VBA Curve Tracer — Vol 4: Using It — Reading Component Curves

4.1 Setting up a measurement

Before connecting a part, set the controls conservatively: choose the voltage range that comfortably exceeds the voltage you want to see, set the Current control low and pick a Current Range multiplier that keeps you inside the device’s safe dissipation, select BJT or FET, set Polarity (N for NPN/N-channel, P for PNP/P-channel), pick the number of Steps and a Step Output magnitude, and give yourself some Step Delay if you are working at high power. Then bring the Current and Voltage controls up gently and watch the curves grow. The single most important habit with any curve tracer is current limiting — it is genuinely easy to blow the part up, and the whole point of the adjustable limit and the per-step delay is to stop that.

On the scope, X-Y mode with a reasonably deep record (Versteeg suggests using a large sample memory, e.g. tens of thousands of points, and a modest bandwidth limit — a 20 MHz limit helps) gives a clean, low-noise trace.

4.2 Reading a bipolar transistor

A BJT produces the classic fan of output curves: horizontal axis V_CE, vertical axis I_C, one curve per base-current step. Reading it:

• The steps. Set, say, 50 µA per step. The first curve is I_B = 50 µA, the second 100 µA, the third 150 µA, and so on up the family.
• Gain (hFE / β). Pick a curve, read its collector current, and divide by that curve’s base current. Versteeg’s worked example: a curve sitting at I_C ≈ 10 mA for I_B = 50 µA gives hFE ≈ 10 mA / 50 µA = 200. Evenly spaced curves mean constant gain; curves that bunch up at the top show gain falling off at higher current.
• The knee and saturation. At low V_CE the curves rise steeply out of the origin (the saturation region) before flattening — that knee is V_CE(sat). For a 2N3904 at I_C = 10 mA with I_B = 1 mA, Versteeg measured V_CE(sat) ≈ 250 mV (datasheet ~200 mV).
• The Early effect. The “flat” active region actually tilts gently upward — collector current creeping up with V_CE. That slope is the Early effect and its extrapolation gives the Early voltage; a steeper slope means a softer, lower-output-impedance device.
• Datasheet reality check. The blog’s 2N3904 example lines up the display against the datasheet’s hFE-vs-I_C spread (roughly hFE 40 min at 0.1 mA, 100–300 at 10 mA, 30 min at 100 mA) — exactly the sort of verification the instrument is for.

4.2.1 Breakdown tests

By connecting only some of the leads you can pull out the different breakdown voltages. Base open (C and E connected) gives BV_CEO — Versteeg’s 2N3904 broke down near 60 V against a 40 V spec. Base shorted to emitter gives BV_CES; emitter open (C and B connected) gives the higher BV_CBO. These are high-voltage, potentially destructive tests — use the 200 V range with a tight current limit and watch for snap-back.

4.3 Diodes and LEDs

A diode traces as a single curve. In the forward direction (anode to C, cathode to E) you see the conduction knee — for a 1N4148 Versteeg reads it starting around 500 mV and climbing to ~750 mV as current rises. Flip the Polarity switch to see the reverse region: the same 1N4148 showed reverse breakdown around 180 V (spec is 100 V minimum — real parts often exceed it). Zeners are read the same way in reverse: the curve stands up at the Zener voltage, and you can read how it moves with current — one example sat at at least 10.2 V at 100 mA, illustrating the finite dynamic resistance. LEDs show a forward knee that depends on colour; at a 40 mA test current Versteeg records roughly red 2.2 V, green 2.2 V, blue 3.4 V, white 3.4 V — the same physics that makes a red LED usable in a simple constant-current dropper where a blue one needs more headroom.

4.4 FETs

FETs are traced as a drain-current family stepped by gate voltage rather than current, so switch the device type to FET and set the Step Output in volts. The wrinkle is where the interesting action sits:

• Enhancement MOSFETs turn on above a gate threshold, so use the Offset to place the zero step near V_GS(th) and step from there. A small-signal part’s threshold is typically a volt or two (positive-going for N-channel, negative-going for P-channel — Versteeg’s small-signal example is a P-channel LP0701 with V_GS(th) ≈ −1 V), on a modest current range with the limit at ~50 %. Watch temperature and use step delay for thermal protection.
• Depletion-mode JFETs are normally-on: with gate and source shorted they pass their full I_DSS, and you close them with a negative V_GS. Because the internal step generator’s offset range is limited (the default FET offset is around ±2 V), deeply-pinched JFETs may need external gate bias to sweep fully. Example parts from the blog: a 2N4391 with I_DSS ≈ 60 mA and V_GS(off) ≈ −4.7 to −5.1 V, and a 2N5754 with I_DSS ≈ 3 mA and V_GS(off) ≈ −1.5 V.
• Breakdown (BV_DSS). As V_DS climbs, drain current eventually runs away at avalanche — for the small P-channel LP0701 in Versteeg’s examples that onset appears around 16.5 V (its rated BV_DSS).

4.5 Matching pairs

The front panel’s DUT Select switch and the two Left/Right device positions exist for matching. Trace one candidate, note its family, then flip to the other under identical settings and overlay the curves. Parts whose families sit on top of each other — same knee, same step spacing (same hFE or I_DSS), same slope — are a matched pair. This is the everyday use for building differential input stages and complementary output pairs, and it is far more informative than matching on a single hFE number, because you are matching the whole curve.

4.6 Limits and gotchas

• You can destroy the DUT. Curve tracing deliberately runs devices near their limits. Always set the current limit first; use Step Delay at high voltage/current to cut average dissipation and avoid thermal “blooming” of the curves. Versteeg’s power-device examples (e.g. an MJL3281A) explicitly cap current — running without a limit “could have killed the transistor.”
• Range interactions. High-power, high-gain parts can rail the outputs on the ×10 current setting — use ×1. The blog notes an MJL3281A traced at 100 µA/step read hFE ≈ 90 at 10 V on ×1, but ×10 drove the output into its limit.
• The step offset is not a calibrated voltmeter. For precise gate-bias work, confirm the actual offset/step voltage with an external DMM rather than trusting the dial.
• JFETs need negative steps the internal generator does not fully provide; expect to add external bias for deeply-pinched parts.
• It needs your scope. The VBA has no built-in screen — the measurement quality depends partly on the scope’s X-Y performance and your record/bandwidth settings.
• It is an analog instrument. There is no data logging, cursor readout, or automated parameter extraction — you read values off the scope graticule. That is the trade for its simplicity, immediacy and buildability.

Used with a little discipline, the VBA turns a scope you already own into a proper component characteriser — the tool that tells you not just whether a part works, but how well, and whether two of them are truly alike.