Introducing The Biped, a 9v Biphase workalike using 4049 and simplified LFO.

[nelson]

Hi Folks

I want a mutron biphase clone, I dont want to spend a fortune on 14 optocouplers, I dont want to run it from a bipolar 15v supply and I dont want to use 1meg rev log pots.

So, I have drawn this up. It uses the same filtering and phase stages as the Bi phase but has a simplified and modified "workalike" LFO and uses 2 X CD4049 for the VCR.


I am posting this here to get suggestions on this one. Obviously it is a big build and I am no expert so I would love it if someone could look over it to make sure there are no errors and to suggest improvements.

http://www.4shared.com/file/1883493/ccb10fe/the_biped.html


Thanks
Paul

[Eb7+9]

Nice work Paul,

the one thing that jumps at my eyes is the 100uF cap at the output of your Vb source - you don't need it because the Zout of your op-amp is way lower than anything your cap will provide - it doesn't help in that respect .... actually, in that configuration it can make some op-amps heat up ... try finding the National Semi app note on running op-amps on single supplies - a small valued cap is often stuck on the voltage devider side in front of the op-amp to shunt high-F noise, not after it ... everything else looks really groovy in there except for the front op-amp whose inputs seem reverse labelled, not sure about the mix op-amp too ... minor typos ...

~JC

[nelson]

Nice work Paul,

the one thing that jumps at my eyes is the 100uF cap at the output of your Vb source - you don't need it because the Zout of your op-amp is way lower than anything your cap will provide - it doesn't help in that respect .... actually, in that configuration it can make some op-amps heat up ... try finding the National Semi app note on running op-amps on single supplies - a small valued cap is often stuck on the voltage devider side in front of the op-amp to shunt high-F noise, not after it ... everything else looks really groovy in there except for the front op-amp whose inputs seem reverse labelled, not sure about the mix op-amp too ... minor typos ...

~JC

Thanks JC,

Your right, I have updated the schematic correcting what you pointed out.

Paul

[Processaurus]

Wow, that sounds awsome, but the link isn't working for me...

[nelson]

Wow, that sounds awsome, but the link isn't working for me...

http://www.4shared.com/file/1883493/ccb10fe/the_biped.html


Sorry, link changed since I updated it.

[Unbeliever]

In my experience the square wave output from that kind of LFO needs some attenuation (compared to the triangle wave) if you want them to have the same max sweep. Looks good though.

[bioroids]

You may have to readjust the bias to accomodate variations in the supply voltage. In my experience little variations have a lot of inpact on the bias setting, so I'd recomend to run the LFO, Vref and bias from a fixed 5v supply.

Luck

Miguel

[nelson]

You may have to readjust the bias to accomodate variations in the supply voltage. In my experience little variations have a lot of inpact on the bias setting, so I'd recomend to run the LFO, Vref and bias from a fixed 5v supply.

Luck

Miguel

I think I read your previous post on the bias sensitivity.

Adding a 5v regulator would free up an opamp for something.

Any ideas?

Envelope control?

In my experience the square wave output from that kind of LFO needs some attenuation (compared to the triangle wave) if you want them to have the same max sweep. Looks good though.

Let them vary.

This one should fit in a 1590DD. 

[Unbeliever]

Well, if you are still taking suggestions - and if you haven't done so already - search through the archives on '4049 phaser' to find tips on avoiding distortion and noise. By running off a lower voltage and using the 4049s you're losing a lot of the 'charm' of the biphase IMHO - high headroom, low noise. A 12-stage phaser it might be, but it's a bit of a distance from the Biphase 'sound'.

[bioroids]

I'm still working on a phaser with the 4069, and it's true you chave to carefully balance the noise with the distortion.

On the bias stuff, I'm running a 78L05 regulator and get various Vrefs from there.

One fixed at more or less, 3v which replaces the standard V1/2 (this is the bias to the opamps and to the source of the Nmosfets in the 4069/49).

Then I run the bias pot between the 5v and the 3v.

The LFO is run of the 5v supply also, with it's own Vref to avoid coupling the LFO noises to the audio.

This is a very stable setup for input voltages from 7v to 12v (maybe more). The opamps are not biased at exact V1/2 but that is not really important, as long as you have enough headroom.

