Heathkit IP-32 — Vol 6: Cheatsheet & the Generational Pair

6.1 How to Use This Volume

This volume is the bench laminate — the page taped inside the lid, not the page you read cover to cover. Everything here is a synthesis of Vol 1 through Vol 5; nothing new is derived. Where a row needs more context than a table cell can hold, it carries a detail: Vol N § pointer back to the volume that explains it.

Read Vol 1–5 once, slowly, before you ever apply mains power to an IP-32. Then keep this volume open while you work. The Danger callouts in this volume are not decorative — the IP-32’s raw B+ bus sits at roughly 600 V DC even though the regulated output tops out at 400 V, and there is no active current-limit circuit protecting the B+ output at all (only the bias line is current-limited). That combination — lethal voltage, no B+ fold-back, and 60+-year-old paper/electrolytic parts — is why this cheatsheet exists.

⚠ Danger — Every number, step, and warning in this volume assumes a competently refurbished unit with a verified-good chassis ground and a 3-A rear fuse fitted. If you have not yet worked through Vol 5 (Calibration & Refurbishing), do not use the “quick-steps” below as a substitute for it.

---

6.2 Quick Specs — Outputs & Performance

Table 1 — 2. Quick Specs — Outputs & Performance

Parameter	Spec	detail
B+ output	0 – 400 V DC, regulated, continuously variable	Vol 1 §2, Vol 3 §2
B+ current	100 mA continuous · 125 mA intermittent max	Vol 1 §2, Vol 3 §2
B+ current limiting	None — protected only by the 3 A rear fuse and the 6L6 pass-tube dissipation limit	Vol 2 §4, Vol 4 §3
Raw B+ (internal, pre-regulation)	≈ 600 V DC no-load on the doubler caps — lethal, not user-accessible in normal operation	Vol 2 §3, Vol 3 §3
Negative bias (C−)	0 to −100 V DC @ 1 mA, variable	Vol 1 §2, Vol 3 §2
Bias current limiting	Internally current-limited (this is why there is no bias current meter)	Vol 3 §5
Filament output	6.3 V AC @ 4.0 A, two binding posts	Vol 1 §2, Vol 3 §2
Load regulation	± 1 % no-load to full-load, 100–400 V DC	Vol 1 §3
Line regulation	± ½ V output change for a 10 V line change	Vol 1 §3
Ripple / jitter / noise	< 10 mV	Vol 1 §3
Output impedance	< 10 Ω, 1 Hz – 1 MHz	Vol 2 §3
Minimum bleeder load	≈ 3 mA, always present (4 × 27 kΩ / 2 W bleeder)	Vol 2 §4, Vol 3 §4
Mains input	105–125 V AC, 50/60 Hz; rear 3 A fuse	Vol 1 §2
Dimensions / weight	13″ W × 8½″ H × 7″ D; 12 lb (16 lb shipping)	Vol 1 §2
Era / kit price	1962–1967; kit ≈ $56.95	Vol 1 §1

⚠ Danger — The absence of a B+ current-limit circuit is the single most important fact on this page. Short the B+ output and the only things standing between the pass tubes and destruction are the fuse and however fast you let go of the leads. This is not a “current-limited bench supply” in the way a modern switcher is. Treat every B+ connection as if it were live mains.

6.2.1 Active-device complement (for fault-finding at a glance)

Table 2 — 2.1 Active-device complement (for fault-finding at a glance)

Device	Qty	Role	detail
6L6	2	Series-pass regulator tubes, wired in parallel	Vol 2 §2
6BH6	1	Control / error amplifier	Vol 2 §2
0A2	2	Gas voltage-reference tubes (150 V each → −150 V / −300 V rails)	Vol 2 §2, Vol 3 §3
6X4	1	Bias-supply rectifier tube (full-wave)	Vol 3 §3
Silicon diodes	4	B+ voltage-doubler rectifier	Vol 3 §3
Selenium rectifiers	2	Screen-supply half-wave rectifier (wear item)	Vol 3 §3, Vol 5 §2

There are no transistors anywhere in the IP-32. The silicon parts are all rectifier diodes; every gain/regulation device is a tube. See Vol 2 for why this matters (the “solid-state” label some catalogs and secondary sources attach to this unit refers only to the B+ rectifier, not the regulator).

