EH Soul Preacher opinions (part-2)

[markm]

Okay fellas,
I have been working on this circuit and in doing numerous searches on the forum here, I have discovered there is a "correct" version of the Soul Preacher and an "incorrect" version as well......or maybe one is more correct than the other.
In looking both of these over, I think this one which is everywhere on the net is the "bogus" SP;


This one here seems to be more logical however, having never in my life seen nor heard a real EH Soul Preacher, I couldn't honestly say.
Mark Hammer mentioned that there was possibly an R.G. schematic of this circuit......maybe this one is it?
Maybe this one appears to me the correct schem?



Any opinions?
And R.G., if you read this post, is the hand drawn schem yours?

[R.G.]

Maybe it's just late here, but it looks at first glance like the same circuit, different drawings. Parts values and connections seem to match.

What was it you saw different?

[markm]

Thanks for the response R.G,
I at first thought it was the same as well so, I overlooked the hand drawn schem.
But on a second look, it appears as though the diode from the PNP transistor is not connectectd to pin 7 of the IC.
On closer inspection however, I thought beforehand the MosFET was hooked up differently but now I see it is in fact the same.
Also, the connection from pin 1 of the IC to the 2 10K resistors on Pins 5 & 6 is before the 10uf Tant cap in the hand drawn schem and after the cap in the other.
*sigh*
So I guess it's that connection and the connection on the bottom diode huh?
Maybe I'm confused 

[markm]

....sorry, not a MosFET but a JFET.....too many comp projects going on 

[Mark Hammer]

Actually, I suspect the hand-drawn one is MORE correct.  The 2nd diode you noted as "needing" to be tied to pin 7 would be redundant as drawn in the "cleaner" drawing - it is tied to the same spot at both ends.  In the hand-drawn version, one diode takes it feed from A2 and the other takes its feed from the output of A1.  Since A2 is set up as a unity-gain inverting stage, and since the diodes have the same orientation, that makes a full-wave rectifier.  I don't know what it makes the circuit shown in the "clean" drawing.

So here is some "detective strategies" to tuck away for future reference, for those just starting out.  The sidechain, or level-detecting portion of the compressor, can use a full-wave or a half-wave rectifier.  That means that it can sense the average level using only the positive (or negative) half-cycles as "information".  This is simple and cheap to arrange (see the Dr Q for an example), but the problem is that it provides a sort of sporadic indication.  That "envelope ripple", as it is called, produces an audible form of distortion that people rarely find useful or desirable.  You CAN smooth this out and make it seem more even, but at a cost of responsiveness - i.e., you can't get smoothness of response and ability to recognize quick changes in the same simple circuit.

By using both half-cycles (positive and negative) as the information source, the level-detection tends to be somewhat smoother.  This occurs for two reasons.  First, the "ripple" is essentially doubled in frequency, so whatever ripple remains tends to be fast enough that we don't notice it quite so much.  Second, it requires smaller capacitor values to smooth out that remaining ripple such that we don't have to pay too big a price, in terms of responsiveness to change, to produce the smoothing.

So, how DO you produce full-wave rectification?  There are many ways, but two methods that come up frequenctly are illustrated here and in the Dynacomp/Ross.  In the classic Dynacomp rectifier circuit, we see a single transistor with equal-value resistors on the emitter and collector.  This is used for producing what is called a "phase-splitter".  Here the signal available at the emitter pin is 180 degrees opposite to the signal available at the collector pin, and both signal levels are equal.  If you stick a same-orientation diode on each of those outputs, you end up with the two interpolated half-cycles being summable as if it were a single-direction-swinging AC voltage that shows overall typical level.  Smooth that out, and away you go.

In other instances, the strategy is to take a second "copy" of the input signal to be level-detected, invert it with a unity-gain op-amp, and sum that with the noninverted version.  So, the two half-cycles are emanating from the same single device, as they  are with a phase splitter, but the outcome is the same: you get the positive half-cycles of one copy, and the flipped-over negative half-cycles of the other copy.  Voila, full-wave rectification.

The advantage the SP has over the OS is that the OS uses the single-diode half-wave approach (more ripple), while the SP uses the full-wave approach.

Make sense?

[markm]

Ya know....
even if I don't end up with a very successful build from this, I have learned quite a bit over this past week and it's made me use my noggin a little bit more than normal.
I agree that this is very educational for lots of us here on the board unfortunately, this thread has been relatively ignored it seems 
Thanks for the info Mark and I too beleive the Hand Drawn schem is the way to go.
I may just get ready and give it a whirl. 

[R.G.]

Actually, I suspect the hand-drawn one is MORE correct.  The 2nd diode you noted as "needing" to be tied to pin 7 would be redundant as drawn in the "cleaner" drawing - it is tied to the same spot at both ends.  In the hand-drawn version, one diode takes it feed from A2 and the other takes its feed from the output of A1.  Since A2 is set up as a unity-gain inverting stage, and since the diodes have the same orientation, that makes a full-wave rectifier.  I don't know what it makes the circuit shown in the "clean" drawing.

OK, I *was* sleepy.

Yes, the hand drawn one is definitely correct, and the cleaned up version incorrect, exactly as Mark has proposed, and for that reason. I missed that line connecting the diode to pin 7. This is indeed a full wave rectifier of the crude sort. It loses not one but two diode drops in rectification. But it works for the purpose, I guess.

