EH Soul Preacher UPDATE!

[markm]

Hi folks,
It looks as though it is quite possible that we will after all have an actual working and fully functional EH Soul Preacher Compressor! 
I have the working layout posted here http://aronnelson.com/gallery/album15 in my gallery.
It needs a few more "tweaks" before I would consider it verified but, it does work and it sounds Damn Good if I say so myself.
It's very Orange Squeezer like of course but, with more Comp control and significantly MUCH smoother.
Quick Attack and Quick release.
Clean, clean, clean as a whistle......quite transparent and did I mention how clean it is 
I have more work to do on the power supply as I am only getting 8.3 volts through the 9V side and 3.87 through the 4.5V side.
A little resistor tweaking should help and I'm open for some suggested starting points.
This could become a highly reccommended build from me.....maybe 
Please take a look if you have the chance.
   Thanks!

[TELEFUNKON]

so what`s the difference schematicwise?

I don`t understand layouts.

[Eb7+9]

... xclnt !

[bluesdevil]

Wow, great job on the layout!! I'm looking forward to building this one when the final voltage tweaks are figured out. I've been tempted to build a few of the simpler comps for quite a while, but glad I held out for this one.

[infinitemonkey]

That looks nothing like the original. The traces are all straight and you haven't left room for the gigantic blobs of solder.

Good job figuring this out. I'm glad to hear you've got a version that sounds good. There seem to be really mixed reviews for this pedal. Opinions seem to be very strong in one direction or the other. Opinions vary of course, but for the reviews to be so polarized I have to wonder if there were variations in the circuit.

Noise seems to be one of the biggest complaints in the negative reviews. With this version being clean, maybe a comparison could lead to a nice mod for the factory pedals too.

[markm]

so what`s the difference schematicwise?

I don`t understand layouts.

Here's an early "corrected" schem from me.....
Unfortunately, as I moved on through this project, I neglected to record some of the changes and I will have to go back and look at what I've done so far but, this should give you an idea;

Wow, great job on the layout!! I'm looking forward to building this one when the final voltage tweaks are figured out. I've been tempted to build a few of the simpler comps for quite a while, but glad I held out for this one.

Thanks VERY much!!
This has been a real project for me.....I've put it aside a few times and "gave up" a few more  and today, I got to leave work early and had an hour or two to spend on it and in that time, made some real headway and came up with a working version.
A "christmas miracle" perhaps?!!

... xclnt !

It is pretty darn cool but, still has some slight issues that at least should be more or less "easier" to fix. (I Hope!)
BTW, Thanks very much again infinitemonkey, without you, this would never have come this far! 

[alfafalfa]

Mark,

Why did you change the order of the 220K resistor  and  the 100 n cap between Q1 and IC 1a ?
And  is there a sub for the bc 309 ?

Alf

[markm]

Those are flip-flopped beacuse on the original EH board, the cap goes before the 220K resistor.
(Mark Hammer you were correct on this!  )
The trannies I have been using to sub for the BC239 are 2N3904. They seem to work fine but, I may try others and report back.
My first concern is getting the power supply correct. If I am unable I have considered simplifying the PS as it does seem like overkill with the original style PS.
An interesting note also is the 20K resistor that is shown in the schem is just a jumper on the original board.
I have a 20K right now in my test circuit and plan on seeing what different values bring in this position.
Maybe it's some of the improvements brought to this circuit but I must say, I really think this circuit could give the O/S some competition for being the ultimate transparent DIY comp....Personally I think it's that good. 

[markm]

My first concern is getting the power supply correct. If I am unable I have considered simplifying the PS as it does seem like overkill with the original style PS.

After spending a good couple of hours experimenting with the power supply on this circuit, I think it is safe to say that the voltages shown on the schem will NOT work with the transistor used in my Layout.
I managed to get 8.99V and 4.5V out of the power supply as indicted on the schem but when the voltages are corrected the circuit distorts and gates.....obviously a biasing issue.
So, inturn when using the indicated resistors on the schematic which produce voltages of 8.3V and 4.1V the circuits remains clean with no gating and sounds quite good.
My conclusion is, like it or not fellas  that the schem values should be used and the circuit should be powered at the voltages I have indicated.
To me this does not make sense however, who am I to question success right??
I thought in bringing the voltages up, the circuit would perform even better but, alas, it did not.
I suppose I could have gone on and messed with the biasing of the rest of the circuit but, why??......it works and works quite well indeed!
One more thing I may try is powering up the original EH board and measuring power supply voltages however, it is 30 years old and I'm sure there is quite a bit of drifting going on within the components.
The 20K resistor shown on the schem but subbed as a jumper on the original board is fairly essential to get a good level out of this circuit.
I subbed the 20K with a jumper and lost about half of the volume it produced and gained a whole lot more noise.
This in my theory is perhaps some of the reason the Soul Preacher has a reputation of being noisey. With a 20K as indicated on the schem it performs with gain to spare and no noticable noise.....I mean surprisingly Quiet!
So the 20K is in the new layout as well.
I will be posting this in the gallery as soon as it is finished.
Thanks for following me this far and I hope I haven't lost some of you or put you to sleep through this whole ordeal. 

[markm]

The Layout in the gallery is now Verified.
You can find it here http://aronnelson.com/gallery/album15 both in PCB and Perf formats! 
Enjoy!

[tungngruv]

Thanks for taking this on Mark.

[swt]

Hey mark!! i think i can get those bc239 in here. if you want some i can buy them and send them to you. consider this a way to say thanks to your contribution, and make mine. let me know.

[markm]

Interesting.
I will PM you.

[markm]

Thanks for taking this on Mark.

