Guitar pickup internal preamp

[AudioWizard]

Hi all,

I'm in the process of designing a preamp for home-made guitar/bass pickups. I've come up with a few schematics, mainly some based on a couple JFETs, and others on opamps.

The goal, of course, is to design something with the lowest power possible. Turns out this is not trivial!

JFET-based preamps are usually the most satisfying ones: very few parts, extremely good linearity, and the second harmonic is favored instead of the 3rd that most opamps tend to "favor".

But I can't seem to manage to get a power draw less than 300-400 µA if I want a low output impedance.

One thing I'm wondering about is how EMG, for instance, manages to make internal preamps that draw as little as 80 µA, and still have an output impedance of "only" about 2k! I'm not sure about the gain those have, but I'm guessing it must be of several dB's, since these pickups have a high output level and comparatively fewer coil turns than most passive pickups, so some gain is definitely needed here.

Anyway, if possible, I'd like to avoid 1) opamps, in favor of JFETs, and 2) having to rip one of those EMGs apart. ;-)

And by the way, if I can make something "better" than EMG, why not.

Any ideas welcome!

[tschrama]

Maybe EMG uses a BJT as a buffer/amplifier, set at 80uA?

[Giaime]

I think they use a jfet and another jfet set as a fixed current source limited to that value. I've seen that with logic devices used as linear amplifiers (CD4049): if you connect the Vcc directly to a psu, they will suck much current. Many people avoid this using a jfet set as a current source. Sorry I don't know more... but when you develop something tell me 'cause I'm also interested in guitar on-board preamps...

[AudioWizard]

Originally posted by Giaime
I think they use a jfet and another jfet set as a fixed current source limited to that value.

That's basically what I've done, with two J201's.

If I want some gain with that, I have to design an inverter, not a mere buffer. So basically I have a resistor between Vcc and the main JFET's drain, and a current source between its source and ground (made off another J201).

The problem with that design is that the output impedance is grossly defined by the drain resistor. And the lower this resistor, the lower the gain you can get...

I know about the "micro-amp" JFET design, with which you can get pretty high gain, but every time I simulated one of those, the linearity was awful. Not well suited for audio stuff.

[azira]

What about an SRPP/Totem pole kind of design? Pretty popular tube circuit that 2 JFETs can easily mimic. I should think that it would give fairly low output impedance and it's a active bootstrapped current source so you should be able to set Iq as you desire.
--
Danny