[sdiy] Noninverting precision half-wave rectifier?

[Jerry Gray-Eskue]

<<Did it give the lopsided output I described?>>

If I followed your description correctly, yes. This makes sense when the
second stage of the full wave rectifier is considered, actually it is
detailed in the Fig 7-9 for the Positive and Negative case.

It is more interesting when one of the diodes is reversed, but I think it is
then just a case that you described as:

<<...a (non-inverting)precision diode ... high input impedance but suffers
from slew-rate distortion at high frequencies (depending on the opamp
used)...>>

The output is half wave rectified but the output behind the diode is hitting
the rail leading to the slew rate issue you described. This distortion is
very apparent at 2 kHz on a scope, I was using a TL074CN.

- Jerry


-----Original Message-----
From: David G. Dixon [mailto:dixon at interchange.ubc.ca]
Sent: Saturday, September 26, 2009 10:00 PM
To: 'Jerry Gray-Eskue'; Synth-diy at dropmix.xs4all.nl
Subject: RE: [sdiy] Noninverting precision half-wave rectifier?


Yes, the inverting configuration you had there is indeed the classic
inverting half-wave rectifier shown in Figure 7-2, page 134 of C & D.  The
non-inverting configuration you had there is something else entirely;
specifically, the first half of a high-impedance absolute value circuit
(opamp "A" of Figure 7-9, page 141), which produces the +1/-2 gain response
I mentioned, but which, when fed to the second half of the circuit (opamp
"B" of Figure 7-9, page 141), gives full-wave rectification at unity gain.

Did you try breadboarding the non-inverting configuration?  Did it give the
lopsided output I described?

By the way, you can use a DPDT On-On-On switch (soon to be available in
Taiway mini-toggle format from Small Bear Electronics) to get +, +/-, and -
polarities from a precision half-wave rectifier.  You can attenuate this
signal with a pot of the same value as the input resistor, wired as a
variable resistor in the feedback loop.  I'll try and draw the schematic
below (don't laugh!).  I'll let you figure out which switch positions give
which polarities:


In----RRR----+----------------------------------+---|
             |                                  |   |
             |                          /O---|  |   |
             |                         /     |  |   |
             |               |---|>|--O  O---)--|   |
             +-|(-)\         |               |  |   R
               |op amp-------+               |  |-->R
             +-|(+)/         |               |  |   R
             |               |---|<|--O  O---|  |   |
            GND                        \     |  |   |
                                        \O---)--|   |
                                             |      |
                                             |------|---> Out