Doc #72 — Micro-Hydro Design and Construction

---

RECOMMENDED ACTIONS (BY URGENCY)

First month (Phase 1)

1. Identify existing micro-hydro installations nationwide through the skills census (Doc #8) and contact with rural communities, Federated Farmers, and electricity distribution businesses. Some installations feed power into the local grid; these are documented by lines companies. Private off-grid systems may be undocumented.
2. Locate and secure micro-hydro equipment stocks — turbines, generators, controllers, and penstock pipe held by micro-hydro suppliers and installers (primarily small businesses in rural areas).
3. Classify experienced micro-hydro installers as critical-skills personnel (Doc #1). NZ has a small community of micro-hydro specialists; their knowledge is disproportionately valuable.
4. Begin flow measurement at candidate sites where micro-hydro is most likely to be needed — communities at the end of long distribution feeders, remote farms, areas with aging distribution transformers.

First year (Phase 1–2)

5. Complete technical knowledge capture from experienced installers — site assessment methods, turbine selection, penstock design, controller wiring, commissioning procedures.
6. Publish standardised micro-hydro design guide derived from this document and installer knowledge — a practical field manual for site assessment and construction.
7. Establish regional micro-hydro construction teams based in areas with highest stream density and greatest grid vulnerability (West Coast of South Island, Northland hill country, Wairarapa, Coromandel, East Cape).
8. Inventory available PVC and polyethylene pipe nationally — this is the primary penstock material and is available in significant quantity from plumbing and irrigation suppliers.
9. Begin construction of first micro-hydro installations at the highest-priority sites, using commercially available turbines and generators where stocks exist.

Years 2–5 (Phase 2–3)

10. Develop local crossflow turbine manufacturing capability — establish fabrication procedures at one or more machine shops, building from documented designs. First units should be tested at accessible sites before wider deployment.
11. Train penstock fabricators in steel pipe welding for sites where PVC pipe is unavailable or insufficient diameter.
12. Develop generator capability — initially rewound salvaged motors (Doc #95), progressing to purpose-built generators using NZ copper wire (Doc #70).
13. Establish electronic load controller (ELC) production — simple power electronics circuits that can be assembled from salvaged components.
14. Scale construction to 20–50 new installations per year across NZ, prioritising communities with grid vulnerability and suitable stream resources.

Years 5–15 (Phase 3–4)

15. Expand to larger community-scale installations (20–100 kW) as fabrication capability matures.
16. Develop small dam and weir construction capability for sites where run-of-river flow is marginal but storage would improve reliability.
17. Integrate micro-hydro with local mini-grids in areas where national grid distribution has degraded beyond repair.
18. Ongoing training of new micro-hydro technicians through apprenticeship programmes (Doc #156).

---

ECONOMIC JUSTIFICATION

Cost of construction

A 10 kW micro-hydro installation — sufficient to serve a small farming community of 5–10 households with basic electrical needs (lighting, refrigeration, water pumping, workshop power) — requires approximately:

Component	Person-weeks	Materials
Site assessment and survey	0.5–1	Measuring equipment (improvised if necessary)
Intake construction	1–2	Concrete, stone, timber, steel screen
Penstock installation (200 m typical)	2–4	PVC pipe (existing stock) or fabricated steel pipe
Turbine (if fabricated)	2–4	NZ steel plate and bar, welding consumables
Generator (rewound motor)	1–2	Salvaged motor, NZ copper wire, bearings
Electrical — controller, wiring, switchgear	1–2	Salvaged electronic components, copper cable
Powerhouse construction	1–2	Concrete, timber, roofing
Total	8–17	—

If a commercially manufactured turbine and generator are available from existing NZ stocks, the fabrication time drops significantly — total effort is approximately 4–8 person-weeks, dominated by civil works (intake and penstock).

Comparison with alternatives

Grid-connected power: Where the distribution grid remains functional, micro-hydro is unnecessary for basic supply. The economic case for micro-hydro arises where grid reliability degrades — which is expected for remote rural distribution lines as transformers fail (Doc #69) and line maintenance capacity diminishes. A community that loses grid supply and has no local generation loses refrigeration, water pumping, workshop capability, and lighting — a severe economic and quality-of-life impact.

Diesel or petrol generation: Fossil fuel generation is available during Phase 1–2 while fuel stocks last but is not sustainable. A micro-hydro system, once built, generates electricity from water flow indefinitely with minimal ongoing inputs. The breakeven against fuel consumption is rapid — a small diesel generator consuming 2–3 litres per hour costs approximately 50–75 litres per day of continuous operation. Within weeks, the fuel consumed exceeds the total energy cost of micro-hydro construction.

Solar panels: NZ has significant installed solar capacity, and panels last 25+ years. But inverters have shorter lifespans (Doc #73), and solar output under nuclear winter conditions — reduced sunlight due to stratospheric aerosols — may decline by 30–70% in the first 1–2 years, diminishing over 5–10 years.² Micro-hydro provides reliable baseload power day and night, and NZ’s increased rainfall under some nuclear winter models may actually improve stream flows in some regions.

Wood gasification for electricity: Gasifier-driven generators (Doc #56) are an option but require continuous fuel supply and labour-intensive operation. Micro-hydro, once installed, runs unattended for days or weeks at a time with only periodic maintenance.

Breakeven assessment

The construction investment of 8–17 person-weeks for a system that operates for decades is among the best returns on labour investment in the entire recovery programme. A 10 kW system running at 50–80% capacity factor (depending on seasonal flow variation and maintenance downtime) produces approximately 44,000–70,000 kWh per year — equivalent to the electrical consumption of 5–12 NZ households under reduced-consumption recovery conditions.³ The system pays for its construction labour within the first year of operation in avoided alternatives (fuel consumption, lost productivity from power outages, or the labour cost of manual alternatives to electrically powered equipment).

---

1. NZ’S MICRO-HYDRO HISTORY AND CURRENT STATUS

1.1 Historical context

New Zealand’s electrification began with micro-hydro. One of the earliest public electricity supplies in the Southern Hemisphere was a hydro-powered installation at Reefton on the West Coast of the South Island in 1888, generating approximately 50 kW for street lighting and gold mining operations.⁴ Through the late 19th and early 20th centuries, small hydro schemes proliferated across NZ — powering mines, sawmills, dairy factories, and rural towns. Before the national grid reached rural areas (a process that extended from the 1920s through the 1960s), many farms generated their own electricity from streams on their property using small Pelton or crossflow turbines driving DC generators. Hundreds of such installations existed at peak, though exact numbers are uncertain.⁵

As the national grid expanded, most farm micro-hydro systems were abandoned — grid power was cheaper and more reliable. But the underlying resource has not changed. NZ’s combination of high rainfall (600–6,000+ mm per year depending on region), steep terrain, and dense stream networks makes it one of the most naturally favourable countries in the world for micro-hydro development.⁶

1.2 Current installations

NZ has an uncertain but meaningful number of operating micro-hydro systems. These fall into three categories:

Grid-connected small hydro: Installations up to several hundred kW that sell power to the local distribution network under distributed generation agreements. These are documented by lines companies and the Electricity Authority. The Electricity Authority’s EMI database records distributed generation by fuel type; the exact count of hydro installations at micro-hydro scale is uncertain because installations are reported in various size bands and some are classified with larger schemes. The skills census (Doc #8) is the appropriate mechanism to establish accurate national numbers.⁷

Off-grid farm systems: Private installations serving individual farms or small groups of properties not connected to the grid, or used as backup when the grid is unavailable. These are poorly documented nationally — many were installed decades ago by the landowner or a local contractor and have no formal consent or connection agreement. Some are still operating; others are decommissioned but may have usable equipment.

Demonstration and community systems: A smaller number of installations built for educational, community energy, or sustainability purposes. These include systems at polytechnics, tramping huts, and eco-villages.

1.3 Existing NZ suppliers and expertise

NZ has a small but established micro-hydro industry. Companies such as PowerSpout (based in Kaiwaka, Northland) design and manufacture micro-hydro turbines domestically — PowerSpout produces Pelton and Turgo turbines from locally sourced components including injection-moulded runners and permanent magnet generators.⁸ Other suppliers and installers operate in Canterbury, Otago, and the West Coast. The total number of experienced micro-hydro practitioners in NZ is probably in the low dozens — small enough that these individuals should be identified and their knowledge captured as a priority.

---

2. WHY MICRO-HYDRO MATTERS FOR RECOVERY

2.1 Grid resilience

The baseline scenario assumes the national grid continues operating (Doc #67). This is the most likely outcome for the grid’s backbone — major generation, the HVDC link, and main transmission lines. But the distribution network that delivers power to individual farms and communities is more vulnerable. NZ has approximately 30,000–50,000 distribution transformers (Doc #69), many serving small numbers of customers via long rural feeders. As these transformers age and fail — and replacement becomes progressively more difficult — some rural areas will lose grid supply.