---

6.3 Front-Panel Quick Reference

Table 3 — 3. Front-Panel Quick Reference

Control / indicator	Function
OFF / STANDBY / ON rotary switch	STANDBY keeps filaments warm and kills B+; ON enables B+
B+ OUTPUT pot	Sets 0–400 V DC output
C− OUTPUT pot	Sets 0 to −100 V DC bias output
METER SWITCH	Selects whether the single voltmeter reads B+ (0–400 V) or C− (0–150 V range)
D.C. OUTPUT VOLTAGE meter	Reads whichever rail METER SWITCH selects
D.C. OUTPUT CURRENT meter	0–150 mA full scale, reads B+ current only (no bias current meter)
Pilot light	Mains-on indicator
Binding posts (×7)	B+, C−, two filament (6.3 V AC), chassis ground, two COMMON

---

6.4 Operating Quick-Steps

This is the condensed form of Vol 4’s operating procedure. Follow it in order every time — the STANDBY warm-up step is not optional housekeeping, it protects the 6L6 cathodes from cold-start emission stress and lets the 0A2 tubes strike cleanly before B+ is applied.

Table 4 — 4. Operating Quick-Steps

Step	Action	Why
1	Confirm B+ OUTPUT and C− OUTPUT pots are fully counter-clockwise (minimum) before power-up	Prevents an unexpected full-scale output the instant B+ comes up
2	Set rotary switch to STANDBY, not ON	Filaments and 0A2 tubes warm/strike; B+ stays off
3	Wait ~30–60 s for tube warm-up (longer if the unit has been idle for months)	Cold cathodes + cold gas tubes give erratic early regulation
4	Set rotary switch to ON	Enables B+ through the series-pass 6L6 pair
5	Set METER SWITCH to B+, bring B+ OUTPUT up slowly while watching the voltmeter	Confirms regulation is tracking the pot, not jumping/oscillating
6	If using bias, set METER SWITCH to C−, bring C− OUTPUT to the desired negative value	Bias is current-limited internally — safe to approach directly, but still lethal voltage
7	Connect the load (breadboard/tube circuit) with the supply in STANDBY, never live	One-hand rule — see §6
8	Bring supply to ON, monitor the D.C. OUTPUT CURRENT meter continuously while the circuit runs	No B+ current-limit exists — you are the current limiter
9	To shut down: B+/C− pots back to minimum → STANDBY → OFF, in that order	Avoids transients into the load; lets filaments cool before full power-down
10	After OFF, wait, then verify both B+ and bias rails have discharged before touching anything inside the case	Bleeder is only ~3 mA — caps stay charged for a while; see §6

⚠ Danger — Never adjust wiring, clip leads, or reach toward the binding posts or chassis with the supply in ON. STANDBY removes B+ but the filaments and pilot light stay lit — do not mistake STANDBY for “safe.” Full safety requires OFF and a verified discharge (§6).

---

6.5 Refurb Quick Order

Full detail in Vol 5. This is the sequence, not the procedure — do not skip Vol 5 the first time through a given unit.

