Ray Newman wrote:

>Forrest,
>
>Exactly what quantity per:
>build (or run) ?
>  
>
Generally I build about 25-30 of any given assembly at a time, but 
that's more related to how many I can easily fit on the amount of 
workspace I have right now (which will be increasing shortly).   I have 
one unit that I do 44 of (4 rows of 11 fit in the area I work with), and 
another I do 24 of (4 rows of 6). 

For the most part, I'd be comfortable actually doing runs of around 100 
at a time.

>month ?  year ?
>  
>
The unit I sell the most of  do about 50/month of on average.   I've got 
another product coming which I expect to double or triple that.

>More options as quantities rise.
>  
>
I realize that.  Unfortunately I'm at the low end of the scale, or so it 
seems.

>Potting works the best for VERY low builds if you design pc board with all the connectors on one end.
>  
>
Is the idea that that end is "up" in the mold?  I.E. you build the board 
such that everything which needs to stick out is on that end, then you 
dunk the board such that that end is still out of the epoxy?

My main concern about the potting/encapsulation is things like "what 
happens when I have a capacitor decide to blow apart", and "will 
component X fail (due to heat, or just the epoxy, etc).

I've got a couple of "wart in the middle" type of apps which the 
encapsulation seems ideal for...  I.E there's no displays, there's no 
moving parts, just a couple of connectors (and/or cables sticking out), 
assuming that the tantalum caps don't turn it into a small epoxy bomb.

That said, I've also got some stuff though which needs a bit more 
interaction (buttons, leds, lcd's, etc) which makes me think that 
potting isn't the end all be all answer.   Not to say I don't want to 
play with the potting/encapsulation since I think it's an interesting 
option for many things I plan to do.

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