Please help me clip this thing

[KarenColumbo]

I recently had some hours spare time again and stumbled upon this circuit:
http://www.tubecad.com/articles_2002/SRPP_Deconstructed/page24.html

I breadboarded it with J201s and what I had resistor-wise (as close as possible) and it worked at first try. While the "effect" is extremely subtle it somehow colours the sound of the guitar quite well imho, it even begins to clip the signal a bit. Which is probably not intended. What it does is making the guitar a bit more airy or transparent.

Of course now I wonder

• where to put some LP/HP combinations to maybe tinker with a "Tone" pot.
• And of course maybe there is some room in there to add clipping diodes or drive the second jfet pair over the cliff.

I grounded two 914s (at least I think they are) in the path from between the first jfets drain and the 7k5 resistor to the coupling cap right before the 300R at the lower right jfet. This just attenuated the signal quite a bit. Duh. Well ... at least I tried.

Or should I just drive the whole thing with a gain stage beforehand? Or drive a gainstage with this circuit?

I hope I don't get on your nerves with my constant fishing in the dark ...

What would you do?

[Ice-9]

Have a read at R.G.'s Geofex site on the mu-amp and the srpp variant for Jfets, this should give a good understanding of what you need to look at.

[Kipper4]

Well if it needs a booster it ain't doing it's job.
I didn't look at the doc, so I'm assuming it's a shunt regulated push pull.
You should be able to get the boost enough to overdrive any thing that follows it.
Why do you want a tone stack?
Let me go look at the docs.

Edit
Here's the ROG page

http://www.geofex.com/Article_Folders/modmuamp/modmuamp.htm

[amptramp]

If your output impedance is sufficiently high, the SRPP stage will clip because you are varying the current between the upper and lower FET and the difference has to go to the load.  If the load impedance is too high, one of the ET's will saturate and you get clipping.  You can design a tone stack with various levels of impedance so that you can clip or not.

You should not need a gain stage to drive it - it already is a gain stage.  The clipping threshold is set by the load impedance.

[PRR]

> clipping threshold is set by the load impedance.

And the supply voltage. Give it less to work with, it'll clip sooner.

[KarenColumbo]

Now that I tinkered with it - and put a ROG Fetzer deluxe after it, I kind of fell in love with what it does to my guitar.

From what I gathered so far, I built a special variation of a

Phase Splitting (Long Tailed Pair) Push-Pull Amplifier.

This is, as far as I can see (not very far yet), typical power amp topology. Which is probably why I should NOT let it clip the signal. Not that i want that now, considering it really sounds very nice - it brings the guitar signal to life somehow. The dull cheap EMGs really sparkle when I switch it in.

There's still 2 problems:

1. It it quite noisy - I have to somehow tone down some aspects of it. I'm still reading into the chapter "negative feedback" to compliment hiss and other landfill out of the signal. Could be the breadboard - at least partially, I had much quieter things on the board. Perhaps I can play with R12 (40k) and make it a trimpot?
2. I really have to pad down/attenuate the input signal somehow. Buffer? "Sensitivity Pot"? What would you do? The 500k pot variable resistor at the input injects a lot of hiss and hums when I touch it.

Thanks for bearing with me - my learning curve gets steeper

Here's the board:



Here's the schematic:

[KarenColumbo]

I'd like to add: I'd really like to avoid any op amps in here - just for learning purposes.

I added a few things:

• The back-to-back LEDs (red) are taken from here: http://runoffgroove.com/ginger.html

. They are said to avoid clipping and retain the sound

• The fetzer valve at the end of the circuit was done according to the respective ROG specs - I even measured the 5457 and chose Rd and Rs (R19 % R20) accordingly
• I made R6 2k2 to try and push the thing over unity gain.
• C8 should be 22n - I took the wrong cap out of the box (very similar optics, those "creamies") - could not hear any difference, though
• C4 and C13 were suggested in the excellent Fetzer Valve article to emulate miller capacitance, I think
• The TubeCad article didn't specify any capacitance values at all, so I took what I had at hand without exercise - hence 22u and 100u for those coupling polcaps

Updated the schematic with a series resistor at +9V