Micro-hydro provides a hedge against this outcome. A community with its own generation maintains electrical capability independent of the distribution network. This is particularly relevant for:

• Remote farming communities at the end of long feeder lines, where a single transformer failure can leave an area without power
• West Coast South Island communities, where the distribution network traverses difficult terrain and is vulnerable to storm damage that may not be repaired promptly under reduced maintenance capacity
• Communities near NZ’s 25,000+ permanent streams⁹ that have the resource to generate their own power

2.2 Enabling new productive capacity

Micro-hydro can bring electricity to locations not currently served by the grid, enabling:

• New agricultural processing (milking, refrigeration, grain milling) in areas currently too remote for grid extension
• Workshop capability (welding, grinding, machining) at distributed sites — a 10 kW micro-hydro system can power a useful workshop
• Lighting and communication for communities that develop in locations chosen for agricultural productivity rather than grid access
• Water pumping for irrigation or domestic supply

2.3 Integration with the national system

Where micro-hydro sites are near functioning grid infrastructure, surplus generation can be fed back into the distribution network — a meaningful contribution, especially if many installations collectively add several MW of distributed generation. This is not a replacement for the national hydro system (approximately 5,300 MW — see Doc #65) but it supplements it at the local level and reduces the load that must be carried by major transmission and distribution infrastructure.

---

3. SITE ASSESSMENT

3.1 The fundamental equation

Micro-hydro power output depends on two variables: head (the vertical drop the water falls through) and flow rate (the volume of water per second). The theoretical power is:

P = ρ × g × H × Q

Where: - P = power (watts) - ρ = water density (1,000 kg/m³) - g = gravitational acceleration (9.81 m/s²) - H = net head (metres) — the vertical drop minus friction losses in the penstock - Q = flow rate (m³/s)

In practice, turbine and generator efficiency reduce the electrical output. A well-designed system achieves 50–70% overall efficiency (water-to-wire). A useful rule of thumb:

Electrical output (kW) ≈ 5 × H (metres) × Q (m³/s)

This assumes approximately 50% overall efficiency, which is conservative for a well-designed system and achievable for a locally fabricated one.¹⁰

Example: A stream with 30 metres of head and 50 litres per second (0.05 m³/s) of reliable flow produces approximately 5 × 30 × 0.05 = 7.5 kW of continuous electrical output. This is sufficient for a small farming community’s essential needs.

3.2 Measuring head

Head is measured between the proposed intake location and the proposed turbine location. Methods:

Level and staff: The most accurate method for moderate distances. A dumpy level (optical instrument) and graduated staff measure elevation changes over successive sightings. Many NZ farms have access to a dumpy level; surveying instruments will also be available from construction companies and territorial authorities.

Clinometer and tape: A clinometer (angle-measuring device) and a measured distance along the slope give head by trigonometry. Less accurate than a level survey but adequate for preliminary assessment.

GPS or barometric altimeter: Modern GPS receivers provide elevation data, though vertical accuracy is limited (typically ±3–10 metres). Adequate for identifying promising sites but not for final design.

Water-filled tube method: A long transparent tube filled with water acts as a simple level. The difference in water height at each end, accumulated over successive measurements, gives total head. Slow but requires no special instruments.

3.3 Measuring flow rate

Flow rate must be measured — not estimated from visual impression, which is unreliable. Methods:

Weir method: Build a temporary weir (a V-notch or rectangular notch in a board placed across the stream) and measure the depth of water flowing over it. Standard weir equations relate depth to flow rate. This is the most reliable method for small streams.¹¹

Velocity-area method: Measure the stream cross-section and the water velocity at multiple points. Velocity can be measured by timing a float over a known distance (multiply average surface velocity by a correction factor of 0.75–0.85 to approximate mean velocity across the section — the lower value for rough, shallow channels and the higher value for smooth, deep channels).¹² Less accurate than weir measurement but workable for larger streams.

Container and stopwatch: For very small flows, divert the entire stream into a container of known volume and time how long it takes to fill. Only practical for flows under about 10 litres per second.

3.4 Seasonal flow variation

This is critical. NZ streams vary significantly in flow between wet and dry seasons. A stream that provides 100 litres per second in winter may drop to 20 litres per second in a dry summer. The design flow for micro-hydro must be based on the flow that is reliably available for the majority of the year — typically the flow that is exceeded 70–90% of the time (Q70 or Q90), not the average or flood flow.¹³

Flow measurement over a full year is ideal. Under recovery conditions, this may not be possible before construction is needed. In that case:

• Measure flow during the driest period available (late summer in most NZ regions)
• Ask local residents about stream behaviour — farmers who have lived on the land for decades often know which streams are reliable
• Design conservatively — size the system for a flow lower than measured, to provide margin for drought years

3.5 NZ regional flow patterns

NZ’s rainfall distribution varies enormously by region:¹⁴

Region	Annual rainfall (mm)	Stream flow characteristics
West Coast, South Island	3,000–6,000+	Very high, year-round. Many excellent micro-hydro sites.
Fiordland	3,000–8,000	Extremely high rainfall but much terrain is very steep and remote.
Central North Island (Waikato, Bay of Plenty)	1,200–2,000	Moderate, seasonal. Good sites exist but require careful assessment.
Canterbury Plains	600–800	Low rainfall, but mountain-fed streams provide flow.
Northland	1,000–1,600	Moderate. Lower head available (gentler terrain) but many small streams.
Otago	400–1,500 (varies with elevation)	Highly variable. Mountain streams excellent; lowland streams marginal.
Taranaki	1,500–2,500	Ring of streams radiating from Mt Taranaki. Good micro-hydro potential.
Wairarapa	800–1,500	Moderate. Eastern hill country has suitable sites.

3.6 Nuclear winter effects on precipitation

Global nuclear winter modelling predicts significant changes to precipitation patterns, but the specific effect on NZ is uncertain. Possible outcomes include:¹⁵

• Reduced total precipitation if global cooling suppresses the hydrological cycle — this would reduce stream flows and micro-hydro output
• Increased precipitation in some regions if changes in atmospheric circulation bring more moisture to NZ’s western ranges
• Shift from rain to snow at moderate elevations as temperatures drop — this changes the timing of stream flow (more spring snowmelt, less winter rain-fed flow) but may not reduce annual total
• Greater inter-annual variability — some years wetter, some drier, with less predictable patterns

The practical implication is that micro-hydro design under nuclear winter conditions should incorporate more conservative flow assumptions and, where feasible, include small storage (a pond or weir) to buffer short-term flow variation. Sites with very small catchments are more vulnerable to precipitation changes than sites on larger streams fed by extensive mountain catchments.

---

4. TURBINE TYPES AND SELECTION

4.1 Overview

Turbine selection depends primarily on head and flow rate. NZ sites span a wide range of conditions, and different turbine types are suited to different combinations:

Turbine type	Head range (m)	Flow range (L/s)	Best for	NZ fabricability
Pelton	20–300+	1–50	High head, low flow — mountain streams	Good — can be cast or machined
Turgo	15–150	5–100	Medium-high head, moderate flow	Moderate — more complex runner shape
Crossflow (Banki-Mitchell)	3–100	20–1,000	Wide range of head and flow — very versatile	Excellent — fabricated from steel plate
Francis	5–50	50–5,000	Medium head, higher flow — larger streams	Poor — requires precision casting
Propeller / Kaplan	1–10	100–10,000+	Low head, high flow — river sites	Poor — requires precision casting and sealing

4.2 Pelton turbine

The Pelton turbine uses one or more high-velocity jets of water directed at cup-shaped buckets on a wheel. It is an impulse turbine — the water pressure is converted entirely to velocity before hitting the runner. Pelton turbines are highly efficient (80–90%) at high heads and low flows, and are the standard choice for mountain stream sites with 50+ metres of head.¹⁶

NZ applications: NZ’s mountainous terrain, particularly in the South Island, provides many high-head sites. Pelton turbines are already the most common type in NZ’s existing micro-hydro installations.

Local fabrication: Pelton runners can be fabricated by: - Casting (Doc #93) in bronze or steel — the bucket shape is curved but not geometrically complex. Pattern-making is the main skill required. Sand casting is adequate for small turbines. - Machining from solid steel billets — wasteful of material but produces accurate bucket shapes. Within NZ machine shop capability for small runners (up to ~300 mm diameter). - Welding individual buckets to a hub — less efficient than cast or machined runners but functional. Each bucket can be pressed or forged from steel plate and welded to a disc.

The nozzle (which forms the jet) requires precise machining — a poorly made nozzle wastes energy. Adjustable spear valves for flow control are machined components well within NZ capability.

4.3 Turgo turbine

The Turgo is a variant of the Pelton that uses a different bucket shape, allowing the water jet to enter from one side and exit from the other. This permits a smaller runner diameter for the same output, allowing direct coupling to a higher-speed generator without a gearbox. Turgo turbines are effective at medium-high heads (15–150 m) and tolerate slightly higher flows than Pelton turbines of the same size.

