Tube Testing & Measurement

Emission testers (e.g., Sylvania 140, Precision 912) are the simplest devices. They tie the grid to the plate, operating the tube as a diode, and measure total cathode current at a low, non-realistic voltage. The result is a go/no-go meter reading. Their main flaw: a tube that tests "good" in diode mode may perform poorly under actual amplifier conditions because grid control is never evaluated.

Mutual conductance (Gm) testers represent a major step up. Instruments like the Hickok 539C apply an AC signal to the grid and measure the AC plate current to determine Gm in micromhos at a specific, realistic operating point. They also test for shorts, gas, and leakage between all elements. The Hickok method uses a bridge circuit to read Gm directly on a calibrated scale, with roll charts listing settings for thousands of tube types.

Curve tracers (uTracer 3+, Sofia) are the gold standard. They sweep plate voltage while stepping grid voltage, plotting complete Ia/Va characteristic families. Every aspect of a tube’s behavior is revealed: linearity, emission saturation, grid current onset, and matching potential. The uTracer connects via USB and captures a full set of curves in under a minute using short high-voltage pulses to minimize dissipation.

Military testers (TV-7, AN/USM-118, AVO VCM-163) combine rugged construction with good accuracy. The TV-7 was the standard U.S. military Gm tester for decades. The AVO Valve Characteristic Meter was the British equivalent, capable of plotting basic curves. Calibrated military units remain highly sought after and command premium prices.

A Gm tester applies a small AC signal (typically 5 kHz on Hickok instruments) to the control grid while the tube is biased at a specific DC operating point. The resulting AC component of the plate current is measured and divided by the applied grid signal to yield Gm in micromhos (\u00b5mhos). For a 12AX7 at its typical test point, a healthy tube reads around 1600 \u00b5mhos — meaning a 1V grid swing produces 1.6 mA of plate current swing.

Crucially, Gm varies with the operating point. A tube biased at high current will show higher Gm than the same tube biased at low current, because the plate curves are steeper in the high-current region. This is why the Hickok roll chart specifies exact plate voltage, screen voltage, and grid bias for each tube type. Testing at a different operating point gives a different Gm number that cannot be directly compared to the published spec.

Reading a Hickok involves consulting the roll chart for your tube type, setting the selector switches to the prescribed positions, inserting the tube, and reading the meter after the warm-up period. The meter typically has three scales: a GOOD/BAD emission scale at the top (for quick go/no-go), a Gm scale in micromhos in the middle, and a gas/leakage scale. The Barkhausen equation ties the three fundamental tube parameters together: amplification factor µ equals Gm times plate resistance rp.

A curve tracer sweeps the plate voltage from 0 to typically 400–500V while stepping the grid voltage in fixed increments (e.g., −1V steps). The resulting family of curves plots Ia vs Va for each Vg value. Reading these curves reveals everything about the tube. Steep curves indicate high Gm. Evenly spaced curves mean linear amplification — each volt of grid change produces the same change in plate current regardless of where you are on the curve. Crowding at the top indicates emission saturation.

The uTracer 3+ workflow is straightforward: connect the tube via an adapter socket, set the sweep range and grid steps in the software, click "Measure," and the device fires a series of short high-voltage pulses (each lasting only milliseconds to minimize heat). Within 30–60 seconds, you have a complete set of curves overlaid on screen. The software can export CSV data for spreadsheet analysis and overlay curves from multiple tubes for direct comparison.

When comparing an aged tube to a new one, three differences are immediately visible. First, the maximum plate current is lower — the aged tube’s depleted cathode simply cannot supply as many electrons. Second, the curve spacing narrows, indicating reduced Gm. Third, the curves flatten earlier (saturation onset at lower current), meaning the tube’s linear operating region has shrunk. A tube with kinks or discontinuities in its curves should be discarded regardless of its Gm reading, as these non-linearities cause audible distortion artifacts.

