Owner app — jobs, billing & your customer book
LESS GUESSWORK. MORE UPTIME.

Workbench

Find the fault. Follow the evidence. Get back to work.

🗣️ Ask Remy

AI HANDS
Tell me what you need — I'll take you there and fill it in. Try "Add a job: Mike Reynolds, DD15, health scan, tomorrow 10 AM at the Love's" or "Put coolant 195 and idle 650 on the Martin job." I fill and navigate — the final Save tap stays yours.
–jobs today
–scan income this week
–still owed

Kit payment coverage — $54/wk

…

🔍 Possibility Search — Check First

OUR LIBRARY FIRST
Type what the truck is doing, in your own words. Our guide answers FIRST — the possibilities in check order: free looks, then simple tests, parts talk last. If the guide doesn't have it, AI is the second line. Works offline for everything in the guide.

🤖 AI Diagnosis Brief

NEW
Paste fault codes from the tablet and get a plain-English brief — what each code means, likely causes, first checks, and whether the truck can keep driving. Falls back to the built-in code library when offline.

Owed to you

Today's jobs

Coming up

Rechecks due

DIESEL DIAGNOSTIC AI

Jobs

DIESEL DIAGNOSTIC AI

Add job

Met someone on the road? Log the job yourself — no booking page needed.
Tap a day, then pick a time.
DIESEL DIAGNOSTIC AI

Report inbox

Reports you email to reports@dieseldiagnosticai.com land here. Open one to read it, then file it onto its job — from the job, Send report texts it to the customer like always.
DIESEL DIAGNOSTIC AI

Customers

DIESEL DIAGNOSTIC AI

Calendar

Month

● Booked · Scanned · Report sent · Paid — tap a day to see its jobs.

Jobs

DIESEL DIAGNOSTIC AI

Ledger — completed scans

–Total charged
–Total collected
–Still owed
–Collected this month
Collections — last 6 weeks
DIESEL DIAGNOSTIC AI

Code Library — Detroit DD15 (starter set)

Search by SPN number (e.g. 3216), a word (e.g. NOx, DEF, boost), or a system. Works offline — the library lives in the app.

Full library — all makes (reference lists)

Public fault-code lists harvested from Diesel Repair (repair.diesellaptops.com) — Detroit first, then every other make we could reach. Search by SPN, SPN + FMI (e.g. “2659 0”), or a word in the title (e.g. EGR). Our own DD15 notes merge in first, marked ★. Reference only — the tablet's report is the final word on a truck.

FMI reference — the second number

Every fault code is SPN + FMI. The SPN says WHAT, the FMI says HOW it failed.

FMI Key — complete 0–31 (cross-reference)

Starter library — always confirm against the tablet's report and the truck's actual symptoms before quoting a repair.
DIESEL DIAGNOSTIC AI

Know Your Engine

What your engine calls for — oil, coolant, filters and intervals — from the manufacturers' own published tables (Detroit, Cummins, PACCAR and Caterpillar), so you walk into any oil-change shop already knowing your numbers. Works offline.

🔎 ESN Decoder — Detroit DD

FACTORY TABLES

Read the engine serial number off the pad on the left front of the block. Paste the full ESN, or just the first 6–7 characters. This decoder answers only from Detroit's published model tables — a prefix that isn't in them gets an honest “not published,” never a guess.

Your engine
Truck built
March 2015 is the split: Detroit deleted the coolant filter, and the DD fuel module went from three elements to two.
Every figure comes from published factory maintenance tables and fluids specifications — Detroit/DTNA, Cummins, PACCAR and Caterpillar (full citations in the research records, engine-specs-detroit.md, engine-specs-cummins-research.md, engine-specs-paccar-research.md, engine-specs-cat-research.md and engine-specs-series60-research.md). Where the manufacturer doesn't publish a number, this page says so instead of guessing. Confirm part numbers by engine serial number (ESN) before ordering.
DIESEL DIAGNOSTIC AI

Brakes — See What the Inspector Sees

Drums, shoes, chambers and slack adjustment — the good, the bad, and the published numbers that condemn them. Air brakes work the same on every truck, whatever engine is up front. Works offline.
Brakes — what the inspector sees. Universal chassis education — this set is not tied to any engine. Every limit below is the published one: 49 CFR §393.47 and the CVSA out-of-service criteria — the same lists a roadside inspector works from. The 20% rule: when 20% or more of a rig's service brakes are defective, it is parked — on a 5-axle tractor-trailer, 2 defective brakes do it. Some findings park it alone: steering-axle defects, a cracked spring-brake housing, a cracked drum, no parking brake at all. The same findings also fail an annual inspection (Part 396, App. G).

Panel B1 — One wheel end: how a stop happens

Chamber to drum — the application path
All trucks
ONE DRIVE-AXLE WHEEL END AIR DOES THE PUSHING → 1 CHAMBER BRACKET ON AXLE Pushrod + clevis 2 3 Slack adjuster Camshaft + tube 4 Shoes + linings 6 S-cam + rollers 5 7 Brake drum max dia. stamped on it 8 Anchor pins + return springs THE FOOT VALVE METERS THE AIR — HARDER PEDAL, HARDER PUSH
1Brake chamber — air fills the can and shoves a rubber diaphragm; that push is the muscle of the whole stop. The type number is the diaphragm area in square inches (Type 30 = 30 in²)
2Pushrod + clevis — the steel rod carries the push out of the chamber; the clevis straddles the slack arm, joined by a clevis pin and cotter pin
3Slack adjuster — the lever arm splined on the camshaft. Manual: an adjusting screw sets the clearance. Automatic: a sealed body self-adjusts as linings wear $40–$90 manual · $80–$200 automatic, each · Roundabout — market prices, Oct 2026. Not a quote.
4Camshaft + cam tube — the lever twists this shaft; it runs in its tube from the slack adjuster to inside the drum
5S-cam + rollers — the S-shaped cam head turns between the two shoe rollers and spreads the shoes apart
6Brake shoes + linings — two crescents on the spider; the lining is the friction face that presses the drum (Panel B3)
7Brake drum — the cast-iron bell the linings press against; friction does the stopping. Its maximum diameter is stamped on the drum itself (Panel B2)
8Anchor pins + return springs — the shoes pivot on the anchor pins; the springs pull them back off the drum when the air releases

The chain: pedal → chamber → pushrod → slack adjuster → camshaft → S-cam → shoes → drum. Matching rules (§393.47(b),(c)): same chamber size and same slack-adjuster length on both ends of an axle — and never mix standard-stroke and long-stroke chambers on one axle. A caged spring brake = no parking brake at that wheel — see Panel B4 before touching one.

Panel B2 — Drum: good vs bad

Judged on cracks, heat, surface and size
All trucks
✓ GOOD ✗ BAD FRICTION SURFACE MAX DIA — STAMPED 2 1 fine heat-check web · even color FRICTION SURFACE INSET: CRACK THROUGH THE WALL 4 3 5 BELL-MOUTH — MEASURE AT SEVERAL DEPTHS root dia. mouth dia. — flared wider 6 drum micrometer largest reading governs
1Heat checking — a web of short, fine hairlines from heat cycles. NOT a defect by itself — CVSA warns inspectors not to confuse heat checks with real cracksGOOD
2MAX DIA stamp — the drum's wear limit is cast or stamped into the drum. It condemns at a number, not a look: replace past original diameter + .120″ (a rebore may finish at no more than + .080″) $100–$300 each · Roundabout — market prices, Oct 2026. Not a quote.GOOD
3Crack to the edge, opening on application, or through the wall — a standalone out-of-service item (CVSA drum section). Through the wall = replace immediatelyBAD
4Heat spots — blue/straw hard patches in ONE zone, not spread evenly. That spot overheated again and again: grab, judder, uneven wear. Manufacturer-condemned — replaceBAD
5Deep scoring — grooves you can catch a fingernail in. Scoring never improves, and it eats new linings. Metal-to-metal contact on the surface is a defective conditionBAD
6Bell-mouth — the open mouth flares wider than the root; shoes wear at an angle. Measure diameter at several depths with a drum micrometer. No published numeric limit — judged by measurement, never by eyeballBAD

Out-of-round is checked with the same micrometer at several clock positions — no universal Class-8 number is published for it, so none is printed here. Shop practice: drums are serviced in axle pairs so both sides bite alike.

Panel B3 — Shoes & linings: good vs bad

Thickness, cracks, completeness, attachment, contamination
All trucks
✓ GOOD ✗ BAD MEASURE AT SHOE CENTER 1 full lining · rivets buried 4 2 3 5 FAILED WHEEL SEAL ≈1/16″ play 6 THE RED LINES — LINING LEFT AT THE SHOE CENTER STEER 1-PIECE — MIN 3/16″ STEER 2-BLOCK — MIN 1/4″ DRIVE/TRAILER — MIN 1/4″ RETURN SPRING / HARDWARE intact 7 broken / missing A whole segment missing is defective outright.
1Full-thickness lining — measured at the shoe center; that is where the regulation measures it. Rivet heads stay buried under friction materialGOOD
2Lining at the red line — STEER: below 3/16″ (one-piece continuous) or below 1/4″ (two-block) is illegal. DRIVE + TRAILER: below 1/4″ is defective. Or worn into the wear indicator, if the lining is marked $60–$110 per wheel end · core charge applies · Roundabout — market prices, Oct 2026. Not a quote.BAD
3Cracked lining — an edge crack or void over 1/16″ wide, or any edge crack over 1½″ long, is defectiveBAD
4Missing chunk, rivet showing — a missing piece that exposes a rivet or bolt head is defective — and judge the drum it has been chewing (Panel B2)BAD
5Oil / grease-soaked lining — a failed wheel seal is the usual source. The oil is IN the friction material: contaminated linings are replaced, not cleaned and keptBAD
6Loose segment — a lining segment that moves ≈1/16″ or more on the shoe table is defectiveBAD
7Broken / missing hardware — return spring, anchor pin, cam roller, the shoe itself: every one is a named defective part in the CVSA listBAD

Not the same finding: grease on the lining edge, the back of the shoe, or old stains with no sign of fresh leakage do not condemn a lining. It is fresh, wet contamination of the friction faces that does.

Panel B4 — Brake chamber: good vs bad

Same can, same angle — learn the pattern
All trucks
⚠ SPRING BRAKE — STORED ENERGY. The power spring inside a spring brake chamber is compressed with tons of force and stays loaded even with every air line off. NEVER open, cut, or take the clamp band off a spring brake chamber — it is a non-serviceable unit; replace the whole assembly. Caging is a recovery move, not a repair — it leaves that wheel with no parking brake; chock the wheels before and during. Never cage a chamber with structural damage — caging it may cause serious injury or death; replace the complete assembly. Work from the SIDE of a spring brake chamber — never in front of it or behind it. Don't drop one. Don't set spring brakes on overheated drums and walk away — a cooling drum can crack or warp. Caging turns and torque vary by maker — follow the chamber's own instructions (one maker's bulletin example: ~22–23 turns, max 74 ft-lb — an example, not a universal spec).
✓ GOOD ✗ BAD TYPE 30 12 1 8 9 2 3 4 7 band hidden 6 5 1 9 8 2 3 10 7 4 6 5 stroke in limit · band hidden band SHOWING = over stroke = 11 ≠ same size + same stroke type, both ends of the axle mixed = the truck pulls; STEER mismatch = out of service Paint and surface color alone condemn nothing — holes, cracks, leaks and looseness do. STROKE LIMITS BY TYPE — PANEL B5
1Spring-brake housing — GOOD: smooth, intact, color only. BAD: a crack or hole in the spring section is a standalone out-of-service item (396.3(a)(1)) $35–$100 each for the 30/30 combo unit · Roundabout — market prices, Oct 2026. Not a quote.
2Service housing + clamp seam — GOOD: seam even all the way around. BAD: cracked or rusted through — it fails by the leak and looseness it causes (the standalone holes/cracks line is written for the spring section)
3Air tightness + diaphragm — GOOD: silent at full application. BAD: an audible hiss at the clamp or fittings is a defective brake; a torn diaphragm leaks AND takes that wheel's braking away
4Pushrod — GOOD: straight, traveling in line, threads engaged, locknut set. BAD: bent or cocked at an angle — it corrupts the stroke measurement and side-loads the diaphragm
5Clamp band + mounting — GOOD: band seated evenly, chamber solid on its bracket. BAD: a loose air chamber, or a missing mounting bolt, is a named defective condition
6Clevis + pins — GOOD: pin through, cotter pin in place. BAD: a missing clevis pin or cotter is a named defective part; an egg-worn, sloppy pin is a tell you can see and feel
7Pushrod boot — GOOD: intact, sealing the rod entry. BAD: torn or missing — a wear tell, not a citation: grit and water walk straight in on every stroke. No published reject line names the boot
8Air lines at the chamber — GOOD: routed without chafe, fittings tight. BAD: chafed through the reinforcement ply, or bulging under pressure — out-of-service hose items
9Caging bolt — GOOD: parked in its side holder, dust cap on — that wheel HAS a parking brake. BAD: screwed in and protruding — the spring is held off and that wheel has NO parking/emergency brake. Get it repaired now
10Stroke behavior — GOOD: pushrod travel inside the limit for its type (Panel B5), both sides of the axle near-equal. BAD: travel past the readjustment limit — the colored over-stroke band shows
11Axle matching — GOOD: same size and same stroke type at both ends of the axle (§393.47(b)). BAD: standard on one side, 3″ long-stroke on the other — force curves diverge and the truck pulls. Mismatched sizes on the STEERING axle are standalone out-of-service
12Long-stroke ID (good can) — square air ports, a square cap embossment (0.5″ high = the 3″ rated-stroke version), stamped type + stroke data. Rated stroke runs about ½″ LONGER than the adjustment limit — different numbers; know which can you have before you measure

Read the pair, not the part: every BAD tag above is the published basis — a defective brake under the 20% rule, a failed annual inspection, or a standalone out-of-service item. Two defective brakes on a 5-axle rig park it.

Panel B5 — Adjustment: measure it, don't guess it

Applied stroke is the legal measurement — mark it, apply it, read it
All trucks

Before any stroke check: chock the wheels and release the spring brakes. Never adjust on a hot drum and call it final — a heat-expanded drum reads tight, then drags when it cools.

