Simple tremelo schematic... but does it work?

[SolderFodder]

I found this schematic:



posted at: http://www.hawestv.com/mtv_FAQ/FAQ_fetpreamp5.htm#inputs

It claims to be a FET version of a Gibson tube design.  It's incredibly simple (while figuring out what's input and what is output may be confusing) and low in parts.  Does anyone know if this works?  Are there any others like it that you have built?

[CynicalMan]

My first guess is no. Firstly, tube and FETs are not interchangeable. Secondly, with a JFET this would probably either run out of headroom and distort, or it would just add the LFO waveform onto the output. I doubt there would be any tremolo.

Also, you probably don't want this running into a speaker. If you were to try it out, I'd throw in a trimpot as a drain resistor, remove the LFO from the gain stage, and set the trimpot for 4.5V on the drain.

This is probably more what you're looking for: www.runoffgroove.com/eatremolo.html

[Taylor]

I think the lower left looks like a phase shift oscillator. That seems about right for being a tube tremolo emulation. I don't get why it goes to a transformer and speaker though.

The EA tremolo is a pretty simple transistor-based trem and a lot of people like it. There's also a project on this forum called the Hearthrob or Heartthrob that I like and is similarly simple. IMO it would be easier to start with those than to modify the schem you posted to work as a stompbox.

Edit: cynicalman beat me.  But same advice.

[R.G.]

It claims to be a FET version of a Gibson tube design.  It's incredibly simple (while figuring out what's input and what is output may be confusing) and low in parts.  Does anyone know if this works?

It's possible it works - depending hugely on what's outside the circuit as shown. Function is simple.

Q3 is a phase shift oscillator, as noted. Q4 is an amplifier, driving the speaker through that 1K to 8 ohm transformer. There is presumably some other circuit not shown in that picture that this works with. Otherwise, no reason to have a "from 47K feedback resistor", or "From Q2". So it's not a complete circuit. It works by modulating the bias on Q4. There are tube circuits which work kind of like this, I guess.

The stuff it has to work with that's not shown kind of make the incredibly simple and low in parts less appealing, as it probably won't do much of anything by itself, and what it goes with may not be either simple or low in parts.

I did a sample reading of some of the JFET stuff on the site. There are some technical mistakes in the commentary, along with some correct stuff. Be cautious.

[SolderFodder]

Here's the leading circuit:



Note that on the site the guy is giving this as a ballpark example of a solid state conversion of a Gibson GA-1RT.  He evidently has not tested this design.
You're kind of right, after you account for a gain stage, it's no simpler than a EA Tremelo.  I was however puilding a JFET amplifier already and was thinking about throwing this in.  I don't want something no one has ever built though.

[PRR]

It's not likely to work as-is.

I doubt that Q3 has gain of 24 with that small 1K5 load, so it won't oscillate. It certainly could, with more appropriate values.

> I don't get why it goes to a transformer and speaker though.

Q4 serves the same function as the 6V6 or EL84 in a Champ or Epi Jr. That's the "power" output stage. If you find a hot MPF102 it might run near 5mA-10mA and deliver 0.01 Watts or more. It will be quite audible, but you can talk over it without going hoarse.

(If the oscillator worked) the gain of Q4 WILL change with oscillator swing. But the oscillator voltage swing is WAY bigger than Q4's bias. It probably slams from cut-off to bottomed, with some distorted signal between, probably obscured by saturation in that $1 push-pull transformer misused as SE. And tremolo inside the NFB loop may be self-defeating.

It worked on the tube amp his drawings are based on, because those guys TESTed and tweaked. He's picked a fairly funky amp to use for an example.

There's bits and crumbs of misleading writing on several pages I sampled. The bigger power amp is liable to blow MOSFETs; doubling the MOSFETs at same supply and load does not double power. Some good thinking, but take it all with a heavy grain of salt.

[thomasthebuzzard]

try the tinytrem. works good and simpler then this.

[amptramp]

This kind of thing usually goes on the input stage as in the Univox U-65 in the lower left of the schematic:

[CynicalMan]

Going a bit OT here, but the problem with the Univox and especially the tinytrem is that both have low input impedances The Univox isn't too bad, but the tinytrem has a 6k input impedance, which will make your guitar sound like you have cotton balls in your ears unless you have a buffer before it.

IMO, if you want a simple transistor trem, go with the EA, if you want op-amps go with a cut-down Tremulus Lune. From there, you can look at wackier stuff like the full Tremulus, the Pulsar, 4049 trems, quadrature LFOs, and other fun LFOs like pseudo-random and double LFOs.

One thing that can work is modulating the bias on JFET sources. I experimented with this a while back, but I was just starting out designing and there were ticking and LFO problems. I should take a look at that again.

[SolderFodder]

try the tinytrem. works good and simpler then this.

I followed your suggestion.  Yes.  That is definitly a good option if i can find a LDR in that range.

[MetalUpYerEye]

+1 for the EA Tremolo. I built a Tiny Trem and while it was a very simple, neutral sounding circuit, mine clicked like mad with the switching options available for that setup. Shortly after I finished my EA that had been sitting on the shelf for awhile and am really happy with it. It has a really nice sound, very good basic tremolo and absolutely cooks in booster mode (if you add the booster mod).