Craig Anderton Bi-phase? Inverting an LFO...

[akestromer]

Hi!
I just finished building craig andertons phase shifter project from the amazing epfm book.
Now I was thinking, what if I build another one and turn it into (sort of) a bi-phase.

I'd like to be able to sync them inverted or out of phase or whatever to get that stereo sweep thing going.
I was thinking sending the lfo from one of the phasers into an op-amp and inverting it there. But I don't know enough about circuit design to figure out how to do that.

Any help on this would be greatly appreciated!

Hope this makes any sense at all!  

Thanks!

[akestromer]

Yeah, so basically what I'm asking is: how do you invert an lfo?

...there's probably not a single simple answer for that but, thought I'd try!

[armdnrdy]

Why don't you take a look at the Bi-Phase schematic?

The basis of what you are asking about....all on one page!

[R.G.]

See my posts on matching LFOs to circuits. LFOs don't exist in a vacuum. They have to be massaged to be the right minimum, maximum, and middle DC levels for the circuit they drive.

Inverting an LFO is easy in theory: just bias an inverting amp at the middle voltage of the peak-to-peak of the LFO and out comes the inverted waveform. What's hard is making that middle voltage be correct, and making it not change with speed and depth changes with the LFO it's inverting. And then matching the inverted LFO to its driven circuit.

Inverting is easy. Figuring out what and how to invert is hard.

[R.G.]

And this one came up just as I clicked "send". It's much more interesting aurally if you don't do an inverted LFO, but do a second LFO that's identical but in quadrature (shifted 90 degrees) to the original LFO.

[akestromer]

I've been googling all over the place to find ideas. Looked at some yusynth schematics too he has an adsr with an inverted output, and a bunch of other great things.
I managed to get my yusynth lfo to control the phaser. Maybe I should just make a different lfo and get a squarewave too!

Anyway, thanks for all the ideas! Now I have some more researching to do!

/Åke

[njkmonty]

Im probably not the one to listen to , but i just happened to be starting to have a go at a biphase build, and was trying to work out all the lfo stuff.

is this the reverse switch section of  the phasor B sweep sync?   

[Mark Hammer]

Craig uses the CLM6000 for both the phase-shifter project AND the tremolo project, and also uses a 4049 invertor chip as LFO in both of those.  One can simply tack on one of the unused invertor sections to the triangle output of the LFo.  In fact, some 16 years ago, Saskatchewan's own Dean Hazelwanter did just that, to produce a "stereo" version of the Anderton Tremolo, with the two tremolo sections counter-modulated (left louder as right gets quieter, and vice versa).

here you go:


One could easily implement that LFO with a pair of opto-isolator-based 4-stage sections.

[armdnrdy]

Excuse me if it appears that the student is pointing out a mistake by the teacher!  

The Anderton Phaser in EPFM doesn't use a 4049 for the LFO. It uses a RC4739.

If phase stages were added to Hazelwanter's LFO as is...it would just produce a medium to fast phaser effect.

The LFO would have to modified to achieve a rate and depth more suitable for a phaser.

I still believe that the OP would benefit from studying the Bi-Phase design to see how the things he wants to achieve were implemented.

The OP wants to sync and invert the LFO...and mentioned adding a square wave to the LFO.

All of that is included in the Bi-Phase design.

[akestromer]

Fantastic! Thanks guys!
The phase shifter from the electronic projects for musicians book doesn't use 4049's but you could definitely use that lfo for it! And that way get square wave too! I might try that, thanks!

One cool thing about the epfm phaser is that you can very easily disconnect the lfo and control the LED sections of all four clm6000's (or in my case nsl32's) with an outside cv source. Like the envelope follower from the same book or an lfo or adsr or something. I just don't know how much the optocouplers can deal with before you fry 'em.

[akestromer]

Armdnrdy: yeah, thanks I've been looking at it. I think the mutron is a little too complicated. I was just thinking there could be an easy solution. Since the anderton phaser already has a lfo that just needs to be magically turned around. And you can easily access it on the "N" pad on the pcb.
Well sometimes you get lucky. Sometimes you don't.
But I think the tremolo lfo should work since it uses clm6000's.

[armdnrdy]

Do you have a bread board?

Since you have the phaser built already...it might be a good idea to bread board the Hazelwanter LFO so that you can experiment with component values to better tailor it for a phaser.

You can disconnect the Anderton phaser's LFO and run leads to the bread board.

[Mark Hammer]

Excuse me if it appears that the student is pointing out a mistake by the teacher!  

The Anderton Phaser in EPFM doesn't use a 4049 for the LFO. It uses a RC4739.

If phase stages were added to Hazelwanter's LFO as is...it would just produce a medium to fast phaser effect.

The LFO would have to modified to achieve a rate and depth more suitable for a phaser.

I still believe that the OP would benefit from studying the Bi-Phase design to see how the things he wants to achieve were implemented.

The OP wants to sync and invert the LFO...and mentioned adding a square wave to the LFO.

All of that is included in the Bi-Phase design.

I think the "teacher" has to stay after school and write out lines on the blackboard, after which he needs to go back to teacher's college and get re-certified!    Yeesh!

After a while, all those older Anderton circuits get blurred together in memory.  I must have seen the tremolo GP article in my mind and mistook it for the phaser article from an earlier issue.

Of course, you are entirely correct: Craig uses a dual op-amp LFO for the phaser.

As for adapting the 4049-based Tremolo LFO for phasing, you are also correct in noting that it won't go slow enough.  Probably god for a Uni-vibe, but not slow enough for a phaser.  But I imagine that is easily addressed by simply making the Rate pot a higher value, like 250k, or even 500k. Since the LFO rate is determined by how quickly that 10uf cap charges up, slowing that charge-up time involves increasing the resistance of R14 + P5.  When P5 is at min resistance, you reach fastest speed.  And since that fastest speed is good enough for tremolo, it is also good enough for phasing.

I guess the one thing I don't have much insight into is whether the current drive of a single invertor section is sufficient to push 4 LEDs, and whether those LEDs should be wired up in series, or in parallel.

[armdnrdy]

I think the "teacher" has to stay after school and write out lines on the blackboard, after which he needs to go back to teacher's college and get re-certified!    Yeesh!

I was just shocked...and amazed! I was in disbelief! No...it just can't be! 
It was kind of like finding out that your dear mother lied to you! 

I guess the one thing I don't have much insight into is whether the current drive of a single invertor section is sufficient to push 4 LEDs, and whether those LEDs should be wired up in series, or in parallel.

Unanswered questions such as these are why internet searches, this forum, and my bread board have become my best friends! 

[LaceSensor]

Have a look at the Lovetone doppelgänger , it uses inverting LFO and I'd based on the Anderson project.