Biasing the output of an LFO

[mdh]

I think maybe I'm in over my head with what I'm attempting at the moment, but I'm hoping there's an easy answer.  I'm trying to build a phaser based on R.G.'s PWM Phaser in his ASMOP article on GEO.  However, I'm a little lacking in my time for learning uController programming, so I thought that for starters I would just go with an analog LFO and a clock, like the analog envelope follower and clock in the MXR envelope filter.  So far I have the clock from the MXR EF and the LFO from the Small Clone on a breadboard (see schem below; the image here should link to a bigger/clearer image).  The output of IC6D goes to the control pin of a 4066.  Both the clock and the LFO are oscillating, but the DC bias of the LFO output is too low to make the marriage work.  I pulled the LFO output from pin 9 of IC6 and made a buffered voltage divider with a pot and an op-amp to see what range of voltages at pin 9 gave me a good variation in the resistance of the 4066.  I found that the range 3.8-4.7V would sweep from 100R (basically on all the time) to 100k (presumably that would be off 99.9% of the time).  This leads me to believe that if I could make my LFO sweep about 4.3V or so, I'd be golden.  At maximum depth, the LFO seems to sweep from about 0.9V to 2.5V (that's based on the best readings I can get from my DMM since I don't have a scope).  The other problem I see is that the DC bias of the LFO moves toward ground as the depth is reduced.

So, since I've now totally blown it and revealed my ignorance, can anyone tell me how to bias the output of my LFO to cover that range?  Is there another sort of LFO I should be using, or can I modify the Small Clone LFO to behave as I would like?  Thanks in advance!

.

[slacker]

You could move the LFO bias and change the depth by using an inverting opamp stage. Remove the depth control and take the output from pin 1 of the LFO through a 10k resistor to the inverting input of the opamp. Put a pot, say 47k wired as a variable resistor from the inverting input to the output. This works like a depth control when the pot is at minimum resistance you get no output and increasing the resistance increases the size of the LFO waveform.
Then connect a pot, 100k will do, across the supply as a voltage divider and wire the wiper through a 4k7 resistor to the positive input. This lets you move the bias voltage up and down, so you can set it to sweep around 4,3 volts.

Hope that makes sense.

Actually you might be able change the bias of the LFO by changing the value of R7. Making it smaller looks like it would increase the bias voltage.

[mdh]

Ooh, excellent, that did the trick!  Thanks!  Just to be clear, is the "depth control" variable resistor in the negative feedback loop of the extra inverting stage?  That's where I put it, anyway, and in very preliminary tests, it does seem to mess with the output bias a bit.  In any case, at least at maximum depth I'm now getting a very nice amount of variation in the duty cycle going to the control terminal of the 4066 (and a blinky LED), though it may not yet be as smooth as I'd like.  Next I get to add the phase shift stages and actually hear how it sounds.  God I love this place.  Alas, I'm probably not going to get more time to work on it today.  Time to prepare the garden for tomatoes.

If there's any way to accomplish the same goal without adding the extra op-amp, I would be interested in hearing it, though.  E.g., could I do this with a simple transistor stage?

[slacker]

Yeah the "depth control" is between the inverting input and the output, that's what I meant to write 

[slacker]

I ran the LFO through LTspice and found that it swings between about 0.5 and 2.7 volts which is pretty similar to what you got.
Changing R1 to 330k and R7 to 390k moves the bias point upwards and the LFO swings between about 2.5 and 4.7 volts, so that might do what you want. You could try just increasing R1 but that makes the wave form asymmetrical with the up slope being steeper than the down slope.

[mdh]

Thanks for doing that.  Still seems like the original depth control will be a no-go, though, because the center of the sweep will move toward ground as the depth is turned down.  I really need to get a simulator of some sort running.  I did find some more time to play around yesterday, and got together an audio path with two phase stages, but I can't tell for sure if it's phasing, because there's a big ker-thunk from the LFO.  I think that has to do with the new 3rd stage of the LFO, b/c I used an LM358 for the dual, and just added a 741 for that extra stage.  Plus, varying the depth control in the feedback path of the 3rd stage can make the ker-thunk better or worse.  I'm not sure if there's an error in the audio path (it really kind of sounds like a trem with some possible phasing in the background at the moment) or if it's just leaking through the power supply.  I think I'll be reading up on LFO designs in my spare time this weak, but I'm having a blast!