Need some help with MOSFET oscillator design

[rathmann]

I've decided to have a go at my notion of the ultimate signal wiggler amp. Here are the high points:

Three output channels - Center is totally dry running a pair of push-pull tubes; right and left amps are single ended incorporating both Magnatone-like vibrato and Fender "harmonic" trem-like tremolo oscillating out of phase.

I'll be using 6BM8s for the outputs and reverb driver.

I want to use mosfets to run the lfo to save space and provide higher performance and more reliable oscillations. I amassed a frankenstein-like amalgum of circuits proposed by R.G. using the MOSFET Follies, Solid State Replacement of Fender Tremolo, and Single Pot Speed Control for LFOs articles.

Ideally I will have separate depth/intensity controls for the pitch shifting vibrato modulation and the frequency selected amplitude modulation (tremolo). The dilemma is that I'm not sure where to do this. The posted schematic represents one idea of putting the intensity controls in parallel after the oscillator, follower, and phase splitter. Here I'm using a circuit similar to the one specified for eliminating the dual pot speed control by driving vactrols to attenuate both phases of the oscillator signal.

But maybe this would be simpler to just run two followers and phase splitters after the oscillator each with a single pot intensity control.

I'm a total noob when it comes to designing with solid state devices, and I'm only somewhat experienced with tubes because I've looked at a bazillion schematics to get a handle on common circuit concepts. I'm not totally sure what kind of output the MOSFET oscillator will generate and so I'm not totally clear on how much to attenuate to get to the 4volt swing to drive the vactrols of the vibrato.

Any thoughts or feedback would be appreciated. Thanks much!

[rathmann]

Not sure why the images aren't displaying. Here are direct links.

http://music-electronics-forum.com/u...4-albums53-280
http://music-electronics-forum.com/u...4-albums53-279

[Don Moose]

It looks like Q1 is set up by that string of Zeners to regulate to about 200V, which I wonder of you really need. Mr. Keen wanted to keep his MOSFET stuff as close to the fender designs as possible to keep tubeheads comfortable ... I think.

The R part of the LDRs is going to be facing the high voltage, there is zero need to have the LFO and LEDs face more than 12V (even 5V should be enough). Look for some solid-state LFO designs of the googles.

Bear in mind, though that I'd one of the ... shall we say 'moderately qualified' around here, and we need to hope one of the smart guys drop in soon.

Hope this helps!

[rathmann]

Thanks for the reply Don.

I'm not sure what kind of swing is required by the bias mod tremolo. That's why I went with the tube replacement mosfets in the design, since it's essentially the brownface trem times two. My discussions with Speedracer a number of years back led to the conclusion that the harmonic trem was a little touchy and sensitive to bias conditions and oscillator swing.

I'm not averse to trying something different, and it really does seem to be a bit of overkill for the LEDs. I just figured that if the higher voltage circuit was necessary for the tremolo, it shouldn't be asking too much to tap off a few volts for the LEDs.

[Don Moose]

The LEDs are about current (up to about 20mA), and a smallish Vf (generally 2 to 3V).

Getting a divider to take 200V down to 3, and be able to supply 20mA is going suck off a lot of B+ capacity that you could use elsewhere and require relatively high-wattage parts. Since you have (in your schematic) a +12 rail, use that instead.

(LEDs I am qualified to talk about)

Hope this helps!

[rathmann]

yeah, I was planning to run the relays and possibly the first couple of preamp stages on 12vdc. It makes sense then to use this for the LEDs.

I'm still not sure what's required for the harmonic trem. I guess I could potentially have two LFOs and have the vibrato and tremolo completely independent of each other. Hmm...

[shzmm]

Well for what it's worth, I also built the R.G. Keen mosfet tremolo with a vactrol. I noticed two things (with a bunch of help from the experts) that may be able to help you.
First was that the final mosfet wasn't necessary for me as it was used in the original design to drive the neon bulb. I placed my led section between the 100k drain resistor and the drain/feedback connection. This may not pertain to you as you are using the last section to alternate the phase of the output.
Secondly, that 1M/33k divider that Mr.Keen mentions sets the operating voltage for the oscillator mosfet? I'd recommend putting a trimpot in place of the 33k as both of the circuits I made needed an adjustment.
Other than that it works great-- I used both .01 and .02 value caps in the feedback path. .01 makes it almost too slow although I like that, .02 is a good balance. I'd be interested in how .04 works...

[Tilman]

Hi Rathman,

I know this thread is pretty old but I try to figure out how to build a magnatone like vibrato myself (without usig the varisters).
Have you ever tried out the two stage phase shifter design you are using in your circuit here ?
If so, does it do the pitch shifting vibrato similar to the old Magnatone amps ?

Best Regards!
Tilman