[sdiy] Signals leaking into the PSU?

[Scott Bernardi]

I breadboarded it, and it did work with 0 - 5v, but that may be because my
transistors were mismatched.

I have done the red-green LED thing on an LFO.
https://www.bernacomp.com/elec/og2/og2_lfo.gif

On Fri, Feb 24, 2023 at 3:16 AM Neil Johnson <neil.johnson71 at gmail.com>
wrote:

> Hi Scott,
>
> > Here's a way to keep a constant current driving a flashing LED using a
> differential pair.  It only takes one extra resistor and transistor
> compared to the standard transistor
> > I whipped up a page for it:
> > https://www.bernacomp.com/elec/og2/constant_current_led.html
>
> Thanks - an interesting solution.  My only concern would be driving it
> from 0V to 5V.  At 0V both bases are at the same voltage so the
> current will split between the two transistors depending on their
> matching.  If they were both perfectly matched then the current would
> split equally between the two paths, so the LED would see half of the
> "on" current.
>
> Driving it from a bipolar signal is fine, for example, for a -5V  to
> +5V input you get a sinusoidal-ish LED current (according to sims
> anyway ;) ) for lovely smooth on-off transitions (probably makes the
> tone warmer too, more analogue-y richness with a mellow undertone,
> like melting butter oozing over hot crumpets.... errr..... nice.....).
> And you could extend this to drive two LEDs (say, red and green) for
> extra blinky goodness.  Or you get to choose the polarity of the drive
> (inverting as you have drawn, or non-inverting if the LED is in the
> other collector).
>
> Cheers,
> Neil
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20230224/0692770c/attachment.htm>