anelok: the Y-Box paint job, part 3: action

Werner Almesberger werner at
Sun Nov 17 12:38:27 EST 2013

Ron K. Jeffries wrote:
> The *Anelok Y *accessory case looks very nice!

Thanks ! :)

> 1a Full size USB A connects Anelok Y to a [optional?] wired keyboard for
> entering password and related information into Anelok. An alternative is to
> enter data using the jog wheel on Anelok, But that is rather slow and
> tedious in comparison to using a full keyboard.

"optional" as in "bring your own", yes.

> 1b The full size USB is also used to connect Anelok to a PC for injecting
> passwords as needed.

Hmm, there may be a bit of confusion. Here's a picture:

Blue arrows are communication, red arrows are power.

To connect Anelok directly to a PC you either use RF (IEEE 802.15.4)
or a Micro B to Full A cable. That cable is the one also used to
power the Y-Box, which in turn distributes power to keyboard and

When using the Y-Box, you need another cable, the Micro A to Micro B

> 3 Second Micro USB (female? male?)

Cables always end in plugs. With the exception of USB dongles,
devices only have receptacles.

The male/female analogy gets a bit complicated with these connectors.
Most would be some sort of transvestite, looking on the outside one
way but the contacts actually going the other way. And then there's
the A/B role ...

> ** [I am not clear how this is done to avoid various security compromises.

Which one of these "this" ? :)

> Your *Anelok* project is moving at a brisk pace. Good stuff! Assuming your
> goal is an Indiegogo funding campaign,

Yes, Indiegogo looks like a suitable place. I'm watching the funding
campaign of the Neo900 project with great interest. They semi-split
it into R&D and production funding, which is closer to real financing
needs than the Kickstarter approach. Indiegogo is more flexible and
should be able to accommodate such cases as well.

> what are your thoughts about case
> manufacturing process for in volume:
> 1,000
> 10,000
> 100,000 and up <--- this is the goal, right? ;)
> Milling plastic and paint might (?) be adequate up to low hundreds?

Up to one or two dozens, I could do it with my mill. If I can figure
out a formula that produces a decent yield, I could make about two
case part sets per day, at a total machining cost of about USD 20
(material and wear).

Sanding, polishing, and painting (two passes) may take about an hour
per part set. The consumables should be fairly cheap. Not sure if the
machines involved have significant wear.

For larger numbers, I'd definitely have to outsource. There's also
the question of how many kits I could get through customs before they
think I need to spend a few small fortunes on an export license and
get slapped with a ton of extra regulations.

So the further scale would depend on the capabilities of the place
that does the medium-volume production. Around 1000 units, basic
injection molding becomes an option, maybe even a little earlier.

High-quality injection molding would require significantly larger
numbers and a wad of money. I'd also have to outsource the final CAD
work since it would probably be too difficult to adapt my crude
processes to what industry uses.

I may want to look into gravity resin casting, though. That would be
a quite accessible process but with numerous limitations, including
major shrinkage when curing and fumes (think "Breaking Bad", pilot

Casting and molding would also allow for more flexibility in the
design since the two sides of each piece are separate parts of the
mold and can thus be easily machined.

If gravity casting works for this kind of product, it may be suitable
for making quantities in the dozens at a higher speed and lower cost
than milling.

> You are a shining example of how to create open hardware using open
> processes.

Heh, thanks ! :) 

- Werner

More information about the discussion mailing list