1 · MARK IT 2 · APPLY IT 3 · READ IT 1 MARK AT THE CHAMBER FACE BRAKES RELEASED 2 RESERVOIR 90–100 PSI 5 TRAVEL = STROKE FREE STROKE — PRY METHOD 3 Meritor 0.5–0.625″ in service MANUAL SLACK — WRENCH WORK 4 DEPRESS THE LOCKING SLEEVE, THEN TURN — VERIFY BY MEASUREMENT
1Mark-and-measure — brakes released: mark the pushrod at the chamber face. Full application: read how far the mark traveled. That travel is the applied stroke — the legal measurement
2Know your pressure window — CVSA counts defects at 90–100 psi reservoir; the annual-inspection method measures at 80–90 psi; DTNA's shop procedure uses 100 psi in the tank with an 80–90 psi application. Printed side by side because the sources differ — not merged
3Free stroke (pry method) — lever the slack until the linings just kiss the drum and measure. Meritor publishes 0.5–0.625″ for a brake in service (an industry roundtable gives 3/8–5/8″ generally — it varies by OEM)
4Manual-slack wrench work — depress the locking sleeve fully, turn the hex to light drag, back off per the maker — then verify by measurement, never by feel. The sleeve must re-engage when you let go
5At the limit is NOT a defect — under ¼″ past the limit = a half defect; ¼″ or more past = a full defect; two halves count as one defective brake in the 20% rule. A chamber type not in the table: stroke must not exceed 80% of the rated stroke marked on the chamber
Chamber type (clamp-type)Standard-stroke limitLong-stroke limit
Type 201¾″2″ (2½″ when the chamber is the 3″ rated-stroke version)
Type 241¾″2″ (2½″ when the chamber is the 3″ rated-stroke version)
Type 302″2½″
Type 36 — CONTESTED2½″ per the text of 49 CFR §393.47(e) — vs 2¼″ per the CVSA roadside chart (older CFR reprints and DTNA's Gunite table agree with CVSA). Both numbers are published; an inspector works from the CVSA chart, the CFR text is the law. We print both — we don't pick one.—

The full published tables — Types 6–36, Bendix DD-3 (2¼″), bolt-type and rotochamber — live in 49 CFR §393.47(e). DD-3 and rotochamber limits run off a different measuring point (from the chamber face) — type matters before you measure.

AUTOMATIC SLACK OVER THE LIMIT? HANDS OFF THE WRENCH — FIND THE CAUSE. Bendix, in its own words: “Do not manually adjust the adjuster… none of these [causes] will be remedied with a manual readjustment.” DTNA tells its own techs the same thing — manual adjustment “is likely masking a mechanical problem. Adjustment is not repairing.” Hand-adjusting also works against the adjuster's back-off clutch and can damage it. The only sanctioned manual adjustments on an automatic slack are at installation and at a brake reline. What to do instead: run the mechanism test, lube it (NLGI #2), check clevis pins, cam bushings and rollers, verify free stroke and applied stroke — fix the foundation brake, or replace the adjuster. EMERGENCY only: a manual adjustment as a last resort to move the vehicle — logged on the post-trip report, repaired as soon as possible. And for manual slacks: adjustment restores clearance, not lining — one that needs constant re-adjusting is reporting worn parts.
DIESEL DIAGNOSTIC AI

Clutch — Adjustment, or Done?

The sandwich, the three numbers, which truck you have, and how you know it's gone — with Eaton's published adjustment specs and roundabout prices on the job. Works offline.
Clutch — the disconnect switch between the engine and the transmission. It lives inside the bell housing — the cast bowl bolted between the back of the engine and the front of the transmission — so you judge it by pedal, gauge and fault codes, not by looking at it. Universal drivetrain education — this set is not tied to any engine. The adjustment numbers are Eaton's published ones (CLMT1351 / CLMT1279). This set answers the money question every driver eventually faces: is it OUT OF ADJUSTMENT (cheap) — or WORN OUT (real money)?

Panel C1 — The Sandwich

The clutch pack, exploded — engine left, transmission right
All trucks
EXPLODED SIDE VIEW — THE PARTS, IN ORDER TORQUE FLOWS → 1 2 3 4 5 6 7 8 TRANSMISSION FACE BELL HOUSING — THE CAST BOWL YOU CAN SEE (DRAWN CUT OPEN) GHOST — SINGLE-DISC VARIANT ONE DISC · NO PLATE SPRINGS CLAMP THE PACK TO THE FLYWHEEL — THE BEARING RELEASES IT
1Flywheel — the heavy steel disc bolted to the back of the crankshaft; the engine-side friction face the discs squeeze against. The ring gear on its rim is what the starter cranks
2Pilot bearing — the small bearing in the flywheel's center; it centers and supports the nose of the input shaft. A cheap part behind the same labor as the clutch itself — it gets replaced during clutch jobs
3Front clutch disc — a steel plate with friction facings on both sides, splined to the input shaft. Squeezed, engine torque crosses into the transmission; released, it spins free and no power crosses
4Intermediate plate — the floating plate between the two discs (twin-disc only). Two discs share the work — double the friction area without a bigger diameter
5Rear clutch disc — the second driven plate. The Class 8 line-haul standard is the 15.5″ twin-disc — torque capacity and heat capacity are why
6Pressure plate — the spring-loaded clamp bolted to the flywheel, spinning with it. Its springs do the squeezing — and on a manual-adjust clutch, the adjustment mechanism lives in here
7Release bearing + yoke — the pedal's hand inside the housing. Pedal down → the yoke (fork arms) works the bearing → the bearing pulls the pressure plate off the pack. It is built to work while you shift — not to ride loaded full-time
8Clutch brake — the small friction disc on the input shaft at the transmission face. Pedal hard to the floor at a standstill, it stops the input shaft spinning so a gear goes in without clash. It only works in the last bit of pedal travel — that squeeze is what Panel C2 protects

You never see any of this on an assembled truck — you see the bell housing, the inspection cover on its bottom, and (on a manual) the pedal. Angle-spring, pull-type is the classic Eaton manual design; diaphragm, push-type is what the automated actuators work — see Panel C3.

Panel C2 — The Three Numbers

Eaton's adjustment specs — gap, squeeze, free play
Eaton-published
1 · RELEASE BEARING → CLUTCH BRAKE GAP MANUAL WINDOW 0.500″–0.560″ 1 0.490″ 0.500″ 0.560″ Self-adjust at setup + Navistar service spec: the floor drops to 0.490″ — the top stays 0.560″. 2 · CLUTCH BRAKE SQUEEZE — THE FEELER TEST 2 BEARING BRAKE 0.010″ CLAMPED AT FULL PEDAL PEDAL TRAVEL → FLOOR coming up, the gauge must free up ½″–1″ off the floor 3 · YOKE CLEARANCE + PEDAL FREE PLAY 3 1/8″ GAP yoke clearance — set at the upper pedal stop ≥1″ pedal free play: 1 inch or more 4 INTERNAL FIRST, LINKAGE SECOND WRONG ORDER → GEAR CLASH · SLIPPING · BURNING (EATON)
1Bearing → clutch brake gap — the master number, measured through the inspection cover; this is the number shops (and warranty) argue from. Manual-adjust (14″ + 15.5″): 0.500″–0.560″. Self-adjust at setup, and Navistar's service spec: 0.490″–0.560″ [Eaton CLMT1351 / CLMT1279; Navistar IK1100025]
2Clutch brake squeeze — a 0.010″ feeler gauge between the bearing and the brake must clamp at full pedal, and free up when the pedal is ½″–1″ from the floor. No clamp, or clamp in the wrong place in the stroke = gears clash going into gear from a stop [Eaton CLMT1351 / CLMT1279]
3Yoke clearance + pedal free play — 1/8″ (0.125″) between the yoke tips and the bearing's wear pads, set with the upper pedal stop; in the cab that reads as 1 inch or more of pedal free play. On a mechanical-linkage truck, shrinking free play = wear asking for its internal adjustment — the pedal is the wear gauge [Eaton CLMT1351 / CLMT1279]
4The order — internal adjustment before linkage adjustment, always. Eaton's warning, in substance: incorrect adjustment can cause transmission gear clash, and slipping and burning of clutch components. A shop that only winds the linkage to make the pedal feel right is masking the real setting

Shop shortcut for #1: the Eaton/Freightliner go/no-go tool A02-12419 — green end 0.50″, blue end 0.56″. How a manual is set (so you can recognize correct work): Kwik-Adjust — a helper holds the pedal fully down, the adjuster is turned through the inspection hole; clockwise moves the bearing toward the transmission. Then yoke clearance, then the feeler test, then free play — in that order.

Panel C3 — Which truck do you have?

Manual-adjust · self-adjust · automated · automatic
4 different answers
SOLO WEAR TAB (SCHEMATIC) LIFE LEFT AT THE MARK = DONE 1 EATON AMT — DASH WARNING ⚠ FC 706 SPN 2983 · FMI 17 CLUTCH WEAR WARNING 2–6 weeks BEFORE it fails 2
1Solo wear tab — schematic. The tab rides on the upper cam and mirrors its movement as the facings wear. Mid-travel = life left; at the replace mark = the clutch is telling you it's done. (Advantage Self-Adjust adds a second tab, so one shows wherever the clutch stops.) You read this tab — you never adjust a Solo: Eaton's install procedure prints “NOTICE: Do not reset the clutch.”
2FC 706 (SPN 2983, FMI 17) — Eaton's automated-transmission clutch-wear alert, published as arriving 2–6 weeks before failure: time to schedule the clutch instead of stranding on it. The follow-up inspection looks at the yoke, rollers and release bearing — a seized bearing can damage the diaphragm spring fingers
(a) Manual-adjust(b) Self-adjust — Solo(c) AMT — DT12 · UltraShift · Endurant(d) Allison automatic
Clutch pedalYesYesNoNo
Dry launch clutchYesYesYes — a real dry clutch, worked by an actuatorNo — a hydraulic torque converter
Who adjustsYou / the shop, by hand — the adjuster lives in the pressure plateThe clutch itself — two sliding cams track the wear. DO NOT RESET (Eaton)Nobody wrenches it — the computer learns it (clutch learn in DiagnosticLink / ServiceRanger)Nothing to adjust — no clutch adjustment exists
Wear gaugeYour foot: pedal free play — plus the bearing gap underneathThe wear tab on the camThe dash: fault codes (FC 706)—
You can checkFree play by hand; the squeeze with a feeler gaugeTab position through the inspection opening; that free play existsDash warnings + codes onlyFluid + codes
Repair pathAdjust first (cheap) → replace the kit when spec can't be reachedRun it to the tab, then replaceClutch unit and/or actuator + a software relearn — a shop with the softwareTransmission service — fluid/filters; internal repair if it ever comes
HYDRAULIC-LINKAGE TRUCKS: the pedal CANNOT tell you the wear. Eaton's own engineering letter: the hydraulic system keeps compensating, the driver loses the sensation of free pedal, and the clutch can go straight to slipping. On a hydraulic truck, read the tab and measure the gap — not your foot. That is exactly why Eaton recommends the self-adjusting clutch behind hydraulics.

DT12 fine print: no wear tab and no reset — the pressure plate self-adjusts inside, the facings are organic, single- or double-disc by powertrain, and the clutch is always replaced as a complete unit. Wear state is a DiagnosticLink question, not a look-under-the-truck question.

Panel C4 — Out of adjustment vs worn out

The money distinction — same symptom family, very different bills
Cheap test first
ENGAGES LOW ENGAGES AT THE TOP 1 2 WHERE IT GRABS TELLS YOU WHERE THE FACINGS ARE
1Engages low / drags — too much free play or incomplete release: the discs never fully let go, gears clash at stops, and the clutch brake gets hammered doing a job it only does at the floor
2Engages at the very top — the facings are nearly gone and the adjuster has no travel left to give. If it can't come into spec, it's done — the rule is at the bottom of this panel
OUT OF ADJUSTMENT — cheap to fixWORN OUT — the real money
Free play shrinking toward zero (mechanical linkage) — wear has moved the bearing; the clutch is asking for its internal adjustment. Normal wear compensation, not failure TEST: FREE PLAY BY HAND RPM climbs, road speed doesn't — slipping under load, worst in a high gear on a grade at gross weight. The facings can no longer hold engine torque TEST: HIGH-GEAR PULL
Gears clash going into gear from a stop, fine once rolling — the clutch brake isn't squeezing, so the input shaft keeps spinning at the stop TEST: 0.010″ FEELER Engages at the very top of the pedal with no adjustment left to give — the internal adjuster is at the end of its travel because the facing is used up TEST: GAP GAUGE
Engagement point migrating noticeably over weeks — the adjustment interval arriving, not necessarily a dying clutch TEST: GAP GAUGE Chatter / shudder on launch that adjustment doesn't change — warped or heat-damaged facings, broken damper springs, oil contamination TEST: ADJUST FIRST
Too much free play after linkage work or wear elsewhere — the clutch may not fully release (drag) TEST: YOKE 1/8″ Burnt smell after ordinary driving, and a release-bearing growl that changes with the pedal pressed vs released — the bearing lives in the same labor circle as the clutch, so it's a replace-the-set conversation TEST: LISTEN + PEDAL
It all started right after a clutch job done in the wrong order — linkage wound up before the internal setting was right (Panel C2, #4) TEST: INTERNAL FIRST The manufacturer's own verdict: Solo — the wear tab at the replace mark. Eaton AMT — FC 706 on the dash. No pedal symptom required TEST: READ THE TAB / CODE
THE RULE — ADJUST FIRST. It's the cheap test. If the bearing gap comes back into spec (0.490″–0.560″), the squeeze verifies, and the slip's gone — it was adjustment. If it can't come into spec, or it slips in spec — it's done, and no amount of adjusting puts friction material back on a disc. Navistar's warranty bulletin makes the same point from the other side: a clutch replaced without first being adjusted into spec can be ruled not warrantable (IK1100025) — adjust before you condemn.

Keep driving on it and the cheap problem eats expensive parts: no free play = the release bearing rides the pressure-plate fingers full-time; it overheats, and a seized bearing can tear the fingers up. Too much play = drag at every stop, hammering the clutch brake until even a perfect adjustment can't give you a clean engagement.

Panel C5 — What the job costs

Roundabout prices — the price is the heads-up
PARTIAL — market

Gary's rule: a roundabout price beside the part is what makes a driver actually check his truck — it shows what neglect costs. Every figure below is a market figure, PARTIAL by nature: repair-cost guides and shop lists dated 2025–2026, checked Oct 21, 2026. A heads-up, never a quote.

