General compressor questions

[David]

1)  I found out the hard way that the 2 10M resistors on the input of the Flatline have nothing to do with voltage division and can't be randomly substituted unless you increase or decrease the capacitor proportionately.   :oops:   Thanks for that nugget, R.G.!  What do these resistors actually do?

2)  Could the "updated" Anderton compressor be rigged to run off a single-polarity supply?

Finally,

3)  In the Flatline, is the LED part of the rectifier, or being driven by the rectifier?  What would happen if the diode bridge were removed, and the op-amp just drove an LED like the Nurse Quacky does?  Also, this is what the uncompensated op-amp in Anderton's compressor is driving.  Anderton then uses half the audio op-amp to restore the gain on the audio signal.  I'm missing something.  Hollis and Anderton both have rectifiers and optoisolators.  Hollis doesn't have a recovery stage and Anderton does.  Are these personal design touches, or specialized electronics knowledge I don't have (yet, heh-heh!   :twisted: )?

4)  Can somebody please, PLEASE tell me how to implement that cussed diode / capacitor / LED bridge on breadboard?   :?

5)  Will a Flatline even WORK on breadboard?

Mark Hammer, I know you've got to have built this thing at least once in order to come up with your Punchline!  Would you please impart a little of your wisdom so I can get this thing up and running?

[Mark Hammer]

Actually, I didn't build it.  The "Punchline" conversion is an on-paper/in-principle design that WILL work, but how it would sound I couldn't say.

The 10m resistor on the input is essentially for pop-free switching, by providing a path for the cap to drain off when the circuit is bypassed.  My guess is that it can actually be anything within reason, 1M, 2M2, 3M3, 4M7, etc. as long as it is noticeably higher than the output impedance of whatever comes before it and is low enough to actually LET the cap drain before you switch it again (e.g., 22M *might* need a little too much time).

The second 10M resistor simply serves as current limiting resistorfor the bias/reference voltage coming from the junction of those two 10k resistors.  It too can probably be reduced a bit without suffering much.

The LED is being driven by the output of the rectifier, in a sense, although the true EE's here will probably note that this is not strictly accurate.  At the very least, it doesn't hurt to think of it as being driven by the rectifier.  The cap further smoothes out the minimal flicker that might occur.

Anderton has a gain recovery stage because he wanted one.  Hollis omitted one because the Flatline was part of a short-lived bout of very minimalist effects designs he came up with (where the hell IS Hollis these days?).  You could very easily add one on, just like I did for the modified Orange Squeezer posted over at JD Sleep's site.  Some folks like to think of compressors as *level* controllers, where the intent is to keep the level from exceeding some range, and others think of them as *dynamics* controllers, where the intent is to alter the amount of contrast between pianissimo and fortissimo while still being able to produce boost.

Quality of rectification has an impact on what happens as the note trails out.  With poorer rectification, the ripple in the envelope signal becomes audible in whatever it is you are doing to the sond, and it can sound like distortion or a slight "burr".  You can fix that somewhat by having a longer decay/settling time for the envelope follower, but that means the device won't respond to signal changes as quickly.  Remember, compressors don't just handle single notes with big gaps between them.  They respond to flurries of notes as well, and if the envelope follower is still working its way back from the last big impulse it took on, then the gain will still be turned way down and you won't hear ANY of the initial pick attack on subsequent notes if they arrive too soon.  The diode bridge and LDR permit John to use a smaller value capacitor for smoothing purposes so, in principle, anyways, it should be a little more capable of preserving a little of the dynamics of rapid-fire notes than the Anderton one would be.  Of course that also depends on the LDR chosen.  Some are faster than others.

Could the "Whisper" be run off a single battery?  Ask Jack Orman.

[David]

Thanks for the quick, insightful answers, Mark!  Unfortunately, now you've provoked more questions.  Sorry, but I just have to ask...

Posit that I take the "front end" of a Flatline.  Posit further that I graft Anderton's rectifier to it and use the other half of the dual op-amp (if necessary) to do gain recovery.  I would thus arrive at a single-polarity version of Anderton's original design, wouldn't I?

[Mark Hammer]

In principle, yes.

T'wer me, I'd do what I've already done, which is build the modified Orange Squeezer I noted earlier.  The component values I list are for a simple gain of 2 in the gain recovery stage, but there is no reason why the feedback resistor in the recovery stage couldn't be upped to increase potential output level.  The OS has a nice sound, runs just fine off a single 9v (which seems to be your main objective from the flavour of your comments), and the most exotic thing in there is the pair of FETs.  A pretty easy perfboard build, although there are layouts posted for single and dual versions.  The dual version that someone posted (forget where) could be easily modified to accommodate a gain recovery stage just by picking your pads appropriately and ignoring others.

[RickL]

If by "Punchline" you mean the mod to change the Flatline from a compressor to  an expander, I've done it and it works. Its been a while since I've played with it but IIRC it has a bit of a noise gate feel to it. Soft notes aren't heard at all and anything over a certain threshold kind of jumps out. This is about what I expected from my understanding of what an expander is supposed to do. If you want more details I can probably dig it out and have another listen.

I found that switching between compressor and expander modes I had to adjust the controls a bit to get usable sounds, again, what I expected. I didn't find the expander mode overly useful for guitar but for the cost of a switch and a little time it was worth the bother.