The Mold Design Review: What to Settle Before Steel Is Cut

Most buyers get two formal looks at a tooling program: the quote, and the first samples. Between those sits mold design, and it often passes without the buyer in the room.

That is understandable. Mold design is the supplier’s engineering. But a specific set of decisions in that window are free to change during it and expensive after, and several of them only you can make, because they depend on how the product is used, not how it is molded.

Settle those decisions in that window. Pair with T1 and T2 trials for the review that happens after.

The Approval Most Buyers Never Ask For

Molders building appearance parts to automotive interior standards commonly send the customer a marked-up drawing or 3D data showing where the parting line, gates, weld lines and ejector marks will fall, and wait for written approval before cutting steel. On some programs it is a formal gate.

Outside that segment it often does not happen unless you ask. Ask.

What you buy with an afternoon of review is the chance to say “not there” while “not there” is still a CAD operation. All four of those marks are inevitable somewhere, and all four are negotiable as to where, which is precisely the kind of decision that depends on knowing which face the customer looks at, information the mold designer does not have.

It also catches a class of problem that is invisible from any single part: mating parts whose witness lines do not line up. Two adjacent trim pieces, each with a perfectly acceptable parting line, meeting at a joint where one line sits a couple of millimetres above the other, reads as a defect on the assembly while neither part is out of spec. Nobody finds that unless someone looks at both drawings together.

If the supplier does not offer this review, asking tells you something either way. A shop with the answers ready has thought about your appearance surfaces. A shop that has to go and work it out has not, and better now than at T1.

What Has to Be Decided Before, Not During

Four inputs are yours to provide, and providing them late is what turns mold design into rework.

Which faces are cosmetic. Not “the front”, a marked-up view showing A-surfaces, secondary surfaces, and assembly-only faces, with the installed viewing direction. This single input constrains where the parting line runs, where gates can sit, and where ejectors can push. Without it, the mold designer infers it from the model, and the model does not know which way the product faces. See cosmetic surface classes.

The texture, if there is one. Eastman’s mold design guidelines are direct that the texture pattern should be decided early so the proper draft angles and contours can be incorporated, because the draft bill is large. Their conversion is 1° to 1.5° of extra draft per 0.001 in of texture depth, against typical texture depths of 0.0025–0.0030 in. That is 2.5°–4.5° added to faces that may have been drawn vertical. Settling the grain after the faces are drawn reopens the part, not just the tool.

Which dimensions actually matter. If everything on the drawing is critical, nothing is, and the supplier cannot tell where to concentrate control. A short list of genuinely critical dimensions with datums is worth more than a fully toleranced model.

The resin, at grade level. Not the family. Gate size scales with the resin, Eastman notes that tooling designed for one material may need a different gate size for another’s viscosity, and filled grades change shrinkage, wear, and achievable tolerances. A grade change after mold design is a redesign in disguise.

Questions Worth Asking During Design

These are not audit theater. Each one has a specific answer that tells you whether a decision was engineered or defaulted.

“Where will the weld lines land, and does any of them fall on a loaded feature?” Answerable from a fill analysis before steel exists. On a filled grade this is a strength question, published data puts weld-line tensile strength in 30% glass-filled PA6 at roughly 60% of the base material, with elongation under 5%. See moldflow for buyers.

“What mold base grade is quoted?” DME’s published plate tolerances put finish-ground plates at 0.0007 in/ft flatness and 0.0005 in parallelism, against 0.002 and 0.003 for rough ground, a six-fold difference in parallelism. Since the halves close across those surfaces, plate grade is part of what bounds the achievable parting-line step. It is also a cost line that is invisible unless you ask.

“How is the tool vented, and who optimises it?” Vent depth is set conservatively by moldmakers as a matter of course, because a mold delivered slightly over-vented flashes and needs welding to correct, so the final adjustment is expected to happen at the molder, and frequently never does. See mold venting.

“Is a family tool being proposed, and how will it be balanced?” Eastman’s guidance is that family molds should be avoided where possible because balanced flow is difficult, and that where they are required, balancing should be done by varying runner diameters, not gate size. “We’ll adjust the gates at T1” is the answer that means balancing has been deferred to the point where the options are worst.

“Which cooling approach is planned for the tall cores?” Cores are where cooling fails first, and the published consequence is sticking and difficult ejection, a symptom usually blamed on draft.

