[ExI] 3d printing a Curta

[Tomasz Rola]

On Sat, 16 Feb 2013, Mike Dougherty wrote:

> Given a decent "spec" like this (or better if necessary)
> http://www.vcalc.net/cu.htm
> 
> How difficult would it be to print one of these amazing devices, even
> with some assembly required?

>From wiki, they say some curious owners tried to disassemble their curtas, 
this was fairly easy, but the other way required specific knowledge:

[ http://en.wikipedia.org/wiki/Curta_calculator#Cost ]

"The Curta Type I was sold for $125 in the later years of production; the 
Type II was sold for $175. While only 3% of Curtas were returned to the 
factory because of repairs under warranty,[2] a small, but significant 
number of buyers returned the Curta in pieces. Many purchasers attempted 
to disassemble the Curta. Reassembling the machine was more difficult as 
assembly required intimate knowledge of the orientation and installation 
order for each part and sub-assembly, plus special guides designed to hold 
the pieces in place during assembly. Many identical looking parts, each 
with slightly different dimensions, required test fitting and selection as 
well as special tools to adjust tolerances."

And I suspect you'd need the actual factory blueprints. It was a 
mechanical marvel in times when lots of people would like to have one. Yet 
nobody cloned the thing, repeated design process, nada, null. Maybe it was 
too difficult to replicate outside Curta's head. I mean, there were plenty 
of mechanical calculators, only they required a desk.

However, my closest relation with curta is 1. reading on the net 2. 
spotting one in the hands of half-mad physicist in "Lost". And I must 
admit it, I'm not a biggest fan of purely mechanical comps - too much 
trouble with tolerances. I guess curtas have to be clean to deliver, so in 
the field there may be a problem.

[...]
> Are there any legal / IP restrictions from doing something like this?

Don't know, but the design had been patented.

> What other nerd toys would you 3d print just for kicks?

Eh. I am so trivial. I'd try to make some kind of electromechanical 
computer, something that would be easy on tolerances because it'd use 
switches and the like. Wait, core memory would be cool to have, too.

I wouldn't do Turing machine, because it is so awesomely useless in 
practical applications :-).

But I am long time fancying myself with idea of mechanical lambda 
calculator - if only I knew how to do this _and_ posessed 3d printer, 
yep, could be very cool. Especially playing with recursion on it, heh. 
Lambda goo.

Also, there was whole class of analog computers, using water and parts to 
enable users playing with models described by complicated differential 
equations, like MONIAC, differential analyser and others:

[ http://en.wikipedia.org/wiki/MONIAC_Computer ]

[ http://en.wikipedia.org/wiki/Differential_analyser ]

Of course there were also analog computers using electric current but I 
guess those would be hard to print.

Some slide rules were cool:

http://en.wikipedia.org/wiki/Slide_rule

And last but not least, abacus:

[ http://history-computer.com/CalculatingTools/abacus.html ]

I still have my old "Russian abacus", as they call it on the page. But my 
variant has only 10-beads rods in it, so maybe it's not Russian. Perhaps 
it is Polish abacus :-).

Regards,
Tomasz Rola

--
** A C programmer asked whether computer had Buddha's nature.      **
** As the answer, master did "rm -rif" on the programmer's home    **
** directory. And then the C programmer became enlightened...      **
**                                                                 **
** Tomasz Rola          mailto:tomasz_rola at bigfoot.com             **