Table 5 — 5. Refurb Quick Order

Order	Task	Detail
1	Discharge every electrolytic manually through a resistor before touching anything inside the case, even if the unit has been unplugged for days	§6 below, Vol 5 §1
2	Visual inspection — bulging/leaking cans, cracked wax/paper caps, discolored resistors, corroded binding posts, oxidized STANDBY switch contacts	Vol 5 §1
3	Recap the electrolytics — main filter (70 µF/350 V → ≥100 µF/450–500 V typical replacement), plus the 40 µF and 20 µF cans, confirmed against the manual parts list before ordering	Vol 5 §3
4	Check the ~0.047 µF coupling cap between the 6BH6 control grid and the 6L6 cathodes — a leaky example is the classic IP-32 fault and breaks both the zero-adjust and 400 V-adjust behavior	Vol 5 §4
5	Check/replace the two selenium screen rectifiers — known wear item; restorers commonly substitute silicon diodes plus a series resistor to make up the lower forward drop	Vol 5 §3
6	Verify pass-tube type — confirm 6L6GC (30 W dissipation) is fitted, not a metal 6L6/6L6GA/6L6GB (~19 W), which can be over-dissipated at high-current/low-output-voltage settings	Vol 5 §2, Vol 3 §2
7	Check 0A2 reference tubes for drift/wear — they set the regulation reference point; replace if regulation or set-points read off	Vol 5 §2
8	Re-zero the panel meter mechanically, then verify the R24 (400 kΩ, B+) and R25 (150 kΩ, C−) 1 % multiplier resistors before trusting the analog reading — a reported unit read 380 V when true output was 400 V	Vol 5 §5
9	Re-zero and re-span the B+ regulation using the two internal trimmers in the R10–R14 divider chain against a known-good external DVM	Vol 5 §5
10	Inspect/clean the STANDBY switch contacts and AC lead routing if hum or intermittent B+ is present	Vol 5 §4
11	Bring up on a variac with a DVM and DMM ammeter in-line, not straight to full mains	Vol 5 §1, §6 below

⚠ Danger — The main power transformer in this design is somewhat under-rated for the family. Do not sustain over-current conditions during bring-up or fault-finding — transformers for this unit are hard to source, and cooking the primary turns a component-level fault into an unobtainium-part problem.

---

6.6 HV Safety DON’Ts

These are absolute — not “best practice,” not “usually fine.” This unit generates and stores lethal voltage internally even though its front-panel output tops out at a comparatively modest 400 V. See the hub-wide HV discipline document for the full bench-safety policy this rule set inherits from.

Table 6 — 6. HV Safety DON'Ts

DON’T	Why
Don’t assume the supply is safe because it’s unplugged or switched OFF	The bleeder draws only ~3 mA — filter caps hold a dangerous charge for a meaningful time after power-off. Manually discharge every electrolytic through a resistor before working inside.
Don’t work inside the case with only the internal bleeder to rely on	3 mA through 4 × 27 kΩ is a minimum-load bleed path for regulation stability, not a rated safety-discharge circuit. Discharge actively.
Don’t touch the chassis or binding posts with two hands, or with one hand while grounded, above ~50 V	One-hand rule: keep one hand in a pocket or behind your back on anything above ~50 V so a shock path can’t cross your chest.
Don’t probe live at the raw-B+ node (≈ 600 V) or anywhere on the doubler caps	This is well past the threshold for cardiac arrest through incidental skin contact — it is not “just a jolt.”
Don’t exceed 100 mA continuous / 125 mA intermittent on B+	There is no active current-limit — you are relying on the fuse and pass-tube dissipation margin, and pushing past rating risks the 6L6 pass tubes and, downstream, the under-rated power transformer.
Don’t load the C− (bias) output beyond 1 mA or short it deliberately	It is internally current-limited by design, but that limiting resistor/pot is a documented weak point on the sibling SP-2717A’s equivalent circuit — don’t rely on the limiter as a load-testing feature.
Don’t connect or disconnect a breadboard/load with the supply in ON	Always drop to STANDBY (or OFF + verified discharge) before touching leads.
Don’t trust the panel meter’s zero point without checking it	A 5 % low reading has been documented on this family (380 V shown for a true 400 V output) — re-zero mechanically and verify the multiplier resistors before trusting a critical measurement.
Don’t treat the selenium rectifiers or old electrolytics as inert curiosities	Aged selenium rectifiers can emit an acrid, mildly toxic odor if they fail; old electrolytics can vent or rupture. Handle failed originals as hazardous waste, not shop trash.
Don’t skip the variac/current-limited bring-up after any repair or long storage period	Bringing a 60+-year-old unit straight to full mains after a repair or years of storage is how a single bad cap becomes a shorted transformer.

