AstroSafariAtlas · Field Manual

§ 03 · Diagnostics / Engine / OBD-I

OBD-I Code 35 With Hot No-Start or Flooding Symptoms: 1994 Astro/Safari 4.3L

Community-reported hot no-start and flooding pattern on a 1994 Chevy Astro / GMC Safari 4.3L with stored OBD-I Code 35. Ranked hypotheses, methodical test order, and common misdiagnoses — no parts-cannon.

Community-reported symptom pattern1994 primary4.3L V6 (verify injection hardware)System · Engine management / fuel / ignition / idleSkill · IntermediateFuel / fire hazardEvidence · Community case + service-info verification required
Origin of this case

This entry originates from a Facebook owner-community post about a 1994 Astro/Safari 4.3L that starts and runs cold but will not restart hot, with a stored OBD-I Code 35. The owner reports the fuel pump and coolant temperature sensor have already been replaced without resolving the fault. The pattern is common enough to catalog, but nothing below is confirmed for this specific vehicle without the tests described.

Engine identification caveat

Before touching the intake, verify the exact fuel-injection system on this VIN. Not every 1994 4.3L Astro/Safari uses the same injection hardware. The under-intake CPI/CSFI "spider" and any internal fuel-line "nut kit" only apply when that hardware is actually installed. Confirm by underhood emissions label, VIN decoding, and a visual check before ordering parts.

What Code 35 means

On GM OBD-I, Code 35 generally relates to Idle Speed Control / Idle Air Control (IAC) system performance — the ECM sees actual idle behavior that does not match what it commanded. The exact definition, pinout, and diagnostic procedure vary by application and must be verified against GM service information for this vehicle configuration before drawing conclusions.

Why the stored code may be secondary

The primary complaint is a hot no-start with flooding-style symptoms. Code 35 can be a downstream effect of an IAC that cannot control an already-abnormal idle — for example, when raw fuel is entering the intake during heat soak, or when a vacuum/airflow baseline is off. Diagnose the running condition, not just the stored code.

Ranked diagnostic hypotheses (not ranked parts)
  • Internal fuel leakage under the upper intake
    Plausible — test-dependent

    CPI/CSFI assembly, pressure regulator, or internal supply/return lines where actually equipped. Consistent with rapid pressure decay and rich hot-restart, but requires visual and pressure-decay confirmation.

  • IAC valve or bypass passage fault
    Plausible — directly related to the stored code

    Sticking pintle, carbon-choked passage, electrically open/shorted circuit, or IAC unable to reach commanded position. This is the failure family most commonly associated with Code 35 itself.

  • Excessively rich hot-start condition (fuel-control side)
    Plausible — requires sensor and pressure data

    Bad sensor input (CTS wiring, MAP), fuel-pressure control fault, or leaking injector enriching the mixture during heat soak.

  • Heat-soak ignition failure
    Plausible only if spark is missing/unstable during failure

    Distributor pickup, ignition module, coil, cap/rotor, wiring, or loss of reference signal only when hot. Confirmed by spark/reference tests during the actual failure window.

  • Vacuum leak or base-airflow fault contributing to Code 35
    Contributor — not a standalone explanation for flooding

    Intake gasket, brake booster, PCV, or throttle-body contamination changing idle airflow enough to set Code 35, even if not the primary cause of the hot no-start.

Interactive decision flow

Reusable Atlas flow · Hot No-Start After Warm Shutdown, 1985–1995 Astro/Safari

Step 1 · Confirm the symptom

Record cold-start behavior, how long the engine ran, how long the hot soak was before the no-start, cranking speed, presence of raw fuel odor or black smoke at the tailpipe, and whether cracking the throttle open changes anything. Do not skip this — the answers steer every branch below.

Field checklist (printable)
  1. Symptom confirmed (cold OK, hot no-start, soak time, throttle-open behavior, spark/odor notes).
  2. All OBD-I codes retrieved and written down — not diagnosing from Code 35 alone.
  3. Injection hardware identified by underhood label and visual inspection.
  4. Fuel pressure measured key-on, cranking, running — against verified spec for this configuration.
  5. Shutdown pressure-decay recorded at 5, 10, and 20 minutes.
  6. Internal-intake leak inspection performed (only if CPI/CSFI-type hardware is actually present).
  7. IAC connector, pintle, and throttle-body bypass passage inspected/cleaned per procedure.
  8. Spark and reference signal verified during an actual hot no-start attempt.
  9. Vacuum, EGR, throttle-plate, and base-airflow baseline confirmed.
  10. Repair verified over multiple heat-soak restarts; codes cleared and rechecked.
Common misdiagnoses
  • Replacing the fuel pump without measuring pressure and shutdown decay first.
  • Replacing the coolant temperature sensor without checking sensor data and wiring.
  • Replacing the IAC solely because Code 35 is stored.
  • Replacing the distributor based on anecdotal similarity, with no spark or reference-signal test during the actual hot no-start.
  • Assuming a vacuum leak explains a rich / flooded hot-restart symptom.

Field notes

  • OBD-I data is limited — record observations manually as your own "freeze frame" (temperature, soak time, cranking behavior, pressure at that moment).
  • Do not condemn a component that only "seems" bad after the van sits — reproduce and test during the failure window.
  • Keep an extinguisher and ventilation ready any time you are testing near a suspected raw-fuel leak.

Related Atlas articles

Some of these are placeholders queued for the research pipeline — they will fill in as the knowledge graph grows.