Simple compressor found on youtube

[11-90-an]

OptoLim v1.1



- Corrected some wrongly written values, 5.1M -> 5.6M
- Component value changes, 470k ->220k, 10k pot -> 5k pot
- Added 2.2k resistor in series with 1u cap to reduce fizz from envfollower

UNVERIFIED perf layout:



UNVERIFIED vero layout: (my first vero, so there might be tons of mistakes... keep alert! )

[snk]

Wow, that was fast!
Thank you, 11-90-an !
I have other projects to finish first, but this one goes on top of the "todo soon" pile 
Is it suitable for bass instruments (I am no guitar player)?

[duck_arse]

when you say slow LEDS are best .........

write the revision number on the drawn circuit each time, so's we don't confuse. and nice work.

[11-90-an]

Wow, that was fast!
Thank you, 11-90-an !
I have other projects to finish first, but this one goes on top of the "todo soon" pile 
Is it suitable for bass instruments (I am no guitar player)?

Thanks! I discovered that making layouts is fun and since today is a holiday for me, I decided to do it now. Of course, it's not verified, so build at your own risk...!

I don't have a bass, but I think it will do fine if your increase the all the 1uF caps to say, 2.2uF, except for the one beside the LED... but if you do you would probably have to increase the 2.2k beside the LDR (in the vero layout) to maybe 3.3k... It would do fine with the standard 1uF's...

[11-90-an]

when you say slow LEDS are best .........

write the revision number on the drawn circuit each time, so's we don't confuse. and nice work.

Good point... forgot to do that, sorry...
And thanks too..

EDIT:
Updated it, and it has quite the trash resolution, since there is a lack of light and it's nighttime already. I'll retake tomorrow morning.

[PRR]

If your TRIM pot is turned all the way, you have full supply voltage through the transistor base, which will burn it.

You clearly never need more than 0.7V here, if that.

[11-90-an]

If your TRIM pot is turned all the way, you have full supply voltage through the transistor base, which will burn it.

You clearly never need more than 0.7V here, if that.

Yep. I need to write more detailed instructions... I'll update them later.

EDIT: layouts updated with more instructions

[duck_arse]

do you trust people taht much? PRR would want you to fit stop resistors so the burn can't happen, I'm sure. he might not care, but still.

[11-90-an]

do you trust people taht much? PRR would want you to fit stop resistors so the burn can't happen, I'm sure. he might not care, but still.

I *actually* tried it when I was BB'ing... no magic smoke...

My excuse would now be "lack of space" but I digress...

[duck_arse]

nuh-uhh, youse gots no excuse - I can see a dozen empty holes on your perf [if I'm looking the right one], you only need one resistor added. and speaking of wrong - where does the 9V connect to on the perf, just that empty pad?

[Steben]

If your TRIM pot is turned all the way, you have full supply voltage through the transistor base, which will burn it.

You clearly never need more than 0.7V here, if that.

What's wrong with relying on some healthy control on one's actions?

[Fancy Lime]

The bias trimmer is not only dangerous, as Paul pointed out, it is also unnecessary. Replace it with a 220k or so Resistor from base to supply in plus two antiparallel silicon diodes from base to ground. That way the transistor will idle at the brink of turning on, which seems to be what you want. I personally would leave the diode that points down (cathode to ground) and the resistor to supply out completely, to get a threshold of 0.7V. The diode that points up should be there to protect the base from going too far below the emitter. In that case, also add a resistor from base to ground. 220k should work.

EDIT: Have a look at the envelope detector of the MXR Dyna comp for comparison: https://www.electrosmash.com/mxr-dyna-comp-analysis

You can get away with this static bias because you are using a BJT. BJTs have a very predictable turn on voltage of about one silicon diode drop between base and emitter. The original youtube circuit used a MOSFET. Now these things definitely need a trimmer because their turn on voltage is quite variable, so you never know if any individual MOSFET will turn on at 0.5V or at 1.9V or whatever.

Andy

[11-90-an]

The bias trimmer is not only dangerous, as Paul pointed out, it is also unnecessary.

