The itch

A grille clip is a plastic fastener about the size of a thumbnail — three quarters of an inch wide, half an inch long, three eighths tall. It holds the muntin bars of a window grille onto the sash of an Andersen Narroline double-hung window. There are a dozen of them in every window, they are invisible when they work, and they are the only thing standing between a tidy grille and a set of wooden bars sliding around behind the glass.

Andersen discontinued them. The windows did not go anywhere — Narroline units are in a great many houses and will be for decades — so the part became one of those things that is both essential and unobtainable. Original model numbers W8713 and 200-238, if you are here because you typed one of those into a search engine.

What I tried

The first one took five days and three revisions. window_clip_B.stl is dated 12 September 2016; w_clip_B_v3.stl is dated the 17th. Then the design changed shape — every file after that is named _plug, because the part grew the round peg that seats into the sash.

What followed was not a product launch. It was four styles over two and a half years, each one appearing when somebody turned up with a window the existing clips did not fit:

Style	First file	Last revision	Revisions
B	12 Sep 2016	11 Dec 2016	3
A	6 Nov 2016	—	1
C	19 Feb 2018	1 Mar 2018	2
D	1 Dec 2018	3 Mar 2019	3

B came first and A second, which is only confusing until you know the styles are named after the window, not the order I made them. Style D took three revisions across fifteen months — the longest I spent on any single clip.

3D

3D

3D

3D

How it turned out

The clips were the first thing MadeParts proved, and the reason it exists: low-volume plastic parts that traditional manufacturing cannot touch at a sane price. A tooling budget for a few hundred units a year makes no sense. A printer does.

In 2020 the clip got its own storefront at windowgrilleclips.com — four styles, printed to order, no inventory. Four years after the first print, and three and a half after the last design change. It was never a plan; it was a background experiment that kept getting mail.

1. Nobody needed a company. They needed a sixty-cent part that nobody makes.
2. The gap between "I printed one" and "anyone can buy one" was four years, and almost all of it was not design work.