar circuits for a few years. YMMV of course.

[glennfin]

I'd use sockets for those IC's. There's a good chance that repeatedly touching unprotected floating pins/connections on a Cmos device will damage the device.

(1) It abuses cheap and readily available ICs via the simplest and perhaps stupidest means possible: directly poking IC pins.

Is this going to be a static-discharge problem?

[alka]

Thank you for sharing!

[FetidEye]

yes, thanks! Lunetta stuff is cool!

[Repeater]

There's a good chance that repeatedly touching unprotected floating pins/connections on a Cmos device will damage the device.

Generally that's the conventional wisdom, but this is a special case...

The only "floating" input in this circuit is the Signal Input (pin 14)... but it's not a digital logic gate! There's a linear amplifier inside that tries to amplify whatever you plug into it up to logic levels to drive the phase comparators. Generally, most people would want to amplify signals externally instead of using this feature of the CD4046. Don Lancaster writes about this in the CMOS Cookbook: "Linear operation causes extra amplitude-variation sensitivity, jitter, tearing and generally poor noise immunity." I believe that this irregularity is the source of much of the interesting behavior in this circuit.

Anyway, you don't need a pull-down or pull-up resistor there.

[emmaker]

CMOS does not like static and can blow up really easy when zapped.

I'd put a low value resistor (50-100 ohms) in series with the touch pads along with protection diodes to the power rails used by the chips.

[mskala]

CMOS does not like static and can blow up really easy when zapped.

I'd put a low value resistor (50-100 ohms) in series with the touch pads along with protection diodes to the power rails used by the chips.

If, as Repeater says, these chips are able to withstand static anyway, then it doesn't matter. But if static were a serious issue, then a series resistor that size would be no protection at all. Static charges are in the range of a few thousand volts. A 100 Ohm resistor would pass a few tens of amperes for as long as the charge lasted, and that's far more than the chip can withstand. In order to be useful for static protection, a series resistor would have to be in the hundreds of kOhm, bringing the maximum current down to what the chip's internal diodes can support. Maybe a little less if you're adding diodes of your own, and you trust them to switch faster and have less voltage drop than the chip's own.

[emmaker]

along with protection diodes to the power rails used by the chips

If, as Repeater says, these chips are able to withstand static anyway, then it doesn't matter. But if static were a serious issue, then a series resistor that size would be no protection at all. Static charges are in the range of a few thousand volts. A 100 Ohm resistor would pass a few tens of amperes for as long as the charge lasted, and that's far more than the chip can withstand. In order to be useful for static protection, a series resistor would have to be in the hundreds of kOhm, bringing the maximum current down to what the chip's internal diodes can support. Maybe a little less if you're adding diodes of your own, and you trust them to switch faster and have less voltage drop than the chip's own.

The resistor is not the protection part, the diodes are. The resistor just current limits what goes to the diodes.

Just because they say the chip is 'protected' that doesn't mean that it's a good idea to not protect the inputs. In some cases the chip degrades slightly each time it gets zapped but still works until it gets zapped that 100th (or whatever) time. Also from the On Semi datasheet there is the however clause.

This device contains protection circuitry to guard
against damage due to high static voltages or electric
fields. However, precautions must be taken to avoid applications
of any voltage higher than maximum rated
voltages to this high−impedance circuit. For proper operation,
Vin and Vout should be constrained to the range
VSS ≤ (Vin or Vout) ≤ VDD.
Unused inputs must always be tied to an appropriate
logic voltage level (e.g., either VSS or VDD). Unused outputs
must be left open.

[mskala]

along with protection diodes to the power rails used by the chips

If, as Repeater says, these chips are able to withstand static anyway, then it doesn't matter. But if static were a serious issue, then a series resistor that size would be no protection at all. A 100 Ohm resistor would pass a few tens of amperes for as long as the charge lasted, and that's far more than the chip can withstand.

The resistor is not the protection part, the diodes are. The resistor just current limits what goes to the diodes.

Indeed, but that's the point - because the voltages are so high, a resistor that size does not meaningfully limit the current. It would have to have a very much larger resistance before it would make any difference compared to just the diodes alone.

[glennfin]

"transorbs" are your friend.

[glennfin]

I count 10 unconnected/floating pins... they don't have to be "inputs" to be susceptible to an ESD event.

There's a good chance that repeatedly touching unprotected floating pins/connections on a Cmos device will damage the device.

Generally that's the conventional wisdom, but this is a special case...

The only "floating" input in this circuit is the Signal Input (pin 14)... but it's not a digital logic gate! There's a linear amplifier inside that tries to amplify whatever you plug into it up to logic levels to drive the phase comparators. Generally, most people would want to amplify signals externally instead of using this feature of the CD4046. Don Lancaster writes about this in the CMOS Cookbook: "Linear operation causes extra amplitude-variation sensitivity, jitter, tearing and generally poor noise immunity." I believe that this irregularity is the source of much of the interesting behavior in this circuit.

Anyway, you don't need a pull-down or pull-up resistor there.

[ehafh]

thanks for sharing, this looks neat.
might have to build one or two!

[igormpc]

The additional $.50 in parts buys you stereo operation and opportunities for cross-modulation.

how can i do it?

[Repeater]

Let's get empirical. I wonder how many people who are certain that this circuit will fail have actually built one and confirmed their suspicions. So far, there have been four successful builds and no CD4046s harmed. My original unit has been working for 2 months with almost daily usage. I'll update this post and keep a tally as long as people respond.

