Surfy Bear harmonic tremolo kit/pedal

[jubal81]

Ordered a board this morning to check it out. It's JFET-based, but I there's no schematic or much details about the circuit topology. Demo sounds very nice, though.


He also has completed pedals for sale. The new site also says he's going to start selling pedal version of the SurfyBear reverb. Pics look very snazzy.

[thesmokingman]

its a good sounding tremolo ... has me thinking hard about housing that in a head unit along with their reverb kit I already have.

[midwayfair]

Not much about the topology, but you ca tell a bit from the board.

Looks like there's a quad op amp and only one transistor (other than the matched JFETs, which are the variable resistors).

You need one stage for LFO, one for the LFO buffer, one for each of the audio stages, and that leaves one as either an inverter -- which is probably necessary because there aren't enough diodes to do a 4.7V reference to hold the JFETs at half the way RG did it -- and/or as a buffer half-volt reference. There's a volume control, so there's either a make-up gain stage or the stages are just more than 1x gain to begin with. You can probably get away with not having an input buffer, but I'm not sure you could get away with not having an output buffer, so it seems most likely that the last gain stage is an inverting mixer stage, so I would guess that the pedal inverts phase.

If you want a starting point, I would look at something like the way Boss does FETs but with a parallel audio stage. And duck_arse's thread on DIYSB where he (and I) couldn't get them to work in a harmonic tremolo.

[somnif]

http://surfybear.weebly.com/blog/harmonic-tremolo-using-jfets

Here he is describing the thought process behind the circuit (in vague terms, no schematic). Pretty much as you described it (at a surface level).

[midwayfair]

http://surfybear.weebly.com/blog/harmonic-tremolo-using-jfets

Here he is describing the thought process behind the circuit (in vague terms, no schematic). Pretty much as you described it (at a surface level).

Actually it's fairly different. He says he's using the NPN transistor for the LFO and the JFETs are the audio path, with the LFO directly altering their bias. His JFET stage (he says it'a a long-tail pair) is actually the opposite of the Cardinal's: The sources are tied together. He'd have to feed their gates a buffered LFO DC output to attenuate the signal if he really is using the FETs as the audio path.

The op amp then is going to be a make-up gain/mixer stage (which you definitely need if you're using a long-tailed pair), some sort of inverter and buffer for the LFO output, and then you have an extra stage for whatever.

The surprise is that he can get a square enough wave to do the blackface mode out of the NPN LFO.

EDIT: Here's a diagram that shows you where you'd feed the bias signal on the tail. It's the junction of the gate resistors and rtail.

[jubal81]

Fired it up on the test rig today and it sounds really great. Surprised to see the marking on the opamp sanded off ...


Initial thoughts:


The bypass (dry through) signal is absolutely dead on - no coloration.
I do prefer the tight response of the JFETs to the optos, but it's a lot more subtle difference than I expected.
The LFO doesn't get all that slow - the downfall of the phase shift oscillator.
There's also a mild thumping in some settings.
Very quiet operation
Going to look at using a log pot for depth to dial in subtle settings better.

[jubal81]

Finally got around to changing the depth pot to 100KC and it's a huge improvement. Much more usable to dial in subtle settings and for whatever reason, I can't get any LFO thump at all any more.


Very nice, though I'm still not in love with the LFO.