MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

I want to add a full-wave rectifier to my setup!
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author I want to add a full-wave rectifier to my setup!
Tronketz
What would the circuit look like if I wanted to take the output of a 10vpp waveform, go into a jack --> rectify --> output 10vpp wave?

applause fun fun!... thumbs up
Tronketz
oh shoot, and I then I want to offset it it to +5/-5 after it's rectified
Tronketz
oh and I found this awesome video tutorial on AC/DC current: http://www.youtube.com/watch?v=cyhzpFqXwdA
zomgw00t
Check out the circuits that make up the Tellun Dunsel.
DGTom
google

Jim Patchell Ideal Diode
Neutron7
do you want precision or diode distorted rectifier? if you want precision, you could use one of these. http://sound.westhost.com/appnotes/an001.htm.

then you need to amplify and offset, which you could do with another op amp from a quad TLO74, (if you dont mind the output being inverted, which sounds exactly the same)

you could use the last one, to make a half wave rectifier, and have a switch between full and half wave.

if you want diode distortion then just make a buffered rectifier with a unity gain op amp, then 4 diodes in classic rectifier formation, and then offset/amplify with another op amp. you could reduce input gain while increasing output gain to get more and more diode distortion.
Tronketz
Thanks Neutron7 for those alternative ideas!
Tronketz
Hey guys, congradulate me, something awesome just happened to my future self!!!

I just grabbed the last seat in a Hands-on electronics workshop at the college I'm attending! I thought I would never get in and this is my last semester! YEEEEEAAAH!!! SlayerBadger!

By the end of the year I won't be asking so many questions, I'll be building some circuits! Dead Banana (explosion imminent)
Tronketz
fuck really... I really spelled congratulate wrong? very frustrating EDIT BUTTON MUFF!!!! EDIT BUTTON!!!! cry
KNYST
Pardon for digging up this old thread...
I have been looking through various schematics of full wave rectifiers, and together with some of them, there are component recommendations like:
"Use of high speed diodes, lower resistance values and faster opamps is recommended if you need greater sensitivity and/ or higher frequencies."
"...resistors should be 1% or better"
"Schottky (or germanium) diodes will improve low level and/or high frequency performance..."
http://sound.westhost.com/appnotes/an001.htm

I want to use the rectifier for both cv and audio, so accurate rectification at about 0.01 - 15,000Hz would be nice.
What would be a fast, suitable Op Amp? Is something "normal" as a TL072 too slow?
What is a fast, suitable diode? Schottky diodes, but there are over 8,000 kinds of them at mousers...

Some input would be much obliged!
Navs
From my experience with saw to triangle converters, the state of your input can influence the quality of the rectified results.

If you have a simulator (if not, grab Falstad), in a rectifier circuit mess with the resistor values first to see what happens when these are not matched. Then play with the input, changing its amplitude and offset.

If you want to build a rectifier for use with any input, you could include amplitude and offset control.
Eric G
What format do you want?
Dotcom has one.
CLee
KNYST wrote:
Pardon for digging up this old thread...
I have been looking through various schematics of full wave rectifiers, and together with some of them, there are component recommendations like:
"Use of high speed diodes, lower resistance values and faster opamps is recommended if you need greater sensitivity and/ or higher frequencies."
"...resistors should be 1% or better"
"Schottky (or germanium) diodes will improve low level and/or high frequency performance..."
http://sound.westhost.com/appnotes/an001.htm

I want to use the rectifier for both cv and audio, so accurate rectification at about 0.01 - 15,000Hz would be nice.
What would be a fast, suitable Op Amp? Is something "normal" as a TL072 too slow?
What is a fast, suitable diode? Schottky diodes, but there are over 8,000 kinds of them at mousers...

Some input would be much obliged!


Your link never went anywhere for me, just spun.

I copied this,



from here: http://www.globalspec.com/reference/67834/203279/4-9-op-amp-applicatio ns-using-diodes

In this schematic A2 is an inverting amp for negative voltages and non-inverting for positive. The better matched all the resisters are the more accurate the rectification will be. You want unity gain for any state. If the 3 resistors around A1 don't match it will have different gain for positive and negative voltages. Same goes for A2, you want the gain as an inverting amp to be the same as when it's a non-inverting voltage follower.

