There’s a clean way to think about a molding machine: it’s a cash generator. You buy material, mark it up, sell machine time and the expertise to convert it, and the whole thing turns into dollars per thousand parts. During a changeover, that generator produces exactly nothing.
Put real numbers on it. A $5,000 purchase order for 25,000 parts on a 4-cavity tool at a 20-second cycle works out to about $144 of sales per production hour. Think of the press as a taxi that earns $144 an hour whether it’s driving or sitting at the curb with the meter off. If your quote carried a $200 setup charge, you’ve budgeted roughly an hour and twenty minutes from shutting down the last job to making acceptable parts on the next one. Run past that, and the changeover is no longer covered — you’re losing money on a press that’s “busy.”
That’s the frame that makes setup reduction a business decision instead of a housekeeping chore. This article doesn’t reproduce historical WJT Associates material; it applies the same shop-floor logic, alongside the well-established SMED method, to the question every changeover really asks: how fast can we get from the last good part to the next good part?
Measure the right interval
Most plants measure setup wrong. They start the clock when the new mold is craned in and stop it when the operator walks away. The interval that actually costs money is wider than that:
Setup time is the gap between the last good part of the old job and the first good part of the new job.
Everything in that gap is non-earning time, including the parts most shops don’t count: pulling the old tool, staging the next one, finding the right water hoses, hunting for the setup sheet, drying the resin you forgot to start, and the qualifying shots it takes to dial the new job back into its window. Time the whole thing, honestly, on a few real changeovers. The number is almost always worse than the setup charge assumes — and the surprises are where the savings are.
The core idea: internal versus external setup
The SMED method (Single-Minute Exchange of Die, from Shigeo Shingo’s work in the auto industry) rests on one distinction that applies directly to molding. Every task in a changeover is one of two kinds:
- Internal setup — work that can only happen while the press is stopped and open: removing and mounting the tool, connecting it, the qualifying shots.
- External setup — work that can be done while the previous job is still running or while the next one is warming up: staging the mold, pre-drying resin, kitting hoses and clamps, pulling the setup sheet and the right inserts.
The entire game is converting internal work into external work, and shrinking what’s left of the internal work. Every minute of setup currently done with the press stopped that could have been done while it was still running is pure lost money.
| Internal setup (press stopped, no money) | External setup (press still earning) | |
|---|---|---|
| Definition | Can only be done with the machine open | Can be done in parallel, before or after |
| Examples | Mount tool, connect water/hydraulics, qualify the shot | Stage tool, pre-dry resin, kit hoses/clamps, pull setup sheet |
| Goal | Make it as short as possible | Move as much work here as possible |
| Cost if neglected | Direct lost production every changeover | Hidden — looks “free” but stalls the changeover |
Where the time actually goes
When you watch real changeovers instead of trusting the setup charge, the same time-wasters show up again and again. None of them are exotic. All of them are fixable without capital.
| Time-waster | Why it costs | Fix |
|---|---|---|
| Hunting for tooling, hoses, clamps, inserts | Internal time spent searching instead of mounting | Kit everything for the next job in advance — external setup |
| Resin not dried in time | Press sits waiting on a dryer | Start drying before the previous job ends |
| No setup sheet, or set points only | Technician rebuilds the process from memory every time | Enforce setup sheets with acceptable ranges, not just targets |
| Re-finding the process at the press | Qualifying shots drag on because there’s no documented window | Document the window once; qualify back to it, don’t rediscover it |
| Mismatched fittings and clamps | Improvised fixes mid-changeover | Standardize clamping and quick-disconnects across the press class |
| Crane and labor not staged | Tool mount waits on people and equipment | Schedule the changeover crew like a pit stop |
The pattern is consistent: most of what makes a changeover slow isn’t the physical tool swap. It’s the searching, the waiting, and the re-discovering — and almost all of it can be moved off the critical path or eliminated with documentation.
A practical reduction sequence
You don’t need a consultant or a kaizen event to start. You need a stopwatch and the willingness to watch your own changeover honestly.
- Time a real changeover, start to finish — last good part to first good part — and write down every step with its duration.
- Tag each step internal or external. Be strict: if it could be done while the press is running, it’s external even if you currently do it stopped.
- Move every external task off the critical path. Stage, dry, kit, and document ahead of time so the press is never waiting on preparation.
- Attack the remaining internal time. Quick-change clamping, standardized water connections, and a documented process window are the three biggest levers.
- Standardize the new sequence and train to it. A faster changeover that only one technician can do isn’t a real gain — write it down and make it the method.
The first pass usually cuts the worst changeovers substantially, because the biggest wins are the dumb ones: a dryer that should have been started an hour earlier, a tool that was three aisles away, a setup sheet nobody could find.
Don’t over-engineer it: the common-sense check
SMED, lean, and JIT are all worth doing, and all of them are ultimately about eliminating waste and protecting profit. But it’s easy to bury a simple changeover under a program when what it needed was common sense. Before you invest in quick-change hardware, ask whether the slow changeover is really a hardware problem or just a preparation and documentation problem — because the second one is free to fix.
It’s also worth remembering what the setup charge connects to upstream. The time a changeover takes, the cycle the job was quoted at, and the cavitation of the tool are all baked into the price before the first part is made. A changeover that runs long doesn’t just cost the floor time; it quietly eats the margin that quote assumed.
For broader, vendor-neutral coverage of changeover and processing efficiency in molding, the technical resources at Plastics Technology (ptonline.com) are a useful reference.
FAQs
What counts as setup time — when does the clock start and stop?
The clock starts at the last good part of the job you’re ending and stops at the first good part of the job you’re starting. That includes pulling and staging tools, connecting water and hydraulics, drying resin, finding the setup sheet, and the qualifying shots. Measuring only the physical tool swap dramatically understates the real cost, because most of the lost time hides in preparation and re-qualification, not in the mount itself.
What’s the difference between internal and external setup?
Internal setup is work that can only happen with the press stopped and open — mounting the tool, connecting it, running qualifying shots. External setup is work that can be done while the previous job is still running or the next one is warming up — staging the mold, pre-drying resin, kitting hoses and clamps, pulling the setup sheet. The core of SMED is converting internal work into external work so the press spends less time stopped.
How much setup reduction is realistic without buying equipment?
Often a lot, because the biggest early wins are organizational, not mechanical. Staging tooling, pre-drying resin, kitting everything for the next job, and enforcing a documented setup sheet move large chunks of work off the critical path for zero capital. Quick-change clamping and standardized connections come later, once you’ve eliminated the searching, waiting, and re-discovering that usually dominate a slow changeover.
Why does a documented process window speed up changeovers?
Because without one, the technician rediscovers the process at the press every time — running qualifying shots until the parts look right, working from memory. With a documented window and a setup sheet that lists acceptable ranges, the changeover qualifies back to a known target instead of searching for it. That turns the most variable, longest part of many changeovers — getting the new job into spec — into a short, repeatable step.