NZ applications: PowerSpout manufactures Turgo turbines domestically, demonstrating that the technology is within NZ’s manufacturing capability.¹⁷

Local fabrication: More complex runner geometry than Pelton — the cups are asymmetric. Casting or 5-axis machining is ideal but not essential; approximations welded from steel plate are less efficient but functional. A welded Turgo runner can reasonably be expected to achieve 55–70% hydraulic efficiency versus 75–85% for a cast or machined commercial unit — a meaningful penalty that reduces electrical output by 10–20% compared with an equivalent commercial unit on the same site. If fabrication capability allows, a crossflow turbine (Section 4.4) is more reliably fabricated to consistent performance than a welded Turgo approximation.

4.4 Crossflow (Banki-Mitchell) turbine

The crossflow turbine is the most important type for NZ’s recovery because of its simplicity of fabrication, wide operating range, and tolerance of varying flow conditions. Water enters through a rectangular inlet, passes through curved blades on the outer rim of a cylindrical runner, crosses the open interior, and passes through blades on the opposite side before exiting. The runner is essentially a drum made of curved steel blades welded between two circular end plates.

Efficiency: 70–85% for well-manufactured commercial units, somewhat lower than Francis turbines at their optimal operating points but higher than what is achievable from locally fabricated units (see Section 13.3). Crossflow turbines maintain acceptable efficiency across a wider range of flow rates than Pelton or Francis turbines — an important practical advantage for NZ streams with seasonal variation.¹⁸

Operating range: Head of 3–100 metres, flow of 20–1,000+ litres per second. This covers the majority of NZ micro-hydro sites.

Local fabrication — the key advantage: A crossflow turbine can be fabricated entirely from: - Mild steel plate (available from NZ Steel, Doc #89) — for the casing, end plates, and inlet guide vane - Mild steel strip or bar — for the runner blades, which are cut, bent to a specified radius, and welded between end discs - Shaft and bearings — standard steel shaft, commercially available bearings (or plain bearings fabricated in a machine shop)

The fabrication process requires: - Cutting and bending steel plate (shear, press brake, or hand work) - Welding (MIG, TIG, or stick welding — all available in NZ workshops) - Basic machining (lathe work for the shaft and bearing housings) - Balancing the runner (important for smooth operation and bearing life)

A competent fabrication shop (Doc #91) can produce a crossflow turbine in 2–4 person-weeks. The design is well-documented in public-domain sources, including the original work by Banki and subsequent development by VITA (Volunteers in Technical Assistance) and various development-agency programmes.¹⁹

Critical dimensions: Runner blade angle (typically 30° to the tangent), blade spacing, inlet width relative to runner length, and runner diameter all affect performance. These parameters are calculable from head and flow data using standard design equations. The design is forgiving — a turbine with imperfect dimensions still works, though at somewhat lower efficiency.

4.5 Francis turbine

The Francis turbine is a reaction turbine where the runner is fully immersed in water. It is the dominant type in NZ’s large hydro stations (Doc #65) but at micro-hydro scale it presents manufacturing difficulties — the runner has complex three-dimensional blade shapes that require precision casting and finishing. Francis turbines are more efficient than crossflow turbines at their design point (85–93%) but less tolerant of off-design flow conditions.²⁰

NZ recovery applications: Existing manufactured Francis turbines from pre-war stocks should be used where available. Local fabrication of Francis runners is not realistic with NZ’s current workshop capability — the blade geometry is too complex for welded fabrication and requires investment casting or precision machining beyond what is practical for micro-hydro scale.²¹

4.6 Propeller and Kaplan turbines

Propeller turbines (fixed-blade) and Kaplan turbines (adjustable-blade) operate at low heads (1–10 m) and high flows. They are suited to river sites where a weir or low dam creates a modest head across a large flow. The Kaplan’s adjustable blades maintain efficiency across varying flow, but the mechanism is complex.

NZ recovery applications: Low-head sites are less common in NZ than medium- and high-head sites, because NZ’s terrain generally provides natural head. Where low-head sites exist (river weirs, existing irrigation drop structures), propeller turbines are effective but difficult to fabricate locally — the blade profile requires hydrodynamic shaping and precision balancing. Salvaged propeller turbines from existing installations are the most practical option.

4.7 Turbine selection for NZ conditions

For new installations under recovery conditions, the recommended approach is:

1. High head (>50 m), low flow (<50 L/s): Pelton turbine — locally castable or machinable
2. Medium head (10–50 m), moderate flow (20–200 L/s): Crossflow turbine — locally fabricable from steel plate. This is the most common NZ scenario and the highest priority for local manufacturing.
3. Medium-high head (15–150 m), moderate flow: Turgo turbine where commercially available units exist
4. Low head (<10 m), high flow (>200 L/s): Propeller turbine from salvage if available; otherwise consider a crossflow at reduced efficiency or a weir to increase effective head
5. Very low head (<3 m): Generally not economical for micro-hydro unless flow is very large. Consider water-powered mechanical drives (mills, pumps) instead of electrical generation.

---

5. PENSTOCK DESIGN AND CONSTRUCTION

5.1 Function

The penstock is the pipe that carries water under pressure from the intake to the turbine. It converts the potential energy of head into pressure that drives the turbine. Penstock design must balance cost (longer and larger pipes cost more) against efficiency (friction losses in the pipe reduce effective head and therefore output).

5.2 Design principles

Diameter selection: The pipe must be large enough that friction losses do not consume a disproportionate share of the available head. A common design target is that friction losses should not exceed 5–10% of gross head. For a given flow rate, this determines the minimum pipe diameter. Standard hydraulic flow equations (Darcy-Weisbach or Hazen-Williams) provide the relationship — tables for common pipe types are included in engineering references (Doc #17).²²

Rule of thumb for PVC pipe: For flows of 10–100 litres per second and pipe lengths of 50–500 metres, typical penstock diameters range from 100 mm to 300 mm. Oversizing the pipe is always preferable to undersizing — a slightly larger pipe has lower friction losses and therefore produces more power, while the material cost difference is modest.

Pressure rating: The pipe must withstand the static pressure of the water column (head × 9.81 kPa per metre) plus surge pressure from water hammer (pressure spikes caused by rapid flow changes, such as sudden valve closure). A safety factor of 1.5–2.0 on static pressure is standard practice. For a 50-metre head site, the static pressure is approximately 490 kPa (71 psi) — well within the rating of standard PVC pressure pipe.²³

Water hammer: Rapid changes in flow (closing a valve quickly, sudden turbine load change) create pressure waves that travel up the penstock at the speed of sound in the pipe material — approximately 300–500 m/s in PVC (depending on pipe diameter and wall thickness), 900–1,200 m/s in steel. Surge pressures can exceed static pressure by a factor of 2–5 if valves are closed too quickly. Mitigation: close valves slowly (minimum closure time proportional to pipe length and wave speed), install a pressure relief valve at the turbine, and ensure pipe pressure rating includes surge margin.²⁴

5.3 PVC and polyethylene pipe

PVC pressure pipe is the preferred penstock material for micro-hydro in NZ, and it was the default choice for the majority of existing installations. NZ has substantial stocks of PVC pipe in plumbing and irrigation supply chains — distributed across rural NZ through merchants such as Placemakers, Mitre 10 Trade, and irrigation suppliers. Major NZ pipe distributors (including Iplex Pipelines and Marley New Zealand) imported and warehoused significant volumes pre-war.²⁵

Advantages: Light weight (easy to transport and install on steep terrain), smooth bore (low friction losses), corrosion-resistant, available in diameters from 50 mm to 500+ mm in various pressure ratings, and joined with solvent cement (no welding required).

Disadvantages: PVC degrades under prolonged UV exposure (must be buried or shaded), becomes brittle at very low temperatures (relevant under nuclear winter), and cannot be manufactured in NZ. Existing stocks are finite — once depleted, alternative penstock materials are needed.

Polyethylene (PE) pipe is an alternative, available in NZ in agricultural irrigation sizes. PE is more flexible than PVC, tolerates freeze-thaw better, but has higher friction losses per unit diameter and is available in fewer pressure ratings.

5.4 Steel pipe — the long-term solution

As PVC stocks are depleted, steel pipe becomes the sustainable penstock material. NZ Steel (Doc #89) produces steel plate and coil that can be formed into pipe:

Fabrication methods: - Rolled and welded pipe: Steel plate is rolled into a cylinder and seam-welded. This is standard practice in NZ fabrication shops. Pipe diameters of 100–500 mm are achievable. - Spiral-welded pipe: Steel strip is wound helically and welded along the spiral seam. More material-efficient than rolled pipe but requires specialised equipment.

Corrosion protection: Bare steel pipe corrodes in contact with water and soil. Protection options include: - Bitumen or tar coating (producible from NZ petroleum or coal tar) - Linseed oil-based paint - Concrete lining (for larger diameters) - Sacrificial zinc anodes (while zinc stocks last)

Steel penstock has higher friction losses than PVC (rougher internal surface) and is significantly heavier to transport and install. These are real disadvantages, partially offset by the ability to fabricate pipe in any diameter to match site requirements.

