So, I got to breadboarding with those findings in mind, and found that I had a pretty dirty little boost on my hands. I was happy with that and drew up this schematic on KiCAD.
The cap on the source (pin 3 of the JFET) lets AC through, whilst not affecting the DC biasing of the transistor.
I started with a 100K pot for the gain, and a 10K trim pot, but I got better results with an A50K pot and a 20K trim pot.
To bias the JFET, I turned the gain pot fully clockwise, then tweaked the trim pot so that it was at maximum volume. As you adjust the trimpot you’ll find there’s plenty of space where it sounds fine, but when you turn too far the guitar sounds gated and you lose volume when the gain is turned up.
The interesting thing I found when I applied the voltmeter to the drain is that when the gain is turned fully anticlockwise the drain is at 9V, when you get to full gain it drops down to 4.5V. I don’t have an oscilloscope at hand so I can’t be sure, but I think this means that at lower gain levels you get asymmetric clipping, which slowly becomes symmetric as you boost the gain rebiasing the JFET. I will come back to this theory in a later post.
Anyway, it was all sounding pretty good, but I decided to take it further. I decided to add a second BJT gain stage (see below), after the JFET and put it on a switch. RV3 is a trimpot.
I though it might be best to have an internal trim pot as a volume control, cos this thing is plenty loud. I have no idea why anyone would want to turn down, but some might.
The switch wouldn’t be activated when played, but I would still need to make sure that which ever way you played it (with or without the second stage on), there would be the proper pulldown resistors. The trimpot on the second stage does this if it is used. But what happens in the middle? Where does the switch go?
As you can see we have a 100K resistor from ‘out’ to ground in the first part as a pulldown resistor, and we have a 100K resistor from ‘in’ to ground in the second part to help bias the BJT. If we keep both of them then those two resistor in parallel will equal 50K, and this will mess up the BJT. So one of them needs to go! We need the first 100K as a pulldown, so we can’t get rid of it. I solved this with clever switch placement. See below.
When the second gain stage is not engaged, R2 is a pulldown resistor, and when we’re using both gain stages, it acts to bias the second transistor. I was pretty proud of myself for working that one out for some reason.
The Wee Fettle Boost was intended to be the start of a whole line of JFET boosts/ drives/ distortions/ fuzzes/ stickers, badges and mugs. This one turned out far dirtier than I had planned, but that was fine, I just had to change tack. Which you can read about in part 3, as well as the next project – The Evil Twin.