Nice build!
I know I saw someone in a Facebook pedal group with a hand built SLO pedal, judging from the control layout, I’m guessing it was a PPCB board as well.
As someone mentioned elsewhere, wonder if using shielded wires from the switch to jacks would help cut the oscillations?
I got them from Tayda. I did not test - can I do this with a multi meter?
I'm curious about this too, how do you test them? How do you test them, and what values are your looking for? I have a DCA that gives me some insight into hfe, etc, but how do I know if it's a good part?
Gorfida: Search "Lead Solder" and look for my post on resolving SLO oscillations.
Grobbins & Chris:
JFET tests for pedal building:
-
Always socket FETs, MOSFETS, transistors and ICs. Very convenient to perform swaps and prevent damage from direct soldering.
IMHO, PedalPCB products are extremely well-made and are designed to never let go of installed components.
-
Organize. You will be comparing and matching FETs, so give them an identity. I make a numbered grid on a piece of paper and place each
FET in its 'square.' You can notate results in the square.
-
Testing continuity for shorted and open (no conductance) conditions and simultaneously diode forward voltage value, using a multimeter
set to diode / continuity test. If the value from 'G' to 'S' is more than 5% of the value from the measurement of 'G' to 'D' (look up your FET
pinout orientation), the FET will behave asymmetrically in gain and as a switch (the point where the bias on 'G' is so high or low, that it prevents
current from flowing). If a single FET measures within tolerance, the resulting value is now a useful element in matching it to other FETs
which are in tolerance. More:
https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/
-
FET gain. There are multimeters with FET gain capabilities. I do not own one and perform this test using a breadboard and basic power
supply which gives me clean 9V and18V DC. Power supply negative goes to ground. I connect Vcc+ to a 50K linear trim, which
then goes to 'D'. Connect a 1uF cap from 'D', to a 1M resistor to gruond to simulate output load. Connect 'G' (in this order from the
FET to 'signal input) to a 47pF shunt to ground (prevents Radio Frequency Interference), 1M resistor to ground (provides stable reference
to ground), and a 22n in-series to the 'signal input' jack. 'S' connects to a 4K7 resistor, which connects to a removable wire to ground.
Test 1: Beginning with that ground wire removed and the input jack shunted to ground, connect a milliamp meter between that 4K7
and ground (replacing the grounding wire). Power up and turn the bias trim pot until the multimeter on milliamp setting just measures zero
(no current flowing). Reconnect the ground wire to the 4K7 resistor and using a DC voltage multimeter setting with the negative probe on G
and the positive probe on 'S', measure and record the voltage (which should be typically 1/2 of Vcc). If you cannot get the current to read zero,
then the FET is probably bad.
Test 2: (optional) remove the signal input shunt to ground. Plug in a signal generator (a phone app will do). Using around a 1kHz signal and
an oscilloscope probing 'D', increase the signal 'volume' from low to high and observe the voltages and shape for symmetry, especially when
the FET is approaching internal clipping. Notate points where the FET begins to collapse (low signal generator strength), become asymmetrical
and clip. IMHO slightly asymmetrical results can work to great effect in an overdrive, yet too much difference can sound bad and be unusable.
Recorded values will be a second variable in 'matching' FETS, especially if you record a lot of data points to create faux amplification curves.
Discard FETS which are shorted, open, or which drift (won't stay at zero milliamps with a stable Vcc) on Test 1 or are extremely asymmetrical. Many builders I have met buy four times more than their current need, weed out the bad and bag and mark like FETS.
I match FETs in this order: zero-current bias voltage (where two FETS share a bias adjustment), gain (point where the FET hard clips), symmetry (or asymmetry, depending on what I am trying to achieve) and finally G-D / G-S forward voltage symmetry.
MOSFETS can be similarly tests using a low-gain preamp circuit, set up like the 'Prince Albert Drive.'
Hope this helps.
