[sdiy] Negative content on half wave rectifier

[Harry Bissell]

I forget who originally suggested the design idea of using two stages of rectification to
reduce ripple. Its one of our 'music' type people (like Thomas Henry or Craig Anderton although
I'm pretty sure it was not one of them). This was usually used with the NE570 iirc.

Tom hit the nail on the head. This circuit only works with a sine wave (or maybe a random wave)
but NOT with a guitar which is essentially a sawtooth wave (with sharp harmonics, or should I say
inharmonics).  Rectify or double rectify a guitar wave and you don't get a doubling of the
ripple frequency.

The best ides if you want an octave up is to seriously filter the guitar wave before you try and
rectify it. Which BTW kind of spoils the party anyway...


----- Original Message -----
From: Tom Wiltshire <tom at electricdruid.net>
To: Justin Owen <juzowen at gmail.com>

Cc: SDIY List <synth-diy at dropmix.xs4all.nl>
Sent: Fri, 22 Jul 2011 09:57:38 -0400 (EDT)
Subject: Re: [sdiy] Negative content on half wave rectifier

I'm surprised it's that bad. I've used these circuits for 'octave up' designs and I don't remember seeing anything like that much negative voltage on the scope. That's what you'd expect *without* the op-amp, isn't it?
I remember using germanium diodes to reduce the diode drops as much as I could. It was for guitar signals, so the levels were low. I think I had it producing a reasonable rectification down to a few tens of mV. My trouble was that a guitar signal isn't actually symmetrical around zero, so rectifying it doesn't actually produce a signal with twice the frequency - it might have a bit more second harmonic, but the fundamental remains a significant part of the sound, which rather spoils things if you're after an octave effect. I spent ages trying to pre-process the guitar to get it more symmetrical when rectified.</sidetrack>

T.


On 22 Jul 2011, at 14:43, Justin Owen wrote:

> Hello,
> 
> I was recently warned off using a single 1N4001 as a half wave rectifier so I started playing with op-amp based versions. I settled on the 'Improved Precision Half Wave Rectifier' in Fig. 3 here: http://sound.westhost.com/appnotes/an001.htm
> 
> This also shows up on P191 of Jung's op-amp cookbook.
> 
> Jung quotes that, on an AC signal, you'll see a negative swing on the output of approx. -0.6V (which I guess is an inverted diode forward voltage drop?) - in reality I'm seeing between 0.8V and 1V depending on the diodes I use (1N914s were best).
> 
> Is there a way to completely get rid of the -V content or at least improve these figures - so you're swinging between 0 and V?
> 
> I have tried biasing the +v terminal but, for input summing of AC and DC signals it's not a solution. To compensate for the +1V bias I made the 0-5V CV swing between -1 and 4V but obviously that -1V portion of the CV is now getting rectified and lower voltages don't smooth up and down like they do without any bias (but with the -V content) - they 'bump' on and off, I guess as the diodes turn on/off - so it's causing more problems than it's solving.
> 
> For the record, I can live with the negative content - but it is bugging me.
> 
> Thanks,
> 
> Justin
> 
> 
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy

_______________________________________________
Synth-diy mailing list
Synth-diy at dropmix.xs4all.nl
http://dropmix.xs4all.nl/mailman/listinfo/synth-diy

-- 
Harry Bissell & Nora Abdullah 4eva