Vendor Schedule & Justifications
Schedule of supply · exclusions · acceptance criteria · I/O list · design rationale
Schedule of supply (vendor scope: HP + trolleys + fans)
| Item | Qty | Specification | Acceptance criterion |
|---|---|---|---|
| Heat-pump dehumidifier module | 2 | 50 kW heating @ 62°C supply / 20°C coil; COP ≥ 3.2; R410A or R32 | Performance test at rated conditions |
| Recirculation fan with VFD | 2 | 5,400 m³/h @ 450 Pa, 1.2 kW IE3 motor, food-grade construction | Flow test at rated static |
| Trolley, SS 304 | 16 + 2 spare | 0.825 × 1.400 × 1.830 m, 16 trays @ 105 mm pitch, 4 × 100 mm castors | Dimensional check; 80 kg load test |
| Tray, SS wire mesh | 290 (16 × 16 + 18 spares) | 0.633 × 1.122 m, ~55% open area | Visual; open-area check |
| Step platform (200 mm) | 2 | SS, 1.0 × 0.4 m, anti-slip | Stability test |
| Master control panel + HMI | 1 | PLC + 10" HMI, Modbus to HP modules, batch logging | Function test (stage transitions, interlocks) |
| Sensor set (full schedule, p.7) | 1 set | Per schedule, calibrated certificates | Calibration certificate ≤ 12 months |
| Commissioning & training | 1 lot | 3 days on-site, dry & wet runs, operator training, as-built docs | Successful wet batch within spec |
Acceptance criteria (wet-run performance test)
| Drying time to 5.0% MC | ≤ 16 h (19.7 h target) |
| SMER | ≥ 3.0 kg/kWh (3.6 target) |
| MC uniformity across batch | σ ≤ 0.5% wb |
| Supply temperature stability | ±1.0 K of setpoint at steady state |
| No visible condensation on panels or door | Pass |
| Refrigerant charge stable (no leak) | Pass |
| Audible noise at operator position | ≤ 80 dBA |
BMS interface — I/O list
| Tag | Type | Signal | Modbus register |
|---|---|---|---|
| T-01 / RH-01 | Supply air | 4–20 mA → PLC | HR40001–40002 |
| T-02 / RH-02 | Return air | 4–20 mA → PLC | HR40003–40004 |
| T-03a/b/c | Product core (top/mid/bot) | PT100 → PLC | HR40005–40007 |
| T-04 | Coil outlet temp | PT100 → PLC | HR40008 |
| P-01 | Fan differential pressure | 4–20 mA → PLC | HR40009 |
| F-01 | Condensate flow | Pulse → PLC | HR40010 |
| LS-01 | Door position | Digital IN (hardwired) | DI 0.0 |
| ES-01/02 | E-stops (door + panel) | Hardwired safety chain | — |
| VFD-01/02 | Fan speed control | Modbus RTU | HR40020–40025 |
| HP-01/02 | HP module enable + capacity | Modbus RTU | HR40030–40040 |
| AL-01 | Alarm beacon + audible | Digital OUT | DQ 0.0 |
Performance test procedure (acceptance wet run)
- Instrumentation: calibrated T + RH sensors (±0.3°C / ±2% RH) at supply and return; 3 product core thermocouples per channel (top/mid/bottom tray); clamp-on electrical power meter; calibrated condensate tank (±0.5 L).
- Load: 1,819 kg fresh shelled halves at 10 kg/m² ±5%. Record initial MC by oven test (10 samples, 105°C / 24 h).
- Run: full batch cycle (stage 1 + stage 2). Log T, RH, condensate flow, electrical power at 1-minute intervals via HMI export.
- Sampling: at end of cycle, sample 10 halves from each of 3 tray positions (top/mid/bottom) × 2 channels = 60 samples. Oven-test each for final MC. Compute mean and σ.
- Pass criteria: mean MC 5.0 ±0.5%, σ ≤ 0.5%, drying time ≤ 22 h, SMER ≥ 3.5 kg/kWh, supply T stable ±1 K, no visible condensation on panels, no refrigerant leak, noise ≤ 80 dBA.
- Mass balance: condensate volume vs (initial mass × MC₀ − final mass × MC₁) should agree within ±5%.
- Report: vendor submits test report with raw data (CSV), plots, and pass/fail for each criterion.
Warranty & documentation
- Warranty: 24 months from commissioning (HP compressor); 12 months on remaining equipment.
- Spare parts: 1 set of filters, 1 fan motor, refrigerant service valves, 2 spare trays per trolley.
- Documentation: O&M manuals, P&ID, electrical schematics, refrigerant circuit diagram, HMI batch-export schema, recommended spares list.
Justifications & trade-offs
Why U-flow
U-flow at 10,653 m³/h gives SMER 3.6. The equivalent cross-flow would need ~26,000 m³/h for the same batch and yield SMER ~1.8 — the evaporator churns sensible heat because the air only picks up ~0.005 kg/kg of moisture per pass (huge airflow relative to the moisture load). U-flow's lower airflow raises the per-pass moisture pickup, making the coil work effectively.
Why shelled kernel (not in-shell)
The shell is the dominant mass-transfer resistance in conventional copra drying. Removing it raises the Page drying constant from ~0.10 to ~0.20 h⁻¹ (2.3× faster), bringing drying time from ~38 h to 14 h — making a one-batch-per-day cycle feasible. Loading density on the tray drops slightly (10 kg/m² vs 25 kg/m² for in-shell) but the speed advantage more than compensates.
Why three-stage profile (case-hardening mitigation)
Shelled kernel at 62°C / 20% RH can seal its surface within the first 1–2 h, trapping moisture inside and slowing the remainder of the batch. A cooler first stage (55°C) for the wet phase preserves surface permeability; the 62°C finish completes drying to target MC. Net cost vs single-stage: essentially zero (Page interval timing, not extra hours).
Why staged HP modules
The batch moisture load peaks early (38 kW) and halves by mid-batch. A single 86 kW unit would short-cycle in the second half. Two 50 kW modules give (a) better part-load efficiency, (b) redundancy — one module down = half capacity, not a stoppage, (c) standard 50 kW units are easier to source than a custom 120 kW unit, (d) future scaling — adding capacity means a third module, not replacing the unit.
Why 16 layers, not 20
A 2.6 m room with 20 layers (top tray 2.25 m) is unreachable for manual loading. Reducing to 16 layers (top tray 1.83 m, reachable with a 200 mm step) costs ~20% capacity per trolley, recovered by widening the room to 4.0 m and using 2 abreast. The net capacity (~1.8 t) is preserved while keeping every tray ergonomically accessible — a non-negotiable operational requirement.