B&K 1653A Variac — Vol 6: Cheatsheet (laminate)

One page, both sides, cardstock — every table you need at the bench with zero scrolling back to Vols 2–5.

6.1 How to Use This Sheet

Print, laminate, hang it on the cabinet next to the 1653A. Every row below is a synthesis of Vols 2–5 — the “detail” column tells you which volume to open if a row raises a question this sheet can’t answer. This sheet assumes you have already read Vol 1’s unknown-chassis decision tree and Vol 3’s safety theory at least once. It is a reference for the moment, not a substitute for that first read.

⚠ Danger — This sheet never replaces the one-hand rule, the pre-flight visual inspection, or the abort criteria. If a row and your gut disagree, your gut wins — kill the power and re-read Vol 3 §1 / Vol 4 before touching the chassis again.

Sibling test-equipment doc set for the same workflow: Heathkit IP-32 (regulated HV supply used after ramp-up), Heathkit SP-2717A (scope-side HV), Heathkit TT-1 (tube verification before re-installing a suspect tube), and the shared legal_ethics.md doc for acquisition/handling ethics that sit outside this instrument’s own safety scope.

6.2 1653A Quick-Spec Reference

Pull this table before you assume a number — don’t guess the dial range or fuse rating from memory. Full spec sheet and comparison against bare autotransformers (Staco 3PN1010B, Powerstat 116B): Vol 2.

Table 1 — 1653A Quick-Spec Reference

Parameter	Value	Detail
Output voltage range	0–150 Vac, continuously adjustable	Vol 2 (spec sheet)
Output current rating	2 A continuous (0–130 V); tapers above 130 V within the 300 VA envelope — don’t invent an exact number, read Table 1 in the manual before pushing past 130 V	Vol 2 (spec sheet, derating table)
AC input	120 Vac, 60 Hz, 300 VA	Vol 2 (spec sheet)
Isolation	Built-in series 1:1 isolation transformer, separate from the autotransformer winding	Vol 2 (block diagram), Vol 3 (why isolation matters)
Leakage current (mfr-published, 25°C/50% RH)	< 0.1 mA at the isolated output	Vol 2 — mfr-published figure, quoted directly
GFCI	None built in. The 1653A’s operator protection is isolation + fusing only — plug it into a GFCI-protected bench outlet if you want ground-fault protection	Vol 2, Vol 3 (Isolation vs. GFCI)
Metering	Single dual-purpose analog meter (0–150 V / 0–2 A), function selected by a front-panel VOLTS/AMPS switch — reads one at a time, not both simultaneously	Vol 2 (meter circuits), Vol 5 (calibration check)
Output protection	3 A input line fuse + internal 3 A output fuse (fuse-protected, not a resettable breaker — the breaker is the 1655A). Never bypass or up-rate	Vol 2
Meter accuracy (mfr-published, typical)	±5% of full scale, calibrated at 120 Vac input, 25°C	Vol 2, Vol 5 (verification vs bench DMM)

6.3 Front-Panel Walkthrough (quick)

Full routine-use narrative: Vol 5 §Routine Use.

Table 2 — Front-Panel Walkthrough (quick)

Control / indicator	Set / read this way	Notes
Voltage control (autotransformer wiper)	Fully counter-clockwise / zero before power-on, every time	Never start a session with the dial already up
Line switch	On only after DUT is connected and voltage control is at 0 V	Sequence matters — see ramp table above
Function switch (VOLTS / AMPS)	Select AMPS to watch current through the dwell — this is the primary abort signal; flip to VOLTS only for the end-of-dwell voltage spot-check	Single dual-purpose meter, ±5% full-scale accuracy — it can’t show both at once, don’t chase the last volt or mA
Output receptacle / terminals	Confirm DUT is OFF before making/breaking connections	Standard bench discipline, not 1653A-specific

6.4 Common Failure-Mode Quick Triage (the 1653A itself)

If the INSTRUMENT is misbehaving, not the DUT. Full failure-mode matrix with fixes: Vol 5 §Common Failure Modes.