The connection of the cap at the pin1/5/6 setup is correct in the hand drawn one too.

I'll go back to sleep now... 

[infinitemonkey]

I have an old Soul Preacher kicking around here. It's probably 70s vintage, as I got it out of a junk bin in the early 80s and it was a long way from new then.  It's in rough shape, and a lot of the off board wiring is missing, but if you want I can post a gut shot later this evening.

Hey, look at me. I posted something.

[markm]

You think you were sleepy?
I called the JFET a MosFET and if that weren't enough, I went back in and edited MosFet into my post after having typed JFET in the first place! I'm better now 
Anyway, BIG Thanks R.G. for the reply, always appreciated for sure.

[markm]

I have an old Soul Preacher kicking around here. It's probably 70s vintage, as I got it out of a junk bin in the early 80s and it was a long way from new then.  It's in rough shape, and a lot of the off board wiring is missing, but if you want I can post a gut shot later this evening.

Hey, look at me. I posted something.

That would be GREAT if you could post some pics!
Hey.....that would be a second post for you and 2 in 1 day 

[Mark Hammer]

Welcome on board Mr/Ms Monkey.    Your misery is only about to begin!

RG/markm, I've been staring at that "cleaned up" schematic as much as you have and din't notice the gaff until markm posted the two side by side.  So, just as guilty as the pair of you.

[TELEFUNKON]

9 hours and 50 minutes not being able to wait for a suitable reply
is being called "impatience" in other parts of the world.

[infinitemonkey]

Sorry for the delay. It was a long day, including trip to Gramma's house accompanied by a cranky toddler, followed by a plumbing emergency when we got home. Thankfully the waters didn't get too deep while we were away.

Here are a few shots of the board. I left the file size big for detail. My apologies if you're on dial-up. There are front and back views, as well as an angle view so you can see the resistor hiding under one of the pots. There's also a sort of x-ray view, made possible by the semi-transparent nature of the board, which may help you to sort out where everything is.

http://www.harrythedog.ca/ehsp/ehsp1.jpg
http://www.harrythedog.ca/ehsp/ehsp2.jpg
http://www.harrythedog.ca/ehsp/ehsp3.jpg
http://www.harrythedog.ca/ehsp/ehsp4.jpg

Let me know if you want more pics. Also, I have no plans for the board itself, so if you want it, PM me somewhere I can send it and I'll throw it in the mail on Monday. It's missing some of the off-board wiring, and it's been kicking around in various boxes for the last 20 years, but if you want it you can have it.

Thanks to all for the warm welcome. I've been lurking here for months, trying to soak up some knowledge. I have a long way to go. I've now doubled my post count in one day, though.

[markm]

Thank you VERY much my friend!! 
umm....you have a PM!

[markm]

The saga continues....
For the power supply section I plan on using 2N3904 trannies, for the buffer section a 2N5088, for the JFET a 2N5457 like the O/S
circuit and for the PNP seen in the schem, a 2N3906.
Anyone care to give an opinion on these?

[Mark Hammer]

Gotta love that "vintage solder job"!   D'they use a whole roll of the stuff for that pedal?

Um, I'm not seeing the 220k resistor I was looking for anywhere on that board.  I do see, however, a 200k resistor going to what could be the gate of the transistor in the corner of the board (I'm assuming the "button" tranny in the middle is the one that sums the two halves of the rectifier and feed the gate of the FET).  What, pray tell, is the number on that particular transistor? (the one in the corner)

[markm]

Um, I'm not seeing the 220k resistor I was looking for anywhere on that board.  I do see, however, a 200k resistor going to what could be the gate of the transistor in the corner of the board

I've been looking at the same thing myself Mark.....I think that resistor is perhaps one of the key elements of this circuit similar to the one in the O/S.
infinitemonkey has been kind enough to donate his board to the cause and this way I can get a first hand look at it and possibly put much of the Soul Preacher mystery to rest.
Apologies for my scrutinizing this circuit but, in a strange twisted way, I've been enjoying the challenge I suppose.
Any thoughts on perhaps my transistor choices?
Also, why do you suppose EH went with a PNP tranny in place of the other 2n5457 the O/S uses?

[Meanderthal]

 Cheaper... an fet used to be kinda high tech, a little more expensive. That few cents multiplied by a few thousand adds up.

[Mark Hammer]

Also, why do you suppose EH went with a PNP tranny in place of the other 2n5457 the O/S uses?

As near as I can tell, the SP does use an FET for the automatic attenuator circuit, just like the OS does.  What it omits is the second FET (and trimpot) that serves a support function for the attenuating FET.  All told, the SP was a more expensive pedal to make than the OS.  The bipolar tranny used was to sum the two half-cycles of rectifier information together.  Whatever money E-H might have "saved" from omitting a second FET, they spent on using a full-wave rectifier circuit.  Better choice if you ask me.

[infinitemonkey]

Heh. If I'd seen this a little sooner I could have read the numbers off the board before I sealed the envelope. I dropped it at the post office about two hours ago, so it will likely be a week or two before you can look at it. Hopefully the Christmas glut in the postal system and at the border won't be too bad.