5.5 Historical alternatives — wooden flumes and staves

Before steel and plastic pipe were readily available, NZ micro-hydro installations used timber flumes (open channels) and stave pipes (wooden barrels). These are low-pressure options suitable only for low-head sites or the open-channel portion of a water conveyance.²⁶

Timber flume: An open rectangular or V-shaped channel constructed from sawn timber, supported on trestles across terrain. Suitable for conveying water with minimal head loss over moderate distances. Not suitable for pressurised penstock applications. NZ’s radiata pine and native timbers (totara, macrocarpa) are suitable for flume construction; durability is improved by charring, oil treatment, or use of naturally durable species.

Stave pipe: A cylindrical pipe assembled from wooden staves, bound with steel bands — essentially a barrel in pipe form. Can tolerate moderate pressures (up to 10–20 metres of head depending on construction quality). Historical technology that pre-dates metal and plastic pipe. Labour-intensive to construct but uses NZ-available materials.

These options are significantly inferior to PVC or steel: timber flumes lose water to leakage and evaporation (10–30% losses versus under 1% for pipe), require frequent structural maintenance (annual retimbering of damaged sections), and are limited to low pressures. Stave pipe tolerates only 10–20 metres of head versus 60–120+ metres for standard PVC pressure pipe. Both have service lives of 5–15 years versus 50+ years for PVC or properly maintained steel. They may nonetheless be relevant for sites where pipe is unavailable and the head is low enough for wooden construction.

---

6. GENERATORS

6.1 Options

The generator converts the turbine’s rotational energy into electricity. Options under recovery conditions:

Commercially manufactured generators: Pre-war stocks of purpose-built micro-hydro generators (permanent magnet alternators, induction generators) are the best option where available. PowerSpout and other NZ suppliers hold some stock; additional units may be sourced from electrical wholesalers (Rexel, Ideal Electrical), rural suppliers (PGG Wrightson, RD1), and decommissioned dairy shed and irrigation installations.

Rewound induction motors used as generators: An induction motor operated above synchronous speed by a turbine generates electricity. Three-phase induction motors are extremely common in NZ — every milking shed, pump station, and workshop has them. When driven by a turbine at slightly above synchronous speed (typically 1,500 or 3,000 rpm for a 50 Hz motor) and provided with a source of reactive power (capacitor bank), an induction motor becomes a generator. This is a well-established technique for micro-hydro.²⁷

The required capacitor bank size is approximately 30–50% of the motor’s rated kVA (as capacitive kVAR). Standard power factor correction capacitors, available from electrical wholesalers, serve this purpose. Motor rewinding to optimise performance as a generator is possible (Doc #95) — specifically, rewinding with fewer turns of larger wire to reduce stator resistance and improve efficiency at generator loading patterns.

Vehicle alternators: Car and truck alternators produce DC at 12 V or 24 V. They can be used for micro-hydro, particularly for small systems (under 1 kW) serving battery charging. Limitations: relatively low efficiency (50–65%), low voltage (requires inverter or is used for DC loads only), limited power output per unit (typically 0.5–2 kW), and brush and bearing wear requires regular maintenance.²⁸ Useful as a stopgap but not the preferred solution for community-scale systems.

Purpose-built generators from NZ materials: In Phase 3–4, as copper wire production develops (Doc #70) and winding skills expand (Doc #69, Doc #95), purpose-built generators can be manufactured in NZ. The dependency chain is substantial: copper wire must be drawn to the specified gauge (Doc #70), a winding jig fabricated to ensure consistent coil geometry, stator laminations punched or hand-cut from electrical-grade steel (or salvaged from scrapped motors), windings impregnated with varnish or epoxy resin for moisture resistance (Doc #116 for resin precursors), and the assembled stator tested for insulation resistance before installation. A permanent magnet alternator requires neodymium magnets salvaged from hard drives, speakers, and wind turbines — or ferrite magnets, which are weaker but can be produced from barium or strontium carbonate and iron oxide by a ceramic process (a Phase 4–5 development). Wound-field designs avoid rare-earth magnets but require slip rings and a small DC excitation supply, adding mechanical complexity. Neither option is a Phase 1–2 undertaking; the preferred near-term approach is the rewound induction motor.

6.2 Speed matching

Turbines and generators must operate at compatible speeds. Pelton and Turgo turbines typically run at 500–1,500 rpm depending on jet velocity and runner diameter. Crossflow turbines run at lower speeds, typically 100–500 rpm. Standard induction motors are designed for 1,500 or 3,000 rpm (4-pole or 2-pole, 50 Hz).

Speed matching options: - Belt drive: V-belts or flat belts provide flexible speed ratios and absorb minor misalignment. Available throughout NZ from farm supply chains (PGG Wrightson, RD1, agricultural merchants). Requires periodic retensioning and belt replacement (belts are a consumable). Some energy loss (2–5%). - Gearbox: More efficient than belts but heavier and more expensive. Salvaged industrial gearboxes are available. - Direct drive: If the turbine speed matches the generator speed, direct coupling eliminates transmission losses. This is achievable by selecting runner diameter to produce the required speed — possible with Pelton and Turgo turbines, more difficult with crossflow.

6.3 Voltage and frequency

For standalone systems (not grid-connected), the generator must produce voltage and frequency suitable for connected loads. NZ standard is 230 V single-phase at 50 Hz (400 V three-phase).

Frequency is set by generator speed — a 4-pole generator must rotate at exactly 1,500 rpm to produce 50 Hz. Speed regulation by the turbine governor or electronic load controller (Section 7) maintains frequency.

Voltage is set by the generator’s magnetic field strength and load. For induction generators, the capacitor bank size and load determine voltage. For permanent magnet generators, voltage is proportional to speed. Voltage regulation is typically achieved through the electronic load controller.

---

7. CONTROL SYSTEMS

7.1 The electronic load controller (ELC)

The ELC is the most reliable method of controlling a micro-hydro system, and preferable to a mechanical governor for micro-hydro scale. Its operating principle is that the turbine and generator run at constant speed (and therefore constant power output) regardless of how much useful load is connected. The ELC monitors the generator frequency and diverts any surplus power — the difference between total generated power and useful load — to a ballast load, typically a water heater or resistive heating element.²⁹

How it works: If the useful load decreases (someone turns off a light), generator speed would increase. The ELC detects the incipient frequency rise and diverts more power to the ballast, maintaining constant total load and therefore constant speed. If useful load increases, the ELC reduces ballast power. The ballast power is not wasted — a water heating element provides hot water as a free by-product.

Why this is better than mechanical governors: Mechanical governors (which control water flow to the turbine in response to speed changes) are complex, expensive, and have slow response times. They are standard on large hydro turbines but unnecessarily complex for micro-hydro. An ELC responds in milliseconds, has no moving parts, and can be assembled from standard discrete electronic components available from NZ distributors or salvage.

7.2 ELC construction

An ELC circuit consists of:

• Frequency sensing: A zero-crossing detector on the generator voltage waveform
• Phase-angle control: A triac or thyristor that varies the portion of each AC cycle delivered to the ballast load
• Control logic: Compares measured frequency with the 50 Hz reference and adjusts the phase angle to maintain constant frequency

The total component count is small — a microcontroller (Arduino or equivalent from salvage stocks) simplifies the design to a single programmed chip plus power electronics. Note that programming a microcontroller requires a functional computer, a USB cable, and the Arduino IDE software; once programmed, the chip operates standalone. Analog designs using discrete components (operational amplifiers, comparators, and timing circuits) are also well-documented and do not require a microcontroller or any programming capability — they are preferred as electronic stocks deplete and programming tools become unavailable.³⁰

Power electronics: The triac or thyristor must be rated for the full ballast power — typically 5–30 kW for a micro-hydro system. These components are available from NZ electronics distributors (RS Components, Element14, Jaycar) and are also found in industrial equipment, large dimmer switches, and motor drives. Triacs and thyristors are semiconductor devices that NZ cannot manufacture; the supply is limited to existing distributor stock, import stockpiles, and salvage from industrial control gear. For a 10 kW system, a single triac rated at 40–50 A (e.g., BTA41 or equivalent) is required, plus a heat sink, snubber circuit components (resistor and capacitor), and an optically isolated trigger circuit.³¹

An experienced electronics technician can build an ELC in 1–3 days from salvaged components. The dependency chain for ELC production includes: power semiconductor (triac or thyristor from salvage), passive components (resistors, capacitors from salvage), a control element (microcontroller from salvage stocks, or discrete analog components — operational amplifiers and comparators), a heat sink (aluminium, fabricable from NZ stock), a ballast load (resistive water heater element, widely available), and an enclosure (fabricable from steel or timber). Of these, the power semiconductor and control electronics are the binding constraints — all other components are locally producible or abundantly available. Standardised designs should be documented and construction kits prepared for distribution to micro-hydro installation teams.

