Start from the part, not the machine
The most reliable way to pick a machine is to start from the part and work backwards to the tonnage, then check whether the machine you have found actually delivers that tonnage usefully. Working forwards from "we have a 300-ton machine" produces the classic buying error: a machine that is nominally the right size but too slow to fill the part.
Nominal tonnage is a ceiling, not a specification
A machine rated at 300 kN can hold 300 kN of clamp force, but you should never plan to use more than about 70–80% of it. The remaining margin absorbs the pressure spike at the moment of injection peak, when cavity pressure is highest and the mould is trying to open. A tool planned at 95% of rating will flash, or the tie bars will flex visibly, on the first shot of every batch.
The reference table you can work from
These are typical applications for commonly available machine sizes. Treat them as starting points, not specifications — the resin, wall thickness and geometry move the answer more than most people expect.
| Machine size | Typical part weights | Typical applications | Screw diameter class |
|---|---|---|---|
| 50–90 kN | < 20 g | Cap closures, pipette bodies, small medical components, connector pins | 14–18 mm |
| 90–160 kN | 20–80 g | Bottle preforms (small), cosmetic caps, electrical housings, syringe barrels | 18–22 mm |
| 160–250 kN | 80–200 g | PET preforms, detergent bottles, gearbox covers, chair fittings | 22–28 mm |
| 250–400 kN | 200–500 g | Jerry cans, storage boxes, automotive interior trim, appliance parts | 28–36 mm |
| 400–650 kN | 500 g –1.5 kg | Washing machine components, buckets, crates, panels | 36–45 mm |
| 650–1,300 kN | 1.5–6 kg | Auto bumpers, refrigerator liners, pallets, large housings | 45–60 mm |
| 1,300–2,900 kN | 6–25 kg | Automotive body panels, IBC tanks, structural parts | 60–90 mm |
| 2,900–29,000 kN | 25 kg+ | Automotive bumpers and body-in-white subassemblies, large containers | 90–150 mm |
Shot weight is normally 30–50% of machine capacity for a general purpose machine. Petrochemical and thin-wall applications push towards the lower figure; thick, heavy parts towards the upper.
Working out clamp force yourself
Clamp force is a simple multiplication, but the error is almost always in the shrinkage correction, not the formula. Do it in this order:
Worked example. A 120 cm² tray, PP shrinking at 2%, 4 cavities:
- Area with shrinkage correction: 120 × 1.30 = 156 cm²
- For 4 cavities: 156 × 4 = 624 cm²
- Required force: 624 ÷ 0.75 = 832 kN
- So this part belongs on a 900–1,000 kN machine, not a 650 kN one
Two things people miss here. The shrinkage factor is not optional — PP and ABS at 2% means every dimension grows, and the mould cavity has to be made smaller accordingly. And forgetting the safety margin is what produces a machine that technically satisfies the calculation and then flashes anyway.
Injection unit sizing
Clamp force tells you how hard the tool is held shut. It says nothing about whether the machine can inject the part in time. This is the second half of machine selection, and it is the half most quotations skip.
The ratio that matters is shot volume to machine capacity. You want 30–50%. Under 20% means the machine is oversized and you are wasting capital on clamp force you cannot use; over 70% means no reserve for the plastic that back-flows through the check ring during the hold phase, which shows up as part weight variation.
The failure mode to test for on any quotation
Ask what the fill time is in seconds and what the cooling time is. If the injection unit cannot deliver the shot within roughly a third to a half of the cooling time, the machine is too slow regardless of what tonnage is printed on the nameplate. This is the single most useful question you can ask a supplier before placing an order.
