Two-Stage JFET Preamp/Booster

This circuit makes a nice, clean preamp or a clean booster pedal. It is not designed for overdrive, but is designed to be overdrivable. If overdriven, it provides the kind of soft clipping one may expect from JFETs, but it is not optimized to be generally used as an overdrive effect.

This design started out as an idea for a booster pedal circuit using low-noise JFETs following the principles of the Fetzer Valve(see I wanted plenty of headroom and however many stages I would need to have enough gain for clipping. I did not want clipping, but I wanted all the gain I could have up to the point of clipping. The output levels  from different guitar pickups vary, so I figured a preamp that could reach the limits imposed by the supply voltage using the guitars I have could provide enough clean gain for most guitars. Using an 18 V supply allows for more clean gain than one can get with any 9 V circuit. Of course, that also means the output voltage might be higher than can be safely used with some solid state amplifiers and pedals. This design can easily be scaled to a 9 V supply, but might require more stages or higher gain JFETs that might be noisier than the 2n5457.

I decided to used Fetzer valve based stages due to the supposed tube like performance of the Fetzer Valve design. I am not sure how well the Fetzer valve emulates a triode stage, but I think JFETs generally do a good job in preamp designs. I like the sound clips on the site, and I like the triode-to-jfet conversion designs I've heard elsewhere. For example, I think the Dr. Boogie preamp I built (based on the schematic from absolutely kicks ass. 

One may argue that the Fetzer design might not be the best for a clean preamp/boost circuit. By omitting the source resistor bypass capacitor, the maximum gain of each stage is much more limited. Also, one may question the Fetzer valve's criteria for selecting the source and drain resistors. See for different ideas about triode like jfet stages. Also, by focusing on triode emulation, the Fetzer biasing scheme might not give the most gain. If I ever have more time on my hands, I might try something a little different. At the moment, work, school, and family leave to little time for my hobbies.:-(

My first Fetzer Valve based circuit used MPF102s because I did not have any 2n5457s at the time. With an 18 V supply using the Fetzer valve biasing scheme, I could not get the gain I needed with  two stages. I added a third stage using a j201 to avoid having to add several more MPF102 stages. In addition to a gain pot, this first version had an output volume pot. This design sounded pretty good, but is noticeably noisy, probably due to the j201. I knew I really needed to order some 2n5457's or some other JFETs that would provide the gain I needed in two stages while still being relatively noise free.

After my 2n5457's were delivered, I built the circuit shown above. In simulation (using Multisim 10)  I could see that using a 220pf miller capacitor (for lack of a better term)  in the second stage caused significant loss of treble under 20 kHz, so I omitted it. With the miller capacitor left in the first stage, rf frequencies are attenuated appropriately. Notice that I put the miller capacitor between the gate and source in the first stage rather than the gate and ground like you see in the Fetzer Valve schematic. It probably does not matter, but I'm used to seeing the gate-source configuration in other schematics like the Dr. Boogie schematic. I'll let others debate the pros and cons each configuration. I also did not include a volume pot in this design. If I left the volume pot in, then one could probably use this as an overdrive effect. By turning up the gain, and lowering the volume pot, one could have overdriven output with a peak-to-peak voltage much less than the supply voltage. However, I intended for this design to be a gracefully overdrivable clean booster/preamp rather than an overdrive effect, so I left the volume pot out. 

In the schematic, I did not specify values for the drain and source resistors. The idea is to select values for these resistors according to the Fetzer Valve guidelines for each JFET. One could use trim pots instead, but I've got enough resistors on hand of different values that I can hand pick resistors that are pretty close. 

Perhaps the design could be improved (Well, of course it can. Everything can be improved). There is nothing spectacular about the design. As simple as it is, I'm sure very similar designs already exist. One possible improvement might be to add a trim pot in series with the gain pot. This way, one can tune the maximum gain to be just short of being overdriven with his or her particular guitar.

When I built this circuit, I liked the results. It provides lots of gain(perhaps I'll measure the gain one day). I was surprised by how noiseless the circuit is. When I use it to boost my Valveking Royal 8, I can hear hiss with the Royal 8's preamp turned way up, but that's no surprise. When I do turn it up to the point of being overdriven, it sounds acceptable. I used it as a preamp for my Crate GFX-212 by bypassing the built-in preamp using the effects loop jack. I liked it better than the built-in preamp, though with out tone controls, it is more suitable as a clean booster pedal. With my predisposition for modular design, I prefer being able to use a separate tone control in conjunction with this circuit than having one as a part of the design.

Here are pics of my build of this circuit:

The enclosure is a simple aluminum box I bought at Radio Shack. The gain pot with the way-to-long shaft also came from Radio Shack. After I create and test a PCB layout for this, I'll probably order a proper pedal enclosure with a stomp switch and everything else that is needed for a proper stomp box. I'll probably also used proper decal labels rather than use a sharpie. 

If anyone builds one of these, please let me know what you think. 

UPDATE: To see a treble booster version of this, go to JFET Treble Booster.