Polybutylene terephthalate (PBT) is a semicrystalline polyester produced from 1,4-butanediol and dimethyl terephthalate or terephthalic acid, with a melting point of 223 °C to 225 °C and a rapid crystallisation rate that permits short mould-open and ejection sequences. Its practical advantage is that it reaches high crystallinity quickly in a tool at 40 °C to 80 °C for unfilled grades, but this also makes the moulding process sensitive to actual clamp force, shot-size ratio, and cycle timing. Moving an existing PBT mould from one injection moulding press to another creates three distinct reset problems: the nominal tonnage of the replacement press may not equal the actual force available at the parting line; the same shot volume may represent a much smaller or larger share of a new barrel and alter residence-time distribution; and dry-cycle kinematics may add incremental seconds that shift total cycle beyond the original process window.

Before resetting the mould, the material boundaries should be fixed. PBT must be dried to a residual moisture content below 0.02% as measured by Karl Fischer titration according to ISO 15512:2019 or ASTM D6869. Typical drying conditions are 120 °C for 3 h to 4 h in a desiccant dryer with a dew point of -40 °C or lower. At relative humidity above 60%, dried PBT reabsorbs surface moisture rapidly; hopper residence should not exceed 30 min to 60 min unless a closed feed and dry-air purge are maintained. Molten PBT held at 250 °C undergoes hydrolysis, colour drift, and molecular-weight loss if moisture is present or if barrel residence time exceeds approximately 5 min; excursions beyond 8 min are generally considered the outer limit before weld-line strength and tensile elongation fail.

Why Does Nominal Clamp Rating Fail to Predict PBT Flash on a Replacement Press?

Nominal clamp force is the maximum design loading at full clamp, not an active four-tie-bar measurement. In a direct hydraulic press, force is developed by clamping cylinders and remains relatively insensitive to small mould-height errors after pressure stabilisation; in a toggle press, force is generated through link kinematics, and tie-bar elongation is the physical variable that controls clamping stiffness. The actual tonne count therefore changes with mould height adjustment, tie-bar diameter, free length, link wear, platen parallelism, and thermal expansion of the tool. A mould transferred from an 1,800 kN hydraulic machine to a 2,200 kN toggle machine may receive less than its original clamp if the toggle is not advanced far enough to stretch the tie bars.

Required clamp force for PBT is calculated from the projected area at the parting line and the cavity pressure at gate seal. The expression is F_clamp = P_cav × A_proj / 10, with F_clamp in kN, P_cav in MPa, and A_proj in cm². For a 300 cm² projected area tool and a gate-sealed cavity pressure of 65 MPa to 80 MPa, the minimum force is 1,950 kN to 2,400 kN. A 30% glass-fibre-reinforced PBT running at 260 °C melt and 80 °C mould generally sits in this cavity-pressure band, while unfilled grades may drop to 35 MPa to 50 MPa. If the replacement press reports only 1,740 kN measured at the tie bars, flash is likely along thin vent land and around ejector pins, even if the machine nameplate is higher.

Direct measurement of clamp force uses strain-gauge tie-bar sensors or permanently installed force pillars. A four-tie-bar audit should record each bar’s force, correct for differential thermal expansion, and calculate mean and spread. The acceptance band is normally ±5% around the mean; if one bar is 15% below the others, the platen is not square and the PBT melt may fill the lower-force side earlier, causing asymmetric shrinkage. Tie-bar elongation can be estimated from ΔL = F × L / (E × A_t), where E is 200 GPa for alloy steel and A_t is the cross-sectional area of one tie bar. An 1,800 kN total force across four 100 mm diameter tie bars with 2,000 mm free length corresponds to roughly 0.57 mm elongation per bar. If the measured elongation is only 0.45 mm, actual total force falls near 1,420 kN, which is below the 1,950 kN requirement.

Venting depth for PBT at the parting line is typically 0.01 mm to 0.02 mm for unfilled grades and 0.02 mm to 0.03 mm for glass-filled grades. Flash visible at those vents implies local clamp force deficiency. If flash appears only at the outer cavities, the tie-bar distribution is not flat; if flash appears at every cavity, total clamp force is too low. If the replacement press cannot achieve the required force after mould-height adjustment, the acceptable corrections are to reduce the number of cavities, shorten flow length, lower fill speed to reduce peak cavity pressure, or move to a press with higher measured clamp force. Raising the tool temperature alone does not compensate for insufficient tonnage.

In parallel with clamp calibration, the melt inventory and shot-size ratio must be recalculated from the new screw geometry. Shot volume includes sprue, runner, cavities, and the screw cushion, not simply part mass. The maximum usable shot volume is V_max = π/4 × D² × L_stroke, where D is the screw diameter and L_stroke is the injection stroke above a specified cushion. For PBT, the processed shot should occupy 30% to 75% of maximum shot stroke. Below 25%, the melt residence time distribution widens because most of the barrel inventory remains molten over several cycles; above 80%, shot weight consistency may fall because compression of the melt and non-return valve leakage consume stroke.

Residence time is more important for PBT than for many amorphous resins because moisture damage and thermal hydrolysis act together. Average residence time can be approximated as t_res = (barrel inventory / shot weight) × cycle time. A mould making a 150 g shot in a press with a 40 mm screw and a 350 g barrel inventory has an inventory factor of 2.3 cycles; at 35 s cycle time the average residence is 81 s. The same shot in a 60 mm screw press with a 900 g barrel inventory has a factor of 6.0 cycles and an average residence of 210 s. The replacement press may still be acceptable if the rear barrel zone is reduced to 235 °C to 245 °C and the shop-floor checks confirm no melt-flow-rate rise above 20% compared with the dried pellet baseline.