7.3 Grid-connected vs. standalone operation

Grid-connected micro-hydro uses the national grid as an “infinite bus” — the grid sets the frequency and voltage, and the micro-hydro generator feeds power into it. This is the least complex control arrangement because the grid handles all regulation. Surplus power flows to the grid; deficit is supplied by the grid. No ELC is needed, though a grid-tie inverter or synchronisation relay is required.

Standalone micro-hydro must regulate its own frequency and voltage. The ELC (Section 7.1) provides this function. Standalone systems can also incorporate battery storage (charged from surplus generation) to handle transient loads that exceed the turbine’s output capacity — motor starting surges are the most common example.

Transition planning: Some existing grid-connected micro-hydro installations may need to transition to standalone operation if the local distribution grid fails. This requires adding an ELC and possibly a battery bank — modifications that should be planned in advance for installations in grid-vulnerable areas.

---

8. INTAKE DESIGN AND WATER MANAGEMENT

8.1 Intake structure

The intake diverts water from the stream into the penstock. It must:

• Capture the required flow without obstructing the stream excessively
• Exclude debris (leaves, branches, stones, sediment) that would damage the turbine
• Allow fish passage (Section 11)
• Survive floods without destruction

Typical intake design: A low weir across the stream raises the water level to the intake opening on one bank. A coarse trash rack (steel bars spaced 25–50 mm apart) excludes large debris. A settling basin behind the rack allows sand and gravel to settle before water enters the penstock. A fine screen (5–10 mm mesh) at the penstock entrance excludes smaller debris.

Materials: Concrete, stone masonry, timber, or gabion baskets (wire cages filled with rocks). All can be constructed from NZ-available materials. Steel reinforcing bar is available from NZ Steel (Doc #89).

8.2 Trash rack management

Screen and rack clearing is the single most common maintenance task on a micro-hydro system. In NZ streams — particularly in native bush catchments — leaf litter, twigs, and debris accumulate on the intake screen, reducing flow and therefore power output. During floods, large debris loads can block the intake entirely.

Approach: - Visit the intake regularly (daily during autumn leaf-fall, weekly at other times) - Design the trash rack for manual clearing — a sloping rack allows debris to be raked off rather than pulled through, reducing the effort and time required per clearing visit - Install a coanda screen where the intake design allows it — a finely profiled curved screen that uses the boundary-layer effect to shed debris while passing clean water. More complex to fabricate but significantly reduces maintenance.³² - Consider an automatic flushing intake that periodically opens a gate to flush accumulated sediment

8.3 Flood protection

NZ streams can rise dramatically during heavy rainfall — peak flood flows in steep mountain catchments may be 50–100 times median flow, with smaller, steeper catchments subject to the greatest relative variation.³³ The intake must survive this without being destroyed or accumulating unmanageable debris loads.

Design features: - Set the intake to one side of the stream, not in the main flood channel - Include a bypass or overflow to divert excess water around the intake - Use robust construction — concrete or stone masonry rather than timber at vulnerable points - Accept that the intake may be temporarily overwhelmed in major floods, requiring cleanup and minor repair

---

9. POWERHOUSE AND INSTALLATION

9.1 Powerhouse construction

The powerhouse is a simple structure housing the turbine, generator, and controls. Requirements:

• Dry and ventilated — generators and electrical controls must be protected from water and excessive moisture
• Flood-safe — located above the maximum expected flood level. This is non-negotiable — a flooded powerhouse destroys the generator and controls.
• Accessible for maintenance — the turbine and generator must be accessible for bearing replacement, cleaning, and repair
• Noise management — turbines and generators produce noise (40–70 dB at 1 metre, depending on size, type, and rotational speed — crossflow turbines are typically noisier than Pelton turbines of equivalent output due to the water impact pattern).³⁴ Locate the powerhouse away from residences where possible, or insulate the structure.

Construction: Concrete block, timber frame, or corrugated steel. NZ building practices are adequate. A 10 kW powerhouse is approximately the size of a small garden shed (3 m × 4 m).

9.2 Tailrace

Water exiting the turbine must be discharged back to the stream via a tailrace — an open channel or pipe. The tailrace should be sized to prevent backpressure on the turbine (which reduces output) and should return water to the stream at a point below the powerhouse flood level.

---

10. MAINTENANCE

10.1 Routine maintenance schedule

Micro-hydro systems are low-maintenance compared to any combustion-based generation, but they are not zero-maintenance.

Task	Frequency	Skills required
Intake screen clearing	Daily to weekly (seasonal)	None — any competent person
Visual inspection of penstock	Monthly	General observation
Bearing lubrication	Monthly to quarterly	Basic mechanical
Belt tension check (if belt-driven)	Monthly	Basic mechanical
ELC function check	Quarterly	Basic electrical
Turbine inspection (wear, cavitation)	Annually	Mechanical/welding
Generator insulation check	Annually	Electrical
Penstock pressure test	Every 2–5 years	Mechanical
Bearing replacement	Every 3–10 years	Mechanical
Full system overhaul	Every 10–20 years	Mechanical and electrical

10.2 Common failure modes

Bearing failure: The most common mechanical failure. Caused by inadequate lubrication, water ingress, or misalignment. Prevention: regular greasing with NZ-produced tallow-based grease (Doc #34 — note that tallow grease has a lower drop point than petroleum-based grease, washes out more readily in wet conditions, and oxidises faster, requiring reapplication approximately 2–3 times more frequently than petroleum grease; in the wet environment of a micro-hydro powerhouse, monthly or even fortnightly greasing may be necessary versus quarterly for petroleum grease); sealed bearings where available; proper alignment during installation.

Penstock damage: PVC pipe can crack from impact, UV degradation (if exposed), or freeze damage under nuclear winter temperatures. Steel pipe corrodes if coating fails. Prevention: bury PVC pipe, maintain corrosion protection on steel pipe, inspect regularly.

Generator winding failure: Insulation breakdown due to moisture, overheating, or age. Prevention: keep the powerhouse dry and ventilated; do not exceed rated output; monitor winding temperature. Repair: motor rewinding shops (Doc #95) can rewind micro-hydro generators.

Debris blockage: The most frequent operational issue. Not a failure per se but reduces output and requires regular intervention. Good intake design (Section 8) minimises this.

Water hammer damage: Caused by rapid valve closure or sudden load changes. Prevention: close valves slowly; ensure ELC functions correctly (sudden full-load rejection is the worst case for water hammer).

---

11. ENVIRONMENTAL AND LEGAL CONSIDERATIONS

11.1 Fish passage

NZ’s streams support native fish species — many of which are migratory, spending part of their life cycle in the sea and part in freshwater. Eels (tuna), whitebait species (galaxiids), and others move up and down streams. A micro-hydro intake that blocks fish passage can damage local fish populations.

Mitigation: - Install fish screens at the intake — mesh size small enough to prevent fish from entering the penstock (typically 3–5 mm for juvenile fish)³⁵ - Maintain minimum flow past the intake — do not divert the entire stream through the penstock. A common requirement is to leave at least 50–70% of the natural low flow in the stream. - Provide fish passage past any weir or dam — a simple bypass channel or fish ladder for low weirs - Time construction to avoid peak fish migration seasons (spring for whitebait, autumn for adult eel migration)

11.2 Minimum flow requirements

Diverting water from a stream for micro-hydro reduces the flow available for downstream aquatic ecosystems. Maintaining a minimum residual flow is both an ecological necessity and — under normal legal conditions — a regulatory requirement.

Practical guidance: Design the system to take no more than 30–50% of the stream’s low flow. This ensures that the stream continues to function as a habitat and that downstream water users (if any) are not unduly affected. More aggressive diversion is possible on larger streams where the proportional take is smaller.³⁶ Regional council plans vary in their minimum flow requirements — some require up to 90% residual flow for sensitive waterways. Check the relevant regional plan where the legal framework permits.

11.3 Legal framework — normal and emergency conditions

Normal conditions: Under the Resource Management Act 1991 (RMA), taking water from a stream and damming or diverting a waterway requires a resource consent from the relevant regional council. Consenting processes can take months to years and may impose conditions on flow, fish passage, construction methods, and environmental monitoring. Small takes below regional plan thresholds may be permitted activities not requiring consent, but this varies by region.