⚠ Danger — If you are new to tube-era HV work, do not use the IP-32 as your first hands-on unit without a mentor or without having read Vol 4 (Using It) and Vol 5 (Calibration & Refurbishing) in full. This is unforgiving gear built to a 1962 safety standard, not a 2020s one.

---

6.7 The Generational Pair — IP-32 and SP-2717A

6.7.1 The framing the scaffold got wrong

Jeff keeps the IP-32 and the SP-2717A side by side on the bench, and it is tempting to reach for the obvious story: “the IP-32 is the small one, the SP-2717A is the big one, grab whichever current rating the circuit under test needs.” That story is false. Both supplies deliver the identical 0–400 V DC @ 100 mA continuous (125 mA intermittent) B+ rating, the identical 0 to −100 V @ 1 mA bias rating, the identical four 27 kΩ / 2 W bleeder, and the identical dual-meter front panel. There is no low-current/high-current split anywhere in this family’s published specifications.

The honest framing is generational, not tiered. Both units sit on one unbroken circuit lineage that runs from the 1957 PS-4 through several later models, ending at the 1982 SP-2717A. What changed across that 20+-year span was never the output rating — it was how the regulation reference and rectification were built, as manufacturing moved from an all-tube reference to a solid-state one while the series-pass element stayed tube throughout.

6.7.2 Family lineage timeline

Table 7 — 7.2 Family lineage timeline

Model	Years	Generation	Notes	detail
PS-4	1957–1962	all-tube reference	Electrically identical circuit to the IP-32; only styling differs	Vol 1 §1
IP-32	1962–1967	all-tube reference	This dive’s subject. Kit-built, “Classic II” styling	Vol 1 §1
IP-17	1968–1977	transitional	Low-profile cabinet; selenium screen rectifiers → silicon; added 6.3/12.6 V heater option; 3-wire grounded plug	Vol 1 §1
IP-2717	1977–1982	transitional (kit)	Kit form of the final-generation circuit	Vol 1 §1
SP-2717A	1982	solid-state reference	Factory-assembled Heath-Zenith relabel of IP-2717A; this pair’s sibling	—

⚠ Danger — Do not read “solid-state reference” as “solid-state regulator.” In every model in this family, including the 1982 SP-2717A, the actual series-pass regulation element that drops raw B+ down to the set output voltage is a pair of tube pass elements (6L6 in the IP-32, 6L6GC in the SP-2717A). The lethal-voltage handling characteristics of tube gear apply to both units equally — the SP-2717A is not “safer” for having solid-state rectifiers.