Hot runner systems change the reset calculation because runner volume is eliminated but manifold volume is added. A valve-gated hot runner can reduce shot volume by 10% to 30% while increasing melt inventory in the manifold. For PBT, manifold temperature should not exceed 250 °C, and dead-flow pockets should be eliminated because they extend residence time locally. If the manifold volume is large, a lower manifold set point of 235 °C to 245 °C is recommended, and colour stability should be checked across the first 20 shots after restart.

Representative reset audit values for a 150 g unfilled PBT shot transferred from a direct hydraulic press to a toggle press are given in the comparative matrix below.

Shot Size, Cushion Control and Non-Return Valve Response

After the shot-size ratio is known, the screw cushion should be re-established on the replacement press before cycle drift is evaluated. Cushion is the residual melt volume between the screw tip and the non-return valve seat at the end of holding pressure. A cushion of 2 mm to 5 mm on a hydraulic press or 3 mm to 8 mm on an electric press is common for PBT; values below 2 mm create a risk of decompression and loss of holding-pressure transmission, while values above 10 mm increase dead-space melt volume and reduce the precision of holding-pressure control. The non-return valve must be checked for leakage by rotating the screw forward with the nozzle closed or by measuring screw recovery against a known shot; if the cushion decays by more than 1 mm over five shots, the ring and seat should be replaced before PBT lot-to-lot comparisons can be trusted.

Switch-over position should be recalculated by volume, not copied from the old press. The new screw diameter changes the stroke necessary to deliver the same melt volume. A short-shot progression from 92% to 98% fill at the same injection velocity identifies the position where the cavities fill without packing. Once the cavities are full, holding pressure for a glass-filled PBT typically runs 70 MPa to 90 MPa for 4 s to 8 s, depending on gate freeze. The gate seal time is confirmed by plotting part weight against holding time and selecting the plateau point, not by assuming the previous press value transfers unchanged.

If Cycle Time Shifts After Transfer, Check Plasticating and Opening Kinematics Before Holding Pressure

Total cycle drift on a replacement press is the sum of dry-cycle differences, screw recovery time, cooling time, and any change in mould-open dwell for part removal. The diagnostic sequence begins with the dry cycle: run the clamp, ejector, and core pulls without material and record time intervals with linear sensors or machine timers. A toggle press with a longer clamp stroke and a different hydraulic pump may add 0.8 s to 1.5 s to mould open and 0.6 s to 1.0 s to mould close. On a 25 s original total cycle, an increase of 1.4 s to 2.5 s is a 5.6% to 10.0% drift. This should be corrected through clamp speed profiling and synchronised ejection, not by shortening holding time.

Screw recovery time must be compared with cooling time. If the replacement press plasticates more slowly, the mould may be ready to open before recovery completes, making the cycle screw-limited. PBT screw speed is usually set between 60 rpm and 120 rpm for a 40 mm screw; a 60 mm screw may require 40 rpm to 80 rpm to maintain similar surface speed and avoid excessive shear heating of a glass-filled compound. Back pressure is kept low, commonly 0.3 MPa to 0.7 MPa hydraulic, to avoid fibre fracture and temperature rise. If recovery time is slower, reduce screw speed only until recovery time is just below cooling time; lowering it further increases cycle time without improving melt quality.

Cooling time must be recalculated from part wall thickness and mould temperature. For a 2 mm unfilled PBT wall at 250 °C melt, 80 °C mould, and 140 °C ejection, the cooling estimate is t_cool = (s²/π² α_eff) × ln[(4/π)(T_melt − T_mould)/(T_eject − T_mould)]. With an effective thermal diffusivity of 0.10 mm²/s, t_cool is approximately 5.0 s to 5.5 s. The original press may have produced a shorter cooling time because lower cavity pressure allowed a slightly lower mould temperature; copying the original cooling time onto the replacement press without confirming ejection temperature can produce part sticking, warpage, or a soft surface.

For 30% glass-fibre-reinforced PBT, thermal diffusivity is anisotropic; in-plane fibre orientation causes thermal conductivity along the flow direction to exceed transverse direction. Cooling time calculations should use the thicker cross-flow dimension rather than the nominal wall and should account for fibre orientation. Published data for anisotropic diffusivity of PBT is grade-specific, so infrared thermography on ejected parts is preferred to confirm surface temperature before locking cycle time.

Cycle drift after resetting the mould should not be immediately interpreted as a PBT material problem. The average residence time, melt cushion, clamp force, and dry cycle must first be brought within the target bands described above. If the total cycle still exceeds the original after those corrections, in-mould cavity pressure traces should be aligned between the two presses by adjusting injection velocity and holding-pressure decay so that the gate freeze point occurs at the same time after fill. This is a process-transfer discipline rather than a material correction.

The reset evaluation is not complete until material quality boundaries are verified against the original production lot. A production audit should compare tensile strength and elongation at yield per ISO 527-2:2012, Charpy notched impact per ISO 179-1:2010, and melt flow rate per ISO 1133-1:2022 on moulded plaques. Shrinkage should be checked after 24 h per ISO 294-4:2018. If the replacement press uses an 18:1 L/D general-purpose screw instead of the original 22:1 L/D screw, the melting history is shorter and the melt may show lower homogenisation; a melt-flow-rate increase of more than 20% from the dried pellet baseline indicates resin degradation during the residence audit. PBT should not be purged with polycarbonate or acetal residues; use an HDPE-based purging compound when changing from another polymer, and avoid processing above 270 °C for extended intervals. Published data for specific combinations of PBT grades, tie-bar dimensions, and screw geometries is limited; the numerical values in these reset calculations are representative industrial settings and must be verified on the specific press and mould because clamping force, hydraulic response, and heater-band zoning vary by machine family.