Wattplot, Panel Upcycling Guide

A single Wattplot planter is 8 ft × 5 ft (max). It fits any solar panel that fits in that envelope, new panels from the manufacturer, or old panels rescued from decommissioned arrays that would otherwise be landfilled.

Most decommissioned panels are still functional. They’ve lost 8-15% of their nameplate output over 10-20 years, but they’re not cost-effective for grid-tie anymore. They’re perfect for shade + food + some power in a Wattplot.

Why upcycle?

• ~10M tons of solar panel waste globally by 2050 (IRENA 2016). Most of it is glass, aluminum, and silicon, recyclable, but recycling costs money and the infrastructure is sparse.
• A second-life use case (shade + some power) delays recycling by 10-20 years and recovers ~50-90% of the panel’s original useful energy.
• Most panels are removed from service because the inverter or racking failed, not the panel itself. The cells are usually fine.
• A 12-year-old 250W panel is still a 235W panel. For a shade-and-power application, that’s a free 235W.

The 8×5 envelope

Wattplot’s single-planter size is set by lumber stocking length: 8 ft is the longest common dimension at Home Depot, with zero waste on the long rails. 5 ft is the wide cross-rail cut from one 8-ft board (two cross-rails per board, 6” waste).

Panels can overhang the bed by 0.5” per side (1” total in either direction). The mid-clamps grip the aluminum frame, which sits on the wood rails, overhang is fine and is how the LONGi Hi-MO X10 (97” panel, 96” bed) works.

Anything bigger → chain multiple planters in a row. Two 8×5 planters side by side is 8×10 ft, which fits two 60-cell panels laid out as a 2×1 array.

Common salvage panels (all fit)

These are the named presets in wattplot_params.py. Each one has been validated to fit the 8×5 cap and produce a sensible bed size:

All five are tested in wattplot_params.py, calling apply_panel_preset(name) swaps the panel + bed + derated wattage in one call.

How to use a custom (non-preset) panel

For a panel that’s not in the preset list:

import wattplot_params as P

# Measure your panel (with the aluminum frame, the clamps grip the frame)
P.PANEL['L_in'] = 66.5         # length, including frame
P.PANEL['W_in'] = 39.0         # width, including frame
P.PANEL['thickness_in'] = 1.6
P.PANEL['mass_lb'] = 40.0
P.PANEL['wattage'] = 245       # new nameplate (or your measured Voc/Imp)
P.PANEL['panel_age_years'] = 8 # 8 years old
P.PANEL['panel_bifacial'] = False
P.PANEL['panel_efficiency_pct'] = 17.0

# Then call the helper to resize the bed and derate the wattage
P.apply_panel_preset('longi_620W')  # arbitrary preset just to trigger the helper
# Or call the function inline if you didn't change a preset key:
# (See apply_panel_preset source for the derate + bed-resize logic)

The derate formula is 0.5% per year, floor 70%, typical for crystalline-silicon panels. Polycrystalline degrades slightly faster (0.7%/yr). Adjust panel_efficiency_pct to match the panel.

Lumber math (the cut list scales with the bed)

The cut list is derived from the bed dimensions, not hardcoded. For a bed of B_L × B_W inches:

If B_L = 96" and B_W = 44.6" (the LONGi 620W default), the cut list is exactly what’s in bom.md and the build guide.

If B_L = 65" and B_W = 39" (the 60-cell preset), the long rails are 65” each (cut from 2×6×8ft, 31” waste per board) and the cross rails are 39” (2 per board, 18” waste per board). The diagonal brace is 76” (from 2×4×8ft, 20” waste).

Structural check

The wind-load analysis (analysis/wind_load.py) takes the panel dimensions and computes the wind force, moment, and safety factor on the bed. For a smaller panel, the wind force is lower (the wind force is proportional to the panel area), so the safety factor goes up, a smaller, lighter panel is easier to support.

This means old/salvage panels are often easier to validate than the 620W default. The wind analysis should pass at smaller panel sizes with the same bed structure.

MPPT sizing

A Sunapex 10A is fine for the 60-cell and 72-cell salvage cases (7-8A peak). For the bigger panels, step up to a 20A or 30A MPPT.

When a salvage panel is not a good fit

• Cracked glass. Water will get in and the cells will fail. Don’t use these.
• Delamination. Bubbles or peeling on the front. The cells are degrading; useful life is short.
• Hot spots. If a thermal camera shows >20°C delta on the panel surface under sun, the cells are mismatched and the panel will continue to degrade. Don’t use these.
• No label / no specs. A panel you can’t identify may have different dimensions than expected. Measure it before designing the bed.

A good salvage panel has a clean label (Voc, Isc, Imp, Vmp, Pmax), intact glass, intact aluminum frame, intact junction box, and measures close to its nameplate Voc/Isc under full sun (use a multimeter).

Open question: bifacial upcycling

Most pre-2018 salvage panels are monofacial. Bifacial panels are typically 2018+. If you’re sourcing from a recent decommissioned commercial array, you might find bifacial salvage, and the 10% bifacial bonus is significant for a Wattplot’s annual kWh.

If you have access to bifacial salvage, set panel_bifacial=True in the preset. The function will set bifacial_bonus = 0.10.

Summary

A 12-year-old 250W residential panel salvaged from a rooftop decommissioning is a 235W panel in a Wattplot. Free shading, free power for the controller and grow light, second-life delay on recycling, 235 kWh/year in Phoenix. The single most sustainable way to add solar to a garden bed.