Help me break down this tremolo circuit

[EETStudent]

Hopefully somebody can answer a few questions about this tremolo circuit...

http://www.schematicheaven.com/fende...6g16_schem.pdf

My amp utilizes this circuit almost exactly. There are however a couple differences that I want to question. Maybe someone can follow me

1) Starting from the footswitch jack, immediately there are two .01 caps, one going to the 1M grid resistor and one to the speed control. In my amp these two caps are .02uF, but all else in that stage is the same. What difference would this make?

2)In that same stage, there is a plate to cathode .02 cap. This is not present in my amp.

3) The second is nearly identical, except after the cathode is a 470k resistor going into a .1uf cap to ground. These two components are not present in my circuit. I just have the cathode 470k above ground, and the single .1 coupling cap into the intensity control.

What differences do these adjustments make? Is there room for improvement?

[Enzo]

Look at the overall circuit. You have a triode with a feedback loop from its plate back to its grid. That loop comprises three caps and resistors to ground at each node. Three RC stages if you like. This make an oscillator. The RC time constants determine the frequency. Making those caps larger - or the resistors larger - will slow down the oscillation. it will make the trem slower. I make that change all the time. If an amp has a trem that is still too fast all the way to the slow end, I can slow it further with the larger caps. I have yet to see an amp where the trem speed wasn't useless over about half way up anyway, so slowing the whole thing down is usually a good idea.

Note the 1meg resistor between the second and third caps goes to the cathode. It could also go to ground, and on many amps it does. Look at a number of Fender amps to see the variations on the same circuit.

The plate to cathode cap? You could do without it. It is not present in many of the amps.


In the driver side, the extra series 470k and the extra shunt 0.1 cap are there to tame the level of the thing. All that depends on the amp in question, how the trem oscillator is interfaced to the rest of the amp and how strong it needs to be.

Room to improve? OK, what does it not do well enough now? Improvement implies there is some shortcoming in the current arrangement. In one amp, the trem might be strong wnough to create a whump whump sort of sound, so it needs to be tamed. But in another amp that might not be the case. You can't look at the trem by itself, it is part of a larger whole amp circuit.

[mooreamps]

Hopefully somebody can answer a few questions about this tremolo circuit...

http://www.schematicheaven.com/fende...6g16_schem.pdf

Is there room for improvement?

Plenty. It's a stupid design. It's a text book design, but it's a stupid design.
For young engineering student, I'll spell this one out. As a rule of thumb, for any oscillator circuit, for stability, inputs should be buffered, outputs should be buffered.

So, think about how you would alter this circuit, such that, you take the output off the cathode circuit, instead of the plate, and then write a brief on why this is better.


-g

[Enzo]

Better is such a slippery word.

Just how does this circuit perform the job less well than an enhanced version?

Isn't the existing cathode follower coming out of the oscillator stage a buffer?

Is stability a problem for a 2Hz oscillator?

[mooreamps]

Better is such a slippery word.

Just how does this circuit perform the job less well than an enhanced version?

Isn't the existing cathode follower coming out of the oscillator stage a buffer?

Is stability a problem for a 2Hz oscillator?

Yes, better is a slippery word.

1. For one, I've heard some Fenders with the anoying "tick" coming out of their "tremolo/vibrato" amps. I've never liked that set-up, with no offence to ; and with all due respect to Mr. Leo.

2. I don't believe I've seen any vintage layouts that were actually buffered with a cathode follower. They all appear to me to be taken off the plate circuit.

3. Maybe using the term stability is not actually correct. Stability would imply the 2 hz signal drifting in frequency, which if it went up to 3 hz for example, it's like who would really notice that when playing ? But, it should come up "solid" when turned on, be a strong enough signal to provide the desired effect, and should produce any noise contribution to the loud speaker when used. Now, I know this all sounds all well and good for a guitar amp, where sometimes some things are taken into account as only a small compromise to build something useful and affordable. I'm not suggesting implimenting a 100 dollar circuit to fix a 2 dollar problem. But, I am thinking this is something that could be "enhanced" ecomonically.

-g

[EETStudent]

Well you guys kinda lost me there towards the end, but my questions were answered.

Enzo, I guess I just want to make sure everything is as good as it possibly can be. Maybe it is, but how do I know? The circuit sounds terrific, but it never gets fast enough and its not quite deep enough. Id like a pretty heavy pulse with things maxed out, seeing as thats mostly what i use tremolo for. If I were to speed it up, would I need to change all three of those caps? Any recommendations as to how to increase the depth?