The diodes have a voltage drop across them which will limit the accuracy of the rectifier when the input is around +/- 1.5v. When below that A1 operates open loop. Using schottkey diodes will improve that because they have a lower breakdown voltage. Which to use, I don't know. I'm also confused at the variety of doides.
KNYST
Part from what I already said, maybe I should have specified I'd like to use the rectifier for this only:

I won't even have any pots, just in-out.

@Eric G: I'm not looking for a built module. It will be a part of a banana frac panel w00t
CLee
The circuit I just posted will do that. It's a standard cookbook design.

Craig
KNYST
Thanks Craig.
Yes, I have a couple of full wave rectifier schematics to choose from, most of them are very similar. (I couldn't get that url to work now either)

The tricky part is to figure out what to use. Anyone built a full wave rectifier with good/accurate results?
Preferably over the whole range of an oscillator.
Would be much interested in knowing what specs you used for the components.
Navs
I tried over 10 variations of the rectifier for my application (I had a very tricky input signal!). I solved my issue another way, but the best rectifier I found was the SSL precision circuit described by Elliot Sound.

Do you need the waveform to be rectified and hence doubled in frequency or do you just want to level shift it? If the latter, there are easier circuits. Either just attenuate and offset or use a diode clamp and offset.
KNYST
Navs wrote:
I tried over 10 variations of the rectifier for my application (I had a very tricky input signal!). I solved my issue another way, but the best rectifier I found was the SSL precision circuit described by Elliot Sound.

Do you need the waveform to be rectified and hence doubled in frequency or do you just want to level shift it? If the latter, there are easier circuits. Either just attenuate and offset or use a diode clamp and offset.

OK, I'll check that SSL precision circuit out out.
The latter. I want the frequency intact!

If anyone interested missed the previous posted link:
http://sound.westhost.com/appnotes/an001.htm
I found it a good read with a lot of rectifier variations.
Nantonos
KNYST wrote:

I have been looking through various schematics of full wave rectifiers, and together with some of them, there are component recommendations like:
"Use of high speed diodes, lower resistance values and faster opamps is recommended if you need greater sensitivity and/ or higher frequencies."
"...resistors should be 1% or better"
"Schottky (or germanium) diodes will improve low level and/or high frequency performance..."
http://sound.westhost.com/appnotes/an001.htm



Quote:
Graeme symmetrical rectifier

Note: The above rectifier circuit has been attributed to SSL by Rod Elliot but I first saw it in the 1973 edition of "Applications of Operational Amplifiers," Jerald Graeme, Burrr Brown. It is used in the SSL Phase Meter.


KNYST wrote:

I want to use the rectifier for both cv and audio, so accurate rectification at about 0.01 - 15,000Hz would be nice.
What would be a fast, suitable Op Amp? Is something "normal" as a TL072 too slow?

Probably. I found this comment:
Quote:
Graeme talks about how for small signals the required slew rate is the input signal slew rate times the open loop gain. Which is a lot even at modest frequencies. As a practical matter he then discusses that its limited by SR and he uses that point to introduce the reduced transistion voltage half-wave circuit.


The Burr Brown note PRECISION ABSOLUTE VALUE CIRCUITS. By David Jones and Mark Stitt has a good discussion including suggested op-amps for a high-speed version, and compensation capacitor value (based on the unity gain bandwidth of the second op-amp). They also have variants using instrumentation op-amps to avoid the need for close tolerance resistor matching. They don't give any suggestions for diodes however.

I was planing to test the "High Speed, FET Input" option (opa2134-based), preceeded by an inverting summer of the input signal and a variable DC offset, then a unity gain buffer to drive the rectifier stage from a low output impedance. I was hoping an AC-coupled output buffer would give me a nice, DC-free rectified output signal (sine in gives folded sine wave out, with prominant second harmonic).
KNYST
^
Great post! Lots of interesting stuff to go through.
(and I like your avatar!)

This is what I bought for my rectifier projects:
LT1013 - quite fast and not too expensive
0.1% resistors
regular 1N4148 diodes...

Unfortunately, I have nothing else to share yet, as I haven't even made a single attempt smack!
Navs
KNYST wrote:
The latter. I want the frequency intact!


