Auto-shutoff methods in consumer-grade circuitry

Tutorials, tools, techniques, methodologies......

Moderator: Modulators

User avatar
NoiseWiki
Wiki Bastard
Posts: 3811
Joined: Thu Apr 04, 2019 3:38 pm
Location: Los Angeles
Has thanked: 1084 times
Been thanked: 1249 times
Contact:

Re: Auto-shutoff methods in consumer-grade circuitry

Post by NoiseWiki »

FAP wrote: Fri May 26, 2023 2:45 pm Found a much cooler solution to this that'll give me control of the sample rate(?) to boot: more on that later.
Well I guess that didn't pan out but what were you trying?
User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

Re: Auto-shutoff methods in consumer-grade circuitry

Post by FAP »

NoiseWiki wrote: Mon Jun 05, 2023 10:51 am
FAP wrote: Fri May 26, 2023 2:45 pmFound a much cooler solution to this that'll give me control of the sample rate(?) to boot: more on that later.
Well I guess that didn't pan out but what were you trying?
Okay, so I'm going to address this in two parts: first, the 'sample rate' mod, then how I set up the power/reset circuit which is critical to making this thing function.

The fidelity ("sample rate") mod can be achieved by putting a pot between MIC+ and the right side of C4 on the PCB; in the image below, these are the yellow dots. I've found 20kΩ works best: turn cw for a clear sound similar to the stock effect, turn ccw for a lower-fidelity, slightly delayed sound. The important thing to note here is this only works when the leads for Y2 (the crystal/oscillator) are shorted; in the image below, these are the cyan-colored dots.
vader-fidelity-mod-nutshell.jpg
Now why does this only work when Y2 is shorted? Well for one, the fidelity mod only came about after a different issue was addressed: the time limit.
As stated, the voice-changer effect will shut off automatically after two minutes of use. The timer is started as soon as a connection between VCC (3v) and CHANGER is made; in the image above, these are the pink-colored dots. In layman's terms, this means pressing the big red button on the toy gives you two minutes to do silly voices before you have to press it again to keep doing silly voices.

Through trial and error, I found a way to keep the voice-changer effect ON indefinitely by simply shorting the leads of Y2, however this comes with two major drawbacks.
First, you're stuck in the lo-fi sound mentioned above, until you connect MIC+ and the right side of C4 on the PCB (in which case you can get a sound remarkably close to the way the toy normally sounds).
Second, and more importantly, the toy won't turn on again as long as these leads are shorted; if the toy is turned off or reset, the voice changer effect won't activate again unless the leads of Y2 are connected to the stock crystal/oscillator (herein referred to as just "the crystal"). In other words, the toy expects to "see" the crystal connected in its proper place on the PCB in order to work, and if it doesn't "see" it, it crashes.
Similarly, the toy also won't work if it doesn't "see" the stock electret mic connected as the input (I can only guess this has something to do with expected resistance load, but then again, the toy was never meant to have these sorts of connections made anyways).

So what this means, in essence, is the voice changer effect will only start up if:
1. A connection is made across VCC and CHANGER,
2. The crystal is connected across Y2, and
3. The mic is connected across MIC+ & MIC-.

The good news is, all of these things only need to happen momentarily: after that, the effect can be prolonged indefinitely by once again disconnecting the voice changer nodes, shorting the crystal, and disconnecting the mic.

The image below, while not fully tested, shows how I accomplish this "startup"/reset circuit with a momentary 3PDT foot switch:
vader-power-nutshell.gif
This isn't the simplest of solutions, obviously, but it works. IMO all of this rigamarole wouldn't be so bad if the MIC and SP leads shared GND: unfortunately, they don't. To put it another way, MIC- ≠ SP- ≠ GND ≠ -BATT. This seems to be a common thing with electronic toys, which is understandable: nobody expected some yahoos to stuff their plastic costume accessories into a metal stompbox nearly two decades after the fact. Anyways, that's what the 4PDT is for: to compensate for this discrepancy between input and output negatives when toggling to true bypass.
Post Reply