LDR Sweller

[Kipper4]

Thanks Sam
Pad to Vb works as well.
I tried it without R9 and somehow thought it was better with, I'll try again
R12 is a part of the voltage divider that makes the Vbias?

edit
btw did I say you will have to adapt your playing style if you tickle the stings it wont trigger the half wave rectifier. your going to need to pluck that sucker hard.

Low parts count was not a factor, getting it to work ina fashion was. I'm open to ideas as to how to improve the rectifier so playing style becomes less of a hinderance. More parts if needs be.

initially i had it wired on a dual lm358 single chip as you have suggested guys and it will work.

Demo at the bottom of post 15

[Kipper4]

Looks interesting. If you find the LDR is not filtering the AC noise enough you can change around the envelope detector a bit so that you're always driving the LED with more of a filtered DC current.

As it is my guess is the 1/2-wave ripple will be inaudible

your ideas are welcome for changing around the envelope detector, most welcome.

[deadastronaut]

nifty rich....

[Kipper4]

Try swapping the vactrols led + and - connections for a sort of reverse swell. Kinda........


like this

[garcho]

what's the point of R1 and R8? it looks like it's set up as an inverting buffer, but you have the signal at the non-inverting input. maybe just jump wire the - and output and get rid of the resistors?

you also might be able to get away with removing the attack and pad controls and turning R5 into a variable resistor, as an all-around attack control. just a thought.

[samhay]

>R12 is a part of the voltage divider that makes the Vbias?

Sorry. R8 - 12k, which Garcho also pointed out.

>I tried it without R9 and somehow thought it was better with, I'll try again

If you are using a 358 there, it probably will make a difference - does the op-amp bias quite a lot below Vb (measuring at the output)?

>edit
btw did I say you will have to adapt your playing style if you tickle the stings it wont trigger the half wave rectifier. your going to need to pluck that sucker hard.

What happens if you make R5 bigger?

[Kipper4]

if i make R5 bigger it stops swelling because the vactrol led doesnt go off the same and a note shortly after the first doesnt retrigger the vactrol led.
i stuck a 100k pot in to try it wired as a variable resistor.
after trying it at differant pot settings it still triggers best witha 33k in the NFL.


I got rid of the R1 R8 with no detriment. thanks
I also abandoned the attack control no sweat it still works.
Ive left the decay pot so now im down to a one pot (decay) set up
this will fit in a matchbox WITH a battery soon.

[Kipper4]

im getting 4.9v at the ouput of rectifier ic1b

ive just realised i might get a better response from the rectifier if i rebias the input buffer to a higher Vb
i'll be back

[Kipper4]

rebiasing the input buffer much above 5v and it distorts. At least i know.

[anotherjim]

It's a neat idea to use the LED as the rectifier, but it does mean you limit the potential range of the envelope. Adding the reverse diode has made it a charge pump voltage doubler, but you lose some with the LED's high Vf. I would suggest replacing the LED with another 914 and use other half of the 358 as a voltage follower from the envelope cap to drive the LED. You will still have a sudden death when the envelope falls below the LED Vf, but I suppose the slow response speed of an LDR will cover it up?

To get more positive headroom from the 358 - lower the bias voltage -  somewhere below 1/2 Vcc is halfway for this amp. Less than 4V @9V power.

[Kipper4]

Gret stuff Jim I'll give it a go
cheers

[Kipper4]

You mean like the mutron envelope detector here Jim

http://www.geofex.com/article_folders/ecftech/ecftech.htm


Prolly bout Alf why darn the page

[Transmogrifox]

I have been playing with this in LTSpice and my parts count just keeps climbing in order to get the same effect as you have.  The initial short plucky attack before it decays and swells back in is the coolest part of this little circuit.

It would be pretty straightforward to make it into a standard string swell effect if you drive a BJT off the rectifier and then turn off the LED with the BJT.  What I mean is you would have, say 9V->4.7k->LED->gnd.  Then you connect the BJT collector to the 4.7k to starve current from the LED as the BJT turns on.  You would then drive the BJT from the rectifier through a large (like >1Meg) resistor so you can tie the BJT emitter to ground to be able to completely bypass the LED.

That would be the simple string swell, but you would lose that "plucky" initial attack...but writing this gave me another idea how this might still be done without such an increase in parts count.

[anotherjim]

I was thinking simpler than the Mutron. That's not a bad idea though as it lets you switch control 2 ways and has the LED regulated in the feedback loop, although the last amp config will load the envelope cap times differently from what you have now.
I mean just feed the amp + input and connect - to output as a simple high-z follower. It won't load the envelope cap at all. To reverse the LED action, you could still have DPDT switching so the LED is forward from amp to ground or reverse from amp to +9v - don't forget LED series resistor.

And strictly speaking (hate to add components), the charge pump should have some R from the amp out to limit the currents in the capacitor C12, even if only 100R, otherwise, instantaneous current is theoretically infinite, albeit only for a gnats crotchet of time.

Oh, and the reverse LED snippet just above isn't quite right. You'd want a negative charge pump in that case or the envelope cap won't charge, so the diode to ground is "reversed" again - forward to ground. Cap polarities then need be reversed as well as the LED.


Above shows the charge/discharge paths for a negative pump. The diode to ground path is the essential return/priming stroke of the pump if you will.

Swell if an interesting FX that I haven't really considered until now, so will be interesting to see & hear it polished off.
You can't call it Viagra^tm though - it's registered. Maybe the generic compound name Myacoxaphlopin?

[Kipper4]

Do I have this right please
Jims mod above
Transmogrifox below (after jims mod will this work with these diode orientations, confused,com)
Thanks

[Kipper4]

More likely
Miyocoxaswelling.

[anotherjim]

oops. I forgot to mention you can't have a negative pump with an op-amp follower. The 358 can have input to ground, but not negative. I was suggesting it for the scheme with the LED as the rectifier diode in reverse. So with the extra op-amp driving the LED, D2, D4 and C5 need reversing. The LED on the op-amp output must have a limit resistor too.

Transmogrifix version, only D5 needs reversing. Both ideas don't add much extra, but one or other might suit the effect better.

Consider when Decay pot is zero ohm. The first 358 is driving a short circuit. Find a resistance to put between the pot and ground that doesn't affect the decay range too much, maybe 1k or so.

Swollen Gherkin?

[Kipper4]

Sorry to keep doing this but
Here's how I read your response Jim.
Thanks for putting the time in guys

Have I got this right?

[anotherjim]

That's it exactly. 

[Kipper4]

Thanks Jim I'll bread it up then and try both.