MANUAL CLUTCH JOB — ALL-IN$2,000–$4,500The working range. Published guides span $1,500–$4,500: Fleetbarn 2026 ($1,500–$3,500) · ExpressTruckTax 2026 ($2,000–$4,500) · DispatchFF 2026 ($2,500–$4,000, typically at 400,000–600,000 miles). Under ~$1,500 usually means parts or labor is being quoted separately.
CLUTCH KIT$800–$1,200Discs, pressure plate, release bearing, pilot bearing (published line-item breakdown, 2025). Treat it as the floor for a twin-disc Class 8 kit, not the ceiling — premium brands run higher.
LABOR$800–$2,2508–15 hours at $100–$150/hr. The hours are transmission-out-and-back hours — that IS the job; the clutch itself is cheap compared to reaching it. Cross-check: one shop flat-lists transmission R&I alone at $1,200.
FLYWHEEL$200–$600Resurface or replace. Eaton's procedures require the flywheel inspected and resurfaced/replaced to spec at clutch replacement — a heat-checked, grooved flywheel under a new clutch is how you buy the job twice.
WHILE YOU'RE IN THERE$150–$400Rear main seal, transmission input-shaft seal, pilot bearing, clutch brake. The rear main is a cheap part behind transmission-out labor — skip it now and you pay the same R&I twice. An input-shaft seal leak oils the new clutch: the classic comeback.
DT12 / AMT CLUTCHNo honest number — get a quote firstNo reliable published total exists, so none is printed. Expect it at or above the manual range — the clutch is replaced as a complete unit, the concentric actuator is a separate major part, and the job ends with a software learn. That is reasoning, not a priced range.
ALLISON AUTOMATICNo launch clutch to buyA torque converter does the launching — there is no dry clutch to wear out or replace. Service is fluid and filters; internal work is a transmission-shop conversation, not a clutch job.

Driving on a dead clutch costs more than the clutch: a slipping clutch is a grinding wheel against the flywheel — heat spots, cracking and scoring turn a resurface into a replacement. A clutch that won't fully release makes the input shaft and gear teeth take every stop-and-engage as a clash. And when it lets go entirely, it strands the truck wherever it dies — the replacement then happens at the nearest shop's price, on the tow company's clock, with a load to recover.

DIESEL DIAGNOSTIC AI

Transmission — The Gearbox and the Three Families

What the gearbox does, the parts inside it, the three families (manual, automated, Allison), warning signs, the check-first order, and roundabout prices — from Eaton, Detroit/DTNA and Allison published material. Works offline.
Transmission — the exchange counter between the engine and the wheels. It trades speed for torque in first gear and torque for speed in top gear, and it is where most “the truck won't move” complaints are born. This set starts at the input shaft — the Clutch set covers everything in front of it — and ends at the output flange, where the Driveline set takes over. Three families, one check-first order — air, fluid, plugs and codes, then the shop — and Gary's roundabout prices on the parts: every figure market-sourced and dated, and refused where no honest number exists.

Panel T1 — The Gearbox

Twin-countershaft cutaway — the power path, input to output flange
Eaton manual shown
SIDE VIEW — TWIN-COUNTERSHAFT BOX (SCHEMATIC) 1 2 3 4 5 6 7 8 RANGE POWER: INPUT → COUNTERSHAFTS → THE GEAR YOU'RE IN → RANGE → FLANGE FRONT BOX: NO SYNCHROS — THE SYNCHRO LIVES IN THE RANGE SECTION ART STOPS AT THE OUTPUT FLANGE — THE DRIVELINE SET TAKES IT FROM THERE
1Input shaft — splined to the clutch disc; it delivers engine torque into the box. It stops turning when the clutch brake squeezes at a standstill (Clutch set)
2Main drive gear — the first gear inside the case. It drives both countershafts, all the time the input turns
3Countershafts — two of them, flanking the mainshaft. Torque splits between them — each carries half the load (Panel T4)
4Floating main gears — always meshed with the countershaft gears (constant-mesh). The ones you are NOT using float free on the mainshaft — no bushings, no sleeves
5Sliding clutch — the splined collar that locks one main gear to the mainshaft. That lock IS the gear you are in. Shifting moves these collars (dog teeth) — never the gears themselves
6Mainshaft — the spine of the front box, carrying the floating gears and sliding clutches back to the range section
7Range (auxiliary) section — the smaller second gearbox bolted behind the front box: LO/HI, air-shifted — and the only synchronizer in the box lives here (Panel T3)
8Output flange — the bolted flange at the very back of the box. Power leaves the transmission here, and this set ends here

Drawn as the Eaton twin-countershaft manual (TRSM0503); the DT12 runs one countershaft in an aluminum case (Panel T4). There are no synchronizers in this front box — DTNA's driver material says it plainly: 9- and 10-speed range-shift models “are not synchronized”; the synchro lives in the range section. That is why a missed shift grinds clutching teeth, and why timing matters more than force. Everything behind the output flange — driveshaft, U-joints, differential — is the Driveline set; none of it is drawn here.

Panel T2 — Gear Math

The ratio ladder — one engine, three answers
Published ratios
THE RATIO LADDER — FULLER ADVANTAGE (EATON CUMMINS, PUBLISHED) 1 1ST · 12.80 1,000 lb-ft IN → ≈12,800 lb-ft OUT 1ST PULLS THE DOCK 2 9TH · 1.00 — DIRECT 1,000 IN → ≈1,000 OUT — STRAIGHT THROUGH 3 10TH · 0.73 — OVERDRIVE ≈730 OUT — OUTPUT SPINS FASTER THAN THE ENGINE OUTPUT TORQUE = INPUT × RATIO · OUTPUT SPEED = INPUT ÷ RATIO
1First gear — 12.80, a reduction (ratio above 1.00). The output shaft turns at 1/12.8 of input speed while torque multiplies about 12.8×. That multiplication is the whole reason a loaded truck can walk away from a dock on a grade without cooking the clutch: the engine revs, the wheels barely turn, and the twist is enormous
2Direct — 1.00. Output turns exactly with input: 1,000 lb-ft in, about 1,000 out, same speed. Nothing gained, nothing traded — the through-path
3Overdrive — 0.73 (ratio below 1.00). The output spins FASTER than the engine while torque divides to about 730 lb-ft. At 65 mph on the flat the truck doesn't need dock-pulling twist — it needs the engine turning slowly enough to sip fuel. Hit a grade in top gear and there isn't enough multiplication, so the box (or the driver) downshifts to buy torque back

One real published set — Eaton Cummins' Fuller Advantage automated: 1st 12.80, 9th 1.00, 10th 0.73, overall ratio spread 17.53. A “speed” is one selectable ratio path through the box: the lever or the computer doesn't create ratios, it selects which path power takes. Reverse is its own path, through an idler gear that flips the rotation.

Panel T3 — How 13 Speeds Stack

5 lever positions × range × splitter — sections multiply
TRDR0630
13 SPEEDS = 5 × RANGE × SPLITTER (HI ONLY) 1 2 3 4 SPLITTER LOCKED OUT IN LO — KNOB INTERLOCK LO 1 2 3 4 5 HI 6 7 8 9 10 SPLIT EACH (+S) → 11 · 12 · 13 5 (LO) + 5 (HI) + 3 SPLITS = 13 SPEEDS 10-SPEED: 5 × 2 — NO SPLITTER 18-SPEED: SPLITTER IN BOTH RANGES THE DT12 STACKS THE SAME WAY — WORKED BY AIR, NOT A LEVER
1Front box — five lever positions. This is the section shifted gear-to-gear, by a lever on a manual or by actuators on an AMT
2Range — the second, smaller gearbox behind the front box, with two ratios: LO and HI. It multiplies EVERY front-box ratio — 5 × 2 = 10 speeds. Air-shifted and synchronized: preselect it with the range lever before the lever move, and the change completes as the lever passes through neutral
3Splitter — an overdrive step inside the auxiliary section that splits a gear roughly in half. On a 13-speed it works in HI range only: five LO + five HI + three splits = the published 13 forward speeds (a 3-speed auxiliary behind a 5-speed front). On an 18-speed it works in both ranges
4The interlock — the shift knob physically prevents moving the splitter control forward in LO range, and prevents a range-down with the splitter forward. It isn't a suggestion — the gate is mechanical [TRDR0630]

A splitter shift needs no lever movement: preselect on the knob, then a single clutch movement to break torque — the shift completes when synchronous is reached. Don't double-clutch a splitter shift like a lever shift. And never make a range or splitter shift in reverse, or coast in neutral with the lever out of gear — the box's splash lubrication assumes an engaged path.

Panel T4 — The Twin-Countershaft Split

Why an Eaton carries the load on two shafts
TRSM0503
FRONT VIEW — ONE DRIVE GEAR, TWO COUNTERSHAFTS 1 2 3 CARRIES HALF THE LOAD CARRIES HALF THE LOAD THE UNSELECTED MAIN GEAR FLOATS BETWEEN THEM DT12 CONTRAST — ONE COUNTERSHAFT ONE SHAFT CARRIES IT ALL — LIGHTER ALUMINUM CASE ALL GEARS CONSTANT-MESH — SHIFTING MOVES COLLARS, NOT GEARS
1Main drive gear — turned by the input shaft, meshing with a drive gear on EACH countershaft
2Countershaft gears — the torque splits equally between the two shafts, so each gear set carries HALF the load. Half the load per tooth is why the gears can be narrower for the same strength — Eaton: the design “greatly reduces gear face width” [TRSM0503]
3Floating main gear — a mainshaft gear that isn't selected isn't fixed to the shaft: it floats, centered between the two countershaft gears. The design needs no gear sleeves or bushings on those gears

Folklore check: “floating” names this hardware — the gears float. It does not name a way to shift: float shifting is not Eaton's procedure — the published procedure is double-clutching [TRDR0630]. Power flow through the auxiliary section mirrors the front: mainshaft → auxiliary drive gear → split between two auxiliary countershafts → the selected auxiliary gear → output shaft [TRSM0525].

Panel T5 — Three Families

Same job, three different machines — the same four questions
Manual · AMT · Allison
1 MANUAL 2 AMT 3 ALLISON
1Manual — Eaton Fuller: you shift the lever, your foot works the clutch
2AMT — DT12 · UltraShift PLUS · Endurant: a manual gearbox whose clutch and shifts are worked by a computer, through actuators
3Allison — no dry clutch at all: a torque converter launches the truck, and oil pressure applies the ranges
Manual — Eaton FullerAMT — DT12 · UltraShift · EndurantAllison automatic
Who shifts?You — the lever, double-clutchedThe TCM (computer) — XY / MTM actuators move the railsThe TCM — oil pressure through the valve body
Who works the clutch?Your left foot (see the Clutch set)A clutch actuator — ECA (electric) · LCA / CPCA (pneumatic)
Actuator: $1,500–$2,500 parts + labor · Roundabout — market prices, Oct 2026. Not a quote.
No pedal clutch to work — the converter slips in fluid at a stop
Launch deviceThe dry clutch — the clutch brake stops the input at a standstillThe same dry clutch, feathered by the actuator (the DT12's creep mode imitates a converter at the dock)Torque converter — pump throws oil at the turbine, the stator multiplies torque at launch, and a lockup clutch gives direct drive at cruise
What's inside that wearsSliding clutches, the range synchro, bearingsThe same gearbox parts — plus actuators and the TCMFive wet multiplate clutches + three planetary gear sets — nothing slides in and out of mesh
The driver actually doesPreselect range before the lever move; build 100–120 psi of air firstSelect D / N / R; keep the air built; read the popups and codes — learns and calibrations are shop softwareSelect D / N / R (+ MODE); read the selector — oL OK / oL Lo / oL HI; never idle in D more than five minutes (it overheats the converter)

Folklore check: an Allison gets called “clutchless.” Count the clutches in its row: five wet multiplate clutches, plus the converter lockup clutch. What it doesn't have is a dry launch clutch — the converter does the launching. From the other side: an AMT is not a torque-converter automatic, and its gearbox is not a synchronized one — the DT12's main box is non-synchronized, like the manual's front box (Panel T1).

Panel T6 — Air Is Life

Range, splitter and every AMT shift run on air
58–63 psi regulated
THE SHIFT AIR PATH — PLUMBING, NOT MAGIC 1 2 3 4 5 AIR TANKS 100–120 psi FILTER / REGULATOR 58–63 psi RANGE CYLINDER SPLITTER CYLINDER AMT MODULE TRUCK AIR BUILDS TO 100–120 psi — THE REGULATOR STEPS THE SHIFT CIRCUITS DOWN TO 58–63 psi DRAIN THE TANKS — WATER + OIL IN THE AIR STICKS AMT VALVES AND SLOWS CYLINDERS
1Air tanks — the supply for every shift muscle in this set. Judge NO shift complaint until the gauges show a full build: 100–120 psi [DTNA driver manuals]
2Air filter/regulator — mounted on the transmission; it holds the shift circuits at 58–63 psi (one DTNA edition prints 57–62 — an edition variance, not an error). The regulator is not adjustable and not repairable — replace the unit. Its filter element is a service item
3Range cylinder — the air piston that works the range shift in the auxiliary section
4Splitter cylinder — the air piston that works the splitter
5AMT module — on automated boxes the same air works the shift and clutch actuators (Eaton XY / MTM; DT12 pneumatics). Hissing here after shutdown, or during shifts, is a shift fault in progress — not a cosmetic leak

DT12 — YELLOW POPUP: “Caution: Transmission Air Supply Low.”

The pneumatic controls shift the transmission; with inadequate air, “the quality of gear shifts may be degraded.” DTNA's instruction: wait for air pressure to build — and if it appears while operating, slow down, pull off safely, and correct the problem.

DT12 — RED POPUP: “Warning: Transmission Will Not Shift.”

Severe air loss: the transmission will not shift. Choose the safest place to pull off — “Prepare for the vehicle to be immobile if the problem cannot be resolved.” [DTNA]

DTNA's caution for the manual box, verbatim-grade: a leaking air filter or air lines “can cause slow or hard shifting of the transmission, and eventual transmission damage.” AMT module supply (Eaton): above 90 psi at the inlet, 75–85 psi at the module “L” port — below that, repair the vehicle air system first [PARTIAL: Eaton text via expert reproduction]. An AMT with a truck-air problem is not a transmission problem yet.

