Quick Answer: CT harvesters deliver 1.5-5W on lines carrying 20-100A but fail below 5A or during outages. Solar panels provide 0.5-6W independent of line status but struggle during multi-day storms. For 99%+ uptime, hybrid CT + solar + battery is the only architecture that covers all failure modes. Lifecycle cost analysis favors hybrid after year 3. Compare hybrid systems →
A CT harvester clamped on a conductor carrying 100 A delivers 5+ W continuously — day, night, blizzard. Drop that conductor current to 10 A and you're down to 0.8 W. At 3 A, you're dead.
That nonlinear power curve is the single most misunderstood spec in overhead line monitoring power design. Engineers who size CT harvesting based on average conductor current instead of minimum annual current end up with monitoring nodes that go dark during the exact low-load conditions they were supposed to monitor.
This article goes deeper than the general solar vs CT comparison we published previously — we're assuming you already understand the basic tradeoffs. Here we dig into the physics, quantify the failure boundaries, and lay out 10-year lifecycle costs so you can defend your power architecture choice in a procurement review.
CT Harvesting Physics: What's Actually Happening
A current transformer energy harvester is a split-core toroid clamped around the conductor. Alternating current in the primary (the conductor) induces a proportional current in the secondary winding. The harvested power follows:
P = I² × R_load (referred to secondary), but the usable power is bounded by core saturation, winding losses, and the power management electronics' minimum input voltage.
Key parameters that determine real-world output:
| Parameter | Effect on Harvested Power |
|---|---|
| Primary current (I_p) | Dominant variable — power scales roughly with I_p² at low currents, flattens as core approaches saturation |
| Core material (nanocrystalline vs ferrite) | Nanocrystalline: higher permeability, better low-current performance. Ferrite: lower cost, saturates earlier |
| Core cross-section | Larger core = more flux capture = higher power at same I_p. Also heavier |
| Air gap in split-core | Mandatory for clamp-on installation. Reduces effective permeability 10–50× vs closed core. Precision machining of mating faces matters |
| Winding turns ratio | Higher N_s/N_p = higher voltage but lower current on secondary. Tuned to match power management IC input range |
| Conductor diameter | Harvester bore must match — 23mm to 40mm covers most transmission conductors |
The nonlinearity is the critical point. At low primary currents, the induced secondary voltage drops below the power management IC's minimum operating threshold (typically 0.8–1.5 V depending on the boost converter). Below that threshold, harvested energy is zero — not just low, zero.
The Power Curve: Where It Gets Practical
| Primary Current (A) | Harvested Power (typical) | System Status |
|---|---|---|
| <3 A | ~0 W | Below startup threshold — system on battery only |
| 5 A | 0.2–0.5 W | Trickle charge. Enough for sleep-mode beacon, not continuous monitoring |
| 10 A | 0.8–1.2 W | Marginal. LoRa-based nodes survive. 4G duty cycling required |
| 20 A | 1.5–2.0 W | Operating range for most payloads. From our testing: ≥1.5 W downstream at 5–20 A primary |
| 50 A | 3–5 W | Comfortable headroom. Higher sample rates, 4G continuous |
| 100 A | 5+ W (clamped) | Power management limits output to protect electronics. Excess dissipated as heat in the clamp |
| 500+ A | Clamped at max | Core saturated. Same output as 100 A. No benefit from higher current |
The sharp drop below 50 A is what kills CT-only deployments on distribution feeders. A trunk line carrying 200 A average might dip to 8 A at 3 AM on a mild spring night. That 8 A gets you ~1 W — workable. But a distribution lateral that averages 30 A might hit 2 A during off-peak. That's a hard shutdown.
Solar Panel at Tower Height: Not the Same as Ground-Level PV
Mounting a solar panel on a transmission tower at 30–50 m introduces constraints that don't exist in rooftop or ground-mount applications. Soiling can cost several percent of annual output where panels are never cleaned — more in dusty or high-pollen corridors — and temperature derating can meaningfully cut summer output below STC rating.