6.7.3 IP-32 vs SP-2717A — the real comparison table

Table 8 — 7.3 IP-32 vs SP-2717A — the real comparison table

Parameter	IP-32 (1962–67)	SP-2717A (1982)	Identical?	detail
B+ output	0–400 V DC @ 100 mA cont. (125 mA intermittent)	0–400 V DC @ 100 mA cont. (125 mA intermittent)	Yes	Vol 1 §2
Bias (C−) output	0 to −100 V DC @ 1 mA	0 to −100 V DC @ 1 mA	Yes	Vol 1 §2
Bleeder	four 27 kΩ / 2 W in series	four 27 kΩ / 2 W in series	Yes	Vol 3 §4
Metering	Dual meter (shared voltmeter B+/C−, dedicated B+ current meter)	Dual meter (same arrangement)	Yes	Vol 3 §5
B+ current limiting	None (fuse + pass-tube dissipation only)	None (fuse + pass-tube dissipation only)	Yes	Vol 2 §4
Series-pass tubes	2 × 6L6 (metal-envelope era; 6L6GC recommended for full dissipation)	2 × 6L6GC	Same family	Vol 2 §2
Control / error amp	6BH6	6AU6	No	Vol 2 §2
Voltage reference	2 × 0A2 gas VR tubes (150 V each)	10-zener diode stack (ZD1–ZD10)	No	Vol 2 §2, Vol 3 §3
B+ rectifier	Silicon diode voltage doubler	Silicon diode voltage doubler	Yes	Vol 3 §3
Screen rectifier	2 selenium rectifiers (wear item)	All silicon	No	Vol 3 §3, Vol 5 §3
Bias rectifier	6X4 tube (full-wave)	All silicon diodes	No	Vol 3 §3
Filament output	6.3 V AC @ 4 A	6.3 V AC @ 4 A or 12.6 V AC @ 2 A (≤25 VA total)	No — SP adds a 12.6 V tap	Vol 1 §2
Cabinet / styling	13″ × 8½″ × 7″, “Classic II” dark-gray panel	Low-profile Heath-Zenith styling	No	Vol 1 §1
Assembly	Kit (customer-built)	Factory-assembled	No	Vol 1 §1
Era	1962–1967	1982, Heath-Zenith	No	Vol 1 §1

6.7.4 What that table means on the bench

• Choosing between them is not a current-headroom decision. If a circuit under test needs more than 100 mA @ 400 V from B+, neither supply in this pair delivers it — reach for a different instrument entirely, or parallel loads across both with care and full understanding of what that does to each unit’s regulation loop (not covered in this dive; treat as an advanced/unverified technique).
• Choosing between them is a reference-technology and vintage-fidelity decision. The IP-32 is the correct choice when the point of the exercise is authentically reproducing a 1962-era all-tube regulator circuit, when you want the gas-tube glow of the 0A2 pair as a visual/diagnostic reference, or when working on kit-restoration technique itself. The SP-2717A is the correct choice when you want the more temperature-stable, drift-resistant zener reference and don’t need the selenium rectifiers’ era charm — or simply when the IP-32 is out of service for refurb and the SP-2717A can cover the bench.
• The 12.6 V filament tap is the SP-2717A’s one genuinely new capability. It is a real, if modest, difference: circuits built around 12.6 V heater strings (common in many post-war tube designs) can be powered directly from the SP-2717A without a separate filament transformer, where the IP-32 offers only 6.3 V.
• Fault signatures differ by generation, not by symptom severity. A 0A2 reference tube going gassy or intermittent on the IP-32 and a zener in the SP-2717A’s ZD1–ZD10 stack failing shorted from running hot against the board produce similar symptoms (regulation set-point drift or loss) through different mechanisms — see the fault table below.

---

6.8 Fault Quick-Reference (IP-32)

Table 9 — 8. Fault Quick-Reference (IP-32)

Symptom	Likely cause	Check	detail
No B+ output at all, filaments/pilot lit	Rear fuse blown; STANDBY switch contacts oxidized	Fuse continuity; switch contact resistance	Vol 5 §4
B+ present but won’t regulate / drifts with load	0A2 reference tube worn or gassy	Swap known-good 0A2; check −150 V / −300 V rails	Vol 5 §2
Zero-adjust or 400 V-adjust won’t hold / interacts oddly	Leaky ~0.047 µF coupling cap (6BH6 grid → 6L6 cathodes)	Replace the cap; classic IP-32 fault	Vol 5 §4
B+ reads low/high vs a known-good external DVM	Panel meter zero drift or worn R24/R25 multiplier resistors	Mechanical re-zero; verify R24 (400 kΩ) / R25 (150 kΩ), both should be 1 %	Vol 5 §5
Screen supply low / soft regulation at high B+	Selenium screen rectifiers degraded (wear item)	Measure screen rail (~230 V target); consider silicon + series-R substitution	Vol 5 §3
Hum on B+ or bias output	Filter electrolytics aged/dried; AC lead dress	Recap; check lead routing away from signal/output wiring	Vol 5 §3, §4
Bias output won’t reach full −100 V	6X4 rectifier weak, or R21 bias current-limit resistor drifted	Check 6X4 emission; verify R21 value	Vol 3 §3, §4
6L6 pass tube runs excessively hot	Metal 6L6/6L6GA/6L6GB fitted instead of 6L6GC at high-current/low-voltage settings	Confirm tube type; upgrade to 6L6GC	Vol 5 §2
Transformer runs hot / smells	Sustained over-current condition	Check load current against 100/125 mA rating immediately; the power transformer is under-rated — don’t push it	§6 above, Vol 5 §4

