Sand Mold and Core Printing: A Buyer’s Checklist
Buying sand mold and core printing is two decisions, not one: whether to buy a machine or buy printed molds as a service, and if you buy, which specification actually matters. Six questions decide it — annual mold volume, largest core, tolerance on the functional faces, sand and binder system, post-processing capacity, and who owns process development. Get those wrong and the machine will sit idle regardless of its build volume.
Key takeaways
- The largest core usually sets the machine size, not the largest mold — cores are the constraint people discover too late.
- Throughput is measured in molds per shift, not litres per hour: cure and depowder time is part of the cycle.
- Binder choice follows the pouring temperature, and it is not freely interchangeable on a given machine.
- Post-processing — cure, depowder, coating, assembly — is roughly half the labour of a printed mold.
- Below about 20 molds a month, buying printed molds as a service almost always beats owning a machine.
The six questions that decide the answer
| # | Question | What it determines | Get it wrong and… |
|---|---|---|---|
| 1 | How many molds a year, and how regular? | Machine versus service; one machine versus two | The machine is idle most of the year, or is the bottleneck every month |
| 2 | What is the largest core, not the largest mold? | Build envelope, and therefore price class | You buy a machine that cannot print your cores |
| 3 | What tolerance sits on the functional faces? | Whether as-printed is enough or machining is required | You budget for printing but pay for machining |
| 4 | Which sand and which binder? | Alloy compatibility, gas evolution, sand reclamation | The binder cannot survive your pouring temperature |
| 5 | Who does cure, depowder and coating? | Labour, floor space, throughput | Printing is fast, the rest of the line is not |
| 6 | Who owns process development? | Time to first good mold, and after every alloy change | Every new alloy becomes a new qualification project |
Build envelope: measure your largest core, not your largest mold
This is the mistake that is most expensive to fix. A foundry measures its largest casting, adds a margin, and specifies a build volume from that number. Then the first job arrives with a three-part core set, and the cores do not fit. The cycle-time and cost side of the same decision is covered in printed molds versus traditional sand casting.
In a conventional foundry the core box is separate tooling, so its size is a separate line item. In a printed foundry the core and the mold come out of the same build volume — which means the core now has to fit, and cores can be awkward: tall, thin, and oriented in a way that wastes bed height.
| What you measure | Why it matters | Practical check |
|---|---|---|
| Largest core, including core prints | Cores often exceed the mold in one dimension | Measure the core box, not the pattern |
| Core orientation in the build | Height drives print time more than volume does | Can the core lie down? If not, budget the height |
| Number of cores per mold | All cores must fit in the same bed as the mold | Add every core to the footprint, not just the largest |
| Mold wall thickness plus print allowance | The printed shell is thicker than the pattern | Add 20–30 mm around the casting envelope |
A production machine in this class offers a 1,800 × 1,000 × 700 mm build volume and handles molds of 1.3 t or more. That envelope covers a large share of industrial castings — but it only helps if the cores fit inside it too.
Sand and binder system
The sand and binder pair is not a consumable choice you can change freely after purchase. It determines which alloys you can pour, how much gas the mold evolves, and how much of the sand you can reclaim.
| Sand | Best for | Trade-off |
|---|---|---|
| Quartz (silica) sand | The default for iron, steel and aluminium | Cheapest and most available; thermal expansion at high temperature |
| Ceramic sand | Steel and high-temperature alloys, heavy sections | Higher cost per tonne; better dimensional stability |
| Chromite sand | Chill zones and hot spots | Heavy, expensive, usually used as a facing rather than the bulk |
| Ordner | Pouring temperature | Gasentwicklung | Notes |
|---|---|---|---|
| Phenolic | Steel and iron, high temperature | Lower than furan in many formulations | The usual choice for ferrous work |
| Furan | Aluminium and lower-temperature alloys | 8–15 ml/g at 850 °C in typical printed molds | Good strength, well understood in foundries |
Ask two questions on the spec sheet: which binders the machine is validated for, and what tensile strength the printed mold reaches. A printed mold in this class runs 1.5–2.5 MPa tensile — comparable to a well-rammed conventional mold, and the number you need in order to trust the mold with a heavy section.
Accuracy and surface finish you should expect
| Measure | Typical printed value | What it means in the casting |
|---|---|---|
| Maßgenauigkeit | ±0.3 mm up to 300 mm; ±0.5 mm above | Well inside ISO 8062 DCTG 5–6 for most sizes |
| Schichtdicke | 0,2–0,5 mm | Sets the as-cast surface texture; thinner is slower |
| Minimum wall thickness | 2–3 mm | Thinner than a pattern-and-draft route allows |
| Repeatability | Same file, same mold, every build | No pattern wear, so no drift across a run |
The practical consequence: layer lines transfer to the casting as a fine texture. For a machined or coated surface that is irrelevant. For a cosmetic as-cast surface it means either a mold coating or extra machining allowance — and that decision should be made before the first quote, not after the first casting.
