Welcome the fails file!
Have you ever engaged yourself into a project of nonsense or of no success? And by the time of awareness about the failure - have you then been able to giggle in an amused mood about your own maledicted misfortune, about your betrayed builders brillance or even about your nauseated narcissism? And then, despite the misadventurous outcome, having learned something along this adventure? Yes? Really? Fine! Then share your giggle, your frustration, your suffering or mourning with others. Just share it, not expecting any salvation in doing so, but for the sheer sake of sharing. Or maybe also joint joy. Nothing else.
So this might be the thread for all kinds of these otherwise silently littered projects which for any reason were not successful or were not meant to be successful. Regardless, wheter these builds were fatally flawed, hilarously hopeless, ingeniously idiotic, seriously silly, tragically trashed or otherwise wery wrong … Dont litter them silently, it would be a pity. This might be the place instead where to publish, and then wait and see ... Let others comment and then eveventually lets grin together. This thread is meant to be less serious than recreative and hopefully funny.
I already published such a project which was mainly hilarious and highly prone to fail in another thread (the teggstreme/epoxy midrange dome driver). And here, let me begin with another project of this kind: I was nearly certain that this new, eery iteration would be seriously flawed. But instead of being reasonably renouncing, I just wanted do know what would really happen and built it. Who knows.
So may I proudly present the rubber foam elasticity controlled electrostatic sandwich-folded curtain tweeter: An electrostatically loaded membrane is folded around self-adhesive foam tape. The folds are moved by electrostatic forces from integrated stators, such in a standard ESL. Think of an AMT instead. But unlike in an AMT, the sound should not be emanated by the squeezing and relaxations of the folds distances, thus moving the trapped air, but simply by altering the geometry of the folds free curved surfaces:
I knew, that if it was functional at all, then I would rather have constructed a silentspeaker than a loudspeaker. And as expected, the result was at least a partial mess: There was indeed some sound emanated by this construction. But the sound coming from the transformers was a hefty bit louder instead. And while bilding/folding the sandwich, these free membrane surfaces adhered to the sticky isolation foam tape. So I had to introduce knitting needles to free them again. As you see, one of the plies did not like that.
Lesson learned. Too much resistance, and not enough electrostatic forces.
… And now … Whats your comments, or even better, whats your own story of failed creations?
Have you ever engaged yourself into a project of nonsense or of no success? And by the time of awareness about the failure - have you then been able to giggle in an amused mood about your own maledicted misfortune, about your betrayed builders brillance or even about your nauseated narcissism? And then, despite the misadventurous outcome, having learned something along this adventure? Yes? Really? Fine! Then share your giggle, your frustration, your suffering or mourning with others. Just share it, not expecting any salvation in doing so, but for the sheer sake of sharing. Or maybe also joint joy. Nothing else.
So this might be the thread for all kinds of these otherwise silently littered projects which for any reason were not successful or were not meant to be successful. Regardless, wheter these builds were fatally flawed, hilarously hopeless, ingeniously idiotic, seriously silly, tragically trashed or otherwise wery wrong … Dont litter them silently, it would be a pity. This might be the place instead where to publish, and then wait and see ... Let others comment and then eveventually lets grin together. This thread is meant to be less serious than recreative and hopefully funny.
I already published such a project which was mainly hilarious and highly prone to fail in another thread (the teggstreme/epoxy midrange dome driver). And here, let me begin with another project of this kind: I was nearly certain that this new, eery iteration would be seriously flawed. But instead of being reasonably renouncing, I just wanted do know what would really happen and built it. Who knows.
