A startup is the most dangerous five minutes a mold sees all day. A tool that took months to build and costs more than the press it runs in can be damaged in a single closing stroke — a stray part stuck on the core, a piece of flash on the parting line, an ejector that didn’t return. And the setup that follows is where a process either gets built on purpose or gets dialed in by feel and quietly carries problems into production.
Good setup is a sequence, not a scramble: protect the tool first, then control the variables the plastic actually sees, then prove the first parts are good before you run. Skip the order and you either crush a mold or ship setup scrap. This article lays out that sequence in shop-floor terms.
Mold protection: low pressure before anything else
The single most important habit in setup is low-pressure mold close. Before the machine is allowed to build full clamp tonnage, it closes the last stretch under a deliberately weak, position-limited force. If something is in the way — a part that didn’t eject, debris on the parting line, a sensor cable — the mold meets resistance it can’t overcome at low pressure and stops, instead of crushing the obstruction (and the steel around it) under hundreds of tons.
Three settings make mold protection actually work, and all three have to be right:
| Setting | What it does | Get it wrong and… |
|---|---|---|
| Low-pressure force | The clamp force during the protection window — set as low as will still reliably close the mold | Too high and it crushes debris instead of stopping; a common rule of thumb caps it around a third of maximum clamp force |
| Low-pressure position window | The stretch of travel over which protection is active, matched to the part’s size and height | Too short and the window closes before an obstruction can be caught; too long and every cycle wastes time crawling shut |
| High-pressure changeover point | Where the press hands off from protection to full clamp | Set it so a foreign object inside the window still trips protection — not so early that protection never really engaged |
The protection window has to suit the tool. A tall part with deep cores needs a longer low-pressure stretch than a flat part, because there’s more travel where something could be trapped. Setting the window by habit instead of by the tool in the press is how protection gets defeated on the molds that need it most. The whole point is that the cost of a slightly slower close is nothing next to the cost of a damaged cavity.
What each plastic-side parameter actually controls
Once the tool is mounted and protected, setup becomes a matter of controlling the variables the plastic experiences. The mistake beginners make is treating every knob as interchangeable. Each one has a primary job, and knowing which defect each one moves is what keeps you from chasing your tail.
| Parameter | What it primarily controls | Too low | Too high |
|---|---|---|---|
| Injection speed | How the melt fills the cavity — weld-line depth, gloss, venting behavior | Deep weld lines, dull fill | Burning/flash from air that can’t vent fast enough |
| Injection pressure | Completing the fill before transfer to hold | Short fill, deep weld lines | Over-packed, over-bright spots, flash |
| Hold (pack) pressure | Compensating shrink after transfer | Sink, voids, short weight | Flash, gloss/shadow, ejector blush |
| Inject / hold / cooling time | Filling, packing, and freezing the part | Sink, short fill, sticking, warp | Ejector blush, surface shadow, wasted cycle |
| Melt and mold temperature | Flow and replication | Poor flow, dull finish, short fill | Degradation, long cooling, flash |
| Cushion (screw front residue) | Keeping a pad so hold pressure transfers | Too small — lose hold pressure transfer | Too large — lose effective pressure; aim for a small, stable pad (a few millimeters) |
| Back pressure | Melt homogeneity, temperature, gas venting | Poor mixing, color streaks, trapped gas | Raises melt temp and slows plasticizing — recovery drags |
| Decompression (suck-back) | Preventing nozzle drool between shots | Drool, stringing | Draws air into the barrel — splay and voids |
Two of these deserve a note because they’re routinely abused. Cushion is not a setting you maximize — it’s a small, stable pad (commonly in the low single-digit millimeters) that has to stay consistent shot to shot; a wandering cushion is usually a worn check ring, not a setting to crank. And back pressure is genuinely useful for homogenizing melt and venting gas, but every increase steals plasticizing speed and adds heat, so it’s a balance, not a free lever.
First-article discipline: prove it before you run
The fastest way to contaminate a good production lot is to let setup parts mix into it. A clean setup ends with three non-negotiable steps:
- Match the first parts to the reference. The setter compares the first good shots to the golden sample and the documented process sheet, reusing the established conditions from prior trials or production rather than inventing a process from scratch. Setup should re-establish a known process, not rediscover one.
- Get the first article approved. An in-process quality check signs off the first article against the acceptance standard before the job is released to run. “Looks right to the setter” is not approval.
- Scrap the setup parts. The first shots while the process stabilizes — commonly the first couple dozen, plus everything made during adjustments — get scrapped, not boxed. Their job was to bring the process into the window, and they don’t belong in the shipment.
That last step feels wasteful to a setter under schedule pressure, which is exactly why it gets skipped — and why setup rejects end up in customer lots. A handful of deliberately scrapped setup shots is cheap insurance against a returned production run.
A practical setup sequence
- Mount and protect. Seat the tool, set low-pressure mold protection (force, window, changeover) to the tool in the press, and confirm it trips on an obstruction.
- Establish from the reference. Load the documented process sheet conditions and the golden sample; don’t start from zero.
- Bring the plastic-side variables into the window one at a time, knowing what each one controls.
- Confirm cushion is small and stable and recovery is consistent — a wandering cushion means stop and look at the check ring, not the settings.
- Run the first article, get QC sign-off, scrap the setup shots, then release to production.
The discipline isn’t glamorous, but it’s the difference between a setup that produces a stable, documented, repeatable process and one that produces a tool with a fresh ding and a lot full of mystery rejects.
FAQs
What is low-pressure mold protection and why does it matter?
It’s a setup feature that closes the mold’s final travel under a deliberately weak, position-limited clamp force before full tonnage is applied. If anything is in the way — an unejected part, parting-line debris, a stuck ejector — the mold stops instead of crushing it under hundreds of tons. It matters because a startup is when tools get damaged, and the cost of a slightly slower close is nothing compared to a cracked cavity or a bent core. The protection force, the position window, and the changeover point all have to be set to the specific tool to work.
How do I set the mold protection window correctly?
Match it to the part’s size and height. The low-pressure window should cover the stretch of travel where an obstruction could be trapped — a tall part with deep cores needs a longer window than a flat part. Set the force as low as will still reliably close the mold (a common rule of thumb is no more than about a third of maximum clamp), and set the high-pressure changeover so a foreign object inside the window still trips protection. Too short a window catches nothing; too long just wastes cycle time.
Why does cushion need to be small and stable instead of large?
Cushion is the small pad of melt left in front of the screw so that hold pressure can actually transfer to the cavity. If it’s too small, you lose that transfer; if it’s too large, you lose effective pressure. The target is a small, consistent pad — commonly a few millimeters — that stays the same shot to shot. A cushion that wanders from shot to shot usually isn’t a setting problem at all; it’s a worn non-return valve letting melt slip back, and the fix is maintenance, not a knob.
Why scrap the first shots after a setup?
Because the first shots are made while the process is still stabilizing into the window, and the parts made during adjustments aren’t representative of the running process. If they get boxed with production, setup scrap ends up in the customer’s lot — which can turn into a returned shipment far more expensive than the handful of parts you scrapped. Scrapping the setup shots, after matching the first article to the reference and getting QC sign-off, keeps setup variation out of production.