You are quite welcome Sir!
I may have some slight mods to share on this build as well and will keep ya posted. 

[Meanderthal]

 You've obviously put just a hell of a lot of work into this.... Thank you for that. I have a thing about the old EH stuff... sentimental I guess... but I'm glad you got this up and running, even though the last thing I personally need is yet another compressor, and the old Soul Preacher I had was nothing like what you describe(I wouldn't want another like it, but this one sounds like it's nice!). I just like the idea of bringing new like into that old stuff.

It'd be interesting if you could get the old board fired up, just to make a comparison... I'm wondering if the old one is like the ones Aron and myself used to have.

Merry Christmas!

[markm]

Thanks for the pat on the back meanderthal.
The last thing I need is a comp also but, I truly LOVE compressors and I don't know why.....sick in the head I guess.
I've always been fascinated by comps and I suppose it's because they aren't as straight forward as other effects and compressors often confuse alot of folks.
I like the way this one sounds and like I said, it's obvious from the way it behaves that it's related to the O/S but, smoother.
Here's a work in progress pic and yes....I may just fire up that old board to see how she sounds.
One crucial thing I've discovered in this build is the 20K resistor from the emitter of the PNP tranny, it's jumpered on the original board.  Makes a BIG difference in volume and performance of the circuit!
Anyway, here's the pic;

[Mark Hammer]

I'm finally back home from a week at the in-laws and finally catching up with things.

I echo the praises already justifiably heaped.  This only cements my already great faith in the undertakings of those named "Mark".

A bit of diligence goes a long way, eh?

Note that the 4M7 resistance in parallel with the 1uf cap to ground can be varied to adjust the decay time.  I guess a simple place to start might be simply wiring in a 470k-1M resistor in parallel (along with a SPST toggle) to go between fast/slow decay.

Mark,
Why did you change the order of the 220K resistor  and  the 100 n cap between Q1 and IC 1a ?
And  is there a sub for the bc 309 ?
Alf

Query was directed at the other Mark, but one's as good as another.  The FET is being used like a variable resistor.  In conjunction with the 220k resistor, it forms a kind of volume pot.  When the resistance of the FET gets lower than, say, 1meg, the 220k/FET combo acts like a volume pot turned down.  When the FET resistance starts to get noticeably lower than 220k, the volume pot provides noticeable reduction of the signal being fed into IC1a.  Althought the gain itself does not change the way it does in a Dynacomp or other OTA-based compressor, what you are feeding the gain stage with is reduced via that virtual volume pot (not unlike the way an amp's preamp volume control adjusts what you feed the power...whose gain is also fixed), resulting in less output signal.  The Orange Squeezer, and a number of other compressors and limiters work in similar fashion.

Since the ratio of the 220k resistor and FET resistance is what determines just how much the volume will be turned down, one can make the amount of attenuation gentler by reducing the value of the 220k resistor a bit, such that wide changes in the FET resistance have less impact.  Changing the value of that 220k resistor is akin to changing the "ratio" control on some higher-end rackmount compressors.  You can do the same thing on an Orange Squeezer too, but since it lacks the input buffer stage that the SP does, it is somewhat less tolerant of substantial changes to the value of that resistance I imagine.

One of the things that occurred to me was that if the 220k/FET combo acts just like a volume control, then one ought to be able to use a bypass cap on it...just like a volume control with a "brite" switch...for a little less brightness-killing (i.e., more brightness).  My gut sense is that something like a .001uf cap in series with a 100k fixed resistor, laid in parallel with the existing 220k resistor will provide less attenuation of the highs than the mids and lows.  Note that it will still provide some attenuation when you strum, just disproportionately less in the high end.  Since this "preserved" treble is also going to be a part of what feeds the envelope follower in IC1b, I'm not absolutely certain of the impact on the quality of compression.  Quite frankly, though, it is largely the note fundamentals, and all the stuff below about 1500hz that is providing the brunt of the signal on those peaks that result in volume squashing.  So, even if there is any impact, it will be minor.  If my gut feeling is correct, this should produce a more natural sounding compression, sort of what those who mix some straight signal in with the compressed signal try to aim for.  I'm fully backed up with projects and obligations to attend to so I won't do the experiment myself, but it would be worthwhile if you could attempt it by tacking on the suggested components on the copper side temporarily.  Incidentally, this also lends itself to a variable control for adjustment of "air".  Something like a .001uf cap, 100k pot, and 47k fixed resistor (all in series and placed in parallel with the 220k resistor) might work nicely to adjust the content of what undergoes less compression.  Note that as the parallel resistance gets smaller, there is less compression of what gets bypassed, BUT the corner frequency also goes up.  Conversely, as the corner frequency of the bypassed content goes down, you start to get just as much compression for bypassed as nonbypassed, so you avoid any irritating stridency.  My hunch is that there is likely a nice balance somewhere out there that involves placing a second fixed resistor in parallel with a 250k pot (instead of just a 100k pot), but we'll leave that for another day.

Which brings us to the input buffer.  Though I can't imagine that the BC239 is an atrocious choice for input buffer, neither is it known as a low-noise unit either.  It was simply the default NPN medium-hfe transistor at E-H for a little while.  There may well be other trannies that can yield a quieter input signal.  Note that this applies to the input stage only, and not to trannies in any of the other non-audio positions.

In sum, though, nice work and thanks for the corrections.

The other Mark

[Bernardduur]

Next try the 3080 version??

[markm]

Next try the 3080 version??

Where's the schem?
It's probably a reworked Dynacomp I would think....no? 

[Bernardduur]

Check my thread on the Deluxe Big Muff

The Deluxe is a Soul Preacher (this one 3080 based) and an opamp Big Muff