Through real-world production scenarios and in-depth case studies, this paper reveals underlying contradiction that "parameters cannot be directly copied when producing same product across FANUC/Sumitomo/Haitian machines." It establishes an analytical framework of "equipment differences → parameter logic → process results," helping adjustment engineers shift from "memorizing parameters" to developing core competency of "understanding differences and reconstructing parameter combinations."

 

Essence of injection molding machine adjustment is "matching material rheological properties + mold structure + equipment capabilities through parameter combinations." However, three machine tools from FANUC (Japanese precision), Sumitomo (Japanese precision), and Haitian (domestic high-efficiency) have formed an insurmountable "equipment genetic barrier" due to long-term differences in their technological paths, specifically reflected in three dimensions:

FANUC, represented by its ROBOSHOT series, uses high-response servo motors (±0.1ms level control accuracy), achieving a reproducibility accuracy of ±0.5% in injection volume. Its barrel temperature control module is equipped with a high-precision PID chip, with a single-segment temperature control error of only ±1℃. Combined with a low compression ratio screw (such as CR23:1), melt delivery uniformity is extremely high.
While Sumitomo's DE series is also a Japanese precision machine, it places greater emphasis on "pressure gradient control." Its servo valve response accuracy is slightly lower than FANUC's (±0.5ms level), but its temperature control zones are finer (up to 8 segments in some models), and hot runner temperature fluctuations can be controlled within ±0.5℃. Standard screw compression ratio (e.g., CR22:1) is slightly lower than that of FANUC, requiring more precise positional compensation for melt volume.
Haitian MA series, representing high-efficiency domestic machines, uses a hydraulic system (pressure stability ±1 bar). While its response speed is slower (±10ms level), it excels in load-bearing capacity. Standard screw compression ratio (e.g., CR24:1) is relatively high, suitable for filling thick-walled parts, but due to weak compatibility with mold temperature controller interfaces, actual mold temperature often deviates from set value by more than ±2℃.
Key Impact: Hardware differences directly determine "parameter-sensitive areas" of three machines—FANUC is more sensitive to speed/position accuracy, Sumitomo is more sensitive to pressure gradient, and Haitian is more sensitive to temperature uniformity.

FANUC uses "position-pressure dual closed-loop" control. During injection stage, it prioritizes tracking position curve, switching to pressure control when approaching V/P switching point to ensure consistent filling. For example, when metering position is set to 95mm, system will strictly stop screw at 95mm before initiating holding pressure.
Sumitomo uses a "speed-pressure gradient" algorithm. During injection stage, injection proceeds at a preset speed range, with pressure dynamically adjusted according to melt resistance. This is more suitable for filling complex structures (such as multi-slider, thin-ribbed parts) with balanced filling. For example, when filling deep-cavity parts, Sumitomo will actively reduce end speed to avoid flash, while FANUC may cause slight flash due to position priority.
Haitian simplifies this to a coarse-adjustment logic of "pressure-time". During injection stage, constant pressure is used, and time is set by process engineer's experience. This logic is suitable for standardized products (such as daily necessities), but it is prone to uneven filling for precision parts (such as mobile phone frames).
Key impact: For same product's "filling target" (such as no flash, no missing material), three machines need to achieve it through different parameter combinations—FANUC adjusts position accuracy, Sumitomo adjusts speed gradient, and Haitian adjusts pressure upper limit.

FANUC products come standard with high-precision mold interfaces (e.g., positioning ring accuracy ±0.02mm), and engineers habitually calibrate injection volume regularly (once a month) to ensure metering repeatability.
Sumitomo's hot runner control module has high integration and supports independent temperature control in multiple zones by default; however, mold interface tolerance is slightly larger (±0.05mm), making it prone to temperature fluctuations due to thermal expansion over long-term use.
Haitian, in order to reduce costs, has weaker mold temperature controller interface compatibility. Pressure drop in water pipes often causes actual mold temperature to be lower than set value (e.g., set to 60℃, actual 55℃), requiring additional heating compensation.
Key Impact: Configuration differences lead to "hidden parameter drift"—even when setting same parameters according to manual, actual process window of three machines will gradually deviate due to factors such as mold condition and ambient temperature.

