Jake Anderson wrote:
>>> Set them a problem on paper (a drum
>>> spinning at 3000rpm hits a 13kg steel box at velocity Y how big a
>>> bearing does it need) and they will solve it in minutes.
>>>
>>
>> Does this problem even have a solution? <:-O
>>
>> Vitaliy
>>
>>
> 2 actually.
> a) You work out the acceleration of the 13kg mass is after working out
> the rotational inertia of the drum and from that determine the force
> then select appropriate bearing.
>
> b) You put the biggest one in that will fit and hit a few things with it
> to see how long it takes before the bearings get "crunchy"
> (with a 100mm drum and 60mm bearings we get about 6 events with 5x 1
> minute fights per event (about 5 hits), the fights are sposed to be 3
> minutes but somebody breaks first ;->)

I guess I didn't understand the context of the problem, nor the problem 
itself. You didn't provide all necessary information. ;-)

> Its kinda the NASA/ New Space divide. One is simulate and calculate, the
> other is build and test. I prefer build and test, at least that way when
> your done you have a working product, or a cool explosion, Both Good ;->

It's a constant tradeoff in engineering. I prefer a balanced approach. What 
we normally do, is have several design reviews (as a project is 
progressing). Spending 15 minutes to an hour doing a design review 
(depending on the complexity of the circuit) can usually guarantee that at 
least 90% of mistakes are caught, and the remaining mistakes are probably 
not show stoppers. In some cases, it eliminates the need for an extra 
iteration (for boards, usually means a savings of around $100 and anywhere 
between 3-5 days of lead time).

However, the cost/benefit ratio of a design review grows exponentially. It 
is often better to build a prototype with some mistakes, greenwire the 
changes, test, and then order another set, than to spend several extra hours 
double-checking everything. The former approach has a higher certainty of 
success, and usually means less engineering time (although perhaps more time 
overall).

Best regards,

Vitaliy 

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist