Finalizing the JFET Booster

In this article I’ll continue building a JFET booster. After breadboarding the JFET boost in the last article, it didn’t really sit well with me calling it a booster when it does not actually boost the signal. In this article I’ll just go and do a small redesign or extra design, whichever way you want to call it, to make it actually do the boosting.

So, the first order of business is making it boost. I’ll get to the final schematic, finalize the list of components, and as usual go through the bill of material, preparing the enclosure, final assembly and then I’ll try it out. I’ll focus on design of the circuit and will try to go through the rest as quickly as possible.

Make it Boost the Signal

The amplification of the effect due to parts used is around 0.5, or as calculated on the effect’s page around -6.8dB. This can hardly be called a booster then, it’s more of an attenuator. Let’s make it a booster then before doing anything else.

I did not want to change the design of the circuit, but good thing about it, I could use it as a building block. All I need to do is add a JFET amplifier to the signal chain, this way I can make it a proper booster.

The circuit itself is a relatively straight forward self biasing common source amplifier. So I thought I could just add another stage to the circuit, but this time with slightly different biasing to maximize the amplification.

Basic Circuit

The amplifier is super simple in essence. I removed all the distracting stuff out of the picture (input filter and coupling capacitors namely). I’ll use another BF256B since I already did measurements for it. Here’s the basic circuit:

I need to determine values for R2 and R3 to balance gain and headroom. C1 is there to maximize small signal amplification and it is a customary 22 uF value (to give flat amplification across the full spectrum).

The values I considered using are:

R2 (Ohms)	R3 (Ohms)	Gain (dB)	VD (V)	VS (V)	ID (mA)
900	122	12.6	3.76	0.71	5.8
1600	400	15	3.73	1.32	3.3
2700	1000	17	4.14	1.80	1.8
4000	1500	19	3.69	1.99	1.3
5100	2000	20.3	3.60	2.12	1.1
5220	2700	19.4	4.68	2.24	0.8
7000	2700	22	3.20	2.24	0.8

The table gives the values for resistors I considered using, and their respective operating conditions. All the values are calculated values. Measured transistor parameters I used are: I_DSS about 9.7mA and V_P about -3.16V. I did the calculation and then I went and verified them by running this through SPICE and later on breadbording them.

Some things to note. It’s a dance at this stage, between input headroom, output headroom, keeping the JFET in the correct operating region, desired amplification and distortion.

The smallest distortion is closer to the I_DSS for drain current, but that makes it harder to achieve higher gain and headroom, for example. Also, I have to think about how to control either gain or volume.

Controlling gain/volume

Maximum amplification I can get is roughly 20ish dB. Since Fetzer stage has about -7 dB gain, combining stages that is about 12 to 13 dB. Not huge, but I still wanted to add some way of controlling either gain or volume.

There are basically 3 options to control gain/volume. I could control gain by adding a potentiometer in the transistors source like when I was doing MOSFET booster, or I could do fixed gain and control volume like when I did a BJT type booster. What I could also do is adding the potentiometer to control drain resistance and that way control gain.

This is what I mean:

I didn’t have a pot smaller than 5K so had to adapt my choices a bit (that explains, I hope, an odd choice of 5220 ohms for R2 in the table above).

There’s a 4th way. I could’ve replaced input 1M resistor with potentiometer and control input level and have fixed gain. But I’ll leave it at this for now.

Choosing the right control option

Choosing one is an exercise in compromise. There’s no right or wrong answer I suppose, it’s more of a what I want to achieve with it.

Fixed boost with volume control is good but since you can’t control gain it might clip in a way you don’t want. That can be dealt with volume pot on the guitar for example, but still, might not be good enough of a control in all cases.

Potentiometer in source of the transistor can control gain but the gain law is not great. Nearly all control is in the last 20% of the rotation. Potentiometer in the drain has great control, usable nearly throughout the whole range of motion. That is, usable if you can live with noise it makes when you’re adjusting it since it has DC current flowing through it.

If I look at the gain law between options b and c from the schematic:

The green plot is the actual gain law, and the blue plot is “ideal” law. I compensated it for attenuation of the Fetzer stage. It looks that putting potentiometer into drain gives clearly a better control of the output. I’ll go with that, I just hope the noise when I adjust the pot does not annoy me too much.

