2001–2004
LB7
Early common rail; unique injector/return diagnosis.

2001–current · generation overview
Choose the RPO first: LB7 through L5P do not share one test sheet.
Common reference specifications: These are common reference specifications. Always look up the current OEM procedure and specifications for your exact vehicle, VIN/engine serial number and installed hardware before performing repairs. Model year, engine revisions and replacement parts can change the requirements. For aftermarket fasteners, follow the current instructions for the exact kit and lubricant. Never loosen a common-rail line while cranking or running, reach for a suspected pressure leak, bypass safety controls, or work under an unsupported vehicle.
Pick the hardware first
Duramax spans major changes in injectors, supply, pumps, turbo controls and aftertreatment. Identify the RPO and build date before choosing a specification or component.
2001–2004
Early common rail; unique injector/return diagnosis.
2004.5–2005
Added VGT/EGR and revised airflow.
2006–2007
Revised injectors and controls.
2007.5–2010
DPF-equipped generation.
2011–2016
CP4 and SCR/DEF branches.
2017+
Denso injection—do not reuse Bosch procedures.
Numerical coverage is limited to the applications shown in the selector. The generation overview is not a claim of complete model-year coverage. Mechanical procedures have a separate search below.
Measured pass / fail data
Read the application and test condition on each value; mixed-year chapters still require selecting the correct labeled step. These are field-diagnostic references, not permission to reuse a specification on a different pump, injector, calibration or model year.
Crank / no-start
Target
About 2,500 psi (17.2 MPa)
Test condition
Watch actual rail pressure through cranking
Applies to
LB7, LLY, LBZ and LMM reference
Fuel supply
Target
≤5 inHg WOT; ≤7–8 inHg under load
Test condition
Vacuum gauge at specified fuel-system test port
Applies to
LB7 through LML suction-side systems
Fuel supply
Target
<2 inHg with symptoms
Test condition
Compare clear-line aeration and restriction result
Applies to
LB7 through LML suction-side systems
Sensor plausibility
Target
1.0–1.8 MPa (145–261 psi) LB7/LLY; 0 psi LBZ/LMM
Test condition
Key on, engine off
Applies to
LB7–LMM
Filter head
Target
Must not fall 10 psi to 2 psi in under 1 minute
Test condition
Hand-prime to 10 psi and inspect for external leakage
Applies to
LB7–LMM reference
LB7 injector return
Target
≤5 mL/injector and ≤20 mL/bank in 15 s
Test condition
FICM disabled; commanded pressure 114–135 MPa
Applies to
LB7
LBZ injector return
Target
≤144 mL total; ≤72 mL/bank; ≤18 mL/injector
Test condition
181–189°F; command 17,400 psi at idle; collect 30 s
Applies to
LBZ reference
Balance rate
Target
Within ±6 mm³/stroke
Balance rate is a screening value, not proof of a failed injector.
Test condition
Fully warm, in Drive after loaded road test
Applies to
LB7–LMM field screening
LML injector return
Target
≤3 mL in 15 s
Test condition
Cranking or idle with CH-50378 regulated adapter
Applies to
LML/LGH
LML pressure control
Target
<10 mL in 15 s
Test condition
Cranking with return hose routed and plugged per procedure
Applies to
LML/LGH
LML EGR
Target
≤3 percentage points
Test condition
Command EGR to 20% and compare desired/actual
Applies to
LML/LGH
Unit check: 1 MPa ≈ 145 psi; 100 kPa ≈ 14.5 psi. Keep the scan tool’s displayed unit visible when recording data.
Service-data rule: confirm assembly torque and any test that opens or blocks a high-pressure circuit against current VIN/ESN-specific OEM information.
Manual contents
Each chapter is a test path. Open it, perform the checks in order, and let the result—not the most expensive part—decide the next move.
Test
Record cold/hot, ambient temperature, fuel level, recent repairs and exactly when the fault occurs. Duplicate it without clearing anything.
If it passes
The failure is repeatable and its operating conditions are documented.
If it fails
Intermittent complaint: inspect history, connectors and freeze-frame before a road test with recording enabled.
Test
Run a complete module scan. Save active, pending and history codes, freeze-frame, monitor status and a short cranking or drive log.
If it passes
Codes and data agree with the complaint.
If it fails
Conflicting or communication codes: stabilize system voltage and diagnose network/power faults first.
Test
Sample fuel in a clean clear container. Check oil level and odor, coolant, air filter, intake plumbing and visible harness damage.
If it passes
Fuel is clean, oil is not diluted, coolant is stable and the intake is open.
