Tech page
Tools · Free
FET Playground
Measure a real JFET with a multimeter and a resistor, then design the stage around that exact part — not a datasheet typical. Live schematic, transfer and output curves, load line, frequency response, SPICE export.
Two numbers define a JFET in the square-law model — Idss and Vp — and datasheets give a range that can span five to one. That spread is the thing that actually burns pedal builders: bias a stage from a "typical" number and the part on the bench sits somewhere else entirely. This starts at the part itself: a guided jig flow gets Idss and Vp from a DMM and a resistor, and every designer downstream works from that measured device.
From there it's a live workbench, not a form: a common-source stage with a clickable schematic, transfer and output characteristics with the load line drawn in, symmetric-clipping headroom, and — new in v2 — the frequency domain, with a real source-bypass shelf and Miller effect against a real pickup. MOSFET boosters and buffers are in too, alongside named presets (Fetzer Valve, a SHO-style booster) that open pre-configured with commentary on why they're built that way.
Honest numbers
What is exact and what is estimated
The device model is a square-law fit (real JFETs run an exponent closer to 1.8–2.2), rds is ideal unless you enter one, and the frequency response is small-signal about the operating point, so it says nothing about distortion. Device data comes from Fairchild datasheets where marked verified; the rest is seeded from memory and flagged in the part list. Every approximation is labelled where it's used, not buried in a footnote — and none of it has been checked against a part on a real bench yet.
Reading the workbench
Three things measuring first changes
The datasheet number is a range, not a part
A J201's Idss can vary 5× within one reel — the datasheet spread is drawn on the transfer curve as a band so 'I didn't measure' is visibly a gamble. The guided jig flow gets Idss and Vp from two DMM readings and a resistor; the stage designer then works from your part's actual square law, not the typical one.
The schematic is the interface, not a diagram of one
Click a component in the live schematic, change its value, and the node voltages, the load line and the operating point move with it. That loop — change a value, watch what actually happens — is faster to build intuition with than a table of formulas, and it's the same engine driving both the designer and the teaching panels underneath it.
Bypassed, unbypassed, or somewhere between
A source-bypass cap does not just add gain — it moves a corner frequency, changes headroom, and shifts where the stage clips first. The frequency-response view puts a Bode plot next to the DC operating point so a bias decision and a tone decision stop being two separate calculations.
Also on the bench
The same discipline, tube side
FET Playground measures a real JFET instead of trusting a datasheet typical. Bias Bench does the equivalent job for tube bias — fixed and cathode, push-pull and single-ended, with screen-grid current and PSU sag accounted for.
Free, and staying that way
If this saved you a trip to the bench, chip in
No account, no paywall, no ads — and that's not changing. A one-time contribution covers the bench time, test gear, and hosting that make free tools like this possible.
Tech page
Your Cable Is a Tone Control
Tech page
Building Better Pedals: The Real Role of Components in Guitar Tone
Measured the part and the stage still doesn't sound right on the bench? I take a small number of one-to-one consultations →