Panel T7 — Warning-Sign Card

What it's telling you, the first check, and the call
6 signs
SignWhat it usually meansFirst checkCall
Grind — the SAME shift, every timeA truck fault, not your timing: worn range synchro (grinds on the range change), the clutch not fully releasing, or worn sliding-clutch teeth. Driver error is random; worn parts fail in the same place every timeAir first (Panel T6), then clutch releaseCHECK FIRST
Jumps out of gearWorn sliding clutch first in expert practice — its teeth no longer hold full engagement; also worn detent, bent or worn shift fork, partial engagement. Worst on deceleration and in high range. A safety fault: it can leave you with no drive on a gradeNote when it happens — decel? high range? — and get it seenPARK IT
Won't take a gearAt a standstill: clutch brake or clutch drag — the input is still turning (Clutch set). Rolling, on a manual: linkage/tower or fork damage. On an AMT: air, actuator, or an active fault holding the box in neutral — rarely the gear set itselfAir → codes (Panel T8)CHECK FIRST
Rattle in neutral at idle, clutch outUsually NOT a failure. Idle-speed fluctuation makes the internal gears touch drive-side and coast-side in turn. Eaton's warranty guideline RWRG0046: it is “not a component damaging condition” — and not a warrantable failureRaise RPM slowly: the noise fades between ~700–1,000 RPM. Or partially release the clutch — it goes awayUSUALLY NORMAL
Hiss at the box + air gauge droppingAn actuator, range/splitter cylinder, or air-module leak. On any air-shifted box this is a shift fault in progress: the DT12 degrades shift quality (yellow popup), then refuses to shift (red)Walk the air path; drain the tanksCHECK FIRST
Hot + burnt smellSustained running above 250 °F is outside Eaton's limit (intermittent 300 °F does no harm) — low or wrong oil, sustained overload, a failed cooler. An Allison showing P0218 limits itself to 4th range as protection — that's the box defending itself, not a broken gear setFluid level + the correct spec; cooler airflowSHOP SOON

And the sign that isn't the transmission at all: “slipping” in a gear. Sliding clutches are meshed or they aren't — a geared manual or AMT cannot slip in gear. If RPM climbs and road speed doesn't while a gear is fully engaged, the slip is at the clutch (Clutch set) — or the gear was never fully engaged.

Panel T8 — Check First

The order is the lesson — and the service numbers
In this order
1 · AIR — build 100–120 psi before judging any shift. Listen at the box; drain the tanks (Panel T6)
2 · FLUID — level WITH the lower edge of the fill hole (below). The specified fluid only — never mix types or brands
3 · PLUGS & CODES — harness seated, no chafe, no green corrosion. Read and RECORD the codes — don't clear them to “see if it comes back”
4 · SHOP — everything inside the case, actuator internals, TCM programming, and learns/calibrations are shop work
THE FILL-HOLE RULE (DTNA) 1 2 LEVEL = LOWER EDGE OF THE HOLE SEEN / TOUCHED — NOT FULL IT MUST BE LEVEL WITH THE LOWER EDGE OF THE FILL OPENING
1The fill opening — on Eaton manuals and the DT12, correct level is level with the lower edge of the fill opening, truck level in its normal operating position, checked with a finger or a bent pipe cleaner. UltraShift PLUS reads on a sight glass; an Allison reads on its selector (oL OK / oL Lo / oL HI); the Endurant warns on its own when the level runs low
2Seen or touched is NOT full — DTNA, verbatim-grade: “A lubricant level close enough to be seen or touched is not sufficient.” Low again after topping means find the leak. Overfilling is damage too: it forces oil out through the mainshaft openings, down onto the clutch and clutch brake
FamilyFluid (spec)Interval — conditions ride with the number
Eaton manual + UltraShift PLUSEaton PS-386 syntheticLinehaul 500,000 mi / 5 yr · Vocational 180,000 mi / 3 yr
Endurant HDPS-386750,000 mi / 5 yr — HD, linehaul only · Vocational 250,000 mi / 3 yr
Endurant XDPS-386Linehaul 500,000 mi / 5 yr · Vocational 250,000 mi / 3 yr
DT12MB 235.16 75W-85 (MY2017+) · MB 235.11 75W-90 (older; still approved in service)Efficient Long Haul 400,000 mi / 5 yr (the truck's Maintenance System can extend toward 600,000) · Long Haul 300,000 / 5 yr · Severe 180,000 / 3 yr
AllisonTES 668 recommended · TES 295 / 468 approvedFluid (TES 295/468) 300,000 mi general · 150,000 mi severe · Filters 75,000 mi

Capacities — always finish at the fill hole, never by the jug count: Eaton 10-speed ≈ 11.75 qt · RTLO 13/18-speed ≈ 14 qt · DT12 refill (MY2017) A-box 12.7 qt · B-box 8.5 qt · C/D-box 8.9 qt. No additives, no friction modifiers, no EP gear oils in Eaton/Detroit geared boxes [DTNA]. Wrong-spec DT12 oil damages the box and is not warranty-covered (Detroit service letter 17TS1). Drain warm, and read the magnetic plug: light fuzz is wear paste; chunks are never normal.

Ignore web pages claiming a 150,000-mile DT12 fluid-and-filter change — that's content-farm material, contradicted by DTNA's own brochure (the table above). DTNA publishes no DT12 filter interval: the service filter is a small internal screen, cleaned at service [PARTIAL]. Not driver-serviceable, stated straight: anything inside the case, actuator internals, TCM programming, and the learns/calibrations that follow parts replacement — those are DiagnosticLink / ServiceRanger procedures, done in a shop, often on shop air.

Panel T9 — The Money Card

Roundabout prices — the price is the heads-up
PARTIAL — market

Gary's rule: a roundabout price beside the part is what makes a driver actually check his truck — it shows what neglect costs. Every figure below is a market figure, PARTIAL by nature — trade press, market guides, dealer flyers and verified-expert cases, retrieved Oct 21, 2026. A heads-up, never a quote. Where no honest number exists, the tile says so — that refusal is a lesson too: get the quote before authorizing work, not after.

AMT CLUTCH ACTUATOR$1,500–$2,500Parts + labor for the category — Eaton ECA / DT12 CPCA class (FreightWaves, Aug 28, 2025). Texture: one documented ECA case ran a $899 part and 1.5–2.0 hrs including the ServiceRanger calibration (verified-expert case).
XY SHIFTER — EATON AMT, PART ONLY$800–$1,000The part: “just under $800” from a supplier in one expert thread; “right at $1,000 for the part only” on a dealer invoice in a driver forum. Budget the part from this tile — the installed total is the next one.
XY SHIFTER — INSTALLEDNo honest number — get a quote firstNo published all-in installed figure exists. Get part and labor separated, in writing, before you authorize — not after.
TCM / TRANSMISSION ECU$2,199The genuine Eaton TCM part, retail listing. Dealer replacement quotes run $1,500–$3,000 plus programming/calibration (module-repair vendor).
RANGE SYNCHRO KIT$350–$400Parts — synchro/sliding-clutch kit + auxiliary-section gaskets, dealer-account pricing (verified-expert Q&A). The money is in the labor: auxiliary-section access, in the truck only with special tools — otherwise the box comes out.
MANUAL REBUILD$2,500–$6,000Fleetbarn, “Truck Repair Cost 2026.” The same guide puts full replacement at $6,000–$10,000+.
MANUAL SWAP — INSTALLED$7,000–$10,000FreightWaves, Aug 28, 2025.
REMAN UNIT ONLY — WELLER$3,349 · $4,189 · $4,58910-speed $3,349 · 13-speed $4,189 · 18-speed $4,589. TLG Peterbilt dealer flyer, Feb 2026 — after coupons, with an acceptable core returned. Install labor is extra.
OVERHAUL / REPLACEMENT$8,000–$15,000DispatchFF owner-operator guide, 2026 — typically at 500,000–800,000 miles.
AMT FULL REPLACEMENT$10,000–$18,000FreightWaves, Aug 28, 2025. The labor anchor behind AMT money: DT12 transmission R&I books 8.8 hrs, and a clutch adds 1.4 on top (verified-expert book times).
ENDURANT LCANo standalone price — a claim insteadEaton Cummins' launch claim: replacing the LCA without removing the transmission saves “about $5,000” versus a transmission-out actuator job. Printed as a manufacturer claim, not a price — the saving is the story.
ALLISON REBUILD / REMAN TOTALNo honest number — get a quote firstNo reliable published range was found — marketplace listings and a 2017 forum quote were rejected as evidence. A borrowed number would be worse than none.
TRANSMISSION OIL COOLER — CLASS 8No honest number — get a quote firstLight-duty estimator ranges exist; they do not transfer to Class 8 coolers, lines, and labor — and they will not be laundered into a truck figure.

The tow is the tuition: a box that won't take a gear doesn't limp to a cheaper shop — it stops where it dies, and the repair happens at the nearest shop's price, on the tow company's clock. Heavy hook $250–$600 plus $4–$15 per mile (2026 towing guides) — more for a loaded trailer or an after-hours call. Driving on a known fault compounds it: every clashed range shift wears the synchro you were trying to save, and sustained running above 250 °F turns a fluid problem into a gear-and-bearing problem. The check-first order in Panel T8 costs minutes.

DIESEL DIAGNOSTIC AI

Fuel Injectors — The Deep Dive

Detroit's Amplified Common Rail, taken apart: how the injector boosts its own pressure, one code per hole, how you find the dead one, what a swap really involves, and roundabout prices — from Detroit/DTNA and Bosch published material. Works offline.
Fuel injectors — the part that does two jobs. On a Detroit the pump makes only moderate pressure; each injector boosts it again, inside itself, at the moment it fires — the Amplified Common Rail System (ACRS). That is why these injectors carry two solenoid functions, six-digit calibration codes, and prices that punish guessing. The order that protects your wallet: wires first, proof second, parts last — the tablet names the hole; only a bench proves the injector. Wave-2 Panel A shows where they sit under the valve cover; this set is the deep dive.

Panel I-1 — The Amplifier

One injector, cut away — moderate pressure in, injection pressure out
ACRS — DD platform
ONE DD INJECTOR — CUT AWAY (SCHEMATIC) 1 2 3 4 5 6 NEEDLE AMPLIFIER RETURN ≈900 BAR IN ≈2,200 BAR OUT COOLANT — OUTSIDE THE CUP WALL THE PUMP MAKES MODERATE PRESSURE. THIS INJECTOR BOOSTS IT AGAIN — INSIDE ITSELF. NO RATIO IS PRINTED — NOBODY PUBLISHES ONE
1Electrical connector + harness — both solenoid functions are wired through the injector harness under the valve cover; the front bank (holes 1–3) passes through a 14-pin valve-cover connector. The harness lives in hot oil mist, and it is a repeat offender (Panel I-2)
2Needle control solenoid — commands the nozzle needle: when fuel goes in. Its codes are SPN 651–656, one per cylinder (Panel I-2). Circuit codes (open, shorted) point at the wiring first; a stuck-open needle points at the injector itself
3Amplifier control solenoid — commands the booster piston, and its codes are bank-level: SPN 4257 (holes 1–3) and SPN 4258 (holes 4–6). A clean 651–656 screen does NOT clear the amplifier side
4Amplifier (step) piston — rail-pressure fuel pushes the big end; the small end squeezes a small volume of fuel up to injection pressure. Two jobs in one part: this injector meters AND boosts
5Rail fuel in — from the side — rail → short steel line → cam-housing pass-through → P-tube → the side of the injector, at ≈ 900 bar / 13,000 psi. Side-fed under the cover — not top-fed, and nothing above it drives it mechanically
6Needle, nozzle, and the cup — fuel leaves at ≈ 2,200 bar / 32,000 psi. The tip seats in the injector cup threaded into the head — the cup is the wall between fuel and coolant. Fuel in the coolant is the cup's signature; there is no SPN for a cup

The pressures are pinned by three independent trade sources quoting named OEM staff (CCJ via Freightliner; Truck Parts & Service via Cummins' support director; Fleet Equipment), with Bosch's amplified-injector (HADI) literature behind the design; Detroit's own manuals pin the ACRS architecture. The contrast, in Cummins' published figures: XPI makes full pressure at the pump — 2,600 bar maximum — and the injector only meters it. Not all fuel that enters is injected: control fuel returns to the tank, and a leaking injector can instead dump fuel into the oil — oil level rising on the dipstick, with a diesel smell on the stick, is the driver-visible warning.

Panel I-2 — Six Holes, Six Codes

The code map — one SPN per hole, bank codes above, the harness first
Detroit fault family
THE HEAD, COVER OFF — FRONT (FAN END) AT LEFT 1 2 3 4 4257 · 2797 4258 · 2798 1 2 3 4 5 6 651 652 653 654 655 656 RAIL — DRIVER SIDE CIRCUIT CODE? WIRES FIRST. GROUP + AMPLIFIER CODES ARE BANK-LEVEL — THREE HOLES AT A TIME
1SPN 651–656 — one per hole — “Fuel Injector Cylinder #N Needle Control Valve,” numbered as the library records them. The FMI tells the kind: 5 open / 6 shorted = circuit; 3 / 4 = response out of window; 7 = stuck open. An active 651–656 carries a factory reaction — 25% derate, EGR shut, regen inhibited — the truck is already protecting itself, and a DPF problem queues up behind it
2Bank codes — 4257 / 4258 and 2797 / 2798 — amplifier control valve (4257 = holes 1–3, 4258 = holes 4–6) and injector-group faults (2797 / 2798) are detected on shared drivers, three holes at a time. A bank code with no 651–656 points at the amplifier side or the group wiring — not at one needle
3The harness has its own code — SPN 3597; check it first — chafing at the back of the engine and oil intrusion are repeat offenders. Detroit's published cause list for a group fault runs: a failed injector in the group → the harness run to the 14-pin connector → the MCM — and field experience says the harness is the usual answer
4The rail, and SPN 157 — Detroit's published cause list for 157/16 (rail pressure too low) includes “fuel injector (amplifier or needle) leakage”: one injector leaking internally bleeds the rail down with no 651–656 code at all, because its electrics are fine. A 157 alone justifies neither six injectors nor a pump — the Fuel System Integrity Check separates a lying sensor or quantity-control valve from a system that genuinely cannot hold pressure

Gen 5 adds cold-condition variants of the needle family, SPN 7384–7389 — same drawers, colder mornings. The module names the hole; your job is proving it is the injector and not its wires. The library-verified injector behind these codes (DD15/DD16): A4720700887 — Bosch CRIN 4.2, new 0445120207 / reman 0986435540 — recorded per cylinder at SPN 651–656 in the parts library.

Panel I-3 — Which Hole Is Dead

The diagnostic path — in the order a good shop works
Proof before parts
FIND THE BAD ONE — IN ORDER 1 2 3 4 5 OLD SCHOOL COLD PORT = DEAD HOLE CUTOUT ONE AT A TIME BALANCE +100 STARVED −100 = OVER SWAP FOLLOWS IT? BENCH PROVES IT THE CUT THAT CHANGES NOTHING FOUND THE DEAD HOLE ISB (BALANCE) IS A GHG14-ERA TOOL. GHG17 + SPN 1322: DETROIT SAYS DO NOT USE ISB — FOLLOW THE CODE'S TREE.
1Old school — the cold port OLD SCHOOL — an infrared thermometer on each exhaust port as the engine warms up finds the cold hole. It finds the hole, not the reason: injector, compression and valvetrain all read cold. Cheap, honest, and older than the ECM (expert practice — no factory procedure)
2Cutout — the headline test — DiagnosticLink kills one injector at a time while the engine idles. A healthy hole's cut stumbles the engine; the dead hole's cut changes nothing, because it was contributing nothing
3Balance — contribution (ISB) — the software trims each injector to smooth the idle: +100 = starved (the software has run out of compensation), −100 = over-fueled (fully backed off). A pegged cylinder indicts — it does not convict: Detroit's own flow runs the compression test next, then checks the fuel supply to the injector, before condemning it. Era footnote on the drawing: ISB is the GHG14 procedure; Detroit's GHG17 idle-misfire bulletin says do not use it for SPN 1322
4Swap — the proof step — move the suspect injector to a known-good hole and re-run. Fault follows the injector = injector. Fault stays on the cylinder = the cylinder, or its wires. Skipping the swap is how good injectors get bought
5Bench — the verdict — spray pattern, dribble after shutoff, and return flow at pressure are bench measurements. The scan tells you which hole and whether the wires are guilty; the bench tells you whether the injector itself is good. A shop that condemns an injector on codes alone, with no cutout, swap or compression test, is selling — not diagnosing

“Replace all six” is a sales sentence, not a factory one: no Detroit source retrieved requires it for a single proven failure, and Detroit's procedures condemn and replace injectors individually. A set is an elective call on a high-mileage engine with several holes drifting — the driver's decision, with the numbers in front of him (Panel I-4). Safety before anyone opens anything: wait at least 10 minutes after shutdown before servicing the high-pressure circuit — residual pressure can penetrate skin. Under the valve cover is shop work, not roadside work.

Panel I-4 — The Code on the Part

Calibration codes, the discard pile, and what the job costs
Swap facts
1 2 3 123456 STAMPED ON THE INJECTOR SIX DIGITS — ONE PER INJECTOR DIAGNOSTICLINK THE DISCARD PILE — ONE-TIME USE, NEW EVERY TIME BOLT COPPER SLEEVE SEALS: 3 O-RINGS + WASHER P-TUBE
1The six-digit calibration code — printed on the injector (the digits in the drawing are illustrative). It tells the MCM that injector's measured flow behavior, so fueling can be trimmed hole by hole. Photograph all six codes, matched to their positions, before the old injectors leave the bench — a code entered against the wrong cylinder is its own imbalance
2DiagnosticLink entry — not optional — the codes live in the MCM, and Detroit troubleshooting checks for “improper injector setting” by comparing the codes on the DiagnosticLink display against the codes on the injectors. Skip the entry and the truck still runs — rough, thirsty and unbalanced, on the OLD injector's correction. Detroit's word is “calibration code”; Cummins shops say “trim code” — the same step in the other dialect. A quote without the programming line is an unfinished quote
3The discard pile — hold-down bolt: Detroit's removal procedure says it in writing — “Discard bolt.” It is a stretch fastener, one-time use. Copper heat-protection sleeve: out and discarded; if it stayed in the bore, it gets extracted, never doubled up. Seals: the kit is 3 O-rings + 1 washer, new every time. P-tube: a one-time crush seal (expert-pinned) — budget one with every injector. Reused hardware is how a finished job becomes a no-prime or a fuel-in-oil comeback
NEVER prime by cranking. Detroit, in writing, in two operator manuals: “Never use the starting motor and fuel pump to prime the fuel system. Prolonged use of the starting motor and fuel pump to prime the fuel system can result in damage to the starter, fuel pump, and injectors.” The approved methods: an external priming pump at the filter-module port (DAVCO Shop Pro FXP / ESOC class), or the engine-mounted hand primer — 3 minutes, or 250 strokes. The prime is the last step of the repair, not a formality — a truck that “won't start after the injector job” is usually air, an unsealed filter, or a reused P-tube.

Gary's rule: a roundabout price beside the part is what makes a driver actually check his truck — it shows what neglect costs. Every figure below is a market figure, PARTIAL by nature, sourced and dated Oct 21, 2026. Where the math is shown, it is labeled arithmetic on those inputs — a heads-up, never a quote. DieselDoc's standing rule: price by ESN at the counter.

REMAN INJECTOR — DD15/DD16 · A4720700887$300–$600Per injector. Floor: $299 + $60 core on a marketplace listing. Anchor: $599.95 for a genuine Bosch reman from a named supplier (Dieselogic). Both listings captured Oct 21, 2026 — the marketplace floor carries marketplace risk.
SIX-INJECTOR JOB — PARTS + LABOR≈ $3,500–$7,500Labeled arithmetic on PARTIAL inputs: a reman set runs $2,500–$5,000 and set labor $1,000–$2,500 (Bostech Detroit cost guide). Cross-check: 6 × $599.95 ≈ $3,600 of parts — inside the set range. The job's book time is ≈ 6.7 hr quoted for a 2019 DD15 (credentialed expert). See which input your quote disagrees with.
CUPS — LABOR STRUCTURE9.0 hr first + 0.7 hr each moreExpert-quoted book times: the first cup books 9.0 hr and each additional cup 0.7 hr — 12.5 hr for all six, in-chassis with the head on. Cup parts are small; the money is the labor. No total cup-job price is printed — only the structure, so a quote can be checked against it.
SINGLE INJECTOR — INSTALLED TOTALNo honest number — get a quote firstThe part price is on the tile above, and the access is the same cover-off job whether one injector or six comes out — which is why single-injector quotes look disproportionate. No published DD-specific installed total exists, so none is printed.
NEW OEM INJECTORNo honest number — get a quote firstNo dated US counter listing for a brand-new Detroit/Bosch injector was retrieved. Refused rather than estimated — the reman market is the priced one.
DD13 INJECTORNo price — order by ESNThe DD13 number EA4600700687 is SINGLE-SOURCE in our library — Bosch's application guide, via Seidel Diesel, is the only published source that prints it. A4720700887 carries NO DD13 claim; the homework sheet that said it fits a DD13 was not supported.

While you're in there — each legitimate add-on is cheap because the cover is already off, and each should ride the quote as its own line: P-tubes, seal kits and hold-down bolts for every injector disturbed (the discard pile, above); a valve (overhead) adjustment if the truck is due (2.2 hr standalone book time, expert-quoted); the injector harness, if the codes pointed at it; and fresh verified fuel filters if the failure story involves fuel quality — new injectors behind the filters that killed the old ones is doing the job twice.

DIESEL DIAGNOSTIC AI

Driveline — From the Flange to the Tires

Everything power touches after the gearbox: the shaft, its joints and middle support, the pumpkin, the tandem's power divider, and the vibration sorter — from Dana Spicer, Meritor, and DTNA published material. Works offline.
Driveline — the road power takes home. The transmission hands the twist to a shaft at its output flange, and from there the power has exactly one road to the ground: shaft, joints, pumpkin, axle shafts, tires. The checks in this set are free and take minutes; the parts they protect run to thousands — that is Gary's rule, and it is why the prices ride beside the parts. One warning before the first panel: “carrier bearing” names two different parts — this set says CENTER BEARING (on the shaft) or DIFFERENTIAL CARRIER (the pumpkin), every time, and flags the trap wherever it can bite.

Panel D1 — The Power Path

One road from the flange to the tires — and what each stage does
Single + tandem
THE POWER PATH — SIDE VIEW, FRONT AT LEFT 1 2 3 4 5 GEARBOX CENTER BEARING 3.36 = PULL 2.64 = FUEL TANDEM — THE ROAD FORKS IN THE FRONT PUMPKIN 6 FRONT-REAR REAR-REAR THROUGH-SHAFT ONE ROAD: FLANGE → SHAFT → PUMPKIN → AXLES → TIRES
1Transmission output flange — the handoff. The gearbox hands the driveline a spinning shaft at engine-side speed and torque; the gearbox's own story is the Transmission set. This set starts here
2Driveshaft, U-joints, slip yoke — a hollow steel tube (a tube resists twisting better per pound than a solid bar) that delivers the twist across a distance, at an angle, while the axle moves under it. The U-joints do the bending; the slip yoke — splines sliding mid-shaft — does the length change. Worn splines add slack, part of the clunk on takeoff. Balance weights welded on the tube are factory: a missing one is a vibration cause $200–$500 one U-joint, installed · Roundabout — market prices, Oct 2026. Not a quote.
3CENTER BEARING (on the shaft) — on a longer wheelbase one long shaft would sag and whip, so the shaft is built in two pieces and the middle rests in a rubber-cushioned bearing bolted to a frame crossmember. It holds the middle up; it transmits no power itself $300–$750 center bearing, installed (part ~$100–$250) · Roundabout — market prices, Oct 2026. Not a quote.
4The pumpkin — DIFFERENTIAL CARRIER (the pumpkin) — the last gear reduction and the great splitter, and the last multiplication happens HERE: pinion torque leaves toward the wheels multiplied by about the axle ratio, while wheel speed drops by the same factor. The ratio in plain terms: 3.36 = PULL — the shaft turns 3.36 times per wheel turn, cruise ≈ 1,300–1,400 rpm. 2.64 = CRUISE / FUEL — fewer turns, cruise ≈ 1,100–1,200 rpm, less pull off the line. Gearbox and axle are one decision: effective ratio = axle × top gear — 3.36 × 0.74 OD = 2.41 effective
5Axle shafts + wheel hubs — full-float design: each shaft carries torque and NOTHING else; the truck's weight rides on the axle housing and the hub's two tapered bearings. The shaft unbolts at its flange and slides out without removing the wheel or disturbing the hub bearings — and a broken shaft does not let the wheel leave the truck. Two separate roads meet at the hub: weight on the housing, twist on the shaft
6Tandem — two driven pumpkins. Power enters the FRONT-REAR carrier first, where the power divider lives (Panel D4); a through-shaft carries it on to the REAR-REAR carrier. A single drive axle has one pumpkin and everything behind it just carries weight
⚠ NAME TRAP — “carrier bearing” is two different parts. On the shaft, it's the CENTER BEARING (on the shaft) — key 3 above, the rubber-cushioned support at the crossmember. Inside the pumpkin, the differential case rides on carrier bearings of its own — part of the DIFFERENTIAL CARRIER (the pumpkin), Panel D2. A driver who orders “a carrier bearing” can be handed either one. Say which one you mean, every time.
SLIP YOKE / SPLINE SECTIONNo honest number — get a quote firstNo published market figure for the slip yoke was gathered in our research, so none is printed. The check costs nothing: twist play at the yoke is part of the shake test (Panel D3), and worn splines announce themselves as takeoff clunk.
⚠ BEFORE ANY HANDS-ON CHECK IN THIS SET: Park on level ground. Engine OFF. Parking brake set. Wheels chocked. Hands on the shaft only on a parked, chocked, brake-set truck — keys out of reach.

Panel D2 — The Pumpkin, Opened

Differential carrier cutaway — the gears, the bearings, and the only gauge it has
Differential carrier
THE PUMPKIN, OPENED — FRONT VIEW (SCHEMATIC) 2 1 3 5 4 FLANGE FLANGE CASE RIDES ON CARRIER BEARINGS OIL LEVEL = BOTTOM OF FILL HOLE NO DIPSTICK. NO LOW-OIL LIGHT. THIS PLUG IS THE GAUGE — OIL AT THE BOTTOM OF THE HOLE, COLD
1Pinion + pinion seal — the small gear on the end of the shaft's road; it turns the ring gear and turns the power 90° toward the wheels. The seal where the pinion exits the snout is a differential-oil leak with no gauge behind it: gear-oil sling on the snout, the shaft's rear joint, and the underside above the pumpkin means report it. A new seal on a wobbling pinion leaks again — sometimes the seal is the victim, not the culprit ~$400–$550 pinion seal job, installed — general cost guides, light-duty-biased · Roundabout — market prices, Oct 2026. Not a quote.
2Ring gear + differential case — ring teeth ÷ pinion teeth IS the axle ratio (Panel D1). The case's spider and side gears let the left and right wheels turn at different speeds in a corner. The whine that follows the throttle lives here: gear-tooth contact shifts between the drive side and the coast side, so a worn set sings under pull and changes pitch — or goes quiet — on coast
3Fill plug — the only gauge — oil level sits at the bottom of the fill hole, checked with the axle cold on level ground (Meritor). There is no dipstick and no low-oil light; the driver's daily version of this check is the leak walk (Panel D3's companion checks): pinion snout, carrier face, each drive wheel's inside face. No sight glass on a drive hub — the transparent sight window belongs to steer and trailer oil-bath hubs. Drive wheel ends are fed from this carrier's oil, so their check is a leak check, not a level window
4Axle shafts + hub seals — full-float: the shaft bolts to the hub at this flange and carries torque only (Panel D1, key 5). The hub seal at the hub's inner face keeps this same oil in: a leak shows at the hubcap, the flange gasket, or the fill plug, and oil flung on the inside of the wheel and the brake is the giveaway — oil-soaked brake linings are a condemning defect, so the seal job becomes a seal + brakes job (Brake set). Caught on time it's ≈ 2.5 hr of standard labor — a few hundred dollars done on time
5DIFFERENTIAL CARRIER (the pumpkin) + its carrier bearings — the whole geared unit — pinion, ring gear, case, and the tapered carrier bearings the case rides on — unbolts from the front of the housing as one piece. That is why the shop fix for internal failure is a carrier swap, not an in-truck rebuild. This is also the expensive thing the $20 seals and the fill-plug check are protecting: low oil starves the gear mesh first, then — through the housing tubes — the wheel bearings $1,800–$2,400 reman carrier, part · Roundabout — market prices, Oct 2026. Not a quote. $2,500–$4,000 carrier swap, installed — labeled arithmetic: part + 4–6 hr labor + oil/seals · Roundabout — market prices, Oct 2026. Not a quote.
⚠ NAME TRAP — “carrier bearing” is two different parts. This panel's bearings are the DIFFERENTIAL CARRIER (the pumpkin) bearings — the tapered bearings the differential case rides on, inside the housing. They are NOT the CENTER BEARING (on the shaft) — the rubber-cushioned shaft support at the crossmember in Panel D1. Say which one you mean, every time.
RING & PINION SET — STANDALONENo honest number — get a quote firstNo clean published Class 8 standalone figure was gathered — gear sets are normally priced inside a rebuild or a carrier swap (the carrier chips above). Refused rather than estimated.
AXLE SHAFTNONE PRINTEDNo honest number — get a quote first. No published market figure for a Class 8 axle shaft was gathered in our research. The house rule is flag, never fill: the part exists, the honest price doesn't — yet.