Real-World Derating Factors
| Factor | Impact | Mitigation |
|---|---|---|
| Fixed orientation | Tower geometry dictates panel angle. Rarely optimal. Expect 60–80% of nameplate in best season | Curved panels conformal to housing capture wider angle range |
| UV degradation | Higher UV flux at elevation + no shade. PET encapsulation yellows in 2–3 years | ETFE lamination required for 10+ year service life |
| Wind loading | 100+ km/h at tower height is routine. Panel must not become a sail | Low-profile integration into device housing; engineered brackets rated for wind zone |
| Bird soiling | Tower crossarms are perching spots. Guano = localized shading = hotspots + 20–50% output drop | Mounting position below crossarm; angled surface or integrated housing that birds can't perch on |
| Ice/snow accretion | Complete output loss for days or weeks. No one climbs the tower to clear a 6W panel | Battery sizing must cover the ice period; or use CT as backup during icing events |
| Lightning / surge | Tower is a lightning path. Panel wiring needs TVS diodes and proper grounding | Standard for utility-grade hardware; often missing in lower-cost assemblies |
| Maintenance access | Every panel replacement = a tower climb. $2,000–5,000 per climb in crew + equipment | Design for 15+ year panel life. ETFE + tempered glass + conformal coating on connections |
Solar Output at Tower Height
Using a 6W nominal panel (monocrystalline, ~22% cell efficiency), typical for overhead line monitoring applications:
| Condition | Output (W) | Annual Hours (temperate climate) | Daily Wh Contribution |
|---|---|---|---|
| Clear sky, near-optimal angle | 4.5–5.5 | ~1,200 h | 18–22 Wh |
| Clear sky, poor angle | 3.0–4.0 | ~600 h | 12–16 Wh |
| Overcast / diffuse | 0.5–1.5 | ~1,500 h | 2–6 Wh |
| Heavy cloud / rain | 0.1–0.5 | ~800 h | 0.4–2 Wh |
| Night | 0 | ~4,380 h | 0 |
| Snow/ice cover | 0 | Variable (0–1,000 h) | 0 |
Annual solar yield for a tower-mounted 6W panel in a temperate zone: roughly 3,500–5,500 Wh. A monitoring payload drawing 1W average needs 8,760 Wh/year. Solar alone covers 40–65% of the energy budget — the rest must come from battery reserves charged during surplus periods, or from a CT harvester.
10-Year Lifecycle Cost Comparison
This is where procurement decisions should live. Component cost is noise; lifecycle cost is signal. Well-designed hybrid dual-source power architectures routinely achieve very high uptime in remote grid infrastructure, with lifecycle costs breaking even against CT-only architectures within a few years once tower climb costs are factored in.
CT-Only Architecture
| Cost Item | Year 0 | Years 1–10 | 10-Year Total |
|---|---|---|---|
| CT harvester module | $80–150 | — | $80–150 |
| Power management board | $30–60 | — | $30–60 |
| Battery (7.4V / 10Ah Li-ion) | $40–80 | Replace at year 5: $40–80 | $80–160 |
| Battery replacement tower climb | — | 1 climb at year 5: $2,000–5,000 | $2,000–5,000 |
| Total | $150–290 | $2,190–5,370 |
The tower climb for battery replacement dominates the lifecycle cost. A $50 battery becomes a $5,000 event.
Solar-Only Architecture
| Cost Item | Year 0 | Years 1–10 | 10-Year Total |
|---|---|---|---|
| Solar panel (6W, ETFE) | $30–60 | — | $30–60 |
| Charge controller (MPPT) | $20–40 | — | $20–40 |
| Battery (7.4V / 10Ah Li-ion) | $40–80 | Replace at year 5: $40–80 | $80–160 |
| Larger battery for autonomy (14Ah upgrade) | +$20–40 | — | $20–40 |
| Battery replacement tower climb | — | 1 climb at year 5 | $2,000–5,000 |
| Total | $110–220 | $2,150–5,300 |
Similar lifecycle cost to CT-only. The dominant cost in both cases is the tower climb, not the electronics.
