Hi, Achim,
I've a similar situation as I used star topology of the network wiring
installed in the house. (this is cat5 however)
I solved this by designing a 16 master board that effectively separate each
branches as unique bus.
The 16 masters are in fact two ds2482-800 installed on a single board.
For more flexibility, I grouped the 16 buses on 4 connectors of four buses
using RJ45. Then a 'hub' split the four buses.
In my case two hubs are near the master and the other two are allowing to
start a secondary star topology from other location.
The power (+5V and +12V) is provided to all hubs by a modified PC
powersupply using a 3x1.5mm² orange cable.
The owfs daemon is started as
/opt/owfs/bin/owfs --i2c=/dev/i2c-0:0 --i2c=/dev/i2c-0:1 -m /tmp/ow
Owfs show the consolidated view within a nice folder structure:
*ll /tmp/ow*
...
drwxr-xr-x 1 root root 8 Sep 18 10:15 alarm
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.0
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.1
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.10
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.11
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.12
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.13
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.14
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.15
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.2
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.3
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.4
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.5
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.6
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.7
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.8
drwxr-xr-x 1 root root 8 Sep 18 10:15 bus.9
drwxr-xr-x 1 root root 8 Sep 18 10:15 settings
drwxrwxrwx 1 root root 8 Sep 18 10:18 simultaneous
drwxr-xr-x 1 root root 8 Sep 18 10:15 statistics
drwxr-xr-x 1 root root 32 Sep 18 10:15 structure
drwxr-xr-x 1 root root 8 Sep 18 10:15 system
drwxr-xr-x 1 root root 8 Sep 18 10:15 uncached
If your are interested I can post schematics & board of master and hub.
Pascal
2011/9/18 brucek <bru...@valinet.com>
> If you have an extra pair in the wiring, you could use one pair in and one
> pair out so that the star is functionally a liner network
> bk
>
> -----Original Message-----
> From: ekgnkb3d
> Sent: Saturday, September 17, 2011 7:05 PM
> To: owfs-developers@lists.sourceforge.net
> Subject: [Owfs-developers] increase OW reliability in bad topology
>
>
> Hi all,
> my problem is a little bit off-topic (because not specific to owfs), but
> maybe in this forum are the experienced 1-Wire net experts ;-)
>
> I want to replace in an existing building the 12 analog temperature sensors
> by 1-Wire DS18B20 temp sensors. The used cables are standard telephone
> twisted pair which are shielded together. They are in star-topology, so
> each
> of the 12 cable come from a different room (10m to 50m length each) to the
> central controller. So the complete net have a "weight" of approx 350m.
> The cables shall be reused, because it would be a hell a lot of effort to
> replace them too.
>
> As mentioned in http://www.maxim-ic.com/app-notes/index.mvp/id/148 Maxims
> application note 148 and
>
> http://docs.google.com/viewer?url=http%3A%2F%2Fwww.1wire.org%2FFiles%2FArticles%2F1-Wire-Design%2520Guide%2520v1.0.pdf
> 1-Wire.org Design Guides , the star is the most critical topology for
> 1-Wire
> nets. But what can I do, especially with OWFS, to increase the reliability
> of such a net? I've thought of the following already:
>
> - Using active pull-up bus masters, e.g. DS2482 with FET
> - If using 4 bus-masters, each one have to drive 3 sensors only and
> therefore reduce the weight to approx. 100m each.
> - Slowing down the net to 16kBit/s Standard Mode (how to do this on DS2482
> with OWFS?)
> or even go slower down, to 8 or 4 kBit/s?? (The temperature is only needed
> each minute)
> - Problems usually look like the devices are lost, the search algorithm
> occasionally can't find some devices.
> Is there a way to configure OWFS to retry let's say 100 times until it
> really gives up?
> - Connecting each DS18B20 with 3 wires (+5V and 100nF Capacitor each)
> - Ensuring that the GND and OW-Data wire is a twisted pair of the cable
> - No termination resistor at the end of the cables
>
> any more constructive ideas?
>
> Thanks
> Achim
>
> --
> View this message in context:
>
> http://old.nabble.com/increase-OW-reliability-in-bad-topology-tp32487173p32487173.html
> Sent from the OWFS - Dev mailing list archive at Nabble.com.
>
>
>
> ------------------------------------------------------------------------------
> BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA
> http://p.sf.net/sfu/rim-devcon-copy2
> _______________________________________________
> Owfs-developers mailing list
> Owfs-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/owfs-developers
>
>
>
> ------------------------------------------------------------------------------
> BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA
> http://p.sf.net/sfu/rim-devcon-copy2
> _______________________________________________
> Owfs-developers mailing list
> Owfs-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/owfs-developers
>
--
Pascal
www.brain4home.eu
to subscribe, send a mail to <list-subscr...@brain4home.eu>
------------------------------------------------------------------------------
BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA
http://p.sf.net/sfu/rim-devcon-copy2
_______________________________________________
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