 

Background and Known Conditions
- Product: A brand of mobile phone frame (dimensions 140×70×8mm, average wall thickness 1.2mm, 4-slide block + hot runner injection);
- Material: ABS (melt index 25g/10min, recommended mold temperature 60℃±2℃);
- Objective: All three machines must achieve "no flash, no missing material, and a CPK ≥ 1.33".
Case 1: FANUC → Sumitomo: "Speed Trap" of Frequent Flash
FANUC Initial Parameters (Stable Production):
- Barrel Temperature: Rear Section 210℃ → Middle Section 230℃ → Front Section 245℃ → Nozzle 250℃;
- Injection Pressure: 16MPa (Segmented: 80% Fill → 60% Holding);
- Injection Speed: 120mm/s (3-Segment: 100% Fill → 80% Holding);
- Metering Position: 95mm (Screw diameter 40mm, displacement per revolution 50cm³);
- Results: 500 molds produced continuously, no flash, no shortage, stable dimensions (critical dimension φ10mm hole tolerance ±0.05mm).
Issues with directly copying from Sumitomo DE-50:
- After producing 200 molds, flash appeared on slide parting surface (thickness 0.15mm, out of tolerance 0.1mm);
- Obvious shrinkage marks on the back of product (depth 0.08mm, affecting appearance).
Engineer Analysis and Adjustments:
Sumitomo servo valve response is slow (±0.5ms vs. FANUC ±0.1ms). Original 120mm/s injection speed resulted in a lag in final filling pressure on Sumitomo valve, causing melt buildup and flash. Therefore, injection speed was reduced to 100mm/s (to match Sumitomo's response capability).
Simultaneously, under Sumitomo's "speed-pressure gradient" control, original holding pressure of 60% (9.6MPa) only compensated for initial shrinkage, with insufficient material replenishment at the end due to speed lag. Therefore, holding pressure was increased to 11 bar (≈1.1MPa, Sumitomo pressure unit is bar) to increase end-feedback compensation.
Furthermore, Sumitomo screw compression ratio (CR22:1) was slightly low, resulting in insufficient actual melt volume at original 95mm metering position (displacement per revolution 48cm³ vs. FANUC 50cm³). Therefore, metering position was increased from 95mm to 99mm (95×(50/48)=98.96mm≈99mm).
After adjustment, 500 molds were produced, flash disappeared, shrinkage mark depth <0.05mm, and dimension CPK=1.41 met standard.
Case 2: FANUC → Haitian: "Double Trap of Pressure and Temperature" of Material Shortage
FANUC initial parameters (same stable parameters as above).
Problems from directly copying to Haitian MA3000:
- Material shortage at the end of product (away from gate side) (shortage area approximately 5%);
- Dimensions shorter after cooling (0.2mm, out of tolerance 0.1mm).
Engineer Analysis and Adjustments:
Upper pressure limit of Haitian hydraulic system (25MPa) is lower than that of FANUC (35MPa). Due to system losses (pipeline/valve group resistance), original 16MPa injection pressure only resulted in an effective pressure of 14MPa. Therefore, injection pressure was increased to 18MPa (close to Haitian's upper limit to compensate for losses).
Haitian mold temperature controller interface has weak compatibility, resulting in an actual mold temperature of only 55℃ (FANUC is stable at 60℃). This reduced ABS fluidity led to filling difficulties. Therefore, mold temperature controller was calibrated, and heating power was increased to stabilize mold temperature at 60℃±1℃ (using an infrared thermometer for real-time monitoring).
Haitian's "pressure-time" coarse adjustment logic could not accurately match end-fill, requiring an extended injection time to compensate for insufficient speed. Therefore, injection time was extended from 200ms to 250ms (to ensure melt completely fills end).
After adjustment, 500 molds were produced. Material shortage disappeared, dimensional shortness decreased to 0.05mm, and CPK=1.35 met standard.

Core Conclusions
Fundamental reason why parameters cannot be directly replicated when producing same product across three different machines is:
- Hardware inherent characteristics determine sensitive parameters: Fanuc is sensitive to speed fluctuations, Sumitomo is sensitive to insufficient pressure, and Haitian is sensitive to temperature deviations;
- Software logic determines direction of adjustment: position closed-loop control requires locking metering point, speed gradient requires optimizing number of segments, and coarse pressure adjustment requires extending holding pressure;
- Configuration inertia determines hidden costs: "small problems" such as mold interfaces and mold temperature controller compatibility can amplify parameter deviations.
Core Competency Requirements for Machine Adjustment Engineers:
1. Equipment Difference Sensitivity: Ability to quickly identify "parameter-sensitive areas" of three different machines (e.g., FANUC speed, Sumitomo pressure, Haitian temperature);
2. Process Target Decomposition Ability: Ability to break down macro-level targets such as "no flash, no missing material" into specific actions such as "adjusting speed ranges," "compensating for metering," and "calibrating mold temperature";
3. Systematic Verification Thinking: After parameter adjustments, verification must be conducted through the entire process of "short-shot testing → filling time → dimensional measurement" to avoid piecemeal solutions.