Highland Cloud ForestGabriele Zoppoli · Genova

Deep clean: four years in, two of them with mold building

Four years online — the last two with mold quietly accumulating in a stagnant corner — and the highland cabinet finally got its first deep clean. Triggered by what I'd taken for botrytis on a Dracula and traced back to a single fan that had silently disconnected itself months ago.

4 min read

A Dracula pholeodytes bud was the first symptom — a flat grey halo I’d taken for botrytis around what should have been an opening flower, on a plant that hadn’t shown trouble before. A week of changing nothing didn’t fix it. So I emptied the cabinet.

Black mold tracking up the corrugated insulation in the back-left corner. Plain water and a sponge took most of it off — saprophytic, not a leaf pathogen — but the location *was* the diagnostic clue: that's where the airflow had stopped.
Black mold tracking up the corrugated insulation in the back-left corner. Plain water and a sponge took most of it off — saprophytic, not a leaf pathogen — but the location was the diagnostic clue: that’s where the airflow had stopped.

Tracing the cause

The freezer-evaporator behind the back wall has three small PWM fans pulling air across the cold-plate. Two of them sounded right. The third — back-left — was silent. Pulling the polycarbonate cover plate off, the cable was hanging loose: a faulty crimp had let the connector slip off the pin months ago. The other two fans were enough to keep the cabinet on temperature, so the alarms never tripped; the back-left corner just stopped breathing.

The freezer-evaporator fan array — three PWM fans on a polycarbonate cover plate, pulling air across the cold-plate behind. One had silently disconnected itself. The dead corner was the corner with no airflow.
The freezer-evaporator fan array — three PWM fans on a polycarbonate cover plate, pulling air across the cold-plate behind. One had silently disconnected itself. The dead corner was the corner with no airflow.

Pulling everything out

For the first time since it went online four years ago, the cabinet got fully emptied. Around 70 plants — Heliamphora, Sarracenia, Nepenthes, the dense pleurothallid layer, the hanging Dracula and Restrepia — out onto the dining table. The pump sprayer in the foreground holds the working solution: Physan-20 at 10 mL per 5 L of water, which is what I use for everything cabinet-related. Walls, polycarbonate, the steel evaporator plate, all sprayed and scrubbed.

Most of the black on the walls actually came off with just water and a sponge — a saprophyte, the kind that lives on dust and condensate in a stagnant corner, not something that infects leaves. The Physan was belt-and-braces; the real fix is getting airflow back to that corner.

The full collection on the dining table — first time the cabinet has been completely emptied since it went online. The pump sprayer holds the Physan-20 working solution at 10 mL per 5 L.
The full collection on the dining table — first time the cabinet has been completely emptied since it went online. The pump sprayer holds the Physan-20 working solution at 10 mL per 5 L.

Recrimping the dead fan

With the cabinet bare it took about ten minutes to swap the cable. New crimp, into a known-good lead — the fan span up immediately, ruling out the motor. While the work was open I redid the crimps on all three connectors, on the principle that whatever weakened one will probably weaken the others.

Starting the scrub-down. The cable swap had been a ten-minute job; the bulk of the time was here — Physan and a sponge across every surface.
Starting the scrub-down. The cable swap had been a ten-minute job; the bulk of the time was here — Physan and a sponge across every surface.

Drainage and plant work

The canalina — the condensate channel below the evaporator — is a thin aluminum fold suspended on fishing line at a precise 5° tilt toward the drainage pipe. Water from the back of the cold-plate runs down its length and drains by gravity through a PVC pipe to a tank under the cabinet. With everything out, it got scrubbed and the fishing line re-tensioned.

The canalina — the aluminum fold across the bottom of the frame. Suspended on fishing line at 5° toward the drainage pipe; condensate runs off the back of the evaporator (the serpentine refrigerant-tube pattern pressed into the aluminum cooling plate, visible above) onto it, then by gravity through a PVC pipe to the tank under the cabinet.
The canalina — the aluminum fold across the bottom of the frame. Suspended on fishing line at 5° toward the drainage pipe; condensate runs off the back of the evaporator (the serpentine refrigerant-tube pattern pressed into the aluminum cooling plate, visible above) onto it, then by gravity through a PVC pipe to the tank under the cabinet.

While the plants were on the table, I repotted all the Nepenthes — N. glabrata, pitopangii, tenuis, inermis — and tied each climber up a thin wooden orchid stake with plastic zip ties. They’ve been getting tall enough to need it. The hanging Dracula and Restrepia moved into less stagnant positions; the corner that lost its fan no longer holds a hanger.

Cabinet back online

Reset state: clean walls, polycarbonate shelf back in (the *Heliamphora* tier above the pleurothallid layer below), three fans turning, ready for the collection back in.
Reset state: clean walls, polycarbonate shelf back in (the Heliamphora tier above the pleurothallid layer below), three fans turning, ready for the collection back in.

The polycarbonate shelf is back in — that’s the tier that lets the Heliamphora sit at full intensity above the pleurothallids below. All three fans are turning. The plants went back to their positions, with the high-airflow corners getting the species that had been suffering the most.

(Filed on the terrarium photo timeline under May 2026.)

← All posts

Comments

Thoughts, corrections, trade offers welcome. Comments live on GitHub Discussions; sign in with a GitHub account to post.