LFO Sine and Cosine

[jeff]

 Been a while since I was on the forums. Hope everyone's well.



Can anyone share a circuit to take the typical Fender trem oscillator's sine wave to create a sync-ed cosine wave. Or, in other words, one LFO wave and another LFO wave 90° out of phase to the original?
I wanna try something.
 
Thank you.

EDIT: If I added an extra RC filter then took one signal of the plate, one of the 2nd filter, would those two be 90° out of phase?


P.S. N', Oh yeah, what'd I miss since I've been gone?


Jeff

[AlNewman]

Perhaps this is what you're looking for?  I'm not sure if I understand the question, but the outcome seems similar.

https://el34world.com/charts/Schematics/files/Alamo/Alamo_montclair.pdf

[tubeswell]

The plate of the LFO will be 180 Degrees out of phase with the grid. (The C/R network between the plate and the grid adds another 180 degrees to make the thing self-oscillate.)

[jeff]

Perhaps this is what you're looking for?  I'm not sure if I understand the question, but the outcome seems similar.

https://el34world.com/charts/Schematics/files/Alamo/Alamo_montclair.pdf

Interesting schematic. Not sure how those two stages are wired(cascade?)
Looks like two LFO signals are 180• out of phase. May be wrong. I'm looking to make two signals 90• out of phase.
Maybe I could add a 4th RC then take one signal from plate, one from 2nd filter?

[jeff]

The plate of the LFO will be 180 Degrees out of phase with the grid. (The C/R network between the plate and the grid adds another 180 degrees to make the thing self-oscillate.)

I dig,
Here's an idea:
What, if I added a 4th C/R filter and took one signal off the plate and one off the 2nd filter?

[kagliostro]

Ciao Jeff

I think you'll find interesting if not exhaustive the content of this link

https://www.aggh.net/discussion/index.php?topic=30731.0

About Merlin's Vibrotron you can find detailed info also here
(on the previous thead linked the Vibrotron link is out of service)

https://ampgarage.com/forum/viewtopic.php?t=33565

Quote from our missing friend Roly - R.I.P.
Now where have I seen this Vibrotron before?

With a couple of interesting variations this "VibroTron" is basically the "vib/trem" circuit of a Vox AC30 (and functionally similar to the Fender "Harmonic Vibrato").

{As an aside; the use of a LED to bias some stages is novel and valid, however you need to be aware that the colour of the LED is important since different colours have different cut-in voltages, and thus bias points.  You should also be aware that LED's are photo-sensitive and if you mount the LED where stage lights can get to it they will shift the bias point.}

I'm not going to pretend for one second that this, and Leo Fender's very similar "Harmonic Vibrato", are anything other than the most difficult to understand circuit you are going to encounter in a guitar amp (and possibly the most widely misunderstood and misrepresented).

The input signal is conventionally amplified, then phase-split to produce an inverted mirror-image.

These two signals are then fed to the capacitor/resistor networks in the middle.  This apparent mess of components is in fact two quite similar filters. and this is where most of the misunderstandings occur with the simplistic assumption that they are just low and high pass amplitude filters, somewhat like an active crossover - but it's not that simple.

Normally we employ a frequency-selective filter to change the amplitude of the signal, such as in a tonestack, but frequency-selective networks also have the property of shifting the phase of the signal, or introducing frequency-dependent delay from input to output.

In this case these filters are what are called "all-pass" because the intent here is to employ this frequency-selective delay rather than just simple amplitude attenuation.

{Amplitude variation can be obtained far more simply, and I have to wonder at the implicit arrogance of those who think they are smarter than Tom Jennings and Leo Fender; who don't stop and ask themselves why these two very notable designers put in all those extra valves and components if only just to produce amplitude modulation.  Every time I look at these circuits I can hear Tom and Leo laughing their arses off at the know-nothing dweebs who totally miss the point.}

These two identical filters are certainly tuned to different frequencies, but together with being fed anti-phase by the phase splitter the overall effect is that the outputs are in quadrature, or 90 degrees phase difference, over most of the audio passband, and certainly midband which is our main interest.  THAT is what Tom and Leo were doing here.

The following (final) double triode is a balanced mixer fed by these two output signals, but also by the vib/trem oscillator via its own phase splitter, so the main output is thus smoothly varied (not "switched") between the two filter outputs.

