Anelok: happy anniversary !

Werner Almesberger werner at almesberger.net
Tue Sep 9 16:38:39 UTC 2014


Apelete Seketeli wrote:
> Simulator was a great move, but what about getting some prototypes in
> the hands of (a few) people ?

That's certainly a good idea. But how to do it ?

The complete design information (schematics, BOM, layout, software)
is publicly available, so anyone who wants to make an Anelok PCBA
based on the current design could start immediately.

Well, if anyone plans to do so, please tell me first: I would then
fix a few bugs I found in the design verification. I left them in so
far since that way, the design files correspond to the boards I have
and I don't have to work with two sets of files. But there would be
no point in carrying these bugs forward to new boards.

Making a case would be a bit trickier since my tools generate RML-1
and one would have to adapt the toolpaths to the target machine
anyway. But nothing there is overly hard - it just needs to be done.

Anelok can also be used without case, by not separating the two
boards and gluing a 1 mm piece of plastic on the slider. This would
then be the same setup that I currently have, i.e., like this:
http://downloads.qi-hardware.com/people/werner/anelok/tmp/brd1-oled-0818.jpg

But this of course has haptics that differ dramatically from having
a real case.


I could make some boards and send them out, but that a) takes a lot
of time (maybe two days for a complete board, and with such a
repetitive task there's always the risk of making mistakes, which
could completely ruin a board), and b) it would drain my already
precarious stock of parts.

Running out of parts would be particularly bad if I hit some issue
that forces me to make changes large enough that they need another
board revision. E.g., I have five OLEDs left. They're fragile and in
experiments they suffer some amount of rough handling. One of them
will go to the 2nd board I still have to populate. One may be needed
to replace anything that fails. That leaves three for future
experiments. Oh, and I have one FC-135 (32.768 kHz) crystal left. I
was lucky to have any around at all ...


So my current plan is to bring the design to a state where the main
questions can be answered and where a low-volume production run has
a reasonably low risk of ending in disaster (*). These devices would
be mainly of use for developers and may differ substantially from the
final product. After finishing the design, the next step would be to
define a financing strategy for such a run, and then try to implement
it. Production would be by a contract manufacturer (to be defined),
like we've done with UBB, ATBEN, ATUSB.

(*) E.g., pending unknowns are:

    Will it work on battery ? If yes, how well ?
    Will the slider feel "right" and be sufficiently accurate ?
    Is the Flash large enough for at least basic functionality ?

> I guess producing prototypes may be linked with setting
> up a structure/company around the project ?

Yes, the centralized prototype production approach would need that as
well.

> Overall, the more I read about Anelok, the more I wish to get my
> hands on one. That may just be geek lust though :-).

Great ! I hope lots and lots of people will think like that ;-)

Cheers, Werner



More information about the discussion mailing list


interactive