If you have the time to take 5 minutes to post, you have time to breadboard a circuit with 4 components, right? Perhaps we'll all learn something.

If not, this is a $1 circuit. Thanks for playing!

[Repeater]

The additional $.50 in parts buys you stereo operation and opportunities for cross-modulation.

how can i do it?

The schematic in the first post has two outputs. Pan them left and right or use a pair of amps and you're good to go.

[mskala]

Let's get empirical. I wonder how many people who are certain that this circuit will fail

Don't put words in my mouth. I asked a question. Nobody here is "certain this will fail."

Riding a motorcycle without a helmet is something many people have done without problems, too. That doesn't make it a good idea. And encouraging newbies in particular to do it is likely to draw some concern from experienced motorcyclists who are still alive.

You posted a design that does not conform to industry-standard ESD practices. You say it hasn't failed for you, yet. I'm sure that is true. But it doesn't mean you're exempt from reasonable questions about its reliability.

At least you're not selling them. (EDIT: well, in 2020, now you are.)

[EATyourGUITAR]

I say kill the chips and let them burn as long as you record the audio everything will be fine.

[fluxmonkey]

i propose naming this circuit "doomed to fail" - consider what a catchy name did for th atari punk console. and ultimately, arent we all dtf? nobody lives forever.

to the breadboard, batman!

[glennfin]

+1, and I will add......

I work for a semiconductor manufacturer and I can assure you, the device manufacturer has done far more testing than just building and testing the functionality of their device used in a simple circuit on a breadboard.
ESD, thermal extremes, supply under voltage, supply over voltage, etc, etc, etc.... They have very good reason to make recommendations regarding protection and expected operation... It's called good design practice.

.........than being said, If you have a bucket full of CD4046's and your IC's are socketed, then by all means, take a chance and possibly blow up some chips. As was mentioned, ESD events can have a cumulative effect on a device. The damage caused by an ESD event isn't always obvious right off the bat and just because it hasn't happened yet on an extremely limited sampling, doesn't mean it won't.

Let's get empirical. I wonder how many people who are certain that this circuit will fail

Don't put words in my mouth. I asked a question. Nobody here is "certain this will fail."

Riding a motorcycle without a helmet is something many people have done without problems, too. That doesn't make it a good idea. And encouraging newbies in particular to do it is likely to draw some concern from experienced motorcyclists who are still alive.

You posted a design that does not conform to industry-standard ESD practices. You say it hasn't failed for you, yet. I'm sure that is true. But it doesn't mean you're exempt from reasonable questions about its reliability.

At least you're not selling them.

[alka]

i propose naming this circuit "doomed to fail" - consider what a catchy name did for th atari punk console. and ultimately, arent we all dtf? nobody lives forever.

to the breadboard, batman!

word. although i'm not quite certain I am not immortal.

[Repeater]

Worst case scenario is that you have to replace a chip or two every once in a while. This is about as easy as maintenance gets. Which electrical circuits require no maintenance over long periods of time? Electrolytic capacitors are ticking time bombs and I don't see many people filling up threads with cautions about using them.

Thanks for mskala and glennfin for your cautions about ESD. This is not a product, nor would it ever be in this form. Just something fun to tinker with, akin to a Mimms circuit snippet. I urge you to read about the original Cracklebox and reconsider the goals of this circuit, which are to cheaply and efficiently make fart noises in stereo.

[glennfin]

I understand completely, that's why I mentioned having a chip blow up occasionally might be perfectly acceptable. Comparing ESD damage of unprotected CMOS devices to caps drying up over many years is an apples to oranges comparison.

Following good design practice by including external device protection (including bypass caps on the voltage rails right at the device) costs next to nothing to implement, may actually save you money, is no less "efficient" and teaches the unexperienced about good design practice. Why leave it out?...

This goes for ALL designs using CMOS devices, not just the cracklebox

Worst case scenario is that you have to replace a chip or two every once in a while. This is about as easy as maintenance gets. Which electrical circuits require no maintenance over long periods of time? Electrolytic capacitors are ticking time bombs and I don't see many people filling up threads with cautions about using them.

Thanks for mskala and glennfin for your cautions about ESD. This is not a product, nor would it ever be in this form. Just something fun to tinker with, akin to a Mimms circuit snippet. I urge you to read about the original Cracklebox and reconsider the goals of this circuit, which are to cheaply and efficiently make fart noises in stereo.

[Repeater]

Following good design practice by including external device protection (including bypass caps on the voltage rails right at the device) costs next to nothing to implement, may actually save you money, is no less "efficient" and teaches the unexperienced about good design practice. Why leave it out?...

This goes for ALL designs using CMOS devices, not just the cracklebox

I agree with you, though I think it's pretty typical for schematics not to include bypass (or decoupling) capacitors for the sake of clarity.

This is not a tool to teach circuit design to beginners, nor am I advertising it as such. If that's something that you are interested in working on, more power to you! I've learned primarily through The Art of Electronics by Horowitz and Hill, The CMOS Cookbook by Don Lancaster and Ken Stone's CGS site for synthesizer specific topics... and my breadboards. I would recommend anyone interested in these things to try the same resources for much more nuanced and educational information than I can provide.

I should also mention that this is the beginning of my investigation of this circuit as well. I'm working on a more evolved version that makes use of all the CD4046s outputs and quirks to drive other synthesizer circuits. Happy to hear how others get along with it!