If it fails
Stop for wrong fuel, water, DEF, metal, fuel-diluted oil or coolant loss. Correct the root condition before repeated cranking.
Test
Confirm model year, engine code and whether this truck uses Bosch common rail through LML or Denso common rail on L5P/L5D.
If it passes
The correct wiring diagram, scan PIDs and service procedure are selected.
If it fails
Do not borrow thresholds or torque values from another generation—the fender badge is not a service manual.
Test
Load-test the battery bank, voltage-drop positive and ground paths, then compare measured speed with RPM, cam/crank sync and voltage.
If it passes
Use a charged battery bank and verify a stable RPM PID plus cam/crank sync. Return-test time and pressure change by RPO—select LB7, LLY/LBZ/LMM, LML or L5P first.
If it fails
Repair the weak battery, cable, ground or starter before interpreting pressure data.
Test
Watch RPM, security status and cam/crank correlation and injection authorization while cranking.
If it passes
RPM is believable, synchronization is achieved and injection is authorized.
If it fails
Follow the sensor, tone-wheel, wiring, security or module-power path indicated by the missing input.
Test
Measure filter-head vacuum/air on early systems or electric supply pressure on equipped trucks. Test during cranking, not only key-on. Check restriction, aeration and filter sealing.
If it passes
LB7–LML suction reference: ≤5 inHg at WOT or 7–8 inHg under load; under 2 inHg with symptoms can indicate air entry. Do not apply this to electric-supply L5P/L5D.
If it fails
Work tank-to-pump in sections. A high-pressure pump cannot make up for an empty, restricted or aerated inlet.
Test
Graph desired versus actual rail pressure through the complete crank event.
If it passes
LB7–LMM: about 2,500 psi is required to start. LML diagnosis must compare actual to desired and maintain injector-return pressure above 3 bar/45 psi.
If it fails
Separate inlet supply, pressure control, excessive return/leakage and pump output. One low number does not condemn a pump.
Test
Use balance/contribution screening plus approved return and compression confirmation. If injection conditions are met, verify air shutoff, compression, valve motion and mechanical timing.
If it passes
LB7: ≤5 mL/injector/15 s and ≤20 mL/bank at 114–135 MPa. LBZ enhanced test: ≤144 mL total; replace an injector over 18 mL when its bank exceeds 72 mL. LML: ≤3 mL/injector/15 s.
If it fails
Isolate the failed cylinder/system and prove wiring, injector or mechanical cause before replacement.
Test
Attach the fuel-system vacuum gauge at the specified low-side port. Record the gauge at idle, WOT and under load along with rail pressure. Secure test equipment away from belts and exhaust.
If it passes
The field limits are 5 inHg at WOT and 7–8 inHg under load. Interpret with the actual test condition; move to the air-entry test if restriction is not excessive.
If it fails
Excessive vacuum: isolate filter, collapsed hose, pickup and tank vent. A low reading under 2 inHg with symptoms is only an air-entry clue, not proof.
Test
With the engine stopped, use the hand primer and gauge to establish 10 psi. Time the decay and inspect the low-side fittings for wetness.
If it passes
Pressure does not fall from 10 psi to 2 psi in less than one minute: continue to running aeration and rail checks.
If it fails
A drop to 2 psi in under one minute calls for sectional leak isolation. The result alone does not distinguish a filter-head leak from another path.
Test
Install fuel-compatible clear test sections before and after the filter housing using the low-pressure test procedure. Prime completely, run the engine and observe both sections at the complaint condition.
If it passes
No persistent bubbles after purging: continue to rail-pressure and injector-return testing.
If it fails
Bubbles only downstream point toward the housing, seals, heater or drain. Bubbles already present upstream point toward the tank/line path. Move the test boundary until the entry point is isolated.
Test
After restoring a sealed, primed inlet, graph actual and desired rail pressure during crank. For LB7–LMM the starting reference is approximately 2,500 psi (17.2 MPa).
If it passes
Actual reaches the starting reference and follows desired: investigate injection authorization and cylinder condition if it still will not run.
If it fails
Still below the reference: continue to the RPO-specific return/relief test. Never use the LB7 collection limit on LML or L5P.
Test
Use the LB7 return-test arrangement with FICM disabled and 114–135 MPa test pressure. Collect for 15 seconds. That pressure is approximately 16,500–19,600 psi and is a special test condition, not the normal 2,500-psi starting reference.
If it passes
At most 5 mL per injector and 20 mL per bank in 15 seconds: leakage meets this test's benchmark.