The Defects This Window Is Your Last Chance At

Most cosmetic findings at T1 have a remedy: process tuning, a gate change, a polish. A specific group does not, because the cause is built into the tool.

Sunburst around a hot drop feeding an appearance surface, the heat is where it is. A raised secondary weld line, the ridge left where a flow front stalls and restarts, which texture cannot hide because texture does not hide a bump. Pressure lines and stress marks, both products of the fill pattern the tool creates. And ghosting, which appears only after the surface is grained and is therefore found last.

For those, mold design review is the correction opportunity. After it, the choices are accept, pay for a tool change, or paint.

Which is the real argument for taking the fill analysis seriously. Not because you will read the plots better than the analyst, but because the meeting where someone says “there is a gas trap at the end of that rib” is the last meeting where fixing it is free.

Ask for the Staged Sample Review

One more request, and it costs almost nothing.

Appearance-part molders commonly review in stages: first shots, then a smooth pre-texture sample, then the textured sample. Each stage exposes a different class of problem, and the smooth part is the honest one, flow marks and weld lines that a grain would mask are fully visible on it.

Texturing is close to irreversible: deepening a grain is possible, removing one means polishing the cavity back and re-etching, which changes dimensions. So a look at the part before it is grained is the only look you will get at what the grain is about to hide. It costs a shipment and a week.

A Short Checklist

Before steel is cut, you should be able to answer yes to these:

[ ] Cosmetic surface classes issued as a marked-up view, with viewing direction
[ ] Texture pattern selected, and draft reviewed against it
[ ] Critical dimensions identified (a short list, with datums)
[ ] Resin fixed at grade level, including fill content
[ ] Parting line, gate, weld line and ejector locations reviewed and approved in writing
[ ] Mating-part witness lines checked against each other, not just individually
[ ] Fill analysis reviewed: weld line locations, gas traps, fill balance
[ ] Mold base grade and steel confirmed, with the cooling consequence understood
[ ] Family-tool balancing method stated, if a family tool is proposed
[ ] Pre-texture sample review agreed as a program step

None of these require you to design a mold. All of them are cheaper now than at any later point in the program.

Buyer FAQs

What is a mold design review and who runs it?

The supplier’s review of the tool design before steel is cut. On appearance parts it typically includes a marked-up drawing or 3D data showing parting line, gates, weld lines, and ejector marks for written approval. The supplier runs it. Your job: provide product-use inputs and approve mark locations. If the supplier will not send locations for approval, treat that as a process gap before you authorize steel.

What should a buyer provide before mold design starts?

Four things: which faces are cosmetic and from what viewing direction; the texture pattern, if any; a short list of genuinely critical dimensions with datums; and the resin at grade level including fill content. Each constrains mold design. Each is expensive to introduce later. A late texture decision alone can demand 2.5° to 4.5° of extra draft on faces already drawn. Hold steel until those four are in writing.

Why does texture have to be decided before the tool is designed?

Textured faces need substantially more draft, scaling with depth. Published guidance: roughly 1° to 1.5° of extra draft per 0.001 in of texture depth, against typical depths of 0.0025–0.0030 in. Adding a grain after faces were drawn with minimal taper means redrawing the part, not just re-quoting the tool. Settle texture before CAD freeze on draft-critical faces.

Which molding defects cannot be fixed after the tool is built?

Defects whose cause is built into the tooling: sunburst around a hot runner drop on an appearance surface, raised secondary weld lines, pressure and stress lines from the fill pattern, and ghosting that appears only after texturing. For these, design review and flow analysis are the correction window. After steel, options are acceptance, a tool change, or a coating. Do not defer those findings to “process will fix it.”

How This Guide Was Put Together

The draft-per-texture-depth conversion, the family-mold balancing guidance and the gate sizing note are quoted from one resin manufacturer’s published mold design guidelines. Mold plate flatness and parallelism figures are from a mold base supplier’s published tolerance sheet. The vent-depth practice is from a trade-press article on vent optimisation. The staged review sequence and the marked-up approval drawing describe practices that are common in automotive appearance-part molding; they are not universal requirements, and outside that segment they generally have to be requested.

Disclaimer

PlasticsTechnologyAlliance.com is an independent buyer resource. It does not manufacture parts, design or approve tooling, or certify suppliers. Mold design decisions for your part are your supplier’s engineering responsibility and should be confirmed with them.