The oil itself (Meritor MM 5A): full-synthetic 75W-90 (spec O-76-N) or 75W-140 (O-76-M), GL-5. Synthetic linehaul change interval is 500,000 miles on paper — which is exactly why the LEVEL matters more than the interval: oil that's in there for half a million miles has to stay in there. DTNA's maintenance notice says it in one sentence: “Failure to keep the rear axle filled to the proper level with the recommended lubricant can result in rear axle damage.”

⚠ BEFORE ANY HANDS-ON CHECK IN THIS SET: Park on level ground. Engine OFF. Parking brake set. Wheels chocked. Hands on the shaft only on a parked, chocked, brake-set truck — keys out of reach.

Panel D3 — U-Joint: Good vs Bad

The cheapest part on the power path — and the one whose neglect costs the most
Spicer signs
U-JOINT — GOOD vs BAD 1 2 3 4 5 HEALTHY PURGE TO ALL FOUR SEALS WORN PLAY ✗ SEAL WON'T PURGE THE SHAKE TEST — HANDS ON THE SHAFT
1The healthy joint — a forged cross; on each arm a bearing cap full of needle rollers, sitting in the yoke ears. Seals intact, grease fitting on the cross or a cap (some modern joints are sealed-for-life with no fitting — that's a design fact, not a defect). The service rule, from Spicer: grease with the specified EP grease and purge until fresh grease shows at all FOUR seals — purging pushes old grease and contamination OUT ~$50–$100 the part, typical heavy joint · Roundabout — market prices, Oct 2026. Not a quote. $200–$500 one joint, installed · Roundabout — market prices, Oct 2026. Not a quote.
2Play at the caps — Spicer's deterioration sign: looseness, end play across the bearings. The shake test (key 5) finds it — and do the shake BEFORE greasing, never after: fresh grease masks the very play you're inspecting for
3Rust dust + heat color — rust-colored dust weeping from a cap seal means the seal failed: grease out, water in. A cap discolored blue/brown ran dry and cooked. Spicer's cause statement deserves quoting: “Experience with universal joint failures has shown that a significant majority are related to lubricating film breakdown” — lack of lubricant, wrong lubricant, or failure to lubricate properly and often enough
4The seal that won't purge — if one cap won't take grease to its seal, work the shaft side to side under pressure and relieve the seal per Spicer's procedure. A joint that STILL won't purge all four is a failed joint: replace it. From the driver's seat the same joint reports as a squeak or rhythmic click that tracks road speed (once per shaft revolution), a clunk on takeoff and throttle reversal, or a vibration that grows with speed
5The grab-and-shake test — at each joint, push up-down and side-to-side, then twist back and forth. Healthy is solid: no felt play across the caps, no clunk. Play = the joint reporting itself; note WHICH joint (front / middle / rear) for the shop. Same hands, same stop: twist the slip yoke (Panel D1) and push the shaft up at the CENTER BEARING (on the shaft) — a little rubber give is normal; the shaft wallowing in torn rubber is a done support
RUNG 1 — GREASE GUN, ON SCHEDULE. The purge check costs grease and minutes. This is the rung the whole set exists to sell.
↓
RUNG 2 — ONE JOINT, CAUGHT BY THE SHAKE TEST: ~$200–$500 installed. The second joint on the same shaft costs less — the shaft-out labor is already paid.
↓
RUNG 3 — JOINT IGNORED TO FAILURE: the shaft separates at speed and can disable the truck — the bill becomes the joint PLUS whatever the dropped, flailing shaft struck (yokes, shaft, in the worst published cases the transmission output or the pinion), PLUS a tow, PLUS downtime. No single total is published for that event, so none is printed — the ladder is the lesson.

Spicer on the failure shape: U-joint failures are “of a progressive nature, which, when they occur, generally accelerate rapidly resulting in a mass of melted trunnions and bearings.” Slow, then sudden. The signs are early and cheap; the ending is fast and expensive.

⚠ BEFORE ANY HANDS-ON CHECK IN THIS SET: Park on level ground. Engine OFF. Parking brake set. Wheels chocked. Never work under a truck held up by air suspension alone.

Panel D4 — Tandem & the Power Divider

Locked vs unlocked — two different stories, and the tire ruler that governs both
DTNA driver's manual
TANDEM, TOP-DOWN — THE POWER DIVIDER 1 4 FROM GEARBOX DIFFERENT SIZES FRONT-REAR CARRIER THROUGH-SHAFT REAR-REAR CARRIER 2 3 5 UNLOCKED AXLES FREE TO DIFFER LOCKED DASH LIGHT LIT 6 MATCH TIRES: 1/8″ RADIUS · 3/4″ AROUND
1Power divider + inter-axle differential — housed in the FRONT-REAR carrier: a small differential, same idea as the one between the wheels, except it sits between the two axles. Power enters here first and passes through to the rear-rear. Its repair is carrier work $1,800–$2,400 reman carrier, part · Roundabout — market prices, Oct 2026. Not a quote.
2UNLOCKED — the normal state — DTNA's driver's manual, verbatim: “When the interaxle lock is not engaged, there is differential action between the two axles. The differential compensates for different wheel speeds and variations in tire size. Keep the interaxle lock disengaged when driving on roads where traction is good.” Unlocked, the divider absorbs turns, bumps, and small tire differences — that absorption is its job
3LOCKED — what the switch does — DTNA: “When the interaxle lock is engaged, the differential action between the two axles is locked. Both drive axles now share the power. Both axles, and both sets of wheels, turn together at the same speed.” Power entering the forward axle passes straight through to the rearmost. A red warning light stays lit on the dash the whole time it is engaged — it is a warning, not a badge: the truck is in an abnormal, wear-increasing state on purpose, for traction
4LOCKED + DIFFERENT-SIZE TIRES — the catastrophic half — keep this story separate from key 5; the condition is the tire mismatch. DTNA's general-information notice, verbatim-grade: when the lock is engaged on a tandem with tires of varying diameters, “torque builds up between the drive axles, which prevents the IAD lock from disengaging and leads to a severe increase in axle operating temperature that can cause premature oil failure and result in gear, bearing, or seal damage… Using the interaxle differential lock on a tandem axle with different size tires can cause catastrophic damage to the axle and axle components.” The axles fight in a straight line, the fight never ends, the lock may not release, the oil cooks
5LOCKED ON DRY PAVEMENT — the wear half — a different story from key 4, with a different condition: tires matched, traction good, lock still on. In every turn the two axles travel different path lengths and the chained driveline stores the difference as windup, scrubbed off through the tires. DTNA: the lock “increases drivetrain and tire wear and should be used only when improved traction is required.” The damage is real — but it is wear over miles, NOT the instant break of the mismatched-tire case. Do not merge the two halves
6The tire-match ruler — even unlocked, mismatched tires make the differentials work constantly to absorb a difference that never goes away, and their oil pays for it. Meritor's Parts Failure Analysis Guide: match drive tires to within 1/8″ of the same rolling radius and 3/4″ of the same rolling circumference, and keep the tandem's front and rear axle ratios within ±1% of each other — or “the carrier can overheat, lubricant additives deplete and axle components wear prematurely”
⚠ NAME TRAP — “carrier bearing” is two different parts. The power divider lives in the front-rear DIFFERENTIAL CARRIER (the pumpkin) — its home is this panel. It has nothing to do with the CENTER BEARING (on the shaft), the shaft's middle support in Panel D1. Say which one you mean, every time.
The other lock — axle differential lock (DCDL). A separate switch with its own dash light; do not confuse the two. Where the interaxle lock chains the two AXLES together, the DCDL chains the two WHEELS on one axle together, so a wheel on ice can't take all the power while its partner sits still. It increases turning radius (the truck wants to push straight), it is not for normal-traction driving, and it must never be engaged while a wheel is obviously spinning.
POWER DIVIDER REBUILDNo honest number — get a quote firstNo published market figure for a standalone power-divider rebuild was gathered in our research. In practice it is priced as carrier work — the carrier chip in key 1 is the honest number this panel can print.

Engagement discipline (DTNA Notice): “The interaxle lock should not be engaged on a vehicle with obviously spinning wheels. Engagement at high speed or power can damage the axle(s).” The lock is for poor traction, engaged gently — not a recovery tool for a wheel already spinning.

⚠ BEFORE ANY HANDS-ON CHECK IN THIS SET: Park on level ground. Engine OFF. Parking brake set. Wheels chocked. Hands on the shaft only on a parked, chocked, brake-set truck — keys out of reach.

Panel D5 — The Vibration Sorter

Three questions from the driver's seat sort nearly every driveline complaint
Check first — free
THE SORTER — THREE QUESTIONS 1 FOLLOWS ROAD SPEED? SAME ROAD SPEED, DIFFERENT GEAR STAYS → BEHIND THE GEARBOX MOVES → LOOK FORWARD 2 THE THROTTLE TEST PULL, THEN COAST — DOES IT CHANGE? CHANGES → DRIVELINE IDENTICAL → TIRES FIRST 3 WHEN DID IT START? AFTER WHAT WORK? RIDE-HEIGHT WORK → ANGLES ONE-AXLE SWAP → TIRE MATCH 4 EQUAL WITHIN 1° >3° EATS JOINTS MEASURE AT RIDE HEIGHT — A DROOPED TRUCK LIES
1Road speed or engine speed? — at the speed where the vibration lives, change gear and hold the same road speed. If it stays with the road, it's behind the transmission — driveline or wheels, this set. If it moves with engine rpm, look forward, not here
2The throttle test — the sorter — press the throttle (pull), then lift (coast). Changes, worsens under pull, or vanishes on coast → torque-sensitive → DRIVELINE SUSPECT: U-joint working angles, a worn gear set, shaft slack. Identical on pull and coast → speed-only → tires, wheels, and balance first. Gear noise is load-side/coast-side dependent (Panel D2), and a binding joint's shudder is worse under heavy throttle — the test works because torque is the variable
3When did it start? — right after suspension or ride-height work (air leak, height-control valve off, sagging springs, torque-rod work) → suspect angles before parts. Right after a tire swap on ONE axle of a tandem → check tire match first (Panel D4's ruler). A vibration with a birthday is telling you its cause
4Angles — equal and small (Spicer) — a U-joint working at an angle speeds up and slows down twice per revolution; the second joint, at an EQUAL angle, cancels the first. So the rules: the two angles on a shaft must be small and equal within 1°; any angle over 3° reduces joint life and may cause vibration; and angles are measured at ride height, loaded and unloaded. Ride height IS an angle setting — and no amount of balancing fixes an angle vibration, because the shaft isn't unbalanced
The noisePoints atFirst check
Clunk on takeoff / throttle reversalShaft slack — U-joints, slip splinesThe shake test (Panel D3)
Whine that follows the throttleRing & pinionPull/coast comparison (Panel D2)
Drone / hum from mid-truckCENTER BEARING (on the shaft)Push-up test at the support (Panel D3)
Squeak once per shaft revolutionDry U-joint capPurge test at the grease gun (Panel D3)
Growl at one cornerThat wheel endLeak walk at the hub (Panel D2)

Every check on this card is free — the parts they protect are not. The hands-on checks this card points to (Panel D3's shake test, Panel D2's leak walk) happen parked: level ground, engine OFF, parking brake set, wheels chocked. $200–$500 one U-joint, installed · Roundabout — market prices, Oct 2026. Not a quote. $2,500–$4,000 carrier swap, installed — labeled arithmetic · Roundabout — market prices, Oct 2026. Not a quote.

DIESEL DIAGNOSTIC AI

Batteries & Charging — Dead Battery or Bad Alternator?

From DTNA/Freightliner Powernet (Service Bulletin 54-278), Delco Remy, Prestolite Leece-Neville, East Penn/Deka, EnerSys, and John Deere published material. Prices PARTIAL market, dated Oct 21, 2026. Works offline.
Batteries vs alternator — the argument this set settles. Drivers sit between the steering wheel and the seat and can't tell a dead battery from a dying alternator — so they buy the wrong part. The split is one sentence: the batteries START the truck; the alternator RUNS it and refills them. Two numbers — resting voltage with the engine off, and voltage at the batteries with it running — decide most arguments. This set teaches both, in the order the factory troubleshooting flows use them: batteries first, every time.

Panel ELEC-1 — Two Jobs, One System

The batteries start it — the alternator runs it and refills them
Two jobs
TWO JOBS, ONE SYSTEM 1 3 BATTERY BANK GROUP 31 × 3–4, IN PARALLEL STARTER CRANK BURST 2 4 ALTERNATOR RUNS THE TRUCK + REFILLS THE BANK WON'T CRANK = BATTERY / CABLE STORY DIES RUNNING = ALTERNATOR STORY
1The batteries START the truck — cranking a big diesel is the single hardest electrical job on the truck: a short, enormous burst to the starter, plus the computers waking up. Between starts the bank also buffers load spikes and carries the key-off memory loads. That is the whole job description — storage, not generation
2The alternator RUNS the truck — once the engine is started, the alternator carries the truck's electrical loads and replaces what cranking took out of the bank. Its report card is not how the truck cranks — it's the running voltage at the battery pack (Panel ELEC-2)
3Won't crank in the morning = battery / cable story — slow crank, clicking, or silence on the key points at the bank or its connections first. One honesty note from DTNA: below 32 °F, slow turnover is NORMAL — cold alone doesn't condemn a battery or a starter
4Dies after the jump, or dies going down the road = alternator story — if the engine dies when the booster comes off, nothing on the truck was carrying the load or refilling the bank: it was running on borrowed power. Same verdict for a truck that dies running with the voltmeter falling
Why the confusion is expensive: a dead alternator makes GOOD batteries test “dead” (they're just empty), and weak batteries make a GOOD alternator look bad (it can't lift a pack with a shorted cell to full voltage). So every factory flow starts at the batteries — Delco Remy: “all troubleshooting begins at the batteries” — and refuses to condemn the alternator until the batteries pass.

