Thread for random DIY-related questions

[crochambeau]

If the power lead got grounded while the pot was only presenting a tiny bit of resistance it could easily overshoot the dissipation range of the pot.

9 volts across 10 ohms makes over 8 watts, for example. But the pot would only be seeing the full voltage if there was no load in between it and ground (and ground path is there).

I like feeding a voltage divider into a suitably capable opamp just set to follow the voltage presented to its input, though that is a considerable step up in complexity to what you have drawn, it treats the pot very well as the input impedance to the opamp pretty much eliminates substantial dissipation.

[FAP]

Could you provide an example of this op-amp circuit?

[crochambeau]

Could you provide an example of this op-amp circuit?

Viewed 494 times

You probably do not need either of the limit resistors (R5 & R6) though you could include a limit resistor from ground to insure that the lowest rotation of your pot just barely cuts out, or whatever effect you're after.

Power rails to the opamp are not illustrated (at this point, it's a much larger schematic that does other shit that will muddy the waters here) but Vee goes low and Vcc goes to the supplied power, a rail to rail opamp will insure minimal loss of your original supply.

[crochambeau]

Here's a refined drawing. It's nice to put a series resistor on an opamp for short circuit protection. 10-100 ohms should be fine.

Viewed 493 times

[FAP]

So an update here: I went with the method that involved putting a 2.2kΩ resistor between lug 1 & ground. It worked great initially: did the readings, everything coming up exactly as expected. Within, like, ten seconds, all of a sudden it stops working. No burnt smells, no burnt parts: it simply gave up.

Skipping ahead through many hours of frustrated defeat, I've narrowed it down to the pot. For some bizarre, bullshit reason, the pot only sometimes yields the full 9.0v at full-CW rotation. Most of the time, it reads around 6.5-7.0v at the same position; jiggling the wires occasionally gets it back to 9.0v and the rotation of the pot works as expected, but it only takes one wrong move until I'm back to the erroneous reading.

I also took readings for the resistance of the pot: the total resistance reads as it should (10kΩ), however if I read across lugs 1 & 2, then turn the pot CCW, all of a sudden I go from values within the realm of reason (i.e. 2.2kΩ, give or take a kilohm or two... okay, so maybe not reasonable per se) to resistances far exceeding the total resistance (e.g. ≥14kΩ). Sometimes, it even read as open, though I'm not sure if that's due to apoor connection on my part or some other batshit, esoteric defect that I can't see because I don't have x-ray vision.

Now the solution from a practical standpoint seems obvious—get a new pot—but given that I've apparently burned through three of them already, I'd like to know if there's something fundamental (or subtle) that I'm missing before I keep throwing more parts at it.

[crochambeau]

Now the solution from a practical standpoint seems obvious—get a new pot—but given that I've apparently burned through three of them already, I'd like to know if there's something fundamental (or subtle) that I'm missing before I keep throwing more parts at it.

We'd need to factor in whatever you are starving with the circuit to offer a reasonable diagnosis (file that away in the rhetorical question bucket, I don't actually have the bandwidth to burn on this one). Pots are generally not good at the task of routing high currents. Then again, sometimes a batch of parts is bad, I have experienced bad pots from a reputable source.

I prefer the "treat the pot lightly and manage the current handling with actives" method, and I will die on that hill.

[FAP]

Now the solution from a practical standpoint seems obvious—get a new pot—but given that I've apparently burned through three of them already, I'd like to know if there's something fundamental (or subtle) that I'm missing before I keep throwing more parts at it.

We'd need to factor in whatever you are starving with the circuit to offer a reasonable diagnosis (file that away in the rhetorical question bucket, I don't actually have the bandwidth to burn on this one). Pots are generally not good at the task of routing high currents. Then again, sometimes a batch of parts is bad, I have experienced bad pots from a reputable source.

I prefer the "treat the pot lightly and manage the current handling with actives" method, and I will die on that hill.

I suppose it’ll probably come to that, then. That active circuit isn’t that difficult, I was just hoping there’d be an even easier solution but c’est la vie

To clarify: the reason the first starve circuit failed is because it wasn’t starving anything with the circuit—at least, that’s what I believe happened given the evidence.

[FAP]

Here's a more conceptual, possibly philosophical question: how come different duty cycles of the same frequency sound different? I compared several different duty cycle ratios (in the format of %hi/%low: 50/50, 60/40, 70/30, etc.) of the same frequency and they all sound a little different to my ears. Obviously it's not the frequency that's changing, nor is the amplitude, so what is it that makes them sound different? It's not psychoacoustic/a placebo effect, is it?

[crochambeau]

Here's a more conceptual, possibly philosophical question: how come different duty cycles of the same frequency sound different? I compared several different duty cycle ratios (in the format of %hi/%low: 50/50, 60/40, 70/30, etc.) of the same frequency and they all sound a little different to my ears. Obviously it's not the frequency that's changing, nor is the amplitude, so what is it that makes them sound different? It's not psychoacoustic/a placebo effect, is it?

Harmonics.

And as the frequency goes up we can perceive less and less of them so at say 15 kHz things sound mostly the same, because the ear is only keeping up with the fundamental.

[FAP]

I'm considering building a circuit that would allow me to toggle between true and buffered bypass for a larger project I'm working on (if nothing else, I'd like to have a de-facto testing jig to compare/contrast the two bypass methods). I came across the schematic in the pic below, but I'm just curious: can this really be called "true" bypass if both the input and output must inevitably pass through caps anyways? Furthermore, they're pulled close to the reference voltage "Vr" to mitigate clicking (see this article for reference): wouldn't that have an impact on the output signal as well?

Viewed 9 times

To clarify: this is just the "true" bypass portion of a larger circuit I'm considering. That circuit is too complicated to go into depth in right now, but the basic idea is I'd hook up an effect pedal's mom. n.o. foot switch to simultaneously trigger a 4013 (again, see that article) to in turn trigger the control signal for the 4053. There'd be a mechanical DPDT switch beforehand that would direct the in & out jacks either directly to the pedal's in & out leads, or through the 4053 signal path(s).

Obviously this is a lot of extra parts for a relatively negligible convenience, hence why I'm wondering if it's even worth pursuing if the "true" bypass path here is in reality bogged down with additional components.