Centre Grounding question

[col]

I have a tremelo I'd like to make from an old Babani book which involves 5 Ge transistors. The circuit though is centre grounded with +4.5 and -4.5v. Pretty easy to achieve from 9v but would I be able to use it with the same power supply as +ve or -ve earth pedals?

[gez]

Shouldn't be a problem so long as the ground reference is 'stiff'.  How is the split done in the schemo?  Or more importantly, how are you going to do it?

[col]

Gez,

pm sent with link. I was intending splitting a 9v supply with 2x 100K resistors and 2 caps of 22u to +9v and 0v

Col

[gez]

As I said in my reply, I'd use an op-amp wired as a follower to provide the ground reference.  What you've outlined isn't going to work.

There's also that power amp RG suggested using for a split supply as it doesn't use any bias resistors.  I don't use it so I don't remember the name off-hand (plus my brain is like swiss cheese these days    ).

[gez]

Actually, that circuit could be reworked for single-supply (positive ground) with few changes.  PM sent.

Will the changes to the circuit be enough for Babani not to crush your testicles in a vice if you redraw it and post it, though? 

[col]

Got the PM thanks. Will give it a go as soon as I get time. Will it matter what the two transistors in the oscillator section are as they are not in the audio chain? I suspect that they will have to be closely matched though.  I have some functional 'red spot' Ge s that generate too much hiss to use in an audio situation and that would let me use better ones in the audio path. If it does work I'll post it unless someone thinks I shouldn't as by the time it's finished it will be significantly different to the Babani circuit. Most of the circuits from the early books have already been posted as they were originally electronics magazine articles.

[gez]

Will it matter what the two transistors in the oscillator section are as they are not in the audio chain? I suspect that they will have to be closely matched though.  I have some functional 'red spot' Ge s that generate too much hiss to use in an audio situation and that would let me use better ones in the audio path.

Doesn't matter what they are, and there's no real need for matching.  The trannies act as switches so as long as the Ge devices aren't leaky as hell they should work.  Would be better to use Si though.  They'd need to be PNP as the redrawn circuit is now +ve ground.  Personally, I'd use NPN Si for the whole circuit and flip the supply to standard +9V.  There's no real advantage to using Ge IMO and as the bias method isn't exact for TR4 the extra drop of the Si base emitter junction isn't going matter much.  Just tweak the resistor to base of TR4 until the collector of the last stage gives you reasonable swing.

To avoid tick (the LFO kicks out a square wave), provide separate ground paths for the LFO and audio.  Also provide a separate ground path for TR3.

TR4 seems superfluous to me.  Don't understand why they didn't just hook up TR4's bias resistor (originally 47K) to the base of TR5 instead and pull TR4.

If I understand correctly, the circuit was intended to go between a preamp and poweramp.  It was assumed that the load that follows the circuit is fixed.  In pedal form, that isn't going to happen so probably best to make that 22K on the output bigger.  Or replace it with a 500K pot (volume).