Table 3 — Common Failure-Mode Quick Triage (the 1653A itself)

Symptom	Likely cause	Quick action
Meter reads erratic or stuck (either VOLTS or AMPS position)	Shunt or rectifier-bridge failure in the meter circuit	Cross-check against a clamp meter/DMM before trusting readings
Dead spot or arcing as the dial sweeps a range	Brush wear / carbon dust, or slip-ring/wiper-track oxidation	Do not force through a dead spot — service the brush/track (Vol 5)
Detent/stop feels loose or dial doesn’t hold position	Worn mechanical detent	Cosmetic/usability issue, not a safety abort trigger by itself

6.5 Brown-Out / Margin-Test Quick Recipe

For checking whether a DUT still runs correctly on a sagging line — not a power-up ramp. Full rationale and what a marginal result means: Vol 5 §Brown-Out Simulation.

Table 4 — Brown-Out / Margin-Test Quick Recipe

Step	Action	Notes
1	Bring the DUT up to full nominal line voltage first, using the ramp table above	Never brown-out-test a chassis that hasn’t already been safely ramped once
2	Drop the 1653A output to 95 V, hold, observe DUT behavior	Simulates a moderate brown-out; watch for dropout, hum increase, instability
3	Drop further to 85 V only if 95 V was clean, hold, observe	This is a stress step, not routine use — abort criteria above still apply

6.6 Slow-Ramp Profile by Chassis Class

Dwell = minimum hold time at each step before advancing. Longer dwell on bigger iron because larger filter capacitor banks take longer to reform and larger power transformers take longer to reveal a shorted turn as rising case temperature. Full step derivation, dwell rationale, and the leakage-current curve behind it: Vol 4 (Power-Up Workflow).

The 1653A has one meter, function-switched — set the level with the meter on VOLTS, then flip the Function pushbutton to AMPS before the dwell ends so you’re reading current at the moment that matters. It can’t show both at once.

Table 5 — Slow-Ramp Profile by Chassis Class

Chassis class	Steps (Vac, RMS at the 1653A dial)	Dwell per step	Watch for	Detail
Small AC/DC “5-tube” radio (transformerless, floating chassis)	0 → 25 → 50 → 75 → 100 → 120	2 min	Filament glow at 25 V; case warmth of the filter cap by 75 V; any smell	Vol 4 (ramp table), Vol 3 (floating-chassis hazard)
Large console radio / large iron-power-transformer set	0 → 25 → 50 → 75 → 100 → 120 → 135 → 150	3 min	Transformer hum changing pitch or growing louder; ammeter that won’t settle by minute 3	Vol 4 (ramp table)
Console / table TV (tube type, isolated on the bench)	0 → 20 → 40 → 60 → 80 → 100 → 120	3 min, 5 min at 120 V	HV flyback whine appearing; any hiss/corona; CRT anode area — do not approach until B+ is confirmed present and bled	Vol 3 (X-ray/HV), Vol 4 (ramp table)
Audio power amplifier (tube, output-transformer coupled)	0 → 25 → 50 → 75 → 100 → 125 → 150	3 min	Output-tube plates glowing dull red (bias fault) before B+ is even near nominal; DC offset at the speaker terminals if measured	Vol 4 (ramp table), Vol 5 (routine-use walkthrough)
Transmitter / RF power amplifier (tube PA, multi-hundred-volt B+)	0 → 20 → 40 → 60 → 80 → 100 → 120 → 140 → 150	5 min, hold 10 min at final step	Plate current creeping without drive applied; any corona/arcing at high-voltage tank components; smell of hot wax or varnish	Vol 4 (ramp table), Vol 3 (HV/X-ray callout)

Note — These are starting points, not guarantees. A chassis with known history (you recapped it last month) can skip steps; a chassis with unknown history (attic find, estate-sale box) never skips steps — see Vol 1’s unknown-chassis decision tree for the always-ramp / bypass-acceptable / never-touch call before you even reach for this table.

6.7 Glossary Quick Reference

Fast lookups so this sheet stands alone at the bench. Full derivation of each term: Vol 1 (terms/decision tree) and Vol 2 (circuit theory).

