1 / 3
Deck levelling pad for a sloped zinc flashing — level on top, water still runs underneath — zdjęcie 1
1 / 3

Deck levelling pad for a sloped zinc flashing — level on top, water still runs underneath

Zamów wydruk tego modelu

Obliczymy ostateczną cenę na podstawie wybranego pliku. Sprawdzisz ją i zaakceptujesz przed płatnością.

Zapisane na tym urządzeniu.

Levels a deck board over a sloping zinc flashing and still lets the water run under it. Slope gauges included.

Ocena
0 / 5
Polubienia
0
Pobrania
0
Wydruki
0
Wyświetlenia
11
Oceny
0

Aktualizacja u źródła:

Informacje o modelu

Źródło
Printables
Licencja
Creative Commons — Attribution — Share Alike

A deck board has to sit level on a zinc flashing that slopes. A solid shim does that, and

creates the problem this one is built to avoid: it dams the runoff, so water stands permanently

between the wood and the metal.

That puddle matters more than it looks. Deck boards are copper-treated pine (ACQ, copper azole)

and copper treatments corrode zinc. Wet, in contact, with no air moving through, the runoff

carries copper onto the flashing — the one layer of the whole build you cannot replace without

tearing the sill apart. The board rots from underneath at the same time.

So this pad does two things at once:

- Levels. Bottom face tilted at the slope, top face horizontal.

- Keeps the water running. Two arched channels go straight through the bearing face, so

runoff crosses under the pad instead of piling up against it. Two shallow V grooves split what

would otherwise be two 18 mm bands of sealed contact — a damp sealed band with no air renewal is

textbook white rust.

## Print the slope gauges first — it is not an optional step

slope-gauge-set_4-5-6-7pct.stl is four small flat-topped wedges, engraved 4 / 5 / 6 / 7 %. Put

each on your surface and check its top with a spirit level. The one that reads horizontal is your

slope. Put that number in SLOPE_PCT and print the pad.

1 % of slope is 0.57°, finer than an ordinary bubble vial's graduation. And a pad cut to the wrong

angle is worse than no pad: it bears on one edge only.

⚠ Measure the angle between the surface and the underside of the wood, not the absolute slope of

the surface. Shim the board level first, then read the gap at two known stations along it. Getting

this wrong cost a whole prototype campaign here: 80 mm apart, 17 mm then 23 mm of gap → 7.5 %.

## What is in the set

| File | What it is |

|---|---|

| slope-gauge-set_4-5-6-7pct.stl | the four slope gauges — print these first |

| shim-round_d90-d60_7.5pct_h25.stl | the production part: Ø90 at the zinc, Ø60 on top, 25 mm at the thin edge |

| shim-rect_100x100_7.5pct_h25.stl | 100 × 100 full-footprint variant, 4 ribs, for continuous bearing |

| shim-narrow_w30_7.5pct_h25.stl | 30 mm wide, for a narrow sill |

| terrace_shim.scad | the source — your slope is not mine, you will need this |

The round one has a 5 mm flat shaved off the downhill rim, plus an engraved arrow and the slope %

on the top face: once the pad is fitted and round, that flat is how you still know which way is down.

Point it downhill.

## Fixing it down

Two Ø4.2 through holes, counterbored Ø9 × 4 with a 90° csk, so the screw head sits below the

bearing face and can never punch into the metal underneath.

- Ø4.2 for a 4–4.5 mm wood screw is deliberately a hair tight. On a wood screw the quoted diameter is

the thread crest; the shank is ~3.1 mm and floats, so only the crests graze the plastic. That drag

makes the screw self-retaining in the pad, which matters a lot when you are driving blind from

underneath. It costs no clamping force — the 15–20 mm of thread in the wood wins by an order of

magnitude.

- Stainless only (A2, A4 near the sea). The same copper treatment that attacks the flashing eats

the 5–12 µm zinc plating off galvanised screws; then bare steel rusts and the rust runs down onto

the flashing. Galvanised is fine for removable trials, not for the final fixing.

- After tightening, check the head actually seated: PETG swarf packed under it can hold it proud.

Lay a straightedge across.

- 90° countersunk head, Torx for preference — stainless chews cross-heads.

## Printing

- Sloped face DOWN on the bed. The STL is already in that orientation. That face comes off the

plate as a smooth plane, which is what you want bearing on soft zinc — a stepped face bears on the

edges of its steps and prints them into the metal. The wood side becomes a gentle layer-stepped

ramp; against sawn timber, which varies by more than 0.5 mm on its own, that is irrelevant. The

load path is pure compression through the layer stack.

- No support anywhere, and no brim (~6400 mm² of contact — verified over ~10 discs).

- PETG-CF or PETG. Not PLA. Permanent outdoor compression, in the sun, sometimes wet. PLA creeps

and hydrolyses: it will sag and go brittle in a season. ASA/ASA-CF works too.

- 0.6 mm nozzle, 0.30 mm layers, 3 walls, 22 % gyroid, 5 top / 4 bottom shells, bridges at 15 mm/s,

textured PEI at 80 °C. About 62 g and 1 h 49 per piece at 25 mm — proportional to the height

you need. Solid infill is pointless: the shells and the channels carry the load.

## Your slope, your height

sh

openscad -o pad.stl --export-format=binstl \

-D 'SHOW="puck_print"' -D 'SLOPE_PCT=5' -D 'H_MIN=18' terrace_shim.scad

| Parameter | Here | Note |

|---|---|---|

| SLOPE_PCT | 7.5 | mm of drop per 100 mm of run — the gauges give you this |

| H_MIN | 25 | height at the thin (uphill) edge |

| PUCK_D / PUCK_D_TOP | 90 / 60 | contact area scales with D² — check your bearing pressure before shrinking it |

| PUCK_SLOT_W | 12 | few channels, wide, on purpose: a 17 mm channel does not clog with pine needles, six narrow ones do |

| SCREW_D | 4.2 | do not go to 5.0 — 0.4 mm more radius starts a crack |

| PUCK_TXT | "" | engrave anything you like on the top face |

Aim slightly high rather than low if the board is screwed down: the excess is taken up at the

fixing and contact is guaranteed, whereas a pad that is too low leaves a gap that knocks. (The

opposite advice applies to a board laid loose.)

## Status

Geometry frozen. The slope and the 25 mm height were validated by walking a printed prototype across

8 stations covering the full width of a bay — no station needed its own angle. The run of **16

round pads + 1 narrow** is printed in PETG-CF, fitted, and in service.

CC BY-SA — use it, sell prints of it, modify it. Publish modified versions under the same licence

so the next person with a sloping flashing gets your fix too.

Tagi