Salvage vs. Replace: Evaluating Damaged Production Tooling

A dinged cavity is not automatically a dead tool. A washed-out parting line often is.

Salvage vs replace is a diagnostic, not a feeling after a bad lot. Walk the damage in the order that decides LSR (or HCR) capability: shutoff, then heat, then actuation, then rust in channels, then cosmetics. The first fatal item wins. Everything after that is estimating.

Inspect in shop order

Do this before anyone quotes weld.

1. The land / shutoff.

Raking light. Finger, then a stone, then a dye if you have it. LSR flash lives here. A local ding on a non-sealing face can be stoned or EDM’d. A rolled land the length of the part, or brinelling from a crash close, is a recut or an insert. If the land is gone in a way that changes the part split, you are in replace-cavity territory.

2. Cavity/core faces.

Pits from rust, a wrench kiss, a washout from aggressive cleaner on a polish. Cosmetic LSR (transparent, medical white) is less forgiving than a hidden bumper. Pits print. Welding a polish face is usually a longer argument than an insert.

3. Heaters and thermocouples.

Continuity. Megger if the well flooded. A tool that cannot hold LSR cure temperature is not a mold this week. Heater replacement is salvage. A burned well that needs weld in a thin wall may be replace-insert.

4. Cold runner / valves.

Leaks, stuck valves, cracked manifolds. This is machine repair. Often salvage. If the manifold design was already the flash/waste problem, replacement of *that subsystem* can be the moment you upgrade, not a like-for-like.

5. Slides, pins, guidance.

Bent leader pin, scored gib, cracked slide. Replace the component. If the plate hole is oval, the plate may be the replacement.

6. Channels and rust inside.

Flush. If the tool sat wet, assume scale. Blocked cooling/heating is a process defect later. Salvage by clean/repair unless the plate is rotten.

7. Base vs inserts.

A crashed insert with a healthy base is the textbook salvage: cut a new insert. A cracked plate with perfect inserts is still a new plate. Count what is actually dead.

Photograph everything. If this is a freight claim, those photos are the claim.

When salvage is the adult call

• Spare insert exists; drop it in, re-spot, FAI the cavity.
• Localized ding off the land; EDM/weld/stone with an accepted cosmetic risk.
• Heaters, seals, valves, pins: wear parts. That is PM, billed as salvage.
• Aluminum bridge tool with a rolled shutoff: recut the land or add a steel shutoff insert. Cheaper than a new 103, if volume still fits aluminum life.
• One cavity of eight is dead: block it for an emergency lot, cut a replacement insert. Do not run 7/8 forever and call it the new process.

Salvage must include tryout and a dimensional/visual delta against the last good FAI. A pretty weld that shifts a seal bead is a new defect.

When replace is cheaper than optimism

• Shutoff is gone on a hardened, textured, or optical face and no insert pocket exists. Creating the pocket plus weld drama plus downtime can exceed a new cavity in the same base.
• Parting crash that moved plates. Guidance is the tool.
• Printed LSR insert heat-killed or crushed. Reprint or move to aluminum. Do not salvage polymer shutoffs.
• Class 105 / tired 104 that was already flashing before the incident. The damage is your excuse to stop funding scrap. Replace with the class you should have had.
• Unknown history overseas: rust in channels, undocumented welds, missing electrodes. Sometimes the CAD/electrode package is gone. You cannot salvage archaeology. Recut from a scanned part is a new tool with extra lying.
• Multi-failure: land + water in heaters + bent pins. The stack of salvage quotes will grow. Cap it. Set a rule: if salvage estimates cross a stated fraction of new-cavity cost, stop and replace.

Replacement does not always mean a new Class 103 8-cavity. It can mean a new insert set in the old base, or a 1-cavity emergency tool to keep the line alive while steel is cut.

Drivers, not a price list

Salvage cost moves with hardness (EDM vs mill), finish restore, whether an insert pocket must be created, tryout, and lost production. Replacement cost moves with class, cavitation, whether the base and cold runner reuse, and freight.

Downtime is usually the deciding term. A week of no parts on a frozen SKU can justify a new insert at rush. A dead tool for a seasonal part can wait on a rational recut.

A one-page decision

Write four lines:

1. Fatal feature? (land / heat / plates / unknown)
2. Insert possible without touching the land geometry?
3. Salvage estimate vs new insert vs new tool (same class, same cavities)
4. How the line runs in the meantime (block cavity, RTV emergency, aluminum bridge)

If (1) is land on a non-insertable cavity and (3) salvage is not clearly lower, replace the cavity.

What not to do: stone a land into a new flash spec and ship, weld a transparent A-side “just to try,” or keep running an 8-cavity with two pockets blocked without updating the FAI and the piece-price math.

Put on the maintenance PO: “Inspect land, heaters, slides, channels, insert vs base. Quote salvage vs new insert vs new cavity. No weld on sealing or optical faces without written risk.”

FAQ

When is a dinged LSR cavity still the adult salvage call?

When a spare insert exists, the ding is off the land, or the dead items are wear parts (heaters, seals, valves, pins). Aluminum with a rolled shutoff can take a recut or a steel shutoff insert if volume still fits aluminum life. Salvage must include tryout against the last good FAI.

Why inspect the land before anyone quotes weld?

LSR flash lives on the shutoff. A local ding off a sealing face can be stoned or EDM'd; a rolled or brinelled land the length of the part is a recut or insert. Welding a polish or optical face is usually a longer argument than an insert. Photograph everything if freight is involved.

When is replace cheaper than optimism?

Replace when the shutoff is gone on a hardened, textured, or optical face and no insert pocket exists. A parting crash that moved plates, heat-killed printed inserts, or a Class 105 that was already flashing are the same call. Unknown overseas history with missing electrodes, or stacked salvage quotes that cross a stated fraction of new-cavity cost, also beat optimism.

Can I run seven cavities of an eight-cavity tool forever after one pocket dies?

Block it for an emergency lot, then cut a replacement insert. Do not freeze 7/8 as the new process without updating FAI and piece-price math. Replacement is not always a new 8-cavity 103; it can be a new insert set in the old base or a 1-cavity emergency tool.

What four lines belong on the salvage page?

Fatal feature (land, heat, plates, unknown); whether an insert is possible without touching land geometry; salvage versus new insert versus new tool at the same class and cavities; and how the line runs in the meantime. No weld on sealing or optical faces without written risk.