Test reports

Equipment measured with Cascade, one report per device, in one standard format. Chris asked for this on 2026-10-24: "create a new section of The Correction Log where it compiles test reports of equipment. this is where we can show and archive standard reports once we have an automated standard battery".

The battery is not automated yet, so the archive below is empty. This page sets out what every report will contain, so the first one and the fiftieth can be compared.

The archive

No reports yet. The first will appear here once the standard battery runs automatically.

The conventions every report follows

Chosen by Chris on 2026-10-24, after Amir's at Audio Science Review, adjusted for older equipment. "i think the crosstalk test should be performed at 5 watts. I'll follow some conventions set by Amir in Audio Science Review".

	convention
Receiver or integrated amp	a fixed gain of 29 dB, as ASR (a THX figure, in Amir's words): the volume is set once to the detent nearest 29 dB, then the input is raised until the output reaches 5 W. Chris chose this over a fixed 2 Vrms input on 2026-10-24: "lets do the gain rather than the input"
Amplifier output	5 W into 8 Ω (6.32 V). ASR quotes 4 Ω; most of the equipment here is older and may not be happy at 4 Ω
Setting the volume	with a low input, the knob is turned until the power meter's gain reads 29 dB, then left there for every test. A stepped knob uses the nearest detent, and the report states the gain reached (the Scott 350R: detent 22, +29.6 dB)
Power amplifier (no knob)	the input is raised until the output reaches 5 W
Maximum power	by hand: the knob is turned up with the power meter and the STOP AT THD limit armed
Headline numbers (SINAD, THD+N, signal-to-noise)	a narrow analysis band of 22.4 kHz, as ASR
Distortion across frequency	a wide band of 80 kHz, or 30 kHz for class-D amplifiers, so the 2nd and 3rd harmonics stay in view at high frequencies
Every number	carries its bandwidth, its load and its level

What a report contains

The device and the rig: make, model, condition, the cables, the load, the date, and the instrument.
The setup step: the knob position, the gain reached, and the input that gives 5 W.
The battery, every test at that knob setting: - a dashboard: the spectrum and SINAD at 5 W, 1 kHz; - frequency response, and its −3 dB points; - THD+N against frequency; - THD against level, with the input swept from low up to the 1 % THD point; - noise and dynamic range; - crosstalk in both directions; - SMPTE intermodulation distortion; - power against distortion at several frequencies.
Maximum power at 1 % THD, measured by hand.
Pass or fail against limits written down before the run.
The data: every CSV, with the instrument state it was measured in.

Where this stands

The pieces exist one at a time. Cascade already has the dashboard spectrum and SINAD, frequency response (to 20 kHz, without −3 dB points yet), THD+N against frequency, THD against level and the power meter. IMD runs only as a Developer probe so far. Crosstalk, noise and the multi-frequency power graph are still to build. THD+N, IMD and the generator output were checked against APWIN on a real receiver in October. What is missing is the battery itself: one procedure that runs them in order with the conventions above and writes the report. The plan is in docs/BACKLOG.md ("Standard battery and report").