The truck files the same two reports as codes: SPN 168 = Battery Voltage (the bank's story) and SPN 167 = Charging System Voltage (the alternator's story). Read which one is complaining before buying anything.

Panel ELEC-2 — The Voltmeter Panel

Resting scales by chemistry — and the running band. Bands with sources, never one magic number
DTNA + East Penn
RESTING — ENGINE OFF, SURFACE CHARGE REMOVED 1 FLOODED DTNA + EAST PENN — THE TWO TABLES AGREE 11.8 12.0 12.1 12.2 12.3 12.4 12.6 0% ~20% ~40% HALF ~60% ~80% FULL AGM A SEPARATE SCALE — READS ≈ +0.2 V HIGHER 2 11.8 12.0 12.2 12.4 12.6 12.8 0% ~20% ~40% ~60% ~80% FULL RUNNING — AT THE BATTERIES, ENGINE ON 3 NOT CHARGING 13.2–14.8 V >14.8 ≤ RESTING = NOT CHARGING — JUST DRAINING OVER = SENSE FIRST 4 COLD MORNINGS READ HIGHER — REGULATORS COMPENSATE FOR TEMPERATURE SHOW BANDS WITH SOURCES — NEVER ONE MAGIC NUMBER
1Resting, FLOODED — 12.6 V = full, 12.2 = only half charged, 12.0 = effectively empty, 11.8 = zero. A “12-volt battery” sitting at 12.2 is a half-empty tank. Read it engine-off, after the surface charge is removed (Panel ELEC-4, frame 4). DTNA's and East Penn's tables agree within a tenth everywhere
2Resting, AGM — a DIFFERENT scale, printed separately on purpose — AGM sits about 0.2 V higher: 12.8 = full, and premium AGM (ODYSSEY) reads 12.84 or higher at full. Never judge an AGM by the flooded scale or a flooded battery by the AGM scale — chemistry moves the scale, so the panel prints both and merges neither
3Running, at the battery pack — DTNA's accept window is 13.2–14.8 V (idle, part throttle, wide-open, no load). Running voltage must be clearly HIGHER than resting: a truck “running” at 12.4 V is not charging — it's draining its bank with the engine on. The load proof (DTNA Check 10): at 70% of the alternator's rated amps, pack voltage may sag no more than 0.2 V
4Above 14.8 V = overcharging — check remote sense FIRST, then the alternator (DTNA). On a Delco Remy remote-sense truck, expect 14.0–14.4 V at the batteries — and 14.7–14.9 V at the alternator's own posts is NORMAL: it deliberately runs hotter to overcome cable resistance. Sustained overcharge cooks the bank with the gauge reading “good”

The test gate: below about 80% charge (12.4 flooded / 12.6 AGM), DTNA wants the pack charged to full before any further verdicts — a half-empty bank invalidates every later test. And recharge anything sitting below 75% (East Penn): sitting discharged is what sulfates and kills batteries.

Panel ELEC-3 — Which One Is Lying?

The seat-level signs, side by side — and the third suspect nobody prices first
From the seat
WHICH ONE IS LYING? 1 2 3 BATTERIES ALTERNATOR STARTER + − SLOW CRANK CLICK-CLICK-CLICK DEAD BY MORNING DIES OFF THE JUMP VOLTMETER SAGS LAMP LIT AT IDLE ONE SOLID CLICK LIGHTS STAY BRIGHT NUMBERS GOOD THE TWO NUMBERS SETTLE IT — PANEL ELEC-2
1The battery story — slow, labored crank; click-click-click and no crank (voltage collapsing under load — weak bank or a connection doing the same thing); fine yesterday, dead this morning; and the mirror tell: it jump-starts and RUNS fine, but won't restart after the next shutdown — the alternator carried the truck, the bank can't store a start
2The alternator story — dies when the jump comes off; dash voltmeter sagging as lights and blower come on; charging lamp on or flickering at idle; lights dimming or pulsing, blower wandering with rpm; good batteries dying young (new batteries, dead again in weeks — they test GOOD); and the overcharge tells: batteries always needing water, a rotten-egg smell, hot or bulged cases
3…or the STARTER — the third suspect — one single solid click (or silence), dash lights staying BRIGHT, and both numbers good (rest ≈ 12.6, running in the window). DTNA's order agrees: the starter is only troubleshot AFTER batteries, path, and alternator pass the worksheet. Never let a shop sell the starter first
NEW ALTERNATOR — PART$220–$575 new Delco Remy HD, part only (dealer flyer) · Roundabout — market prices, Oct 2026. Not a quote.
REMAN ALTERNATOR — PART$200–$600 heavy-duty reman, part only · Roundabout — market prices, Oct 2026. Not a quote.
ALTERNATOR — INSTALLED~$500–$1,200 part + labor, installed total · Roundabout — market prices, Oct 2026. Not a quote.
DRIVE BELT — PART$50–$130 serpentine drive belt, part only · Roundabout — market prices, Oct 2026. Not a quote.
CABLES & TERMINALSNo honest number — get a quote firstNo Class 8 price was pinned in our research — and cleaning them costs nothing but time (Panel ELEC-4, frame 2).

Before either part: the belt and the cables. A glazed belt, a dying tensioner, or a failed pulley wears an alternator costume — DTNA checks belt, tensioner, mounting, and pulley BEFORE the alternator is condemned. And if the alternator's posts show good voltage but the batteries don't, the loss is in the cables: more than 0.5 V lost between alternator and batteries at full output is a cable fault, not an alternator (Delco Remy).

Panel ELEC-4 — The Decision

Six frames, in the factory order — each one free or cheap, each one with its verdict
In order
THE DECISION — SIX FRAMES, IN ORDER 1 LISTEN TO THE CRANK NORMAL CRANK = LOOK ELSEWHERE 2 CLEAN & TIGHTEN THE FREE FIX 3 THE JUMP TEST DIES = ALTERNATOR SIDE 4 TWO NUMBERS 13.2–14.8 V = CHARGING 5 BREAK THE PACK NEGATIVE OFF FIRST 6 NAME THE STARTER THIRD SUSPECT EVERY FRAME BEFORE THE PARTS COUNTER
1Listen to the crank — slow and labored → weak bank or bad connection (below 32 °F, slow is normal — don't condemn). Rapid clicking → voltage collapsing under load: batteries or connections. Normal-speed crank that won't fire → STOP: not a battery/alternator complaint — the bank just proved it can do the hardest job on the truck
2Clean & tighten before you condemn anything — wire-brush the terminal pads; tester clips and clamps go on the LEAD BASE of the post, not the threaded stud (DTNA). Retest. A visible share of “dead batteries” die right here, free
3The jump test — the discriminator — jump it by Panel ELEC-6, then: dies when the booster comes off → ALTERNATOR side (it contributed nothing; the truck ran on borrowed power). Stays running but won't restart → BATTERY side (the bank can't store a start). Runs AND restarts → the bank was simply drained — hunt the drain before buying anything: total key-off draw over 1.0 A is a fault to find (DTNA Check 7)
4Two numbers with a multimeter — the decider — REST: engine off, surface charge off first (DTNA's method: blower + lights on for 5 minutes), then read the pack against Panel ELEC-2's scales. RUNNING: same posts, engine on — in the 13.2–14.8 window = charging; at or below resting = not charging; above 14.8 = overcharge, remote sense first. Then load it — high beams + blower on high: a folding alternator lets the number sag toward resting
5Break the pack — if the pack rests low, won't charge to full, or keeps dying as a set: key off, remove the NEGATIVE cables first, isolate each battery, test each one alone. One battery reading ~2 V below its brothers = a shorted cell — the bank-killer the pack average was hiding (Panel ELEC-5)
6Name the starter — third suspect, never first — resting voltage good (≈12.6), connections clean and tight, charging in the window, and the symptom is a single solid click with lights that don't dim → starter, solenoid, or its control circuit. DTNA troubleshoots it only after batteries, path, and alternator pass
✗ DO NOT — the old cable-pull test. Do NOT test an alternator by pulling a cable off a running engine. No OEM document sanctions it, and on a computer-controlled truck the disconnected battery removes the system's electrical cushion while the alternator is live. Frame 4's voltmeter gives the same answer with no risk.

Panel ELEC-5 — The Bank

Four batteries in parallel share one voltage — and one bad one spends it for all of them
Group 31 ×4
THE BANK — ONE RAIL, ONE VOLTAGE 1 ONE SHARED VOLTAGE RAIL 2 + + ✗ + BATT 1 BATT 2 BATT 3 BATT 4 SHORTED CELL 4 EACH BATTERY 61–78 LB — BANK TOTAL 240–310 LB 3 BREAK THE PACK — TEST EACH ONE ALONE THE PACK AVERAGE HIDES THE BAD ONE
1One pack, one voltage — three or four Group 31s in parallel: all positives tied together, all negatives tied together. Still a 12-volt system — but capacity and cranking amps ADD UP, and every battery sits at the SAME voltage at all times. There is no “the good one is at 12.6 and the bad one is at 10” inside a connected pack
2One bad battery drags the whole bank — a shorted cell pulls the shared voltage down about 2 V and steals charge current from its partners around the clock. The pack “won't take a charge” and keeps going dead after full charges. DTNA's first suspect when a pack won't reach 100%: a shorted cell in one battery
3Break it and test each one alone — the pack average hides the bad one; isolated, each battery tells the truth: resting voltage against Panel ELEC-2's scales, and the shop load test — half the CCA rating for 15 seconds, must hold ≥9.6 V at 70 °F (East Penn). Hydrometer reads FLOODED only — never over an AGM: a sealed battery can't be opened, so AGM gets voltage and conductance instead
4Heavy, and honest about it — each Group 31 runs about 61–78 lb; the bank is 240–310 lb of lead and acid. Lift with a strap and your legs, keep it level and upright, and treat a cracked case as an acid spill, not a parts swap (Interstate published weights)
⚠ “REPLACE ALL FOUR” IS SHOP FOLKLORE, NOT AN OEM RULE. No manufacturer source in our research orders replacing the whole set for one failed battery. The published procedure is: test each battery individually, replace what fails — promptly, because a dead one left in the pack overworks the survivors — and match type and age as closely as you can.
GROUP 31 FLOODED — EACH$150–$250 one flooded Group 31 · Roundabout — market prices, Oct 2026. Not a quote.
THE BANK — ×4 FLOODED$600–$1,000 four flooded Group 31s — labeled arithmetic · Roundabout — market prices, Oct 2026. Not a quote.
GROUP 31 AGM — EACH$230–$450 one AGM Group 31 · Roundabout — market prices, Oct 2026. Not a quote.
THE BANK — ×4 AGM~$920–$1,800 four AGM Group 31s — labeled arithmetic · Roundabout — market prices, Oct 2026. Not a quote.

The neglect arithmetic: replacing the one failed battery at $150–$250 flooded, each · Roundabout — market prices, Oct 2026. Not a quote. when the pack-break test finds it is the cheap version. The expensive version is the whole bank after the good ones have been dragged down with it — and a weak alternator quietly sulfating a bank a few months at a time (Panel ELEC-3) is how drivers buy batteries twice before anyone measures running voltage.

Panel ELEC-6 — Jump It Right

The published cable order — and the battery rules that keep a jump from becoming an injury
John Deere order
HYDROGEN — NO SMOKING · NO FLAME · NO SPARKS DONOR (BOOSTER) DEAD (DISABLED) + − + − 1 2 3 LAST CLAMP: ON THE BLOCK, AWAY FROM THE BATTERY ENGINE BLOCK GROUND POINT 4 FROZEN? BULGED CASE = DO NOT JUMP THE LAST CLAMP LANDS ON THE BLOCK — THAT'S WHERE THE SPARK IS DISCONNECT IN EXACT REVERSE ORDER
1Positive to the DONOR positive — the positive (red) booster cable goes onto the booster battery's + post first (John Deere's published order)
2Positive to the DEAD positive — the other end of the positive cable onto the dead battery's + post
3Negative to the DONOR negative — the negative (black) cable onto the booster battery's − post
4Negative to the ENGINE BLOCK — away from the dead battery — the last clamp lands on solid metal of the dead truck's engine, NOT on its negative post. The last connection is where the spark happens, and the battery is where the hydrogen is. Start the truck, let it run a few minutes, then disconnect in exact REVERSE order
⚠ NEVER JUMP OR CHARGE A FROZEN BATTERY. A discharged battery's electrolyte is mostly water and can freeze — a bulged or cracked case is the tell. Warm it to 60 °F first (John Deere). A frozen battery that takes a jump or charge can rupture or explode.
EYE PROTECTION, EVERY TIME. Splash-proof Z87 goggles and gloves for booster work — battery work causes nearly 6,000 serious eye injuries a year (Prevent Blindness). Never lean over a battery while connecting cables or cranking.

Ground cable first, on and off: for any battery service, remove the grounded (NEGATIVE) cable first and reinstall it last (DTNA repeats it on the pack-break step). A jump pack follows the same order — it's a battery too — and confirm both systems are 12 V before connecting. After a successful jump, give the bank real run time or a charger: repeated short runs after jumps are how a drained bank becomes a replaced bank.

DIESEL DIAGNOSTIC AI

Pedals — The Two Surprises Under Your Feet

From DTNA Cascadia driver's and maintenance manuals, Schaeffler/LuK service information, SAE J1939, Detroit and Cummins fault-code lists, and Navistar bulletin IK0800219. Prices PARTIAL market, dated Oct 21, 2026. Works offline.
Two pedals, two messengers. On a modern truck your clutch foot talks to the clutch through fluid, not linkage — so when the pedal dies, what died is the actuation system, and the clutch itself is usually innocent. Your throttle foot talks through a sensor, and that sensor wears out where your foot lives. Gary's framing for the whole set: “things that's going to happen that guys will never see coming.” This set makes sure you've seen them — before a shop sells you the wrong repair.