Then maybe you don't need a rectifier hmmm..... Have a look at the diagram in Nantonos' post - the sine is doubled in frequency. You can test this if you have an envelope follower in your system, like the A-119 which can be used to double frequencies of LFOs etc.


Quote:
I haven't even made a single attempt


Go for it! Depending on how precise you want things, you could use trimmers and a single/ dedicated input source. You may even have to calibrate the rectifier to your PSU. For most jobs, it'll be close enough for rock 'n' roll, but for saw to triangle shaping, for example, you'll want to hone it. One thought I had was making jam-able some of the parameters you would normally trim and forget, like an input DC offset.
EATyourGUITAR
I am also very interested in some of the schematics posted that use only a dual opamp. if you build some of them, report back.

let me know what you think of this.
Nantonos
KNYST wrote:
^
Great post! Lots of interesting stuff to go through.
(and I like your avatar!)

Thanks! I drew the avatar myself a long time ago.

KNYST wrote:
This is what I bought for my rectifier projects:
LT1013 - quite fast and not too expensive
0.1% resistors
regular 1N4148 diodes...


LT1013 has some good properties for CV (low voltage offset of 40μV compared to TL072 6mv, low drift of 0.4μV per month) but speed is not one of them. The slew rate is 0.4V/μs compared to TL072 13 V/μs and OPA2134 20V/μs.

To see the effect of slew limiting, look in the Burr-Brown PDF I linked earlier and compare the 20kHz waveforms with the 2kHz ones.

I agree that 0.1% resistors are sufficiently cheap nowadays that using them rather than 1% anywhere the tolerance is important makes sound economic sense. 0.05% and below are still wildly expensive though.
EATyourGUITAR
Nantonos wrote:
I agree that 0.1% resistors are sufficiently cheap nowadays that using them rather than 1% anywhere the tolerance is important makes sound economic sense. 0.05% and below are still wildly expensive though.


I just started using SMD resistor arrays. they might be %5 to the marked value but they are still matched better to each other. if you look for 100k resistor arrays on mouser, some of them are very cheap.
batchas
I’d like to be able to send bipolar signals like -5v/+5v (sines triangle, square etc), or positive like +10v into an input which is made to accept 0v/+5v.

If I do the full-wave rectification posted in this thread (like this one for instance), then I don’t get what is shown in the question KNYST asked (his post included this illustration).
I get this instead:

And of course with a squarewave at the input, I’ll get a flat line at the output, means not usable.

I made a simulation here: https://www.falstad.com/circuit/circuitjs.html
You can paste this text to see what I mean:

$ 1 0.000005 10.20027730826997 57 5 50
v -16 208 -16 80 0 1 40 5 0 0 0.5
g -16 208 -16 304 0
r -16 80 144 80 0 1500
r 144 80 336 80 0 2000
w 144 80 144 144 0
a 464 192 592 192 8 15 -15 1000000 -0.0003374123035267002 -0.00027201809950069204 100000
O 592 192 720 192 0
d 336 192 336 240 1 0.805904783
d 336 80 336 192 1 0.805904783
r 336 80 464 80 0 2000
r 464 80 592 80 0 2000
w 144 144 144 176 0
w 192 176 144 176 0
w 144 176 144 272 0
w 144 272 192 272 0
w 192 208 -16 208 0
w 288 272 336 272 0
w 336 272 336 240 0
a 192 192 336 192 8 15 -15 1000000 0.0000708580856696644 0 100000
w 336 272 464 272 0
w 464 272 464 208 0
w 464 80 464 176 0
w 592 80 592 192 0
r 192 272 288 272 0 2000
o 0 64 0 4098 5 0.025 0 2 0 3
o 6 16 0 4098 10 0.00009765625 1 1

Not sure the rectification is the right way I should go...
I thought I might need to attenuate to -2.5v/+2.5v then offset by +2.5v, but then what happens with higher signals, like +10v for instance?
So my question is:
Which circuit do I need to add after the CV IN to be able to send bipolar signals like -5v/+5v, or positive like +10v into a CV which is made to accept 0v/+5v, without using pots which need to be set manually.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group