Under emergency conditions: The Civil Defence Emergency Management Act 2002 (CDEM Act) provides for emergency powers that may override normal RMA requirements. Whether micro-hydro construction requires resource consent under declared emergency conditions is a legal question that should be resolved early — ideally in Phase 1, before construction begins. The practical recommendation is:

• For installations on private land using water that is not needed for other purposes, proceed with construction under emergency authority, documenting the installation and its environmental provisions (fish passage, minimum flow)
• For installations that affect other water users or environmentally sensitive waterways, seek rapid consent or exemption through the regional council (which will continue to function, albeit under emergency conditions)
• In all cases, follow the environmental provisions in this document (fish passage, minimum flow) regardless of whether formal consent is obtained — these provisions protect the resource for long-term use, not just regulatory compliance

11.4 Sediment management

Construction near waterways generates sediment that can damage downstream aquatic habitat. Standard NZ erosion and sediment control practices apply:

• Minimise disturbance of stream banks and bed
• Install sediment control (silt fences, settling ponds) during construction
• Time construction to avoid heavy rainfall where possible
• Revegetate disturbed areas promptly

---

12. CONSTRUCTION SEQUENCE — STEP-BY-STEP

12.1 Phase 1: Site assessment and design

1. Identify candidate site — stream with adequate flow and accessible head
2. Measure head — using level and staff, clinometer, or water tube method
3. Measure flow — using weir, velocity-area, or container method. Measure at the driest season available.
4. Survey the route — identify the intake location, penstock route, and powerhouse location
5. Select turbine type — based on head and flow (Section 4.7)
6. Size penstock — diameter based on flow rate and acceptable friction loss (Section 5.2)
7. Select generator — based on turbine output and available equipment (Section 6)
8. Prepare design documentation — drawings, material lists, construction plan

12.2 Phase 2: Materials procurement

9. Source penstock pipe — PVC or PE from stocks; fabricate steel pipe if required
10. Source or fabricate turbine — commercial unit from stock, or workshop fabrication (crossflow)
11. Source or adapt generator — induction motor and capacitors, or purpose-built generator
12. Source ELC components — from electronic component stocks or salvage
13. Procure civil works materials — concrete, timber, steel reinforcing, aggregate

12.3 Phase 3: Civil works

14. Construct intake — weir, trash rack, settling basin, penstock connection
15. Install penstock — trench, lay pipe, backfill. For steel pipe: fabricate, coat, install.
16. Construct powerhouse — foundations, floor, walls, roof, drainage
17. Construct tailrace — discharge channel back to stream

12.4 Phase 4: Mechanical and electrical installation

18. Install turbine — mount on foundations, connect penstock, connect tailrace
19. Install generator — mount, align with turbine, connect drive (direct, belt, or gearbox)
20. Install ELC — mount, wire to generator and ballast load (water heater)
21. Install switchgear and distribution — circuit breakers, wiring to loads
22. Install earthing system — essential for electrical safety. Earth rod driven into moist ground near the powerhouse, connected to all metal parts and the generator neutral.

12.5 Phase 5: Commissioning

23. Test without water — run generator with no water to check rotation direction, belt alignment, wiring
24. Introduce water gradually — open intake valve slowly, check for leaks in penstock and turbine
25. Check no-load speed — generator speed with ELC ballast absorbing all power
26. Apply load progressively — connect useful loads one at a time, confirming voltage and frequency stability
27. Confirm ELC operation — verify that load changes do not cause frequency excursions beyond ±2 Hz
28. Document operating parameters — head, flow, power output, voltage, frequency, bearing temperatures

---

13. LOCALLY MANUFACTURED CROSSFLOW TURBINE — FABRICATION GUIDE

13.1 Design parameters

For a specific example: a crossflow turbine for a site with 20 metres of head and 80 litres per second of flow.

Theoretical power: 5 × 20 × 0.08 = 8 kW (at 50% assumed efficiency)

Runner specifications: - Runner diameter: 200–300 mm (typical for this scale) - Runner length: 200–400 mm (wider runner for higher flow) - Number of blades: 18–24 (equally spaced around the circumference) - Blade curvature radius: approximately runner diameter × 0.326³⁷ - Blade entry angle: 30° to the tangent of the runner circumference - Blade material: 3–4 mm mild steel strip

Casing specifications: - Rectangular steel casing, 4–6 mm plate - Inlet nozzle width equal to or less than runner length - Nozzle angle set to direct water tangentially at the runner

13.2 Fabrication steps

1. Cut two end discs from 6–8 mm steel plate. Bore centre hole for shaft.
2. Mark blade positions on end discs — equally spaced around the circumference at the specified diameter.
3. Cut blade blanks from 3–4 mm steel strip. Each blank is cut to a length slightly less than the runner length and bent to the calculated radius using a pipe or mandrel.
4. Weld blades between end discs. This is the most skill-demanding step — each blade must be positioned at the correct angle and radius, tack-welded, checked, and then fully welded. Jigs are essential for consistency.
5. Machine the shaft — mild steel bar, turned to fit standard bearings. Keyway cut for locking to the runner.
6. Assemble runner on shaft. Key and set-screw. Check for balance — spin by hand and correct any heavy spot by adding or removing small amounts of weld.
7. Fabricate casing from steel plate — cut, bend, weld. The casing forms the nozzle that directs water onto the runner.
8. Install bearing housings on casing sides. Use standard deep-groove ball bearings (6205, 6206, or similar — common sizes available from bearing suppliers) with seals to prevent water ingress.
9. Install shaft and runner in casing. Check free rotation, alignment, and bearing fit.
10. Fabricate inlet guide vane — an adjustable plate inside the nozzle that controls flow rate. Pivots on a steel pin; can be adjusted manually or by a simple mechanical linkage.

13.3 Quality and performance

A well-fabricated crossflow turbine of this type should achieve 60–75% hydraulic efficiency. This is lower than a commercial turbine (70–85%) but entirely adequate for micro-hydro — the practical consequence of reduced efficiency is that the system produces less power from the same water flow, which at micro-hydro scale means a modest reduction in output (perhaps 1–3 kW on the example site above) rather than an economic problem.³⁸

The main factors affecting performance of a locally fabricated unit:

• Blade angle accuracy: Variations of ±3° from the specified 30° entry angle reduce efficiency by 2–5%. Careful jigging during fabrication controls this.
• Internal surface finish: Rough welds and sharp edges inside the casing create turbulence. Grind smooth where accessible.
• Seal quality at shaft penetrations: Water leaking past shaft seals is wasted energy. Standard lip seals or gland packing provides adequate sealing.
• Runner balance: An unbalanced runner creates vibration that accelerates bearing wear. Dynamic balancing is ideal but static balancing (checking for heavy spots by supporting the runner on knife edges and marking the heavy side) is usually adequate for micro-hydro.

---

14. CRITICAL UNCERTAINTIES

Uncertainty	Impact	Mitigation
Nuclear winter precipitation changes	Stream flows may decrease or become more variable, reducing micro-hydro output	Conservative flow-based design. Measure actual flow before committing. Include storage (pond) where feasible.
PVC pipe stock depletion	The preferred penstock material is finite and not locally producible	Inventory national PVC pipe stocks early. Develop steel pipe fabrication capability. Reserve PVC for micro-hydro priority use.
Bearing supply	Standard bearings are not manufactured in NZ and existing stocks will deplete	Plain bearing fabrication from NZ bronze is feasible (Doc #91). Sealed bearings last 3–10 years; plain bearings require more frequent maintenance but are indefinitely producible.
Rare-earth magnet supply for permanent magnet generators	Finite stock, not locally producible	Design for induction generators (rewound motors) as the standard approach. Reserve permanent magnets for small high-value applications.
Skilled workforce	NZ has few experienced micro-hydro installers	Immediate knowledge capture. Training programme. Documented designs enable less-experienced fabricators.
Legal framework under emergency	Uncertainty about whether RMA consents are required	Resolve early in Phase 1. Default to environmentally responsible practices regardless.
Crossflow turbine fabrication quality	First locally fabricated turbines may have lower efficiency than commercial units	Accept reduced efficiency initially. Performance improves with fabrication experience. Even a 50%-efficient system produces useful power.
Generator availability	Stock of suitable motors and generators is finite	Develop rewinding capability (Doc #95). Purpose-built generator fabrication as a Phase 3–4 goal (Doc #70 for copper wire).
Freeze damage to penstocks and intakes	Nuclear winter temperatures may cause freezing in areas that normally do not freeze	Bury PVC penstocks below frost depth. Insulate exposed pipe. Design intakes to tolerate ice formation.

---

15. CROSS-REFERENCES

Document	Relevance to micro-hydro
Doc #65 — Hydroelectric Station Maintenance	NZ’s major hydro system. Micro-hydro supplements, does not replace, national generation. Common turbine engineering principles apply.
Doc #67 — Transpower Grid Operations	Grid architecture and failure modes. Micro-hydro provides resilience when distribution degrades. Grid-connected micro-hydro feeds surplus power to the network.
Doc #69 — Transformer Rewinding and Fabrication	Distribution transformer failure is the trigger for micro-hydro becoming necessary in some areas.
Doc #70 — Copper Wire Production	Generator windings and distribution wiring require copper. Recycled copper is the near-term source; NZ wire drawing is a Phase 2–3 development.
Doc #89 — NZ Steel Glenbrook	Steel plate and bar for crossflow turbine fabrication, steel penstock pipe, reinforcing.
Doc #91 — Machine Shop Operations	The capability base for turbine fabrication. Crossflow turbine construction requires cutting, welding, and basic lathe work.
Doc #93 — Foundry and Casting	Pelton turbine runner casting. Bronze casting for bearings and nozzles.
Doc #95 — Electric Motor Rewinding	Induction motors adapted as generators. Rewinding to optimise generator performance.
Doc #34 — Lubricant Production	Bearing lubrication. Tallow-based grease is a functional but inferior substitute for petroleum grease.
Doc #94 — Welding Consumables	Welding electrodes for steel fabrication — turbine construction, penstock fabrication, intake construction.
Doc #97 — Cement and Concrete	Intake construction, powerhouse foundations, weirs.
Doc #105 — Wire and Fencing	Wire for electrical distribution from the micro-hydro to connected loads.
Doc #73 — Solar Panel Maintenance	Complementary distributed generation technology. Solar is unreliable under nuclear winter; micro-hydro provides baseload.
Doc #157 — Trade Training	Training pipeline for micro-hydro technicians — mechanical, electrical, and civil skills.
Doc #156 — Skills Census	Identifying existing micro-hydro installations, equipment stocks, and experienced practitioners.