Table 6 — Glossary Quick Reference

Term	Quick definition	Detail
”Variac”	Originally a General Radio trademark for a variable autotransformer; now genericized (small-v “variac”) the way “Kleenex” or “Thermos” is	Vol 1
Variable autotransformer	A single tapped winding with a moving brush/wiper — adjusts voltage, but by itself gives NO galvanic isolation from the mains	Vol 1, Vol 2
Isolation transformer	A separate winding pair (here, 1:1) that breaks the direct electrical path to earth ground — the 1653A stacks this ON TOP of its autotransformer	Vol 2, Vol 3
GFCI (Class A)	Differential current sensor that trips around 4–6 mA (nominal 5 mA) of earth-fault current, in roughly 25 ms — general term; the 1653A itself has NO built-in GFCI, see the matrix below	Vol 2, Vol 3
Floating chassis	A chassis whose “ground” reference isn’t earth ground — common in transformerless AC/DC radios and hot-chassis TVs; can sit at up to full line voltage relative to true earth	Vol 3
Cap reforming	The self-healing rebuild of an aluminum-electrolytic capacitor’s internal oxide dielectric under slow, controlled voltage after long storage	Vol 4, Vol 5
Dim-bulb tester	A series incandescent lamp used as a crude self-limiting current limiter — complementary to, not a replacement for, the 1653A’s isolation + metering	Vol 4, Vol 5
One-hand rule	Keep one hand in a pocket or behind your back while probing live circuits, so a shock path can’t cross the heart chest-to-chest/hand-to-hand	Vol 3

6.8 Isolation vs. GFCI — General Safety Education (the 1653A has isolation, NOT a built-in GFCI)

These solve DIFFERENT problems. The 1653A itself provides isolation + fusing only — it has no built-in GFCI. If you want ground-fault protection on this bench, plug the 1653A into a GFCI-protected outlet/receptacle; that GFCI lives upstream of the 1653A, not inside it. Full reasoning and the “isolation can defeat a downstream GFCI” caveat: Vol 3 §Isolation vs. GFCI.

Table 7 — Isolation vs. GFCI — General Safety Education (the 1653A has isolation, NOT a built-in GFCI)

Question	Isolation transformer	GFCI
What it does	Removes the earth-referenced return path entirely	Senses an earth-fault current and trips the circuit
Present on the 1653A?	Yes — built-in series 1:1 winding, downstream of the autotransformer	No — not built in; supply one externally via a GFCI-protected outlet if you want this protection
Does it make a hot chassis “safe to touch”?	No — the chassis can still be at full potential relative to another point in the circuit	No — it only protects the earth-fault path it can see
Can it be defeated by the other?	A floating isolated secondary removes the earth path a GFCI needs to sense — so an externally supplied, GFCI-protected outlet upstream of the 1653A cannot see a fault downstream of the isolation transformer	N/A — depends entirely on where you place it relative to the isolation stage
What it’s for here	Lets a grounded scope probe touch a hot chassis without completing a mains fault loop	Would catch a genuine mains-side ground fault (frayed cord, internal short to case) — but only if you supply it yourself, upstream

Note — The 1655A sibling model adds a leakage-current MEASUREMENT function (a DUT test per UL/OSHA norms) — that’s a measurement tool for characterizing a device under test, not operator GFCI protection, and the 1653A doesn’t have that function either.

6.9 Calibration / Verification Quick Cycle (the 1653A’s own meters)

Full setup diagram and suggested interval: Vol 5 §Calibration/Verification.

Table 8 — Calibration / Verification Quick Cycle (the 1653A's own meters)

Check	Reference instrument	Frequency	Notes
Voltmeter accuracy (function switch in VOLTS)	True-RMS bench DMM (e.g., Fluke 87V/8846A class)	Annually, or after any suspected mishandling	True-RMS matters — a non-RMS meter will disagree on non-sinusoidal loads
Ammeter accuracy (function switch in AMPS)	Clamp meter or DMM in series, at a known resistive load	Annually	Cross-check at both a low and a near-full-scale current point

6.10 Mod Options Quick Reference (optional, not required for safe operation)

Full build notes and rationale: Vol 5 §Mod Opportunities.