So may I proudly present the rubber foam elasticity controlled electrostatic sandwich-folded curtain tweeter: An electrostatically loaded membrane is folded around self-adhesive foam tape. The folds are moved by electrostatic forces from integrated stators, such in a standard ESL. Think of an AMT instead. But unlike in an AMT, the sound should not be emanated by the squeezing and relaxations of the folds distances, thus moving the trapped air, but simply by altering the geometry of the folds free curved surfaces:
I knew, that if it was functional at all, then I would rather have constructed a silentspeaker than a loudspeaker. And as expected, the result was at least a partial mess: There was indeed some sound emanated by this construction. But the sound coming from the transformers was a hefty bit louder instead. And while bilding/folding the sandwich, these free membrane surfaces adhered to the sticky isolation foam tape. So I had to introduce knitting needles to free them again. As you see, one of the plies did not like that.
Lesson learned. Too much resistance, and not enough electrostatic forces.
… And now … Whats your comments, or even better, whats your own story of failed creations?
Attachments
Last edited:
I once designed a motorcycle ignition control unit.
Worked fine on the bench.
However the bike sometimes stuttered when starting.
But once running no problems at all.
You could alter ignition map once above idling to get more power.
Rev limit could be set too.
The plug was pulled on the project before I could find what the starting problem was.
1/ Maybe battery voltage dropped too low on cranking and upset processor?
2/ Maybe ignition was too far advanced on cranking causing kick back ?
We put one on a race bike and the problem was worse with that but once it started it went like stink.
Worked fine on the bench.
However the bike sometimes stuttered when starting.
But once running no problems at all.
You could alter ignition map once above idling to get more power.
Rev limit could be set too.
The plug was pulled on the project before I could find what the starting problem was.
1/ Maybe battery voltage dropped too low on cranking and upset processor?
2/ Maybe ignition was too far advanced on cranking causing kick back ?
We put one on a race bike and the problem was worse with that but once it started it went like stink.
There are exist actually high-frequency isodinamic speaker drivers, and full-frequency heheadphone speackers. They were prodused in Soviet Union about 40 years ago, and similar types are manufactured in our times somewere in Europe (I don't remember exact types). They are fine in quality of sound because film is very thin. The only problem is they don't want to be overloaded.
Link1, in russian.
Link2, in english.
Link1, in russian.
Link2, in english.
Last edited:
I have a bunch of tda1553q car-amp chips in my drawer since the 1990ies and tried to turn them into super-low-thd composite amps. Just to realize that besides the difficulty to use barely documented fixed gain btl amps with internal (unstable) signal ground my motivation was just my urge to recycle anything I have into a magnificent piece of art.
Anyhow, i learned something about completely unstable circuits, breadboarding, simulation, measuring thd and realized that without composite circuit these chips are quite ok and still perfectly usable.
Anyhow, i learned something about completely unstable circuits, breadboarding, simulation, measuring thd and realized that without composite circuit these chips are quite ok and still perfectly usable.
I designed a USB mixer controlled from a Windows PC.
Built up the pcb and with inputs shorted had 1vac of hum on the output !
I found 50Hz was getting in everywhere.
Turned out I had put power supply on same pcb and the grounds were all mixed feeding hum into audio circuit and that was being amplified.
So reworked the pcb having power come into power supply first the on to audio circuit
Audio ground and power ground met at one point on pcb power connector.
Worked a treat and a valuable lesson learned about star grounding.
Built up the pcb and with inputs shorted had 1vac of hum on the output !
I found 50Hz was getting in everywhere.
Turned out I had put power supply on same pcb and the grounds were all mixed feeding hum into audio circuit and that was being amplified.
So reworked the pcb having power come into power supply first the on to audio circuit
Audio ground and power ground met at one point on pcb power connector.
Worked a treat and a valuable lesson learned about star grounding.
In college, I tried to make a guitar synth out of a CD4046 PLL. I kicked that can down the road quite a bit, incorporating frequency sample and hold driven by an envelope follower (the filter of which I eventually morphed from 1st to 2nd order), a digital delay line so there'd still be some samples to hold onto when the note amplitude decay crossed the comparitor threshold from "sample" to "hold", a VCA and VCF also driven by the envelope follower signal and a Mostek MK50240 top octave chip allowing it to play in a different key than what was being played on the guitar, or harmonize.