Where the dry-screw and servo ranges actually differ
You will see this framed as an either-or. It is not, because the two technologies differ in characteristics rather than in capability, and part geometry decides which characteristic matters.
| Characteristic | Servo-hydraulic | All-electric (LX series) |
|---|---|---|
| Available sizes | 90 kN to 29,000 kN — no practical ceiling | 50 kN to 4,500 kN |
| Oil use | Hydraulic oil, requires maintenance and disposal | None |
| Idle power | Consumption continues between shots | Drives sleep between shots, roughly 50–70% saving |
| Oil temperature effect on part | Oil heats over a long run, viscosity drops, cycle drifts | No oil, so the process is far more stable over a shift |
| Injection pressure | Up to 16 MPa on PET applications, enough for preform neck | Typically lower — check against thin-wall PET requirements |
| Repeatability over long runs | Good, subject to oil condition | Better; the reason precision moulders choose it |
| Relative purchase cost at equal clamp force | Baseline | Higher |
So the decision follows from the part. Thin-wall medical or optical parts where dimensions must hold across a long run point to electric. Preform and bottle work needing 16 MPa points to servo-hydraulic. Anything above 4,500 kN has no electric alternative at all, because the physics of clamping force in an electric machine does not scale there.
Where electric genuinely does not pay
High-tonnage structural parts, thick-walled items, and anything with a long cycle where the machine spends most of its time idle anyway. You pay a premium for precision you are not measuring, and the energy saving only matters if you are running a high number of shots per day. If you run one shift and the machine idles for hours, the payback on electric is very slow.
When the answer is a bigger machine
Sometimes the honest recommendation is to go larger than the calculation strictly requires. Three cases justify it:
- Tool sharing. If the same machine will run more than one mould, you want the capacity to take the larger one later without buying again. A machine with 25% headroom is usually cheaper over four years than two machines.
- Cycle-time pressure. If output matters more than capital, a larger machine often produces a lower cost per part because it runs the same tool faster with more reserve. This is the least intuitive but most common economic argument for going up a size.
- Growth in the product line. If a family of parts is likely to grow rather than stay fixed, planning against the largest part in the family avoids a second purchase.
Going the other way is worse. A machine sized for today's part with no reserve will be the constraint on every future quotation, and resizing a machine in place is a major mechanical job.
What to send us for a machine recommendation
- Part drawing or photograph with dimensions, plus material and grade
- Part weight, projected area, and how many cavities you intend to run
- Annual volume and required output in pieces per hour
- Any dimensional tolerance you must hold, and over what run length
- Machine preference — new, used, or either — and any budget range
Send the drawing rather than the tonnage you think you need. We will return the machine class, the injection unit, the fill-time calculation and the reasoning. If a smaller machine will do, we will say so — and if you would be better served by a used machine at that size, we will say that too.
Common questions
What size injection molding machine do I need for my part?
Work backwards from the part. Clamp force = projected area × shrinkage correction (1.20–1.35) × cavities ÷ 0.75. Then separately verify the injection unit can deliver the total shot volume within roughly 30–50% of the cooling time. Clamp force alone does not tell you whether the machine works — a machine can hold the force and still be too slow to fill.
Is a bigger injection molding machine always better?
No. Going larger is justified for three reasons: the same machine will run more than one tool and you want capacity for the larger one later; output matters more than capital, because a larger machine often runs the same tool faster and gives a lower cost per part; or a family of parts is likely to grow. Going smaller with no reserve is worse than either.
Should I buy an all-electric or a servo-hydraulic machine?
It depends on the part, not on which is newer. All-electric gives better repeatability over long runs because there is no oil to heat, plus 50–70% idle energy saving and no oil maintenance. Servo-hydraulic covers the full range to 29,000 kN and delivers up to 16 MPa, which is what PET preforms need. Above 4,500 kN there is no electric alternative. All-electric is the wrong purchase for thick structural parts.
Why does the machine need a 75 percent clamp safety margin?
Because rated clamping force is a static holding figure, while the mould sees a dynamic pressure spike at injection peak — exactly when cavity pressure is highest and the tool is trying to open. Planning at 95% of rating produces flashing or visible tie-bar flex on the first shot of every batch.
How much capacity does a machine need beyond the shot weight?
Normally the shot volume should be 30–50% of machine capacity. Below 20% the machine is oversized and you have paid for clamp force you cannot use. Above 70% there is not enough reserve for the plastic that back-flows through the check ring during hold, which shows up directly as part weight variation.