Table 9 — Mod Options Quick Reference (optional, not required for safe operation)

Mod	What it buys	Caution
Input EMI/line filter	Cleaner mains feed to sensitive DUTs, reduced noise coupling	Add ahead of the existing fuse chain, don’t disturb its wiring
Scope-monitor output tap	A safe, isolated point to watch the output waveform without probing the DUT directly	Must stay downstream of the isolation transformer to keep the isolation benefit
External current-shunt output for datalogging	Continuous current logging during long ramps/reforms	Verify it doesn’t change the ammeter’s own accuracy or trip behavior

6.11 Current-Watch Decision Tree

Read the ammeter at the END of every dwell period, not mid-climb — Function switch on AMPS to do it, since the 1653A’s single meter can’t show volts and amps together. Full runaway-pattern triage matrix with root causes: Vol 4 §Current-Watch.

                         ┌─────────────────────────────┐
                         │  READ AMMETER AT END OF DWELL │
                         └───────────────┬───────────────┘
                                         │
                 ┌───────────────────────┼───────────────────────┐
                 │                       │                       │
        Brief inrush spike,      Current climbs but      Current SPIKES high
        then settles low         is still SLOWLY          the instant you
        and stays flat?          falling at end of         raise voltage,
                 │               dwell (not flat yet)?     even at a LOW step
                 │YES                    │YES              (<40 V)?
                 ▼                       ▼                       │YES
           ┌───────────┐        ┌────────────────┐               ▼
           │  NORMAL    │        │ CAP REFORMING   │      ┌───────────────┐
           │  advance   │        │ NORMALLY —      │      │ DEAD SHORT or  │
           │  to next   │        │ hold this step   │      │ SHORTED XFMR   │
           │  step      │        │ up to 2× dwell.  │      │ ── ABORT NOW   │
           └───────────┘        │ Still climbing    │      │ (see below)    │
                                 │ or not flat at    │      └───────────────┘
                                 │ 2× dwell?         │
                                 │        │YES        │
                                 │        ▼           │
                                 │  ┌────────────┐    │
                                 │  │ CAP LEAKY /  │    │
                                 │  │ NOT REFORMING│    │
                                 │  │ ── ABORT,    │    │
                                 │  │ do NOT push   │    │
                                 │  │ further, log  │    │
                                 │  │ and go to      │    │
                                 │  │ Vol 5 lab      │    │
                                 │  │ reform bench   │    │
                                 │  └────────────┘    │
                                 └────────────────┘

⚠ Danger — A current spike at a low voltage step is the one pattern that means “something is already broken,” not “something is healing.” Do not raise voltage further to “see what happens.”

6.12 Body-Current Physiology Quick Reference (why the one-hand rule exists)

Order-of-magnitude figures commonly cited from IEC 60479-1 body-current research — durations and exact mA figures vary by path, frequency, and individual; treat as approximate awareness levels, not a safety margin to calculate against. Full physiology discussion and the heart-crossing-path rationale for the one-hand rule: Vol 3 §One-Hand Rule.

Table 10 — Body-Current Physiology Quick Reference (why the one-hand rule exists)

Approx. AC current (60 Hz, hand-to-hand)	Typical physiological effect
~0.5–1 mA	Threshold of perception (tingle)
~5–10 mA	”Let-go” threshold region — muscle contraction can make it hard to release the conductor
~20–40 mA	Risk zone for breathing difficulty / involuntary contraction over sustained contact
~50–100+ mA (hand-to-hand, across the chest, ~1 s+)	Ventricular-fibrillation risk zone — commonly cited as the danger threshold for a heart-crossing path

Note — These numbers are exactly why a hand-to-hand path (one hand on the chassis, one hand on a grounded bench fixture) is the path to avoid — it crosses the heart. The one-hand rule isn’t superstition; it’s routing the current path away from the heart if a mistake happens anyway.

6.13 Scope-Probe / Meter Voltage-Rating Quick Reference

Full probe-rating reference table and CAT ratings: Vol 3 §Probe Ratings.

Table 11 — Scope-Probe / Meter Voltage-Rating Quick Reference

You’re probing…	Minimum probe/meter rating to use	Why
Low-voltage B+ rails, filament/heater circuits	Standard 300 V-class probe is typically adequate	Confirm against the probe’s own printed rating, not assumption
Screen/plate voltages on small-signal tube stages	600 V-class probe recommended	Margin above nominal, not just “it probably won’t hit that”
Plate B+ on audio/RF power stages, CRT second-anode-adjacent nodes	High-voltage-rated probe (kV-class, purpose-built)	A 300 V probe on a multi-kV node is a fire/flashover risk, not a “slightly out of spec” risk
Any CRT anode lead directly	Do not probe casually at all — treat as a dedicated HV-probe, trained-procedure task	See Vol 3’s X-ray/HV callout — this is also an X-ray-emission concern on some high-anode-kV tubes, not just a shock concern

⚠ Danger — Matching probe rating to the node’s WORST-CASE voltage, not its expected nominal, is the rule — a marginal supply or a fault condition can push a node well past its “normal” number.