---

6.9 SP-2717A Fault Quick-Reference (for the paired unit)

Because this pair lives on the bench together, a fault-finding session often needs to distinguish “which unit is actually misbehaving” fast. The SP-2717A’s zener-stack/silicon-rectifier generation fails differently from the IP-32’s tube-reference generation even though the symptoms can look alike from the front panel. This table is provided for cross-reference only — the SP-2717A’s own 6-volume dive is the authoritative source; see its Vol 5 for full detail.

Table 10 — 9. SP-2717A Fault Quick-Reference (for the paired unit)

Symptom	IP-32 (all-tube reference) likely cause	SP-2717A (solid-state reference) likely cause
Regulation set-point has drifted off spec	0A2 reference tube aged/gassy	A zener in the ZD1–ZD10 stack has drifted (less common than outright short)
Regulation lost entirely, output pinned or wild	0A2 tube gone soft/dark, or pulled (pin-jumper fail-safe opens the rail)	Control-amp tube (6AU6) shorted — a documented failure mode on this generation
Screen supply weak at high B+	Selenium rectifiers degraded	Silicon screen rectifiers (D1/D2) — check for opens, less prone to gradual degradation than selenium
Reference rail (−150 V / −300 V) shifted after a repair	0A2 swapped for a non-matched tube	A single zener failed shorted in the stack — restorers report zeners run hot against the PCB; standing them off the board helps
Zero-adjust / 400-adjust interaction won’t settle	Leaky ~0.047 µF 6BH6-to-6L6-cathode coupling cap	Equivalent coupling network in the 6AU6-to-6L6GC path — verify against the SP-2717A manual before assuming parts-identical fault
Bias pot feels “scratchy” or burned smell near C− control	R-series bias current-limit resistor degraded	R20 (22 kΩ) current-limit resistor or the R19 (50 kΩ) C− VOLTS pot — a documented weak point if the bias output is loaded or shorted

⚠ Danger — Do not swap tubes or reference devices between the two units to “test” a fault. The 0A2 and the zener stack are not interchangeable reference technologies, and cross-contaminating parts bins between a 1960s kit-built unit and a 1982 factory-built unit is a good way to lose track of which unit has which revision-level parts.

---

6.10 Recap Quick-BOM (IP-32 electrolytics)

Condensed from Vol 5 — confirm every value and voltage rating against the manual parts list before ordering; treat the “replacement” column as commonly reported practice, not a factory spec.

Table 11 — 10. Recap Quick-BOM (IP-32 electrolytics)

Original can	Original rating	Commonly reported replacement	Location / role
Main filter	70 µF @ 350 V	≥100 µF @ 450–500 V	B+ doubler reservoir — sees the ~600 V raw rail, use the higher voltage rating
Second can	40 µF	≥47 µF, ≥450 V on the B+ side	Filter/reservoir
Third can	20 µF	≥22 µF, ≥450 V on the B+ side	Filter/reservoir

Table 12 — 10. Recap Quick-BOM (IP-32 electrolytics)

Other known recap-adjacent items	Note
~0.047 µF coupling cap (6BH6 grid → 6L6 cathodes)	Not an electrolytic, but the single highest-value fault-clearing replacement on this unit — see §5 row 4
Selenium screen rectifiers (×2)	Not electrolytics; wear item — see §5 row 5 for the silicon + series-R substitution practice

⚠ Danger — Every cap in this table sits on or near the B+ doubler chain. Discharge before you touch, every time, with no exceptions for “I just tested it an hour ago.”