If it fails
Above either limit: verify test pressure and repeat the collection before isolating the excessive branch. If required pressure is not achieved, a small collected volume cannot establish a pass.
Test
Bring coolant to 181–189°F (83–87°C), command 17,400 psi at idle, and collect for 30 seconds with the correct return-test apparatus.
If it passes
Total at or below 144 mL and each bank at or below 72 mL meet the stated limits.
If it fails
If a bank exceeds 72 mL, measure its injectors individually; more than 18 mL per injector under the same test is excessive. Do not compare a crank sample against these running limits.
Test
Use CH-50378 or the specified equivalent regulated adapter. Preserve at least 3 bar (approximately 44–45 psi) injector-return pressure. Collect for 15 seconds during the applicable crank or idle test.
If it passes
Each injector returns 3 mL or less per 15 seconds with correct backpressure: proceed to the remaining pressure-control branch.
If it fails
Over 3 mL: verify regulation and repeat before condemning the injector. Open-bottle testing without regulation is invalid for this system.
Test
On LB7–LMM, warm the engine and record balance in Drive only with the truck secured, service brake applied and a trained operator in the seat. Use ±6 mm³/stroke as the in-Drive field screening band.
If it passes
Within the band does not prove nozzle sealing. Pair with return, contribution and compression if smoke, dilution or knock remains.
If it fails
Outside the band: identify the weak/over-contributing cylinder, then distinguish fuel delivery from compression and valve-train faults. Do not order an injector from one balance number.
Test
Measure injector-return pressure with the specified adapter at the complaint condition. Keep the factory regulator or the approved test regulator in circuit.
If it passes
Regulated reference: 4–11 bar (58–160 psi); injector operation needs at least 3 bar (about 44 psi). Proceed with regulated collection.
If it fails
Below 3 bar: correct regulation/leakage before interpreting injector return. Between 3 and 4 bar, or above 11 bar, investigate regulation/restriction using the exact-year test.
Test
Collect each injector for 15 seconds at crank or idle with regulated backpressure; record condition and rail pressure.
If it passes
3 mL or less per injector: pursue remaining control or mechanical causes.
If it fails
Over 3 mL: repeat the test with verified backpressure and isolate the excessive injector.
Test
After injector replacement, perform the specified correction reset and any required coding/learn routines.
If it passes
Reset completes and the original complaint is absent: verify at operating temperature.
If it fails
P026D or roughness persists: verify setup, correction data, return test and mechanical contribution.
Test
Graph RPM, commanded/actual pressure, supply status, key voltage, run permission and air-shutoff position through the stall.
If it passes
One signal changes first and points to a system.
If it fails
If every PID drops together, inspect ignition feed, module power, grounds and network communication.
Test
Repeat the event while measuring filter-head vacuum/air on early systems or electric supply pressure on equipped trucks.
If it passes
Supply stays stable after the engine catches.
If it fails
Check a blocked tank vent or pickup, collapsing hose, filter seal, weak lift pump or air entry.
Test
Compare desired versus actual rail pressure and regulator commands, rail relief leakage and pressure-sensor plausibility as it dies.
If it passes
LB7–LMM: rail relief should not leak at 160 MPa/23,200 psi. LML regulator-2 return must be under 10 mL in 15 s during its specified crank test.
If it fails
If pressure falls first, isolate leakage, regulator/control and supply. If command disappears first, follow the control reason.
Test
Check security, crash/fuel cutoff, PTO/idle shutdown, temperature protection and throttle/air-shutoff commands.
If it passes
No module is requesting shutdown.
If it fails
Resolve the requesting system instead of bypassing a safety strategy.
Test
Log the first cold start and a fully heat-soaked restart using the same PID list.
If it passes
The same system fails in both tests.
If it fails
Temperature-only failure: focus on injector return, regulator/relief leakage, filter-head air entry and cranking speed, starting aids, viscosity and temperature-dependent leakage.
Test
Before starting, compare coolant, oil, intake-air and ambient readings.
If it passes
Sensors agree closely with actual ambient temperature.
If it fails
Test the biased sensor, reference voltage, ground and connector before replacing fuel parts.
Test
Overlay cranking speed with desired versus actual rail pressure; note how quickly actual responds.
If it passes
LB7–LMM: about 2,500 psi is required to start. LML diagnosis must compare actual to desired and maintain injector-return pressure above 3 bar/45 psi.
If it fails
A slow rise points toward drain-back, air entry, inlet restriction, leakage or worn pressure-producing parts.
Test
Command glow plugs, intake heater or grid heater where equipped; voltage-drop the feed and confirm current draw by bank or circuit.
If it passes
Command, voltage and current are present and balanced.