6.14 Voltage Spot-Check Points by Chassis Class

Take these AFTER the ramp completes, meter still through the 1653A’s isolated output, one hand in your pocket. Point-by-point rationale and expected normal ranges for specific circuits: Vol 4 (post-ramp spot-checks), Vol 2 (circuit theory for what each node is).

Table 12 — Voltage Spot-Check Points by Chassis Class

Chassis class	Primary spot-check points	Typical healthy indicator	Detail
Small AC/DC radio	B+ at first filter cap; across the pilot/dial lamp if fitted	B+ present, steady (not drifting); lamp at normal brightness by full line	Vol 4 §Post-Ramp
Large console radio	B+ at filter choke input and output; screen-grid voltage at output tube	Choke-input B+ higher than choke-output B+ by the choke’s IR drop; screen voltage stable	Vol 4 §Post-Ramp, Vol 2 (circuit theory)
Console/table TV	Low-voltage B+ rail; screen/focus voltages if externally accessible; do not probe the flyback/anode lead itself	Rails at nameplate/schematic nominal; no corona sound near the flyback	Vol 3 (X-ray/HV callout), Vol 4 §Post-Ramp
Audio power amp	B+ at output-tube plates; bias voltage at each output-tube grid; DC offset at speaker output	Plate voltages matched between paired output tubes; bias within the tube’s rated range	Vol 4 §Post-Ramp, Vol 5 (routine use)
Transmitter/RF PA	Plate B+ with no drive applied; screen B+; grid bias	Plate current near zero with no drive; bias present before B+ is ever applied	Vol 3 (HV/X-ray), Vol 4 §Post-Ramp

6.15 Hazardous-Materials Quick-Reference

Condensed from Vol 3’s full HAZMAT catalog — identification cues and handling only. Full year ranges, citation basis, and detailed handling procedure: Vol 3 (Hazardous-Materials Catalog).

Table 13 — Hazardous-Materials Quick-Reference

Material	Approx. years found	Where on a chassis	Handling
PCB-bearing oil	Pre-1979 (TSCA ban)	Oil-filled paper capacitors, some power-transformer potting	Do not open/drain; treat leaking units as hazardous waste, not trash
Asbestos	Pre-1980s, tapering after	Wirewound-resistor cores, flash/spark insulation, some heater sleeving	Do not sand, cut, or brush; if friable, bag and stop work
Selenium rectifiers	~1940s–1970s (largely displaced by silicon by late 1960s)	Stacked-plate rectifiers, often finned/painted gray-green	Overheated units release SeO2/H2Se fumes — ventilate immediately, do not inhale the “rotten” smell
Mercury (switches/relays)	Through ~1990s in some industrial/telecom gear	Tilt switches, mercury-wetted relays, some older thermostats on chassis	Do not crush; handle a broken bulb as a mercury spill, not sweepings
Beryllium oxide (BeO)	Ongoing where used — mainly RF power-transistor and diode heat-sink spreaders	White/gray ceramic insulator wafers under RF power devices	Never sand, grind, or crack — dust is the hazard, not the intact part; discard intact if in doubt
Wax-paper / “black beauty” caps	1940s–1960s	Coupling/bypass caps throughout signal and bias circuits	Leaking wax is a mess, not typically a toxic hazard — but treat as failed, replace, don’t reuse
Transformer tar/pitch potting compound	Widespread through mid-20th century	Power and output transformer potting, some multi-section can capacitors	Overheated tar smells sharp/acrid — a strong tar smell during ramp-up is itself an abort trigger, not just a hazmat note
Rubber/cotton-braid wire insulation	Pre-1960s-ish, tapering after	Chassis wiring, especially point-to-point runs on early sets	Brittle insulation cracks and can expose bare conductor — inspect before energizing, don’t just visually skim past it

⚠ Danger — Selenium-rectifier and BeO hazards are both release-on-damage, not present-in-normal- use. Normal slow-ramp operation should never release either — it’s overheating (selenium) or mechanical damage (BeO) that turns them dangerous. That’s one more reason the ramp protocol above matters.