---

6.11 Bench Tools Checklist

What you actually want laid out before powering up either unit in this pair for service work.

Table 13 — 11. Bench Tools Checklist

Tool	Purpose
Variac (isolated, metered)	Current-limited bring-up after any repair or extended storage
DVM rated for the working voltage (≥600 V DC input)	Cross-check the panel meter; the panel meter is not trustworthy until re-zeroed and verified
HV-rated discharge probe / resistor lead	Manual cap discharge before opening the case — never rely on the bleeder alone
In-line DMM or clamp meter for current	Monitoring B+ current continuously during load testing, since there is no active current-limit to protect you from a mistake
Insulated tools, one-hand discipline	Physical habit, not optional gear — see §6
Known-good spare 6L6GC, 0A2 pair, 6BH6, 6X4 (IP-32)	Fastest fault-isolation path — substitution testing beats measurement-only diagnosis on tube gear
Isolation transformer (if bench is not already isolated)	Breaks the direct mains-ground bond while probing live circuits — standard tube-era bench practice

---

6.12 Cross-Reference Index

Table 14 — 12. Cross-Reference Index

Topic	Volume
What the IP-32 is, era, kit price, dimensions	Vol 1 — Overview
Full regulation-loop circuit theory (6L6/6BH6/0A2 interaction)	Vol 2 — How It Regulates
Chassis layout, all rectifiers/filters, metering detail	Vol 3 — Inside This Unit
Step-by-step operating procedure, HV safety in depth	Vol 4 — Using It
Recap BOM, calibration procedure, common failures	Vol 5 — Calibration & Refurbishing
This laminate + the SP-2717A comparison	Vol 6 — Cheatsheet & the Generational Pair (this volume)
Hub-wide HV/mains bench-safety discipline	_shared/legal_ethics.md
The SP-2717A’s own 6-volume dive	Heathkit SP-2717A HVPS

---

Sources

• Heathkit IP-32 assembly & operation manual scans:
  • archive.org — https://archive.org/details/Heathkit_IP-32_HV_Power_Supply
  • vintage-radio.info — https://www.vintage-radio.info/download.php?id=212
  • elektrotanya — https://elektrotanya.com/heathkit_ip-32_regulated_power_supply.pdf/download.html
  • Berkeley (image-only) — https://experimentationlab.berkeley.edu/sites/default/files/NMR/Equipment/Heathkit%20IP-32_Regulated%20Power%20Supply_Manual.pdf
• Bob Eckweiler (AF6C), “Heathkit of the Month #67 — IP-32 Bench HV Power Supply,” OCARC RF newsletter, Aug 2015 — https://www.w6ze.org/Heathkit/Heathkit_067_IP32.pdf
• radiomuseum.org, IP-32 — https://www.radiomuseum.org/r/heath_regulated_power_supply_ip_5.html
• radiomuseum.org, IP-2717A (sibling) — https://www.radiomuseum.org/r/heath_regulated_hv_power_supply_ip_2717a.html
• Heath Company, “Model SP-2717A Regulated High Voltage Power Supply” manual 595-2705-02, © 1982 — https://archive.org/details/Heathkit_SP-2717A_Regulated_HV_Power_Supply
• “Heathkit Zenith SP-2717A Power Supply Restore,” Lazy Electrons, 2018-06-02 — https://lazyelectrons.wordpress.com/2018/06/02/heathkit-zenith-sp-2717a-power-supply-restore/
• IP-32 restoration write-up — https://pudar.net/heathkit-ip-32-restoration/
• Antique Radio Forums, IP-32 repair thread — https://antiqueradios.com/forums/viewtopic.php?t=440
• Chuck Penson, Heathkit Test Equipment Products (styling/production-date reference, cited via Eckweiler)