---

FOOTNOTES

---

1. NZ’s history of small hydro is documented in various local and regional histories. Before the national grid reached rural areas, hundreds of small water turbines provided farm and community electricity across both islands. The exact number is uncertain — no comprehensive national register was maintained, and many installations were informal. See also: Martin, J., People, Politics and Power Stations: Electric Power Generation in New Zealand 1880–1998, Electricity Corporation of New Zealand, 1998.↩︎

2. Nuclear winter climate effects on NZ precipitation are uncertain. Global models (Robock et al., 2007, “Nuclear winter revisited with a modern climate model and current nuclear arsenals,” Journal of Geophysical Research) predict 5–8°C global average cooling from a ~4,400 warhead exchange, with significant but geographically variable effects on precipitation. NZ-specific precipitation modelling under nuclear winter scenarios is limited. The effect on solar irradiance is better characterised — stratospheric soot reduces surface insolation by an estimated 30–70% in the first 1–2 years, diminishing over 5–10 years.↩︎

3. Average NZ household electricity consumption is approximately 7,000–8,000 kWh per year (MBIE Energy in New Zealand statistics). Under recovery conditions, with reduced appliance use but continued essential loads (lighting, refrigeration, water pumping), consumption per household is estimated at 4,000–6,000 kWh per year. A 10 kW system at 70% capacity factor produces ~61,300 kWh/year.↩︎

4. The Reefton Power Station, commissioned in 1888 on the Inangahua River, was among the earliest public electricity supplies in the Southern Hemisphere — contemporary with Tamworth (Australia, 1888) and other early installations. It powered street lighting and mining operations. See: “Electric Light in Reefton,” The Press, 1 September 1888.↩︎

5. NZ’s history of small hydro is documented in various local and regional histories. Before the national grid reached rural areas, hundreds of small water turbines provided farm and community electricity across both islands. The exact number is uncertain — no comprehensive national register was maintained, and many installations were informal. See also: Martin, J., People, Politics and Power Stations: Electric Power Generation in New Zealand 1880–1998, Electricity Corporation of New Zealand, 1998.↩︎

6. NIWA (National Institute of Water and Atmospheric Research) maintains NZ’s rainfall and river flow data. Annual rainfall ranges from approximately 400 mm in Central Otago to over 8,000 mm in parts of Fiordland. NZ has over 425,000 km of rivers and streams (including intermittent watercourses), of which approximately 25,000+ are permanent. See: https://niwa.co.nz/freshwater/nz-river-maps↩︎

7. The Electricity Authority’s Electricity Market Information (EMI) database includes distributed generation data. Accurate counts of small hydro installations are complicated by varying definitions (micro, mini, small) and the fact that many very small off-grid systems are not formally registered. The NZ Electricity Engineers’ Association and regional distribution companies hold the most detailed data.↩︎

8. PowerSpout (https://www.powerspout.com/) is a New Zealand company based in Kaiwaka, Northland, that designs and manufactures micro-hydro turbines — specifically Pelton (PLT) and Turgo (TRG) models — as well as low-head propeller turbines. They are one of the few remaining NZ-based micro-hydro manufacturers and hold practical knowledge of NZ site conditions, installation practices, and turbine performance.↩︎

9. NIWA (National Institute of Water and Atmospheric Research) maintains NZ’s rainfall and river flow data. Annual rainfall ranges from approximately 400 mm in Central Otago to over 8,000 mm in parts of Fiordland. NZ has over 425,000 km of rivers and streams (including intermittent watercourses), of which approximately 25,000+ are permanent. See: https://niwa.co.nz/freshwater/nz-river-maps↩︎

10. The 50% overall efficiency assumption is conservative. Commercial micro-hydro systems typically achieve 55–70% wire-to-water efficiency when properly designed and installed. Locally fabricated systems, particularly crossflow turbines with rewound motor generators, may achieve 45–60%. Using the lower figure for initial design ensures the system meets or exceeds expected output.↩︎

11. Standard weir flow measurement methods are documented in engineering hydraulics references and in NIWA’s hydrological measurement guides. V-notch weirs are most accurate for small flows (1–50 L/s); rectangular weirs for larger flows. Accuracy of ±5% is achievable with careful installation and measurement.↩︎

12. The surface-to-mean velocity correction factor for float measurements varies with channel geometry and roughness. A factor of 0.80 is commonly cited as a general approximation, but values range from 0.75 for wide, shallow, rough-bottomed channels to 0.85–0.90 for deep, smooth channels. See: Herschy, R.W., Streamflow Measurement, 3rd ed., Taylor & Francis, 2009. For micro-hydro site assessment, using 0.80 provides a conservative estimate that is adequate for preliminary design; final design should use weir measurement where feasible.↩︎

13. Flow duration curves — showing the percentage of time a given flow is exceeded — are the standard tool for micro-hydro design. NIWA maintains flow records for many NZ rivers and streams. For ungauged streams (which is the majority of micro-hydro candidate sites), flow must be measured directly. The Q90 flow (exceeded 90% of the time) provides a conservative design basis; Q70 flow is more commonly used as a balance between conservative design and installed capacity utilisation.↩︎

14. NIWA (National Institute of Water and Atmospheric Research) maintains NZ’s rainfall and river flow data. Annual rainfall ranges from approximately 400 mm in Central Otago to over 8,000 mm in parts of Fiordland. NZ has over 425,000 km of rivers and streams (including intermittent watercourses), of which approximately 25,000+ are permanent. See: https://niwa.co.nz/freshwater/nz-river-maps↩︎

15. Nuclear winter climate effects on NZ precipitation are uncertain. Global models (Robock et al., 2007, “Nuclear winter revisited with a modern climate model and current nuclear arsenals,” Journal of Geophysical Research) predict 5–8°C global average cooling from a ~4,400 warhead exchange, with significant but geographically variable effects on precipitation. NZ-specific precipitation modelling under nuclear winter scenarios is limited. The effect on solar irradiance is better characterised — stratospheric soot reduces surface insolation by an estimated 30–70% in the first 1–2 years, diminishing over 5–10 years.↩︎

16. Turbine efficiency ranges are well-established in hydraulic engineering literature. Pelton turbines achieve 85–92% hydraulic efficiency at design point, with 80–90% typical across the operating range. Francis turbines achieve 90–95% at design point, with 85–93% across the operating range. These figures apply to well-manufactured commercial units; locally fabricated turbines will typically achieve lower efficiencies. See: Dixon, S.L. and Hall, C.A., Fluid Mechanics and Thermodynamics of Turbomachinery, 7th ed., Butterworth-Heinemann, 2014.↩︎

17. PowerSpout (https://www.powerspout.com/) is a New Zealand company based in Kaiwaka, Northland, that designs and manufactures micro-hydro turbines — specifically Pelton (PLT) and Turgo (TRG) models — as well as low-head propeller turbines. They are one of the few remaining NZ-based micro-hydro manufacturers and hold practical knowledge of NZ site conditions, installation practices, and turbine performance.↩︎

18. Crossflow turbine efficiency data from multiple sources including Mockmore and Merryfield, “The Banki Water Turbine,” Oregon State University Engineering Experiment Station Bulletin No. 25, 1949; and subsequent development work by VITA, IT Publications, and various development agencies. Peak efficiency of 80–85% is reported for well-optimised commercial units; 65–75% is realistic for locally fabricated units.↩︎

19. The crossflow (Banki-Mitchell) turbine is described in detail in: Mockmore, C.A. and Merryfield, F., “The Banki Water Turbine,” Bulletin No. 25, Engineering Experiment Station, Oregon State College, 1949. Further practical fabrication guidance in: VITA (Volunteers in Technical Assistance), “A Pelton Micro-Hydro Prototype Design” and related publications. The blade curvature radius relationship (R ≈ 0.326 × D) and 30° entry angle are derived from Banki’s original analysis and confirmed by subsequent experimental work.↩︎