The good part is that you don't have to match fets, and you can get pretty wide sweeps. Somehow I can't get them to be too deep, though.

And still experimenting on the noise/distortion issues.

Luck!

Miguel

[nelson]

Well, if you are still taking suggestions - and if you haven't done so already - search through the archives on '4049 phaser' to find tips on avoiding distortion and noise. By running off a lower voltage and using the 4049s you're losing a lot of the 'charm' of the biphase IMHO - high headroom, low noise. A 12-stage phaser it might be, but it's a bit of a distance from the Biphase 'sound'.

Ofcourse, its not an ideal solution or exact repro.

I am hoping accurate biasing and tweaking the divider before the phase stages to get as high a signal to noise ratio as possible will get the best out of this. I understand that this is only a workalike of the biphase and harbour no illusions that I will nail the sound. Still it should be a highly tweakable circuit that doesnt require many expensive parts, matching of components, hard to find parts or its own power supply. It should give a good biphaseISH sound seeing as the signal path is near identical.

I think this version is more accessible to DIY as a guitar pedal than the original.

Guitarists seem to like other 4049 based phasers such as the bad stone.

If optos dont earn ROHS exemption the distortion + noise trade offs will be mandatory. Running a circuit similar to this at 15 to 18v would also help restore the headroom of the original.

It is a workalike and not a clone.

I dont have the desire to clone the biphase exactly, it wouldnt suit my purposes. I am just on a kick for a decent true stereo CV phaser, hopefully people will build it. I am sure you understand infinitely. I dont plan on having the CV section as simple as shown in the schematic.

[Unbeliever]

Sorry, I was looking at your original words - "I want a Biphase clone" - thinking you wanted to keep the original 'tone' of the Biphase. But you are correct - for guitar, it will probably be fine. I guess I'm biased because I like running synths and electric pianos into effects without wanting to worry about them overloading! So, ignoring this bias there is still useful info in the archives from Mike Irwin on such schemes.

[MartyMart]

Nelson, got the linked PDF file but all the "text" is just wierd "blocks" !!
Could be a mac thing ! .....

MM.

[trevize]

wow this is a long awaited project! Can't wait to see it with a layout!

[nelson]

Nelson, got the linked PDF file but all the "text" is just wierd "blocks" !!
Could be a mac thing ! .....

MM.

Open them up in acrobat. For some reason the program I use to make the PDF's doesn't agree with macs preview option.

[nelson]

wow this is a long awaited project! Can't wait to see it with a layout!

I havent started the layout yet. But I plan to.

[christian]

This one should fit in a 1590DD. 

I would have suggested doing the feedback/mixing circuitry differently, but you probably couldnt fit a few extra pots here..
Just some adjustable phase-switchers to mix in out of phase and to feedback out of phase.

Though you can still put +/-feedback there, by using the second opamp as phase-switcher.

I'd also suggest to use a servo mechanism for the CMOS inverters to stabilize the bias, youd loose 1 stage per 4049 and have to use another opamp

This is somewhat to what I had in mind for a phaser project but I'm out to use PWM, theoretically should give you infinite range.

ch.

[nelson]

Hi folks

All the problems with the the CD4049 VCR highlighted in this thread convinced me to just go the optocoupler route.

I think I also got a case of creeping features: The phaser design now has, Square, triangle and sawtooth waveforms. An envelope follower for each phaser with a trigger, a VCA for each phaser, a true blend control, a CV in for rate on each LFO, a CV control of the sweep on each 6 stage phaser and just about all the CV routing options I could think of.


Anyway, here is the schematic.

http://www.4shared.com/file/1937920/4d007e3f/ultima.html


I have had to go the vertical PCB route in order to fit all this circuitry in a 1590DD.

Here is a basic mock up of the potentiometer and PCB placement.

http://www.4shared.com/file/1938117/ae39f9b1/mock_up.html


I am also looking for the keen eyes of all you experienced individuals to highlight any problems.

Any problems you spot that I have overlooked could save me many hours tearing my hair out, so if you see anything, please post.

Thanks

[Unbeliever]

Have you built it yet?

[nelson]

I havent finished designing all the PCB's yet.