With all the circuit active the effect of this is to progressively swing the phase of the signal back and forward by 90 degrees (+/-45deg) - in other words phase modulation.

Franco

[jeff]

Hey, Franco
Been a while. This looks very interesting. I'll need some time to fully digest this all.
I guess my question is more to do with the actual LFO signal itself, not the vibrato circuit it controls, but the actual LFO wave.
This uses a phase inverter to get two LFO signals that are inverted to control the vibrato circuit. I'm trying to figure a way to generate a 3rd LFO signal to control something else that is in sync with the original LFO signal(and its inverted version) but 90° out of phase with either original LFO wave.


For example:
The LFO goes to a phase inverter
One signal turns one tube of the vibrato circuit down as the other turns the other up
Right?
Sig 1 Sig 2
0.       0.
+V.     -V
0.         0
-V.      +V
What I'm looking to do is to generate a 3rd LFO signal that is 90° out of phase with those two signals.


So in theory the 3 LFO waves would cycle:
Sig 1. Sig 2. Sig 3.
0.        0.       +V
+V.      -V.        0
0.         0.       -V
-V.      +V.        0


Dig?

[jeff]

This could be used to control something else that would be in sync, but also on a different on off cycle

[kagliostro]

In the thread I linked there are explanation about how they achieve a 90° signal but the circuit isn't simple, obtain that via some filters and don't acts at the same level on the whole bandwidth, the more, as to obtain the 90° they use a -45° & +45°

I don't think it is simple to go further than that, also to obtain that isn't simple

Franco

[jeff]

In the thread I linked there are explanation about how they achieve a 90° signal but the circuit isn't simple, obtain that via some filters and don't acts at the same level on the whole bandwidth, the more, as to obtain the 90° they use a -45° & +45°

I don't think it is simple to go further than that, also to obtain that isn't simple

Franco

I'll have to read that again.
But to my understanding they are using the filter to get two copies of the input signal so that all frequencies of the input signal are 90° out of phase. I believe that signal filter's low end is only 150HZ. Much higher than the LFO's 6-8Hz.
The LFO is split with a phase inverter to two copies 180° out of phase(plate, cath) to fade in and out the two 90° out of phase signal paths.


Think of it this way:
With an LFO and two LED's you can make the two LED's blink.
With an LFO through a phase inverter you can make two LED's alternate.
I'm looking to get a 3rd LFO signal that is 90° out of phase with the original LED sine wave AND the 180°, inverted LFO sine wave so it's blinking in between the two original alternating LED's


Not using it to blink LED's. That's just the best example I can think of.


Vibrato uses two LFO signals, one of the plate, one of cathode. So essentially, 0° & 180° LFO wave(not talking about guitar signal phase)
I'm looking for a LFO circuit with 0° 90° 180° 270°


Does that make sense?

[glass54]

Hi Jeff
This may be of (intellectual) interest to you. Far from using Tubes designs but has "0° 90° 180° 270° outputs and I know that you are NOT trying to drive LEDs 
https://www.eddybergman.com/2020/04/synthesizer-build-part-27-quadrature.html
A few decades ago, I was a member of the Australian Synthesizer Club, when Moog stuff was WAY TOO EXPENSIVE and eventually the CURTIS Chip appeared (LSI, Synth on a chip!!). I used to design PCBs for a mate who did the schematic designs for Synths modules. My other specialty was knowledge of high end Audio ICs, critical for stable synth designs esp VCOs and VCFs.
Trust this is of interest BUT in a very different rabbit burrow direction and requires good knowledge of ICs.
Kind regards,
Mirek

[jeff]

Thanks,
I'll give it a look.
Cheers

[jeff]

That's it!
I have little experience working with IC's but that's it.
Not sure how well tubes 'n IC's would mix.

[Merlin]

Feed your trem into a cathodyne to split the phase, then into the complex RC network to create two signals 90 degrees out of phase (values must be tweaked to suit the LFO frequency range). Two tubes, lots of RC's, lots of tweaking (use a circuit simulator). This is also called a quadrature oscillator. A great resource is the vibrato edition of stompboxology here:
https://app.box.com/s/fc83561b9625e7f4958f

And also this article on the Wurlitzer vibrato, which my Vibrotron was based on.

EDIT: This thread has branched off to here: https://el34world.com/Forum/index.php?topic=31833.msg350623#msg350623