6.16 Emergency-Abort Recipe

Table 14 — Emergency-Abort Recipe

Step	Action	Notes
1	Turn the 1653A’s output/voltage control to 0 V — do not just flip the DUT’s own switch	Killing at the source is faster than fighting a possibly-fused or possibly-arcing DUT switch
2	Unplug the 1653A’s line cord from the wall	Removes the mains feed entirely — don’t assume the fuse has already opened
3	Let the chassis sit — do not touch it, probe it, or open it up — until filter capacitors have had time to bleed down	Large filter caps can hold lethal charge for minutes after power is removed; see Vol 3 (one-hand rule / physiology) for why “it’s unplugged” is not “it’s safe”

Abort triggers (any ONE is sufficient): current excursion past roughly 2× the expected value for that step · any visible smoke · any audible arcing, sizzling, or corona hiss · any hot-component, ozone, or burning-insulation smell. Full abort-criteria matrix with per-symptom likely cause: Vol 4 §Abort Criteria.

⚠ Danger — Never troubleshoot during an abort. Kill it, walk away, THEN diagnose with the unit fully de-energized and bled down.

6.17 Full-Session Checklist (top to bottom)

The condensed run order — every other table on this sheet slots into one of these phases. Full narrative for each phase: Vol 4 (power-up workflow) and Vol 5 (routine use).

Table 15 — Full-Session Checklist (top to bottom)

Phase	Do this	Table/section on this sheet
1. Pre-flight	Visual inspection of line cord, plug, strain relief; confirm the 1653A’s own line cord ground pin is intact and the bench outlet is properly earthed; confirm the 3 A line fuse and internal 3 A output fuse are both intact (no history of unexplained blows); flip the Function switch to VOLTS and AMPS in turn and confirm the meter needle responds/rests correctly in both positions; voltage control fully counter-clockwise (0 V) before power-on	Front-Panel Walkthrough
2. Chassis-out inspection	Look for bulging/vented caps, leaking wax-paper caps, charred resistors, cracked tube envelopes, crumbling rubber/cotton wiring, transformer tar smell	Hazardous-Materials Quick-Reference (identify what you’re seeing)
3. Connect DUT	DUT switch OFF, connect to isolated output, confirm chassis class	Front-Panel Walkthrough
4. Ramp	Step voltage per chassis class, dwell, read ammeter at end of each dwell	Slow-Ramp Profile by Chassis Class
5. Watch current	Classify what the ammeter is doing at each step	Current-Watch Decision Tree
6. Abort if triggered	Any trigger condition → kill immediately, do not diagnose live	Emergency-Abort Recipe
7. Post-ramp spot-check	Verify key voltages with the isolated meter, one hand in pocket	Voltage Spot-Check Points by Chassis Class
8. Shutdown	Voltage control back to 0 V before switching off; let bleed down before opening the chassis further	Emergency-Abort Recipe (same bleed-down caution applies to routine shutdown)

6.18 Ramp-Profile Chart

Figure 1 — Representative 30-minute slow-ramp voltage profile for a large console radio, plotted as a voltage-versus-time step chart from 0 V to 150 V across seven dwell steps, with each riser and each 3-minute dwell plateau labeled. Source: hand-authored SVG.