Final Schematic

I nearly forgot to include consideration of whether I’ll go with Fetzer stage before my boost stage or after it. Putting the booster stage at the beginning gives better signal to noise ratio, but too much of a boost might cause some extra clipping in the following stages. Again depends on what I want to achieve.

I went with having Fetzer before booster, just because of less clipping. It’s fair to say that I tested the breadboarded effect with a crappy headphones amp (night time testing, the family not super excited about the “noise” I was making). Man, it was a massive difference in sound when I went through a proper amp. At that stage I had made up my mind so take this as you will.

And finally the finished schematic:

Final Few Notes on Schematic

The input stage is pretty much the same circuit that I used in the previous article. I can’t do DC coupling since 2nd stage requires 0V for reference, so C3 is needed as a coupling capacitor there. The 2nd stage is just the circuit I discussed above.

I used 47nF coupling capacitor at output as the original circuit and I replaced the original 100K pot with 100K resistor. I used that resistor to help with the thump on switching the pedal on, originally I thought to leave it out, but the thump when turning it on was too loud. There’s no problem trying things out, playing it for a while and then adjust the design if I don’t like something about it.

The usual paraphernalia:

Final Bill of Material

I went through a great length on choosing components, so I’ll just list out components that I will use here.

Designators	Component	Notes
	27969PSLA	Enclosure
	Protoboard	Veroboard cut to size
J3	Cliff FC681473	DC Power Socket
J1	Switchcraft 112BX (or ACJS-MVS-3S Amphenol)	In – stereo Jack (alternatively use mono jack since I don’t use ring connector)
J2	Switchcraft 111X (ACJM-MVS-2 Amphenol)	Out mono Jack
RV1	5K Linear Pot (P160KNP-0FC25B5K)	TT Electronics 5K linear pot
	Knob	Knob that fits 6mm straight shaft – after cutting shaft to size
S1	SF17020F-0302-21R-L	Taiwan Alpha 3PDT latching foot switch
C1, C3	22nF (0.022uF)	Metal film PET cap
C2	220pF	Ceramic cap
C4	22uF	Electrolytic cap
C5	47nF	Metal film PET cap
C6	100uF	Electrolytic cap
R1, R6	1 Meg	1% 250mW metal film
R2	33K
R3	200 ohms
R4	47 ohms
R5	270 ohms
R7	220 ohms
R8	2.7K
R9	1K
R10	100K
D1	Led	3mm Orange-Red, 2V forward voltage
Q1, Q2	BF256B	JFET Transistor, TO-92, G-S-D pinout

Preparing Enclosure

For drilling, I spent much more time now to get this right. I spent some time getting the outline of the board right and position it. We’ll see how it plays out, but I came up with this:

Here’s the template for convenience:

After the drilling (and also during the drilling) I checked the fit as usual:

This time around, I was trying out some fine liners. I applied primer first, then some acrylic paint, all with a brush. Then I used some fine liners and POSCA pen to do the design. And then did the protective coats of varnish.

I thought to go with something like Alien movies inspired theme (hell, even Fortnite did it). Like JFET bursting out of the chest of a vacuum tube.

Well, saying it out loud like that, it does sound a bit lame but I promise, if I only had better artistic skills it would’ve been great! Anyway, here’s the result:

• 

• 

• 

I should’ve gone with lighter on the pink, but part of this is to see what works well and what does not. More vibrant colours stand out and I should’ve used them more, and with more contrast. Anyway on to the soldering and assembling the pedal.

Following the final schematic I did a layout for this:

Here’s the DIY Layout Creator file that I used for this:

Final Assembly

I’m finally ready to solder everything. I’m not going to go through too many details. I soldered everything as per the layout above going from smallest (or lowest profile) components to biggest/tallest components.

The final soldered board looks pretty much the same as the layout above, and there are few more photos of the steps done:

• 

• 

• 

• 

• 

• 

I just hate this small enclosure, very tight and hard to manipulate wires around, but that can’t be seen from outside 😉. And this is how it looks like after the pedal is finally assembled:

• 

• 

• 

• 

Test Ride

And here’s hot it sounds in action:

I thought the noise made by the pot would not annoy me as much as it did in the beginning, but after a while it stopped being so annoying. In a few more days I might hate it again, who knows.