20. Turbine efficiency ranges are well-established in hydraulic engineering literature. Pelton turbines achieve 85–92% hydraulic efficiency at design point, with 80–90% typical across the operating range. Francis turbines achieve 90–95% at design point, with 85–93% across the operating range. These figures apply to well-manufactured commercial units; locally fabricated turbines will typically achieve lower efficiencies. See: Dixon, S.L. and Hall, C.A., Fluid Mechanics and Thermodynamics of Turbomachinery, 7th ed., Butterworth-Heinemann, 2014.↩︎

21. Francis turbine runner manufacture requires precision casting in stainless steel or bronze, with complex three-dimensional blade profiles that control water flow from radial to axial direction. This is beyond NZ’s current casting capability for hydro-scale runners, though small demonstration units might be achievable. For recovery purposes, Francis turbines should be sourced from existing stocks rather than locally manufactured.↩︎

22. Penstock friction loss calculations use standard hydraulic equations. The Hazen-Williams formula with C=150 for PVC and C=120 for new steel pipe provides adequate accuracy for design purposes. Engineering reference tables (Doc #17) should include pre-calculated friction loss charts for common pipe diameters and flow rates.↩︎

23. Standard NZ PVC pressure pipe (AS/NZS 1477) is rated for working pressures from 600 kPa (PN6) to 1,800 kPa (PN18) depending on class. PN12 (1,200 kPa working pressure) is adequate for heads up to approximately 80 metres with a 1.5× safety factor. PN9 (900 kPa) covers heads up to approximately 60 metres. These are commonly stocked pipe classes in NZ plumbing and irrigation supply.↩︎

24. Water hammer wave speed in plastic pipe: the Joukowsky equation gives wave speed as a function of pipe elasticity and water bulk modulus. For PVC pipe (E ≈ 2.7–3.0 GPa), wave speeds of 300–500 m/s are typical, varying with pipe diameter-to-wall-thickness ratio. For steel pipe (E ≈ 200 GPa), wave speeds of 900–1,400 m/s are typical. Surge pressure ΔP = ρ·a·ΔV where a is wave speed and ΔV is the change in flow velocity — sudden valve closure from 2 m/s flow velocity in a PVC penstock with 400 m/s wave speed produces a surge pressure of approximately 800 kPa, nearly doubling the static pressure at a 50 m head site. See: Chaudhry, M.H., Applied Hydraulic Transients, 3rd ed., Springer, 2014.↩︎

25. Iplex Pipelines NZ (https://www.iplex.co.nz/) and Marley New Zealand (https://www.marley.co.nz/) are the two largest PVC and polyethylene pipe manufacturers/distributors operating in NZ. Both import raw resin and/or manufactured pipe; NZ has no domestic PVC resin or PE resin production. Pre-war stock volumes are uncertain and would require direct assessment of distributor and merchant warehouses via the skills census process (Doc #8). Placemakers and Mitre 10 Trade maintain regional distribution centres with substantial pipe inventory.↩︎

26. Wooden flumes and stave pipes were common in NZ’s early hydro and mining history. The Waihi gold mining operations used extensive wooden flume systems. Stave pipe was used for moderate-pressure water supply before steel and PVC became standard. Construction techniques are well-documented in historical engineering texts, though the skill has largely been lost.↩︎

27. Using induction motors as generators (induction generators) is well-established practice for micro-hydro. The motor is driven above synchronous speed by the turbine; capacitors provide the reactive power needed for self-excitation. The capacitor bank must supply approximately 30–50% of the motor’s rated kVA as capacitive kVAR. See: Smith, N., Motors as Generators for Micro-Hydro Power, Practical Action Technical Brief, 2007.↩︎

28. Vehicle alternator efficiency and output: typical automotive alternators (Bosch, Denso, Valeo) are rated 70–200 A at 12–14 V, producing 0.8–2.8 kW electrical output. Efficiency at rated load is typically 50–65% — lower than purpose-built generators because automotive alternators are designed for compactness and cost, not efficiency. They also require an external DC supply for field excitation (unless modified with permanent magnets), and their brushes and bearings have service lives of 3,000–5,000 hours — requiring replacement every 4–7 months under continuous micro-hydro duty.↩︎

29. Electronic load controllers for micro-hydro are described in: Smith, N. and Upadhyay, D., “Electronic Load Controllers for Micro Hydropower,” Practical Action (formerly ITDG), various editions. The principle — maintaining constant turbine load by diverting surplus to a ballast — is simple and effective. Hundreds of thousands of ELCs are in service worldwide in micro-hydro installations in developing countries.↩︎

30. ELC circuit designs are publicly documented in multiple sources including Practical Action technical briefs, MHPG (Micro-Hydro Power Group) publications, and various academic papers. Both microcontroller-based and discrete analog designs are well-proven. The analog approach avoids dependence on programmable chips, which may be advantageous as electronic component stocks deplete.↩︎

31. ELC power semiconductor requirements: a 10 kW single-phase ELC requires a triac or back-to-back thyristors rated for at least 40–50 A RMS at 230 V (e.g., BTA41-600B or equivalent). These are standard industrial components found in motor soft-starters, industrial heater controllers, and large lighting dimmers. NZ does not manufacture semiconductor devices; the total available stock is the sum of distributor inventory (RS Components, Element14, Jaycar) plus components salvageable from industrial and commercial equipment. For long-term ELC production beyond available semiconductor stocks, thyratron (gas-filled tube) equivalents are theoretically producible but at much lower performance — this is a Phase 4–5 research topic, not a near-term solution.↩︎

32. Coanda-effect intake screens use a precisely profiled curved surface over which water flows as a thin film. Surface tension and boundary-layer effects cause debris to pass over the screen while clean water passes through the fine slots. Developed for municipal water intakes and adapted for micro-hydro. More complex to fabricate than simple bar screens but significantly reduces maintenance. See: Wahl, T.L., “Design Guidance for Coanda-Effect Screens,” US Bureau of Reclamation, 2013.↩︎

33. NZ flood hydrology data is available from NIWA’s flood frequency analysis resources and regional council flood records. Flood-to-median flow ratios of 50–100 are commonly observed in steep, small catchments (catchment area < 50 km²) during intense rainstorms — particularly on the West Coast, Fiordland, and Marlborough Sounds where extreme rainfall events are frequent. Larger catchments show proportionally smaller flood peaks due to storage and routing effects. See: Pearson, C.P., “New Zealand regional flood frequency analysis using L-moments,” Journal of Hydrology (NZ), 37(2), 1998.↩︎

34. Micro-hydro powerhouse noise levels are not widely published in the literature, as they depend heavily on turbine type, speed, head, flow, and powerhouse construction. The 40–70 dB at 1 m range is based on field experience reported in Practical Action technical briefs and ITDG micro-hydro documentation. Pelton turbines running at high head are typically quieter than crossflow turbines due to the enclosed water jet versus the open-air water impact in crossflow design. Enclosure within a concrete or timber powerhouse reduces noise propagation. Practitioners should conduct noise assessment for specific sites if residential proximity is a concern.↩︎

35. Fish screen requirements for NZ waterways are specified by regional councils and informed by NIWA research on native fish passage. The Fish Passage Assessment Tool maintained by NIWA provides guidance on screen mesh sizes appropriate for different NZ fish species. For micro-hydro intakes, a 3 mm mesh screen excludes juvenile fish while passing sufficient water for most installations, though screen area must be large enough to avoid excessive head loss across the screen.↩︎

36. Minimum flow (residual flow) requirements for NZ waterways are set by regional council plans under the Resource Management Act. Requirements vary significantly by region and waterway classification. Environment Canterbury (ECan) regional plans, Horizons Regional Council, and Otago Regional Council all specify different minimum flow thresholds and calculation methods. Under normal conditions, operators should consult the relevant regional plan. The 30–50% guidance in this document is a conservative field heuristic; actual consented takes may be lower or higher depending on the waterway’s ecological sensitivity and existing water allocation. See: Ministry for the Environment, National Policy Statement for Freshwater Management 2020 (amended 2023), which requires regional councils to set minimum flows protective of ecosystem health.↩︎

37. The crossflow (Banki-Mitchell) turbine is described in detail in: Mockmore, C.A. and Merryfield, F., “The Banki Water Turbine,” Bulletin No. 25, Engineering Experiment Station, Oregon State College, 1949. Further practical fabrication guidance in: VITA (Volunteers in Technical Assistance), “A Pelton Micro-Hydro Prototype Design” and related publications. The blade curvature radius relationship (R ≈ 0.326 × D) and 30° entry angle are derived from Banki’s original analysis and confirmed by subsequent experimental work.↩︎

38. Crossflow turbine efficiency data from multiple sources including Mockmore and Merryfield, “The Banki Water Turbine,” Oregon State University Engineering Experiment Station Bulletin No. 25, 1949; and subsequent development work by VITA, IT Publications, and various development agencies. Peak efficiency of 80–85% is reported for well-optimised commercial units; 65–75% is realistic for locally fabricated units.↩︎