Re: [TECH]:: Advice sought on 10 and 000 quantity manufacture of tubular plastic products

At 11:43 AM 2026-04-08, you wrote:

Hi, Russell-

Could you make the tubular section just a simple lopped-off piece
of tube? ie. change the other parts so that the tube doesn't need
any features. There may be a standard size, and in any case
extrusion dies are relatively inexpensive (though hollow ones
are more because they require a 'spider'). Molded caps/plugs are also
available in standard sizes (eg. 38mm OD) or the mold required could be
relatively cheap- given it's a symmetrical thin 2.5D part.
maybe it could be made in a flexible material to allow a momentary
switch? The tube supplier could be asked to cut it to length if the
tolerances are not too tight, and they might do it at little or
no cost.

For the transparent section, maybe epoxy molded in a silicone mold-
(it probably doesn't need a mirror finish) for very small quantity.
Original male part resin printed, female silicone mold made from
that, used to cast a few hundred parts with water-clear epoxy and
a degassing vacuum rig (or live with the bubbles). A few hundred
dollars in supplies and equipment but no hard tooling.

Raw plastic materials are quite inexpensive in China and you could look
at CNC machining for 1k volume, though possibly not quite meeting
the cost constraints.

Injection molds are expensive and you lose a lot of flexibility
and control. You do get full material strength and low part cost.
Best for 50k+.

No switch? Always-on? What about a capacitive sensor that causes it
to go bright when handled? A few-cent MCU maybe also doing the
boost conversion. Might beat PIR for cost and Iq.

Using an alkaline AA battery might make it more like the size of a
lipstick tube and similarly sized/shaped items that women typically
carry (but 1/5 the capacity)-  full-capacity D cells are heavy
and $$ (some appear to be just AA cells in a D body). Though, given
the application, I'm reminded of my landlord in LA who explained
that his massive 5 D-Cell MAGLITE flashlight/torch could perform certain
protective functions in a pinch, as when investigating things that go
bump in the night.

Best regards,
Spehro Pefhany

>The following is from posts to Stack Exchange and Facebook engineerings
>groups.
>I'm sure that there are people here just as capable :-).
>Main concern is suggestion how to manufacture 1000 and 10-20 tubular
>plastic items at low cost.
>I'm asking Chinese and local manufacturers these questions.
>Local manufacture has minimal chance of being viable, sadly.
>
>      Russell
>_______________________
>
>I'm looking for advice on
>
>How to make the following in quantities of:
>
>    - Prototype (10-20 volume)
>    - Small volume (say 1000) <-- this is the main concern
>
>For larger volumes plastic injection moulding is liable to be most economic.
>
>    2.
>
>    Order of cost for labor (labour to me - I'm a Kiwi)
>    3.
>
>    Thoughts on material choices.
>
>In volume production (and maybe 1000 quantity) a mounting "foot" will
>probably be added.
>
>This is a light for use in women's toilets (initially) in refugee camps.
>The light level is VERY low but vastly better than utter darkness, and it
>will operate from a single Alkaline D cell all night for 5 years.
>An optional version with a PIR person sensor allowing brighter light only
>when occupied.
>
>Target cost is $US3 (or less) complete (electronics, housing, battery).
>
>Ideally the device would be highly theft and vandalism resistance BUT this
>is probably unachievable given cost constraints.
>Metal is not an option: ANY metal parts are liable to be removed for
>"recycling" :-( . At least for the prototype and maybe the first 1000
>lights durability and vandal resistance are not a major factor, although eg
>PLA plastic is probably not robust enough.
>------------------------------
>
>3D printing may be acceptable for the first prototypes.
>I have 3D printing facilities. PLA filament and Resin printers at present
>but strength is very poor. ABS filament and carbon fibre filled resin MAY
>be an option.
>
>I have a lathe, milling machine and router and may yet chew the prototypes
>out of tube and sheet.
>
>Diagram shows possible plastic choices but other suggestions welcome.
>
>A, B: Body and end cap ABS PC PE ... Black (ideally) opaque.
>
>C, D: "Lens" (just a light "port") Actual lens is internal.
>
>
>*Diagram is just to convey concept.Dimensions shown are conceptual only and
>need work.*
>Some parts are too thin. TBD.
>
>Please excuse bad diagram - after too long away I'm learning a new graphics
>package and this was quicker :-) :-(
>
>
>*Diagram here <https://i.sstatic.net/GkUWD6QE.png>(Stack Exchange):  *
>
>*And here (Dropbox)
><https://www.dropbox.com/scl/fi/3ldxj6zl4x2sfpak4v4oj/5YL-body-design-conceptual-Cambrian-prototype-dimensioned-V321-FF.jpg?rlkey=r84hgb7mr3c1i2skkd6hhwq1k&dl=0>*
>
>C & D are both transparent.
>*Ideally D is not needed and C fits as a cap on A.*
>
>In volume C&D can be one part but that's harder in small volume.
>
>A-B join can be permanent.
>C-D pinned / grub screws / thread / magic.
>
>Ideally opened maybe twice at start of life.
>Thread may be nice but not if costly/hard.
>
>Adhesive use acceptable.
>
>BONUS if C & D pass IR light well - allowing PIR to work inside and not
>need a custom "port" for its lens.
>
>Acrylic/PMMA - good light transmission.
>Not robust.
>Polycarbonate - slightly less light transmission. Robust. Expensive.
>------------------------------
>
>Manufacture:
>
>3D print
>"Screw machine"
>Other ... ?
>
>Also on: https://www.facebook.com/share/p/14ZCy9T4FVv/
>--
>http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
>View/change your membership options at
>https://mailman.mit.edu/mailman/listinfo/piclist