[Enzo]

Gary

1. Ticking trems on old Fenders is a well knows symptom and is easily cured with a resistor across the trem neon bulb.

2. well

2. I don't believe I've seen any vintage layouts that were actually buffered with a cathode follower. They all appear to me to be taken off the plate circuit.

Then you didn't open the link to the 6G16 schematic in the opening post. Especially relevant since that is the circuit we are discussing.

3. The existing circuit with good caps in it - as opposed to 30 year old ones - comes right up when the switch is hit and runs up and down the speed scale smoothly. Adding another triode to the amp is more than a minor correction, and I don't think the trem would work any "better, " since it already works fine. as to noise, your ticking comes from the neon lamp in the photocell, and would not be affected by a CF stage.

[Enzo]

I guess I just want to make sure everything is as good as it possibly can be. Maybe it is, but how do I know? The circuit sounds terrific, but it never gets fast enough and its not quite deep enough. Id like a pretty heavy pulse with things maxed out, seeing as thats mostly what i use tremolo for. If I were to speed it up, would I need to change all three of those caps? Any recommendations as to how to increase the depth?

You know by listening, if it sounds good then it sounds good.

NOT FAST ENOUGH? Oh dear... I have never seen a trem not fast eniough. Usually by the time the speed knob is up to 5 it is stupid fast. All the way down should give a slow throb in my view - like a Leslie on slow. Most of them only slow down to about 5-10Hz. I' like them down to 1-2Hz. Even with the range lowered like that, the upper end is still way to fast for anything I'd ever do.

You want it faster, then where there used to be .01s in that feedback loop, and you now find .02s, restore them to ,01s.

There is that 470k in series with the .1 cap to the bias on the circuit in post #1. Try reducing it. Tack a 1 meg pot across it and dial it up and down.

Are all those caps fresh now?

[TheTinMan]

Here's a discussion of that bias-vary tremolo circuit that I found really helpful: http://www.el34world.com/Forum/index.php?topic=604.0

Slight detour: if a buffer is a "good thing" for an oscillator circuit, is there a relatively easy way to add a solid state cathode follower to a Princeton Reverb AA1164 circuit? http://www.el34world.com/charts/Sche...ERB_AA1164.pdf

Chip

[R.G.]

Slight detour: if a buffer is a "good thing" for an oscillator circuit, is there a relatively easy way to add a solid state cathode follower to a Princeton Reverb AA1164 circuit?

Sure - use a MOSFET source follower. See:
the MOSFET Heresies
a MOSFET Fender-style tremolo

[mooreamps]

Gary


Then you didn't open the link to the 6G16 schematic in the opening post. Especially relevant since that is the circuit we are discussing.

.

You're right. I did not, and yes it is.
My apologies for not staying on topic.

-g

[tubeswell]

Well you guys kinda lost me there towards the end, but my questions were answered.

Enzo, I guess I just want to make sure everything is as good as it possibly can be. Maybe it is, but how do I know? The circuit sounds terrific, but it never gets fast enough and its not quite deep enough. Id like a pretty heavy pulse with things maxed out, seeing as thats mostly what i use tremolo for. If I were to speed it up, would I need to change all three of those caps? Any recommendations as to how to increase the depth?

To get it deeper maybe you could lower the resistance associated with the depth pot (i.e. making the pot start deeper). When I built my 5G9 I inadvertently left a 1M resistor out here, and was rewarded with super deep CCR type trem, taht was too much because it chirped when dimed. So I put the 1M in and it is more 'normal' now (altho' I am thinking lately "Hmmm, maybe I'll go back to 470k in that position" :-). Of course I guess from this that you could increase any resistor that was in series with the depth pot from the oscillator, and at the same time increase the depth pot range to compensate, so that you could get a depth control that goes from nil to deeper-than-before.

To alter the speed, you can experiment with altering the tapering resisor on the speed pot, to increase or decrease the range of speed adjustment. When I built my 5G9 I built it with a dual ganged 5M RA pot paralleled to give 2M5 RA, but it was only tapered with a 100k resistor to ground, and I was getting too much speed and not enough slowness, so I upped the tapering resistor to 120k (reflecting more the ratio of 100k to 2M that was in the stcock circuit scheme), and it works perfectly.