Single note only (I couldnt afford to build up 6X the same circuitry and purchase a hex pickup), it barely was playable musically and all it ever got me was an "A" grade for an EE elective rolled into my GPA.
I was after pure tones, but it sounded like crap, due to a lot of phase noise, or jitter you could hear disappear immediately when it went into hold mode. Also the PLL came into lock exponentially, which sounded like it took forever to tune up to the note being played; glissandos between staccato notes never seemed to quite make it, unless you turn the lock time down so far that it wasnt much of a glissando anymore. It would lock on pitch if you were lucky, could miss half the time.
Was a true source of endless frustration for me, working on it as I listened to Jerry Garcia's perfect synth lines on Terrapin Station. Eventually, I threw in the towel after graduation and first employ at Digital. I'm certain I could recreate that level of frustration today - X6. It's OK.
Single note only (I couldnt afford to build up 6X the same circuitry and purchase a hex pickup), it barely was playable musically and all it ever got me was an "A" grade for an EE elective rolled into my GPA.
I was after pure tones, but it sounded like crap, due to a lot of phase noise, or jitter you could hear disappear immediately when it went into hold mode. Also the PLL came into lock exponentially, which sounded like it took forever to tune up to the note being played; glissandos between staccato notes never seemed to quite make it, unless you turn the lock time down so far that it wasnt much of a glissando anymore. It would lock on pitch if you were lucky, could miss half the time.
Was a true source of endless frustration for me, working on it as I listened to Jerry Garcia's perfect synth lines on Terrapin Station. Eventually, I threw in the towel after graduation and first employ at Digital. I'm certain I could recreate that level of frustration today - X6. It's OK.
I once designed and built a two-integrator oscillator. It was meant to be a widely tunable, low distortion design.
It surely did oscillate: there was the intended oscillation, parasitic VHF oscillations and low-frequency oscillation of the amplitude control loop. I never got it to work as intended.
It surely did oscillate: there was the intended oscillation, parasitic VHF oscillations and low-frequency oscillation of the amplitude control loop. I never got it to work as intended.
Last edited:
Another example is my first audio power amplifier, which was my own variant of the Blomley design. I got it to produce sound, but the next time I turned it on, within seconds, the emitter resistors went up in flames.
A third example is my direct-drive amplifier for Frank Verwaal's electrostatic loudspeakers, see http://home.kpn.nl/verwa255/esl/esl_root.html It worked fine and sounded very good, but with its 2 kV peak output voltage, the volume was too low for anything but background music.
A third example is my direct-drive amplifier for Frank Verwaal's electrostatic loudspeakers, see http://home.kpn.nl/verwa255/esl/esl_root.html It worked fine and sounded very good, but with its 2 kV peak output voltage, the volume was too low for anything but background music.
Last edited:
I did exactly that in silicon, needing an fsk LO quadrature oscillator for a 402Mhz medband implantable communication chip, as the DAC method was too power hungry
I hope yours didn't have the amplitude loop oscillations and the parasitic oscillations... At least the wiring on a chip is usually short and resistive enough not to cause parasitic oscillations, although you have to be careful wherever signals go on and off chip.
been a while ago, think I used a sampler to take the amplitude. the amplitude control regulated between small pos and neg feedback, as the resitors were all matched. only needed a center freq cal upon chip init to compensate the cap variation.
long time ago I built an electrostatic epoxy coater. it was a 10mm thick PE plastic box with on the bottom some spirals at -60kV, and outside a plate operating at 10kV ac against the inner spirals, to excite the expoxy powder at the bottom. the metal object hanging above the spiral connected at ground potential attracted the epoxy powder. the PE box was constructed using hotair welding, and the weld seams turned very conductive due to the local oxidation of the plastic. did not work .. had to close the curtains to see where it was leaking. 60kV is not funny stuff.. found a PE glue to fix the problem ..