6.19 Depth Index — Where Each Table Came From

Table 16 — Depth Index — Where Each Table Came From

This sheet’s section	Full treatment
1653A quick-spec reference	Vol 2 — full spec sheet + comparison vs. Staco/Powerstat
Front-panel walkthrough	Vol 5 §Routine Use
Common failure-mode quick triage	Vol 5 §Common Failure Modes
Brown-out quick recipe	Vol 5 §Brown-Out Simulation
Glossary	Vol 1 (terms), Vol 2 (circuit theory)
Isolation vs. GFCI general safety education	Vol 3 §Isolation vs. GFCI
Calibration/verification cycle	Vol 5 §Calibration/Verification
Mod options	Vol 5 §Mod Opportunities
Body-current physiology	Vol 3 §One-Hand Rule
Probe voltage-rating reference	Vol 3 §Probe Ratings
Slow-ramp profile	Vol 4 (Power-Up Workflow) — ramp-profile table + cap-reform theory
Current-watch tree	Vol 4 §Current-Watch — runaway-pattern triage matrix
Voltage spot-checks	Vol 4 §Post-Ramp Spot-Checks; circuit context in Vol 2
Hazmat quick-ref	Vol 3 (Bench Safety) — full HAZMAT catalog with sourcing
Full-session checklist	Synthesizes Vol 4 + Vol 5 into one run order
Abort recipe	Vol 4 §Abort Criteria; physiology/one-hand rule in Vol 3
Unknown-chassis routing (not reproduced here)	Vol 1 — always-ramp / bypass-acceptable / never-touch decision tree

Sources

B&K Precision 1653A/1655A data sheet (v053018, 2018; rftesolutions.com mirror) and Transcat-hosted 1653A/1655A instruction manual: confirmed 0–150 Vac output range, 0–2 A output current, built-in isolation transformer, a single dual-purpose analog meter selected by a front-panel VOLTS/AMPS function switch, a 3 A input line fuse plus a separate internal 3 A output fuse, leakage less than 0.1 mA (25°C/50% RH) at the isolated output, and a published meter accuracy of ±5% of full scale (typical, 25°C, calibrated at 120 Vac input) — used to set this sheet’s voltage/current framing. Both documents contain ZERO mention of GFCI, ground-fault, or interrupter, and the manual’s front-panel control list has no TEST/RESET buttons — the 1653A’s operator protection is isolation + fusing only, no built-in GFCI. (The 1655A sibling adds a leakage-current MEASUREMENT function — a DUT test per UL/OSHA norms — which is a measurement tool, not operator GFCI protection, and the 1653A doesn’t have that function either.) NOT used to invent any number B&K doesn’t publish.
UL 943 Class A GFCI trip-threshold research (Mike Holt forums summary, code-authorities.ul.com PDF, csemag.com “UL’s new GFCI classes”): confirmed Class A trips in the 4–6 mA range (nominal 5 mA), with a published response-time curve (~25 ms at 5 mA, tightening to ~20 ms above 300 mA). This is general GFCI-class background for the Isolation-vs-GFCI education table only — it describes an EXTERNAL, user-supplied GFCI-protected outlet, not any component inside the 1653A, which has none.
All ramp-step counts, dwell times, and abort thresholds (“~2×” current excursion) are this program’s synthesized, conservative bench practice — not manufacturer-published numbers for any specific chassis. They are deliberately framed as starting points in the sheet’s own callouts, not hard specs.
Body-current physiology figures (perception ~0.5–1 mA, let-go region ~5–10 mA, ventricular-fibrillation risk zone commonly cited around 50–100+ mA for a sustained hand-to-hand path) were checked against general summaries of IEC 60479-1 body-current research. These figures vary meaningfully by exposure duration, current path, and individual physiology in the underlying standard’s own tables — this sheet intentionally presents them as rounded order-of-magnitude awareness bands, not a calculated safety margin; Vol 3 gives the fuller treatment.
The isolation-can-defeat-a-downstream-GFCI relationship is standard bench-safety reasoning for isolation transformers paired with GFCI protection generally. It applies here only in the sense that IF a user supplies an external, GFCI-protected outlet upstream of the 1653A, that GFCI cannot see a fault occurring downstream of the 1653A’s own isolation transformer — this is general reasoning about isolation transformers, not a manufacturer statement about any component inside the 1653A itself.
Historical hazardous-material year ranges (selenium rectifiers largely displaced by silicon through the 1960s–70s, PCB TSCA ban 1979, asbestos tapering through the 1980s, mercury switches persisting into the 1990s in some equipment) are broad, commonly cited industry ranges rather than precise dated cutoffs — treat any specific unit’s actual construction date as the deciding factor, not the year range alone.

This volume is a pure synthesis of Vols 2–5’s own sourced material; it introduces no new primary claims beyond the two web-verified items above. Any correction made to a fact in Vols 2–5 during review should be mirrored here, since this sheet restates those numbers in compressed form.