If it fails
Diagnose the relay/module, feed, ground or elements. Do not judge the system by dash-lamp time.
Test
Note cold/hot, idle/in gear, speed, load and whether the noise follows fuel quantity or mechanical speed.
If it passes
The complaint can be reproduced in a controlled window.
If it fails
Use temperature survey and recorded scan data to catch the intermittent event.
Test
Run contribution, balance or cutout testing, then confirm with balance/contribution screening plus approved return and compression confirmation.
If it passes
The same cylinder is identified by more than one test.
If it fails
A system-wide pattern points toward fuel quality, pressure control, airflow, timing or voltage—not one injector.
Test
Pair injector evidence with relative compression, crankcase pressure, valve-train inspection and cylinder temperature.
If it passes
Injector evidence is consistent and mechanical checks pass.
If it fails
Low compression or abnormal valve motion requires mechanical diagnosis before fuel-system spending.
Test
Graph desired versus actual rail pressure, charging voltage and airflow while the miss occurs.
If it passes
Pressure, voltage and airflow stay stable.
If it fails
Multiple bad cylinders usually mean a shared supply, control, ground or air-path problem.
Test
Record when it occurs, odor, oil/coolant consumption and whether it changes with cutout, boost or temperature.
If it passes
Fuel, oil and coolant possibilities are separated.
If it fails
Do not assume white means coolant or blue means oil. Verify fluid loss and cylinder response.
Test
Inspect filter/ducts, pressure-test charge air, check pre-turbo exhaust leaks and validate MAF, MAP/boost, EBP where equipped and vane position.
If it passes
Measured airflow and boost match command without leaks or restriction.
If it fails
Repair air, EGR, turbo or sensor faults before judging injector quantity.
Test
Verify starting aids, temperature inputs, compression, injection timing/pressure and contribution.
If it passes
Smoke clears warm and no fluid is disappearing.
If it fails
Eye-burning smoke suggests unburned fuel; sweet vapor plus coolant loss requires cooling-system isolation.
Test
Inspect turbo inlet/outlet and drain, crankcase ventilation, oil level and cylinder sealing.
If it passes
No abnormal oil enters intake/exhaust and crankcase pressure is controlled.
If it fails
Correct turbo, drain, ventilation, guide/seal or ring fault before aftertreatment damage follows.
Test
Save codes and inspect torque-limit reason, aftertreatment status, fuel-pressure protection and transmission requests.
If it passes
No controller is intentionally limiting torque.
If it fails
Repair the initiating condition and complete the required service routine. Do not tune around protection.
Test
Log filter-head vacuum/air on early systems or electric supply pressure on equipped trucks, desired versus actual rail pressure and fuel command during a controlled loaded pull.
If it passes
Supply and actual pressure remain on target as demand rises.
If it fails
A value that collapses under load identifies restriction, aeration, control, leakage or capacity direction.
Test
Test compressor outlet to intake, inspect exhaust before the turbine, and compare MAF, MAP/boost, EBP where equipped and vane position.
If it passes
System holds pressure and command/position agree.
If it fails
Repair leaks, restriction, actuator/geometry or biased sensors before replacing a turbo.
Test
Review differential pressure, soot load, regeneration history and DPF soot/restriction, EGT, DEF/SCR and NOx data.
If it passes
Restriction is reasonable for operating condition.
If it fails
Find the engine-side reason for repeat loading before cleaning or replacing the filter.
Test
Draw fuel from the housing/tank low point into a clean clear container. Inspect for water, gasoline, DEF, growth and metal.
If it passes
Fuel is clean and correct.
If it fails
Stop and document contamination. Follow the OEM extent-of-repair procedure without circulating more debris.
Test
Install the correct gauge/restriction tool and verify filter-head vacuum/air on early systems or electric supply pressure on equipped trucks at idle and under demand.
If it passes
Pressure, volume and restriction stay in range without bubbles.
If it fails
Isolate pickup, lines, heater/housing, filter and pump. Fix the inlet before high pressure.
Test
Command or graph regulator commands, rail relief leakage and pressure-sensor plausibility against desired versus actual rail pressure. Inspect the harness before removing a regulator.
If it passes
Actual responds smoothly and follows command.
If it fails
Perform the model-year regulator, relief and leakage tests; inspect approved locations for debris.
Test
Follow the exact service procedure for balance/contribution screening plus approved return and compression confirmation. Use approved caps, adapters and return tools only.
If it passes
LB7: ≤5 mL/injector/15 s and ≤20 mL/bank at 114–135 MPa. LBZ enhanced test: ≤144 mL total; replace an injector over 18 mL when its bank exceeds 72 mL. LML: ≤3 mL/injector/15 s.