Hybrid CT + Solar Architecture
| Cost Item | Year 0 | Years 1–10 | 10-Year Total |
|---|---|---|---|
| CT harvester + solar panel + power management | $120–200 | — | $120–200 |
| Battery (7.4V / 10Ah Li-ion) | $40–80 | Replace at year 5: $40–80 | $80–160 |
| Battery replacement tower climb | — | 1 climb at year 5 | $2,000–5,000 |
| Total | $160–280 | $2,200–5,360 |
Higher component cost, same lifecycle cost. The argument for hybrid isn't cost savings — it's uptime. Dual-source architectures reduce battery cycling depth, which extends battery life toward 7–8 years and potentially eliminates the mid-life tower climb entirely. That's where the real savings appear: skip one $3,000 tower climb and you've paid for the hybrid electronics 10× over.
The Tower Climb Multiplier
Battery-only systems (no solar, no CT — just a large primary cell or rechargeable pack) seem cheap on paper. In practice, they create a recurring tower climb schedule: every 1–3 years depending on battery capacity and monitoring payload. Over 10 years, 3–5 tower climbs at $2,000–5,000 each = $6,000–25,000 per node. That's the actual cost of "saving" $100 on harvesting electronics.
Hybrid Architectures: Design Patterns
Three hybrid patterns cover most deployment scenarios:
Pattern 1 — CT primary, solar backup. CT harvester provides base power. Solar panel charges battery during low-current periods. Battery bridges gaps. Best for: trunk lines with occasional low-load periods. Our JK platform uses this pattern — CT harvesting from the conductor plus a curved solar panel integrated into the housing, feeding a 7.4V / 10Ah Li-ion battery with regulated DC output.
Pattern 2 — Solar primary, CT supplement. Solar panel is the main energy source. CT harvester tops up the battery at night or during extended overcast. Best for: distribution lines with variable current, tower-mounted (not conductor-mounted) devices.
Pattern 3 — Dual-source with intelligent switching. Power management IC monitors both sources and draws from whichever is delivering more power at any moment. More complex firmware, but maximizes energy capture across all conditions. The GB Icing Monitoring System uses this approach because icing-prone regions simultaneously degrade both solar (snow cover) and CT (ice on conductor reduces effective current flow).
Decision Matrix: CT vs Solar vs Hybrid
| Deployment Parameter | CT-Primary | Solar-Primary | Hybrid | Solar-Only |
|---|---|---|---|---|
| Min annual conductor current >50 A | ✓ Best | — | ✓ | — |
| Min annual conductor current 10–50 A | Risky | ✓ | ✓ Best | ✓ |
| Min annual conductor current <10 A | ✗ | ✓ | ✓ | ✓ Best |
| Heavy icing region (>30 days/year) | ✓ | ✗ | ✓ Best | ✗ |
| Conductor-mounted device | ✓ Best | ✗ | ✓ | ✗ |
| Tower-mounted device | — | ✓ Best | ✓ | ✓ |
| Pre-commissioning install needed | ✗ | ✓ | ✓ | ✓ Best |
| Budget per node <$200 | ✓ | ✓ Best | ✗ | ✓ |
| Uptime requirement >99.5% | ✗ | ✗ | ✓ Best | ✗ |
| Deployment at >1,000 nodes | Hybrid Best — lifecycle cost difference compounds significantly at scale | |||
What to Calculate Before Speccing
Pull your SCADA historian data for the target line segment. You need three numbers:
- Minimum conductor current across a full calendar year — not average, not P95, the actual minimum sustained for >4 hours. That's your CT harvester's worst-case input.
- Worst-case solar deficit period — consecutive days of <2 peak sun hours at your deployment latitude and elevation. That's your battery autonomy requirement.
- Monitoring payload power budget — average draw including communication protocol at target duty cycle. LoRa at 15-min intervals ≈ 0.3W average. 4G at 5-min intervals ≈ 1.2W average.
Those three numbers determine your architecture. Everything else is implementation detail.
Send us your line parameters — conductor current range (annual min/max), mounting type (conductor clamp or tower bracket), and target monitoring payload specs. We'll run the energy balance calculation and confirm which power architecture maintains your uptime target before you commit to hardware. See the JK Overhead Line Power Platform specs if you want to see what a production hybrid system looks like.
