Need help with making a tremolo effect

[MohiZ]

Hi all, I call upon your wisdom again. I couldn't find a nice and simple tremolo schematic to suit my needs, so I thought: why not make one? I don't feel like having to use that LED/LDR stuff, so what I did instead is I used a JFET as a voltage controlled resistor. This is what I've come up with thus far:



This thing is just the LFO and the JFET VCR, which is used to attenuate the signal accorcding to the LFO's signal. The guitar signal comes in through the 22n cap on the top and goes out through the 47n cap. This portion of the circuit is preceded by a buffer, and after this comes a booster to top up the level back to unity.

This thing works quite well, but there's a HORRENDOUS tick in the output signal. No matter what I do, I can't get rid of it. I've tried putting caps in the pathway of the LFO signal and between the drain of the JFET and the signal path. Removing the 1M5 resistor from the drain to gate doesn't help with reducing the tick, but it makes the JFET VCR work worse. I've read that using low-power opamps would help, but I don't understand it. If I cut the wire between the guitar signal and the JFET out altogether and just have the input and output buffers in the circuit, the tick doesn't exist, even though the LFO is still on. So it's not about the power supply?

I also found this: http://www.diystompboxes.com/smfforum/index.php?topic=72527.0
In the schematic it seems a J201 is used as a VCR for the tremolo, but I don't understand how it's possible for it to work. The gate is biased at 4.5V while the source is grounded. Shouldn't the gate be negative compared to the source? That's why I have the source at 9V, and the LFO signal fluctuates between 9V and some value lower than 9V, depending on the bias pot.

I'd really appreciate ideas 

[Mark Hammer]

A great many here find the classic Electronics Australia tremolo (EA Trem) to be a very dependable circuit.  Also uses a JFET as a voltage-controlled element to adjust gain via an LFO.  No audible ticking whatsoever in my own experience.

[SonicVI]

The colorsound tremolo works similarly, the original uses a bjt, but the reissues have a BS170.

[MohiZ]

I tried the EA tremolo, but I find that I want a wider range of speeds than it can offer. The op-amp LFO I've used here can go both very slow and very fast without a change in amplitude. Does the colorsound one have a better range?

[SonicVI]

The colorsound is fairly limited.  Take a look at the Tremulus Lune LFO if you want a lot of versatility.

[Mark Hammer]

I tried the EA tremolo, but I find that I want a wider range of speeds than it can offer. The op-amp LFO I've used here can go both very slow and very fast without a change in amplitude. Does the colorsound one have a better range?

All LFOs, no matter what the topology, use one or a couple of caps to set the range of modulation speeds.  Absolutely no reason why you couldn't have a switch to alter those cap values to a faster range for any tremolo you choose, including the EA Tremolo.

Just to put in a plug, LDR-based tremolos can often have difficulty negotiating fast speeds with any appreciable depth because they never get quite fully off under those circumstances.  A JFET is a far better way to go IMHO because they can respond to faster LFO rates just fine.  You're on the right track.

[MohiZ]

I know that it's possible to change the speed of a given LFO. However, I tried changing the components in the EA Trem LFO but wasn't able to get decent results. I think the problem was that when I tried to make the speed even slower, the rate knob became really sensitive and the LFO would only work with a portion of the rate knob. Overall, I just thought I'd have to make too many components switchable for it to be worth it. Also it would be nice to have a choppy setting etc. There doesn't seem to be that many tremolo designs floating around, and many of them use LDRs, which I can't really get easily. The Tremulus Lune seems very excellent, but uses one of those. That's why I started figuring out my own design for the trem. I know I may be trying to achieve too much, but hey, I think it's interesting.

I actually was able to reduce the ticking a lot by adding a resistor in parallel with the JFET, as I've seen done in the Phase 90, where JFETs are used as VCRs. I'm gonna continue playing with the values and the layout.

Does anyone know why the JFET is set up the way it is in the Magnus Modulus (http://www.diystompboxes.com/smfforum/index.php?topic=72527.0)?

[R O Tiree]

Haven't I seen advice on this recently? When the "square" side of the opamp slams (as far as it can) from one rail to the other, this causes a momentarily large current draw which, in turn, causes the +9V rail to sag by a couple of mV, which is then picked up in the audio path as a "tick" every time the LFO changes state.

IIRC, it was to put a small resistor and a large electro in parallel between +9V and pin 8 and another resistor/cap pair between pin 4 and GND. The resistors displace pins 4 and 8 from the power rails very slightly and the caps provide "reservoirs" for those sudden current spikes.

You could also try using a TL022, or similar low-current opamp?

Finally, the LFO used in the Boss CH-1 doesn't tick. It's very well behaved, in fact.

[R O Tiree]

Ibanez CS-9 chorus LFO is also very well-behaved... Very similar design to the CH-1 LFO.

Here's the CS-9 version:



It uses a TL022, not a TL072 - I just couldn't be bothered to change the designator when I made this schem.

Ouput is taken from the wiper of the "Width" pot (aka "Depth"?)

[Transmogrifox]

I tried the EA tremolo, but I find that I want a wider range of speeds than it can offer. The op-amp LFO I've used here can go both very slow and very fast without a change in amplitude. Does the colorsound one have a better range?

The EA trem variable gain element is a great way to go.  You could easily substitute a different LFO to improve the oscillator range.
This mod helps improve the depth:
http://www.geocities.com/transmogrifox/rEAgenerated_Tremolo.html

You could stick any favorite oscillator in place of the EA oscillator

I find the oscillator from the Peavey Duel 212 amp has a good range to it.  It has nearly the simplicity of the EA oscillator with a little more range.  Schematic here:
http://www.geocities.com/transmogrifox/DUEL-212.PDF
Check out about the 4rth page down.

[MohiZ]

Thanks, loads of great suggestions here! I'll check 'em out.

[R.G.]

Then of course there is the very wide range LFO at GEO: http://geofex.com/FX_images/p90ramp.pdf has one implementation, even set up to drive JFETs as voltage variable resistors.

The VCO style LFO has a very wide range if designed properly, much wider than the pot-controlled ones.

[MohiZ]

I'll have a look at the Phase 90 LFO some other time. I tried out the EA trem with Transmogrifox's mods, and it's good enough for me. I'm going to put a 3P-quardruple-throw rotary switch to change between four different cap settings. Should be sufficient for ultra slo volume swells and ring mod sounds in the other end of the spectrum  Thanks for all the suggestions!

[Cliff Schecht]

Sigh.. I finished a stereo/mono tremolo effect that was using JFET's as VCR's. I've got it where I like it, now I just need to post the dang schematic!