In a push-pull output stage, two or more tubes alternately amplify opposite halves of the signal. If the tubes differ, the output transformer sees a DC imbalance that magnetizes the core and reduces headroom. Matching ensures both tubes draw nearly identical plate current at the operating point. The key parameters to match are: Gm at the operating point (±5% for a good match, ±2% for premium), plate current at the operating point, and ideally the entire plate curve shape for the closest behavior across all signal levels.

Bias matching measures plate current at the actual operating point and is the most reliable single-parameter method. Gm matching adds confidence that the tubes will also have similar gain characteristics. For quad sets (four output tubes), compute mismatch relative to the group average — four tubes at 3% mismatch outperform two perfectly matched pairs with 8% mismatch between pairs. Phase inverter matching is also important: the two halves of a 12AX7 or 12AT7 driving a push-pull stage should be matched for Gm to ensure equal drive to each output tube.

Shorts occur when internal elements touch or when conductive material bridges the gap. Grid-to-cathode shorts are the most common and most dangerous — they can cause a tube to draw runaway current, overheating the plate and potentially destroying the output transformer. On a Gm tester, the shorts test lights a neon indicator for the affected element pair. On a curve tracer, a shorted tube draws current even with the grid driven deeply negative.

Gas contamination means residual molecules inside the envelope have reached a level where they ionize during operation. The telltale sign is a blue or purple glow between cathode and plate (not to be confused with the harmless blue fluorescence some glass types exhibit). Testers detect gas by applying negative grid bias and measuring positive grid current — in a vacuum, no current should flow to a negatively biased grid. Readings above 5–10 µA indicate a gassy tube. Gas contamination accelerates over time and the tube should be retired.

Microphonics describes mechanical vibration of internal elements that modulates the electron flow, producing audible ringing or feedback. The classic tap test — gently tapping the tube with a pencil eraser while the amplifier is on — reveals microphonic tubes through audible pinging or ringing in the speaker. Preamp tubes (12AX7) are most susceptible due to their high gain. Microphonic output tubes are less problematic as the lower gain reduces the effect.

A modern tube testing bench centers on a PC-controlled curve tracer. The uTracer 3+ (around €280 as a kit, €400 assembled) is the most popular choice among enthusiasts and small-scale tube sellers. The Sofia tester is a commercial alternative with a polished software interface and higher voltage capability. Both connect via USB and include software for data acquisition, curve display, tube comparison, and CSV export for spreadsheet-based matching workflows.

Essential accessories include: a regulated heater supply (6.3V and 12.6V), a set of adapter sockets (octal, noval, compactron, rimlock), a safety enclosure for high-voltage testing, and a computer running the tracer software. For serious matching work, a temperature-stable environment helps ensure consistent readings — tube parameters drift with ambient temperature. Some testers also recommend a mains voltage stabilizer since heater voltage affects emission readings.

Not every tube needs to be at 100% to be useful. The decision to keep or discard depends on the intended application. A tube at 75% Gm is perfectly serviceable in a reverb driver, tremolo oscillator, or effects loop buffer where maximum gain and low noise are not critical. The same tube would be a poor choice for V1 of a high-gain guitar preamp where every decibel of headroom and every fraction of noise matter.

The standard grading system used by tube dealers provides a common language for tube quality. NOS (New Old Stock) grade means the tube tests at or above its original published specification — these are unused vintage tubes in their original boxes. Strong means 85–95% of spec. Good means 70–85%. Weak means 50–70%. Bad means below 50% and the tube should be retired from audio service. Some dealers add a "test as new" category for used tubes that still meet NOS specifications.

Building a tube inventory system is valuable for anyone who tests regularly. Record each tube’s type, manufacturer, date code (if visible), test date, Gm reading, plate current at your standard operating point, shorts/gas test results, and any notes about noise or microphonics. A simple spreadsheet works well. When you need a matched pair, sort by plate current at the target operating point and pick the closest values. Retest tubes periodically (every 6–12 months for tubes in active use) to track degradation and anticipate replacements.