[En-Nut-Discussion] RFC: Ethernut Bus System

Dave Smart SmartFamily at mchsi.com
Thu Apr 3 15:23:39 CEST 2003


Solution 2:

I'd cast my vote for solution 2, PC104 connectors. I'm only familiar with
PC104 based on pix in trade mags and some casual reading way back when, but
I envision Ethernut 2 as the bottom board, so would not have pins beneath
it. Thus, Ethernut 2 would be about the same height as today. The additional
boards would have the full up/down connection system, and thereby be
stackable.

I do have concerns about the overall height, but that can be accounted for
in packaging. Today my add-on board is component side down into the Ethernut
board, with components carefully located to not interfere when mated. This
makes a slim sandwich, which would be a bit taller with PC104 design, but
not much.

Additional thoughts - Perhaps smaller "break-out" connectors of the DIL type
as used today to gain access to _sets_ of I/O. What I'm thinking here is
that many apps may only need small subset of the I/O. If there were several
smaller DIL connectors that each had a port, raw and logic power and ground,
and were arranged on even 2.54mm spacing, then a large (64pin) mating
connector could straddle them all for some users needs, but in another app
perhaps a single 16pin mate would be sufficient. (I hope this paints the
same picture in your mind like mine)

perhaps this will show the concept.

* * * * * * *   * * * * * * *   * * * * * * * 	3 sets of 14 pin headers
* * * * * * *   * * * * * * *   * * * * * * *

* * * * * * * * * * * * * * * * * * * * * * * * overlaid with single 48 pin
header
* * * * * * * * * * * * * * * * * * * * * * * *


The other thought when you mention PC104, and again with perhaps enough
knowledge of that system, is to be pin compatible to that environment. The
perhaps standard PC104 options could be used, or if Ethernut ever migrates
to another CPU that is already supported, the breadth of devices in that
environment is quite wide.

Solution 1:
I don't know about the 96 pin DIN, but your comment "large" is a deterrent.

Solution 3:
"unreliable" puts my thumb down.

Solution 4:
I'm always challenged to make things smaller, but perhaps this makes the
cost go up significantly with a single source as well as risk of [lack of]
availability?




-----Original Message-----
From: en-nut-discussion-admin at egnite.de
[mailto:en-nut-discussion-admin at egnite.de]On Behalf Of Harald Kipp
Sent: Thursday, April 03, 2003 6:07 AM
To: en-nut-discussion at egnite.de
Subject: [En-Nut-Discussion] RFC: Ethernut Bus System


Hi all,

As you may know, Ethernut 2 will be plug-in compatible to
Ethernut 1.

For long term planning I'm considering some kind of bus
system for easy add-on board design. I already discussed
this in some private emails, but would like to get more
people involved.

One thing I'd like to discuss are the mechanical connectors.

Solution 1: Using 96 pin DIN 41612 connectors. Disadvantage
is, that they are very large. Advantage is, that housings are
easily available and, because this is a backplane system,
it's very flexible.

Solution 2: Using PC104 connectors. Disadvantage is, that
stacking up boards is limited and component height matters.
Advantage is, that they are 2mm pitch and much smaller than
1" pitch connectors.

Solution 3: Using SIMM connectors. They are looking very
unreliable to me, number of pins and board size is limited.

Solution 4: Using one of those very tiny (down to 1mm pitch)
connectors offered by some manufacturers without second source
and build a backplane system.

What do you think?

Harald

_______________________________________________
En-Nut-Discussion mailing list
En-Nut-Discussion at egnite.de
http://www.egnite.de/mailman/listinfo/en-nut-discussion




More information about the En-Nut-Discussion mailing list