[sdiy] My SMT projects

[Tony Clark]

> For those less storied builders on the list... what Tony is suggesting is
> actually good engineering practice.  Most people recognize the value of 
> good decoupling... this is better than most.

   Thanks for such a great compliment! 

> An engineer seeks to make a product that performs within a given set of
> specifications... should we tighten the noose on Tony (sorry dude - figure of
> speech) he might respond differently. If we demanded a very low 
> cost.... or a very small package... it might be worth seeing if the 
> parts could be deleted without too much effect on the performance...   

   This is very true.  Budget always dictates the level of engineering in 
a project!

> In Tony's case... I don't think he gonna do a
> big production run (but save one for me if you do...)... so the cost is
> probably NOT an object. Likewise the space is his choice.

   The beauty of SMT design is that once a project is debugged and 
finalized, mass production is a snap!  For my projects, there is still 
elements that would need to be debugged and redone in order to make that 
leap into production, but it could be done with enough interest and (of 
course) money.
   BTW, I always do a run of ten PCBs...but you probably don't want to 
know what it would cost you to get a completed hand-built module from 
me!  Heck, _I_ don't like to know what it costs ME!
 
> The moral of the story (didn't know you were a philosopher eh Tony ?) is that
> is is far easier to add components that you think will ensure good 
> performance...  and delete them later... than to try to add them as an 
> afterthought (especially in  SMT).

   You are absolutely right!  For my company's latest project, I left a 
few "extra" resistor pads in various places in the circuit just in case 
there happened to be some last minute design change that the customer 
wanted.   ...And guess what?  They came in handy!  ;)

> OTOH.... some circuits (few imho) have gotten their 'quirky original 
> sound' by design flaws... leaving out too many components and allowing 
> interaction through the power supply etc...

   True, but I don't think that there has been any design that couldn't 
be duplicated with better engineering.  Cost effective?  Probably not, 
but there is something to be said about circuits that behave the way 
you'd expect them to and each unit performs like the next.  I don't feel 
that I should have to pick out an electronic instrument like choosing 
between a set of stradivarius violins to see which has the best tone!

> Over-engineering is always a GOOD thing from an engineering point of view...
> and (double-plus) the accountants will HATE you   ;^P

   Of course it's even better when you can be the engineer AND the bean 
counter at the same time!  ;)  Having to design products for the 
educational marketplace means that I have to do things as cheaply as I 
can, yet still manufacture a product that needs to survive the abuse that 
teenagers are going to put the product through.  It's a good testiment to 
success when a research paper says it will take $3000 worth of equipment 
to perform an experiment, and my company can market that equipment for 
less than $200 retail!
   So I get to be on both sides of the fence, cost-cutting, and 
high-quality.  I like to save money as much as anyone else, but there is 
a saying that I like to live by:

   "A cheap tool is an expensive tool." - Gardening By The Yard HGTV

   Do it right the first time, and you'll be better off in the long run!

   Cheers,

   Tony, the philosopher

-----------------------------------------------------------------------
         The E-Music DIY Archive - New Site Coming Soon!
-----------------------------------------------------------------------

"We wouldn't want to ship something that doesn't work"
   - Carl Stork, general manager, Microsoft Windows division
     Excerpt from EE Times April 2, 2001