[Prev][Next][Index][Thread]

Re: mouse costs

So, you either scrounge parts from old stuff or you buy new ones.

Scrounging and adapting can be cheap but requires additional skills and
ingenuity as well as the knowledge to know what is worth recovering.

I expect it would be unlikely that you would be able to get a controller
this way. 

Surplus dealers occasionally have very good deals such as the
maxon/minimotor etc gearhead motors I have been able to find every now
and then. I notice that one of these was such obviously good value that
the company I had mine from seem to have sold the rest to another dealer
who I saw selling them at twice that price at the Midland Model
Engineering Exhibition.

Unless you are lucky, most people are probably unlikely to come across a
good number of these deals.

Controllers can be purchased as bare boards, ready to go boards or built
up by hand. I expect that ready built controllers are likely to be
expensive or have expensive development tools. Hand wiring a controller
is probably a forbidding task (I dont want to do it!).

Sensor assemblies are not so bad. Suitable parts are easy enough to come
by and there are a number of designs available. Construction for a
simple set of wall sensors should not be too hard.

Development tools may be easy and cheap or terrifying and expensive.
Some builders would be put off by the thought of learning assembly
language, the whole assemble, burn, test, erase game. Probably something
like Turbo C (free now from Inprise) would allow writing and testing on
the PC with execution on a target board.

I am well into ramble mode now - sorry.

What I am getting at is: I think it would be great if there were a
simple kit that would allow builders to put together a mouse which had
some chance of at least wall-following with scope for improvement to
maze solving. You would need a motor/driver set, sensors and a
controller with friendly tools. No EPROMS or other expensive diagnostic
kit. A chassis could be an option. A choice of components for each
subsystem would allow preferences to be catered for.

The bulk of the programming would be up to the user with some key
routines and a monitor provided. 

Cost should be the lowest possible. All that is needed is a good basis
with the capacity for development.

Competitions could include a 'stock' section where kit-built mice could
run against each other. The kit would need work from the builder before
it was really ready to run. 'Proper' work would be needed to make it
competitive against existing mice.

Korea does this and they have enormous interest in the competition.
Availability of a cost-effective starting point would encourage a lot
more folk.

How often have you been asked how to build a mouse and only been able to
give answers which, in hindsight, could only have been discouraging to
the newcomer?

There is no sense in forcing everyone to re-invent the wheel. Give them
a head start. 

Oops, now I seem to be ranting as well.

Best stop now. If folk are interested in this idea, perhaps we could
give it some time at Royal Holloway.

Pete

In article <3.0.6.32.20010321182214.00937800@mailgate.ftech.net>, Duncan
Louttit <duncan@swallow.co.uk> writes
>Hi Everyone,
>
>       O.K. so nobody likes the idea of limiting mouse costs by selling them
>afterwards.
>
>       Does anyone have any workable ideas for encouraging low-cost mice?
>
>       Regards,
>
>       Duncan
>

Peter Harrison
http://www.cannock.ac.uk/~peteh
Follow-Ups: References: