[hpsdr] My $0.02 worth on board size...

Alex harvilchuck at yahoo.com
Thu May 25 06:57:41 PDT 2006


I agree with Mike, it was very well said and summarized.



This is an important topic, we need to settle on what physical options we are going to support.

I would eliminate the plug jacks on the Janus - those things are unreliable and a major pain when they break.
Go with a higher density connector (like the Delta 44), that will give folks the option to fabricate up whatever harness they would like. It solves the problem of not having enough space for connectors.

If we use a 100x185 format with the external connectors on the left, we can slide the ATLAS bus connector to as far as we can to the left (say keep the end of the connector 30mm from the left side of the board).

For those people worrying about "I want to make a small box as possible", my question back to you is how small is small?
What are your requirements? Fitting in a piece of luggage in the overhead bin of ain airplane? Fitting in the space shuttle?
I propose that I could come up with a packaging solution that will meet your needs, but provide the option for a larger card format. 

By using the card cage that Eric is suggesting we are stuck with the 100x100mm format. Yes, it is an open design and anyone can fabricate any size board they wish, but if we go with my proposal we keep the design open to other card sizes.

Yes and card size larger than 100x100 would cost more, but if we can cram more functionality on a card and reduce the total number of cards, then we are still ahead economically.

I'm trying to design CASMIR and I am beginning to doubt that I can cram everything onto 100x100mm.  

Alex, N3NP
 
----- Original Message ----
From: lyle johnson <kk7p at wavecable.com>
To: Mike Niswonger <mike.niswonger at cox.net>; hpsdr at hpsdr.org
Sent: Thursday, May 25, 2006 12:05:01 AM
Subject: Re: [hpsdr] My $0.02 worth on board size...

***** High Performance Software Defined Radio Discussion List *****

Hello Mike!

> ...   The worst case rework on any board that has
> already been laid out will be moving the I/O 
> connectors from the edge opposite the DIN connector
> to the adjacent edge.  Since I'm not designing the
> boards, I'd like to hear some discussion from the
> designers on how greatly it would affect them.  
> While it might be a little painful right now, wouldn't 
> it be worth it for the added future flexibility?

With the small size of the board in all of these 
discussions, 100mm x 100mm, we have severe constraints for 
routing.  Even if we allow the larger 100mm x 185mm size, 
the 100mm x 100mm is still on the list.

Please look at the Janus preliminary layout (on the wiki) 
for the following dicussion.

You see that with the present layout, the signal flow is 
logical between the ATLAS bus and the board IO.  This 
arrangement also provides maximum isolation (distance) 
between ATLAS and the very sensitive analog IO on Janus.

If the connectors were put on the left vertical edge 
(viewing the PCB layout as positioned on the wiki page), 
we have some significant crowding, particularly on the 
lower left of the board area.  There is no room for the 
analog I/O buffers, although we could re-arrange things so 
that the PWM output was lowest on the board and the AKM 
ADC input buffer cirsuitry on the top edge to make room 
for them.

I suspect we'd have to start putting parts on both sides 
of the board.  Not really a problem, just a consideration. 
 I think that, overall, we are forced to less efficient 
use of the limited board real estate with this 
configuration versus the present configuration.

In the case of Janus, even if we went to the large size, 
we'd still be very cramped in the lower left of the board.

If we used the larger 100mm x 185mm board size and put the 
IO on the farther edge, we'd relieve the crowding -- but 
we'd then require that all boards be 100mm x 185mm.  This 
is almost double the cost of the 100mm x 100mm board.  The 
board cost isn't a big driver in all of this, but it has 
impact if you are buying five or six cards for your ATLAS 
bus.

Note also that with the present configuration, one can 
still make larger boards, such as 100mm x 185mm.  The 
limitation is where the IO connectors reside.  With the 
present configuration, a longer board would have more room 
for IO connectors instead of being restricted to the 100mm 
left edge.

Thus, efficient use of PCB real estate and signal flow 
suggest to me that the current arrangement is better 
alternative.  I understand that from case and mechanical 
convenience perspectives, it may not be optimal.

And, certainly, if we are going to change, it is better to 
do so now rather than later :-)

73,

Lyle KK7P
_______________________________________________
HPSDR Discussion List
To post msg: hpsdr at hpsdr.org
Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
HPSDR web page: http://hpsdr.org
Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/




 1148565461.0


More information about the Hpsdr mailing list