If it fails
Confirm the suspect with a second test and correct contamination or compression causes before parts.
Test
Load-test each battery separately; voltage-drop main positive and ground paths during crank and electrical load.
If it passes
Voltage remains stable and drops are low on every path.
If it fails
Repair the exact high-drop connection. Replace batteries as a matched set when required.
Test
Use the diagram. Back-probe powers and ground voltage-drop with the circuit loaded—an ohmmeter alone is not enough.
If it passes
Every feed and ground carries load without excessive drop.
If it fails
Repair fuse, relay, feed or ground before condemning a controller.
Test
Compare to a related sensor KOEO, then verify 5 V reference, low reference and signal while graphing a wiggle test.
If it passes
Signal changes smoothly and agrees with a mechanical reference.
If it fails
Isolate sensor from harness; unplug shared-reference sensors one at a time if the reference is pulled down.
Test
Verify DLC power/ground, measure network bias key-on, resistance power-off, then isolate branches with the diagram.
If it passes
Network values and module list are normal.
If it fails
Locate the shorted/open branch or module by sectional isolation—do not shotgun computers.
Test
With exhaust cold, compare temperature sensors, pressure sensors and ambient inputs before startup.
If it passes
Related sensors agree near ambient and zero differential as designed.
If it fails
Circuit-test the outlier; swap-test only where service information allows.
Test
Command devices while watching position feedback, MAF/MAP response and MAF, MAP/boost, EBP where equipped and vane position.
If it passes
Feedback follows command and airflow changes logically.
If it fails
Inspect sticking, deposits, leaks, actuator calibration and wiring before replacing assemblies.
Test
Review DPF soot/restriction, EGT, DEF/SCR and NOx data, regeneration history, idle time and engine-side codes.
If it passes
Loading matches use and regeneration completes normally.
If it fails
Fix injector, boost, EGR, temperature, thermostat or oil-consumption causes before servicing the filter.
Test
Use the proper scan tool for resets, learns, dosing tests and stationary regeneration. Confirm prerequisites first.
If it passes
Routine completes and monitored values respond as expected.
If it fails
Never reset ash or soot counters unless the specified physical service was completed.
Test
Group voltage/network first, then synchronization, fuel pressure, airflow and aftertreatment. Use freeze-frame to find the first event.
If it passes
One primary fault explains downstream codes.
If it fails
If unrelated modules set low-voltage codes together, return to battery, charging and grounds.
Test
Use exact-year service information for enable criteria, threshold and monitor behavior.
If it passes
The fault can be tested under its actual set conditions.
If it fails
A monitor that cannot run yet cannot be called fixed. Satisfy prerequisites first.
Test
Repeat the original failure window, rerun self-tests and compare before/after recordings.
If it passes
Complaint is gone, monitors run and data stays in range.
If it fails
A cleared light is not a repair. Continue the branch that still fails.
Mechanical reference library
Application disclaimer: These are common reference specifications. Always look up the current OEM procedure and specifications for your exact vehicle, VIN/engine serial number and installed hardware before performing repairs. Model year, engine revisions and replacement parts can change the requirements. For aftermarket fasteners, follow the current instructions for the exact kit and lubricant.
Choose the exact assembly below. Each entry identifies its hardware, preparation, torque stages and exclusions. This is a reference for the listed operation, not a complete engine rebuild manual.
2 matching procedures. Confirm the full application and installed hardware before using a value.
Hardware: Listed Fluidampr damper with new OEM bolts: 890101 / 830111; new torque-to-yield bolt
Verify the exact damper and bolt set. Clean pilot and mating surfaces; confirm correct bolt engagement and installation tooling.
Check pulley alignment, belt clearance and fastener completion before running.
Not for aftermarket high-strength bolt kits or another damper assembly. Never add a torque-to-yield angle a second time as a recheck.
Hardware: Listed Fluidampr damper with new OEM bolts: 830121 / 830141 / 830151; new torque-to-yield bolt
Verify the exact damper and bolt set. Clean pilot and mating surfaces; confirm correct bolt engagement and installation tooling.
Check pulley alignment, belt clearance and fastener completion before running.
Not for aftermarket high-strength bolt kits or another damper assembly. Never add a torque-to-yield angle a second time as a recheck.
Common is a place to test, not permission to replace.
Year, engine or RPO, cold or hot, all codes, freeze-frame, cranking log, fuel sample findings, recent work and which tests passed or failed.
Call (915) 247-2782