Throughput: prints per shift, not prints per hour
Print speed is quoted in seconds per layer, and it is the least useful number on the sheet by itself. What decides the schedule is how many complete molds leave the building per shift.
| Measure | Typical figure | How to use it |
|---|---|---|
| Druckgeschwindigkeit | 18–25 seconds per layer | Multiply by layer count for build time |
| Schichtdicke | 0,2–0,5 mm | A 700 mm tall job is 1,400–3,500 layers |
| Volumetric throughput | up to 153 L/h of mold | Use for comparing machines of different sizes |
| Cure and depowder | Several hours | Add this before you count a mold as done |
| Molds per shift | Derived, not quoted | This is the number to negotiate on |
Ask a supplier for molds per shift on your part, not on a demo cube. A tall core and a flat plate of the same volume can differ by a factor of three in build time, because layer-based processes are governed by height.
Post-processing is half the job
Printing produces a green mold that is not yet ready to pour. The sand 3D printing process covers how the mold is built; four steps follow it, and each one takes labour and floor space.
| Step | Why it exists | Kostentreiber |
|---|---|---|
| Cure | Binder must reach handling strength | Oven or rack time, floor space |
| Depowder | Loose sand must leave every internal passage | Labour hours, the main manual step |
| Coating | Controls surface finish and metal penetration | Consumables plus drying time |
| Assembly and closing | Multiple cores set into the mold | Labour, and the tolerance that matters most |
Depowder is the step most often underestimated. A mold with long, narrow internal passages takes longer to clear than a simple open cavity, and that time does not appear in any print-speed figure.
Buy the machine, or buy the parts?
This decision comes before the specification. Two machines in this class differ mainly in throughput and footprint, not in capability — the J1800 runs eight printheads with up to 153 L/h and weighs about 13 t; the J1800S runs four modular printheads, is slightly slower per layer, and weighs about 7 t with a lower floor-load requirement. Choosing between them is an afterthought if the volume does not justify owning either.
| Ihre Situation | Better answer | Reason |
|---|---|---|
| Fewer than about 20 molds a month | Buy printed molds | Machine cost is fixed; service cost scales with output |
| 20–60 molds a month, regular | Marginal — model both | Depends on mold size and post-processing labour |
| 60+ molds a month | Own the machine | Throughput and lead-time control both improve |
| Occasional very large casting | Buy printed molds | Envelope requirement would force an oversized machine |
| Confidential or prototype work | Own the machine | Sending models out may not be acceptable |
When a service wins
A service wins when your geometry varies more than your volume does, when the largest core would force a machine class you cannot justify, or when you do not yet know whether the printed route fits your foundry at all. Printing one mold externally answers that question for the price of one mold.
When ownership wins
Ownership wins when printed molds become a routine part of production — not an experiment. At that point lead time stops being a purchase decision and becomes a scheduling one, and owning the machine moves it from a queue you do not control into one you do.
Comparing a machine against buying printed molds? Send one of your molds and its core set — we will quote it as printed parts and tell you the volume at which owning a machine starts to win. Sand mold and core printing service · Sand casting service · Sand 3D printer range
Frequently asked questions
What printer makes sand molds for casting?
Binder jetting sand printers. A printhead deposits binder into a bed of quartz or ceramic sand, building the mold and its cores layer by layer at 0.2–0.5 mm per layer. Production machines in this class offer 1,800 × 1,000 × 700 mm build volumes and handle molds of 1.3 t or more.
Wie viel kostet ein Sand-3D-Drucker?
Sand 3D printers sit in the industrial range, well above polymer machines, because the build volume, sand handling and binder system all scale with the bed. A machine specification should be built from the largest core, the required molds per shift and the binder system for your alloys — the configuration drives the price far more than the model name.
Can you print cores as well as molds?
Yes, and that is the main structural advantage. Cores print in the same build as the mold, in one piece, with no core box and no core assembly. A part with five internal passages prints as a single mold-and-core set instead of a mold plus five separately made cores.
What is the largest sand mold that can be printed?
Entry production machines handle 1,800 × 1,000 × 700 mm and molds of 1.3 t or more. Larger formats extend the envelope to several metres in the largest direction. The practical limit is usually the largest core that must fit alongside the mold, not the mold itself.
Do printed molds need coating?
Often, yes. A refractory coating controls metal penetration and improves the as-cast surface, especially where layer lines would otherwise show. Whether you need it depends on the alloy, the section thickness and whether the surface is cosmetic or machined.
