I haven't checked this thread in awhile as I have been busy actually
building and testing some of the things I mentioned here.
I hope this gets to some of the folks needing it.
I have seen lots of mentions on other threads about the cost of NMEA183
multiplexers (around $200).
I have board layout, schematic, and firmware to construct
a 4 to 1
NMEA multiplexer. I built the unit on
Radio Shack perfboard
with point to point soldering in an evening. I have used the unit for 3 years without issue. Total part cost was $15.
The unit does buffering (the only way to do 4x4800 baud in and 1x4800 baud out). It also uses a "high watermark" so when it falls behind due to too much
traffic it starts filtering out "unimportant" messages like
GPS satellite signal quality sentences which are transmitted every second even though the data is not time critical. The units can also be "cascaded" so one
may use two units to get a 7 to 1 multiplexer or three to get a 10 to 1 etc.
The output baud rate can also be configured up to 119.2Kbaud if the output is to a PC rather than another
NMEA device (reduces the buffer time and relieves congestion).
In pictures:
---------- --->NMEA in of another MUX
NMEA183#1 ---->| | |
NMEA183#2 ---->| NMEA |_____|-->PC
Serial Port or USB adapter
NMEA183#3 ---->| MUX | |
NMEA183#4 ---->| | -->NMEA dev (AIS,DSC,Autopilot etc.)
----------
If anyone is interested in building one of these I own all the design info and
software so there is no problem with copyright/licensing.
This is not in kit form, it is a schematic, a layout suitable for point to point soldering,
parts list, and firmware file for a PIC18 processor. I can program a few PICs if anyone is interested and doesn't have
a programmer (PIC programmers go for about $50). All the
parts
are available from digikey.com (cheap) or your local
Radio Shack ($$$).
If there is a lot of interest, I can get some boards printed and put them
with a parts kit and pre-programmed PIC. Cost would depend on volume
but should be $50-$75. (as opposed to $15 to point to point a beta).
Email me at this forum if you are really interested. Let me know if you
feel comfortable acquiring your own parts and point to pointing a beta test version or would only be interested in a PC board and parts kit which you could stuff and solder. Point to Point is now, PC board is only if I get
enough interest or help to fool with it.
I don't mean to sound
stand offish but I am busy with other projects and don't want to do a lot
of
work for things that folks are curious about but never actually build.
Any help would be appreciated. Its seems the technical expertise available on this forum is increasing exponentially.
And my physical design prowess is zero. I bully electrons and throw
code but mounting things in nice boxes with bezels is not my forte.
I mounted mine in an old
aluminum box off something i cannibalized.
(I have been known to use spam cans in a pinch). The holes for the LED's were cut with a pair of wire cutters. So if we do pc boards
and someone wants to do a pretty box then I'd love it.
On other fronts, I do have my realtime browser-web server-message bus-hardware interface infrastructure now working and running my off-grid home tasks. (yes i am doing near realtime browser code, no polling and no apps required) I am
finally ready to start on the open source NMEA2000/CANBUS stack I talked
about ages ago so I can duplicate this infrastructure on my boat.
Remember the idea is to buy transducers from Garmin/Raymarine etc and
use tablets/phones/pc's as the interface units talking to a server on the
NMEA 2000 bus and supporting a local
wifi.
I met my cost targets so far as the entire setup running my off grid house cost approx $75. The CANBUS/NMEA stack will sit on the existing
hardware which is already CAN enabled so my target of an instrument server open source for under $100 should be easily achievable if you don't
include my time. Hope to have it all running before my next run to the islands.
cheerio
gello