On Wed, 2006-10-25 at 10:36 -0700, Shawn Wilton wrote:
> > Olin is dead right about the connector. It is very reliable, and the
> > last standard from bell systems was that
> > the contact point must be gold. I believe it would be hard to find a
> > better connector. My complaint is from
> > the wires that must be used with the RJ-12.
> 
> 
> 
> I have to figure that with the number of complaints this connector has
> received, it is not a good choice.  Sure, it's cheap, available at any
> hardware store and keyed.  But, I'm not the only person complaining about
> the longevity of the connector or it's ability to maintain a good
> connection.  A standard keyed header of some sort would IMO been a much
> better choice.

It is a well known fact that complainers tend to be MUCH more vocal then
those happy with something. As such, saying the connector is not a good
choice simply because of the complaints received on this list is simply
pointless and without any evidence.

That said, the RJ12 chosen for the ICD2 was a matter of trade offs (as
almost every engineering decision is). A "standard keyed header" is MUCH
larger footprint wise on a PCB then the RJ12. Also, the cables are more
expensive to make. It does have the longevity element, and is more
hobbyist friendly. They made the choice to use the RJ12, and that's
done.

Of the two choices, I would have probably chosen the keyed header since
PCB space during development isn't much a concern. Longevity also isn't
much of a concern for me since I don't regularly plug and unplug boards
from the ICD2. A project is plugged into the ICD2 until it works to the
point I can move to the next project, so it can be weeks before the ICD2
is moved to another board. However, in my opinion, calling the keyed
header a "much better choice" is not true, both options have their
benefits.

TTYL

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