Panel PF-1 — The Clutch Pedal Went Dead

Your foot works a pump, not a lever — the hydraulic chain, cab to clutch
DTNA chain
THE HYDRAULIC CHAIN — NO CABLE, ONLY FLUID 1 PEDAL UNIT MAIN CYLINDER RESERVOIR MAX MIN 2 3 SECONDARY CYLINDER AT THE BELL HOUSING 4 RELEASE LEVER → RELEASE BEARING THE LAST MECHANICAL INCH 5 THE CLUTCH ITSELF IS DOWNSTREAM OF ALL OF IT
1The pedal unit + main cylinder — there is NO mechanical connection between your foot and the clutch on a modern hydraulic truck, only fluid. Pedal in → the main cylinder's piston pushes fluid out. The cylinder + reservoir form a hydraulic subassembly that can be removed and serviced without replacing the whole pedal unit (DTNA)
2The reservoir — your #1 diagnostic instrument in this set — MIN and MAX lines are printed on its side for exactly the moment in Panel PF-2; DTNA's procedures service it from under the hood. Fill only with new DOT 4 from a tightly sealed container, and never let the level fall below MIN: below MIN, air enters, “the hydraulic system will not operate correctly, and the clutch could be damaged” (DTNA)
3The hydraulic hose — the single pressure path from cab to bell housing, on quick-disconnect clamps. A leak anywhere along it shows the same way every time: fluid loss and a falling reservoir
4The secondary (slave) cylinder — the worker at the far end: fluid pressure moves its piston, which pushes the plunger rod and the clutch release lever. It carries the bleed valve — the only bleed point in the system
5Release lever → release bearing → clutch — the last mechanical inch. Everything upstream of this point is the ACTUATION system, with its own failures and its own (smaller) prices. The clutch set takes over from here
⚠ THE SIGNATURE — READ IT BEFORE ANYTHING ELSE. FULL reservoir + dead pedal + DRY underneath = the main cylinder is bypassing internally. Its piston seal lets pressurized fluid slip past and back to the reservoir — the fluid goes in a circle, the level stays full, nothing leaks, and pressure never builds. That is an ACTUATION failure. The clutch was never touched. The mirror image: a worn-out clutch announces itself through a WORKING pedal — slipping under load. A dead pedal is actuation's signature.
The other architecture — the concentric slave cylinder (CSC). On Volvo-family trucks there is no external secondary cylinder: the slave is a ring around the transmission input shaft, INSIDE the bell housing, combined with the release bearing. Same job, hidden inside — its leak hides in the housing, and changing it means transmission removal (Panel PF-4 prices that honestly).
DOT 4 — NOTHING ELSE, EVER. DTNA's warning is blunt: the wrong fluid “will damage the rubber parts of the system, causing loss of clutch function.” The whole system holds only about half a quart (0.5 L) — a small leak is a big fraction of it. The fluid is on a two-year replacement schedule: DOT 4 absorbs water with age and corrodes cylinders and swells seals from the inside. The fluid itself is a hazard DTNA prints in bold — it “can cause blindness if it gets in your eyes” — glasses on before the reservoir lid comes off. And the system is self-adjusting: there is no pedal adjustment to “take up” a hydraulic problem.
No pedal at all? On a Detroit DT12 or an Eaton automated truck there is NO clutch pedal and no clutch hydraulics for the driver — Detroit: “shifting and clutch actuation are computer controlled, and there is no clutch pedal needed to operate the vehicle.” When an AMT's clutch actuation fails you get fault codes, a dash warning, or a refusal to engage — the diagnosis starts on the scan tool (Transmission set), never by hunting for a pedal your truck doesn't have.

Panel PF-2 — Which Actuation Part?

Toe-lift, read the reservoir, follow the wet trail — then say the sentence
Free checks
PEDAL ON THE FLOOR — IN ORDER 1 LIFT IT WITH YOUR TOE — FLOPS BACK = ZERO PRESSURE 2 RESERVOIR FULL + NOTHING WET MAIN CYLINDER BYPASSING — FLUID IN A CIRCLE 3 RESERVOIR LOW OR FALLING WET TRAIL: PEDAL UNIT → HOSE → BELL HOUSING 4 LEVEL FINE, PEDAL SPONGY AIR GOT IN — THE CURE IS A BLEED, NOT PARTS
1Lift the pedal with your toe. Does it come up and stay up? Pump it slowly a few times. If it flops straight back down with no resistance at all, the system is building zero pressure — bypassed main cylinder, empty system, or a broken return path. If it feels spongy and springy instead, think air (key 4)
2FULL reservoir + dead pedal + DRY underneath = the main cylinder is bypassing internally. The fluid never left — it just goes in a circle, so the level reads normal and there is no leak anywhere. Confirm with the hold test: hold the pedal down 30–60 seconds and watch it creep. This is the single most valuable finding on the card: an actuation pump failure, not a clutch
3LOW or falling level = the fluid went somewhere — follow the wet trail. In order: the pedal unit / main cylinder (wet at the subassembly or down the firewall), the hydraulic hose (chafe, rub-through, a weeping fitting), and the bell housing (wet at the bottom, or around the secondary cylinder and its bleed valve = secondary seal). On a CSC truck the leak hides inside the housing — dry outside + falling level + fluid at the bottom seam is the tell. Do NOT just refill and drive: the same hole will empty it again, and running below MIN pulls air in behind it
4Level fine, trail dry, pedal spongy = air in the system. Air compresses; fluid doesn't — the pedal stroke gets spent squeezing air instead of moving the release lever, and gears clash going in at a stop. Air got in when the level once ran low, or after hydraulic work. The published cure is a bleed — a two-person manual bleed, or a 15-psi pressure bleed at the secondary cylinder's bleed valve, until the fluid runs clear and free of bubbles — not a parts cannon
What failedThe reservoir saysWhat you find
Main cylinder internal bypassFULLNothing wet anywhere; the pedal sinks slowly on the hold test
Secondary cylinder seal leakLOW, fallingWet at the bottom of the bell housing, around the cylinder and bleed valve
Hose leak or ruptureLOW, fallingThe wet spot is along the line or at a fitting, not at either end
Collapsed hose (one-way valve)May look normalPedal stays down — and pops back up when the line fitting is cracked open and the trapped pressure escapes
Air in the systemNear normalSpongy, springy pedal; incomplete release; the cure is a bleed
Return spring / pedal pivot failedNormal, and dryPedal flops or stays wherever it lands; lift it with your toe and it may work once, then stays down again
Wrong or waterlogged fluidNormalNo single bad morning — the feel degrades over months until one of the rows above arrives “suddenly.” DOT 4 only, changed every two years

What is NOT in this table: the clutch disc, the pressure plate, and the release bearing. A worn-out clutch presents through a working pedal — slipping under load, chatter, engagement at the top of travel (Clutch set). The pedal going dead is the actuation system's signature, and it gets diagnosed here first.

Say it before anyone sells you anything:
“Show me which actuation part failed before we talk about the clutch.”

Panel PF-3 — The Throttle Pedal Lies

The pedal is a sensor — and its track wears where your foot lives
PARTIAL band
THE WORN SPOT — WHERE YOUR FOOT LIVES CRUISE ANGLE 1 BAND = PARTIAL TRADE FINDING 3 REST 0% FULL 100% SPN 91 — CH 1 SPN 29 — CH 2 2 WEAR PRINCIPLE: PINNED (DTE)
1The track wears where your foot lives. On a contact-type pedal a wiper slides along a resistive track with every press — a rubbing contact, and rubbing contacts wear where they spend their time. That principle is PINNED (DTE's published throttle-sensor diagnostics: the wear concentrates where the slider lives). A highway truck's foot lives in one narrow band — the cruise angle, roughly 40–70% — so that is where the track wears first: first a heartbeat hiccup at steady cruise, cruise control dropping out on a grade, then hiccups clustering at the same pedal position. The cruise-band application itself is PARTIAL — expert consensus, printed with its tag on the art, because no OEM document names the cruise band for a specific truck pedal
2Two channels, one pedal — and they must agree. SPN 91 = Accelerator Pedal Position 1, the primary channel the truck fuels on. SPN 29 = Accelerator Pedal Position 2 — the second channel of the same pedal under your one foot, NOT a second pedal; its job is to witness SPN 91. SPN 558 = the low-idle switch (idle validation): the “foot fully off” confirmation the computer cross-checks against a 0% reading. When the traces disagree at the worn band — the notch in the drawing — the computer stops trusting the pedal
3The codes name the wear — including the slow kind. Detroit's published SPN 91 titles include FMI 10 “Rationality Check Failed” (the channels disagree — the classic worn-track catch) and FMI 12 “Drifted Out of Learned Range” — gradual drift is a named, monitored failure, as close as a fault code gets to saying “this part is wearing out slowly.” The response ladder: a fleeting glitch may only log; sustained disagreement brings the lamp and a derate; a channel lost or implausible at idle can put the engine to idle-only / limp-home (Cummins publishes “limp home mode only” for its pedal-circuit fault; Navistar lists “engine speed restricted to idle”)
WIRING IMITATES WEAR — CHECK THE HARNESS BEFORE THE PEDAL. Navistar's APS/IVS bulletin (IK0800219) is explicit: intermittent pedal faults are usually wiring faults. A chafed harness at a sharp edge can drop one channel at one pedal position and look exactly like a worn track — and a new pedal swapped onto a chafed harness sets the same code by Friday. Unplug the pedal connector first: green corrosion, backed-out or fretted pins, water tracks; then follow the harness to its first rub point. And spray cleaner is not a fix: the sensor is a sealed assembly, cleaner cannot put resistive material back on a worn track, and the published repair for a confirmed worn pedal is replacement of the pedal assembly as a unit.
Two honest caveats. (1) Hall-effect pedals have NO wear track — contactless sensing, no wiper, nothing to rub out; their failures are electronic or connector-side and don't follow the cruise-angle script. (2) The free check: on a New Cascadia the pedal pivots at the floor, and an aftermarket floor mat can block the last of its travel — the truck never sees 100% throttle and the complaint is “low power.” Mat out, pedal to the stop, watch the pedal percentage on the scan tool: it must read 100%.

Panel PF-4 — The Money

Actuation money vs clutch-job money — quoted against the same dead pedal
Gary's price rule
ONE SYMPTOM, TWO QUOTES 1 THE ACTUATION FIX CYLINDER + FLUID + BLEED EXTERNAL PARTS TRANSMISSION STAYS IN VS THE CLUTCH JOB TRANSMISSION OUT CLUTCH PACK SWAP EVEN IF IT'S HEALTHY 2 SAME SYMPTOM: A PEDAL ON THE FLOOR THE FREE TEST PICKS THE QUOTE — READ THE RESERVOIR FIRST 3
1The actuation repair — a cylinder (hundreds), DOT 4 fluid (dollars), and bleed labor. External parts; the transmission stays in. The CSC is the honest exception: the part is ordinary money, but it lives inside the bell housing, so its labor is transmission-out labor (tiles below)
2The clutch job you almost bought — the quote a dead pedal attracts when nobody reads the reservoir: a transmission-out clutch pack, priced in the tiles below, for a truck whose clutch may be untouched by the failure. The clutch gets condemned by ITS tests — slip under load, engagement point (Clutch set) — never by a dead pedal
3Part B's version of the same trap — a worn throttle pedal is a few-hundred-dollar assembly plus the learn step; its symptoms (a fade at cruise, a derate) invite an engine-side parts cannon — turbo, fuel system — that costs ten times more and fixes nothing. In both halves of this set, the free test comes before the expensive guess

ACTUATION-SIDE PRICES — part prices unless the tile says otherwise:

CASCADIA CLUTCH CYLINDER ASSEMBLY~$1,370 the pedal unit's hydraulic side, Advance Auto retail · Roundabout — market prices, Oct 2026. Not a quote.The Cascadia pedal unit is proprietary — this is NOT a sixty-dollar part. For contrast, generic master cylinders run $50–$200 in light-vehicle catalogs · Roundabout — market prices, Oct 2026. Not a quote. — that's the floor, not this truck.
SECONDARY CYLINDER — EXTERNAL~$245 WABCO Freightliner-application unit, part only · Roundabout — market prices, Oct 2026. Not a quote.Bolted at the bell housing — no transmission removal to change it.
CONCENTRIC SLAVE CYLINDER (CSC)$225–$270 part only, LuK / FTE listings · Roundabout — market prices, Oct 2026. Not a quote.Ordinary part money — but it lives inside the bell housing, so the labor is transmission-out labor.
DT12 CLUTCH ACTUATOR~$999 Detroit A0032502115 “central slave cylinder,” part only · Roundabout — market prices, Oct 2026. Not a quote.The AMT counterpart: no pedal at all — the actuator IS the actuation system (Transmission set).
ACCELERATOR PEDAL ASSEMBLY$175–$425 HD pedal assembly with sensor · Roundabout — market prices, Oct 2026. Not a quote.Plus the learn step: the computer learns each pedal's endpoints (DiagnosticLink on Detroit — Learn Error, Not Learned, and Learned Range Too Large are all named faults), so a pedal swapped without the learn can set fresh codes on a good part. The learn is part of the job.
HYDRAULIC HOSE / LINENo honest number — get a quote firstNo US retail figure for the Class 8 clutch hose was pinned in our research — refused rather than filled.

THE QUOTE TO REFUSE — until the tests say so:

FULL CLUTCH JOB$2,000–$4,500 manual clutch job, all-in (Clutch set) · Roundabout — market prices, Oct 2026. Not a quote.What a dead pedal gets quoted when nobody reads the reservoir. It may still be the answer — after the actuation system passes its tests, not before.
DOT 4 FLUID + A BLEEDQUART-CLASS MONEYThe cheapest line in this set — and the two-year fluid change is the cheapest insurance in it. Most of Part A never happens to a truck whose fluid gets changed.
The part that failed is the messenger. Test the messenger first.
DIESEL DIAGNOSTIC AI

AI Diagnosis Brief

Paste the fault codes off the tablet. You get a structured brief per code: what it means in plain English, likely causes ranked, first checks, parts likely involved, and a keep-driving caution.
Guidance only — confirm against the tablet report.