Naw, there is a reason of the trimpot. The transistor needs to turn "almost on", yes, but each LED has a different color, and brightness/CLR . In other words, each time you swap out an LED, the bias would have to be re-tweaked. Tried and tested. 

where does the 9V connect to on the perf, just that empty pad?

Fixed that. I'm making v1.2 layout since there are much needed (like Paul's warning) and slightly needed (fizz reduction) improvements...

[Fancy Lime]

Why would you swap out the LED? CLR definitely needs to be adjusted to suit the LED you are using. This will control the brightness and therefore the sensitivity. Why not replace CLR with a resistor + pot in series and loose the 5k pot at the LDRs? That way you have a sensitivity control that also takes care of adapting the circuit to the LED you are using.

Andy

[11-90-an]

I tried it, and I think I know why it isn't working... (your reversed diode and resistor configuration)

The dynacomp is a feedback compressor, I'm quite sure mine's a feed-forward... (envelope is taken from orginal signal, not amplified, or something like that)
So the signal is still "smaller" that is why there is a need for a bias trimpot. It basically makes sure how much the transistor is "on", AND how bright the LED is before encountering the guitar signal (I'll call this "starting brightness"). Or, I can always make the starting buffer into a booster, right? It *would* solve my problem of it being out of phase with original signal...

I was planning to use a 9v - 680k - 100k trim - gnd network to establish a proper bias. Max voltage from that would be around 1.15v. Probably won't burn the transistor out. I can still add that inverted diode for extra protection...  Anyway, I tried it and works really well. A very fine control of the "starting brightness" of the LED.

When I said "swapping the LED" I was citing an example of us DIY'ers having different LEDs, with different "on" voltages, blah blah, you get the point...

I do agree with removing the 5k pot... not really useful, but useful in some ways...
The CLR-pot network would work, but with different LEDs around 10k might not make much of a difference. What do you think of letting it control the amount of guitar signal entering transistor base?

[iainpunk]

Interesting...  Needs a couple more components, but it certainly IS simple!  And it works!  (would not trade it for my Engineer's Thumb, but it does work!)

engineer's thumb, is that like m1 Garand thumb? where you get your thumb between the receiver and bolt while pressing in an ammo clip not forcefully enough

[11-90-an]

From some tests, I noticed that the CLR need not be changed with different LEDs (just for testing pitposes), only the bias has to be tweaked slightly with each LED...

I want to know if this applies with all LEDs... I wanted to test with superbright LEDs, but I happen to have none...

And so I would want to ask, if any of you have any superbrights, could you try throwing this portion of the ckt on the breader...? 

Schem...



Basically just throw in LEDs and adjust the trimmer until LEDs are slightly lit and when guitar/instrument is strummed hard there is considerable sweep...
I'm sorry for any inconvenience/annoyance this may cause...

[Phend]

Question for those who built this compressor...what is your opinion now.
Does it work well enough to box up?
Or is it just a novelty with glitches.

[11-90-an]

Question for those who built this compressor...what is your opinion now.
Does it work well enough to box up?
Or is it just a novelty with glitches.

Erm I'm not sure anyone built is except me... it works, but I would probably want to improve it a bit more... maybe try to figure out the attack/release/decay mechanics, but for you, probably use the new envelope follower that I posted on the post just above yours, @Phend, and the same boost circuit as the latest rendition... remove the 5k limit pot and connect the - wait, maybe I should just stop blabbering and provide a schem, right? Here...

Does this work? https://imgur.com/c8lRu7T

HTH

[Phend]

Thank you very much 11-90-an.
Just got 3 bread boards in the mail.
Did not have any before so I just put my boxes (copied circuits) together and hoped for the best.
Now I would like to do some experimenting, so I will give your circuit a try.
I am just learning about these effects in general.  I do know a little, like today it dawned on me that the guitar generates AC in a many many frequencies, the input cap and output cap remove some.
Plus the output cap stops any DC from going to the amp, that is what turned on the light bulb.
Now to get into the middle of the circuits and try to Visualize what is going on.
Thanks again
Paul