Searching \ for '[ee] Bus vs star network wiring...' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=bus+star+network
Search entire site for: 'Bus vs star network wiring...'.

Exact match. Not showing close matches.
PICList Thread
'[ee] Bus vs star network wiring...'
2005\08\22@200642 by William Chops Westfield

face picon face
Sigh.  I'm an idiot.  For several years now, in the background, I've
been working on the ultimate sprinkler controller.  The basic building
block is a networked relay controller, using a sort of primitive rs232
ring network to allow infinite numbers of modules to be connected to
the controller.  The master controller can either be a dedicated
board with display and such, or a PC-type box.

You would think, having worked in computer networking for 20+ years,
and having watched with some amusement how star wiring topologies
like 10baseT, or even AUI-based ethernet "hubs" have replaced the
more intellectually elegant bus and ring wiring topologies due to
easier management, that I would have built a hub-based network.

But NO, I made each relay controller have an "in" and "out" connector,
at rather substantial expense in PCB complexity, and not much actual
elegance in terms of interconnect layout.  And then I need a hub-like
board to do power distribution, fusing, and connector conversion
anyway.  Grr...

(on the plus side, it looks like the modules will be compatible with
either hub or direct-ring topologies, so I can continue building,
coding, and testing with the existing prototype...)

BillW

2005\08\22@201606 by Marcel Duchamp

picon face
William Chops Westfield wrote:
> Sigh.  I'm an idiot.  For several years now, in the background, I've
> been working on the ultimate sprinkler controller.  

I have one of those.

It works like this:
wife: "Turn on the sprinklers!"
me: "Yup."

2005\08\22@210643 by William Chops Westfield

face picon face
On Aug 22, 2005, at 5:17 PM, Marcel Duchamp wrote:
>
> It works like this:
> wife: "Turn on the sprinklers!"
> me: "Yup."
>
This is California where it doesn't rain in the summer, so you need
to water everything.  the house came with 36 zones of sprinklers,
controlled by three separate 12-zone controllers (which seems to
be near "as big as they come.")  They don't coordinate.  Their
configuration is baroque (probably limited by the 256bytes of
ram in an 8051) and entirely volatile.  I can fit a schedule
that lists every freaking minute of the year in a $10 flash
card these days, and a used PC or palmtop costs less than the
price of single 12-zone controller at the gardening store (even
neglecting that they seem to multiply like coat hangers around here.)
It's all STUPID.  I can do better than that.  It may take me 10
years of spare time, but I'm sure I can do better :-)

BillW

2005\08\22@224531 by Timothy J. Weber

face
flavicon
face
William Chops Westfield wrote:
>
> It's all STUPID.  I can do better than that.  It may take me 10
> years of spare time, but I'm sure I can do better :-)

I'm supposed to be doing a wireless irrigation control system geared
toward farmers in a few months.  The guy who wants it has a similar
attitute - apparently farmers spend a HUGE amount of time and money on
irrigation control, and what they do doesn't make a lot of sense (to him).

Wireless seems important for my case because running cables is a
nightmare on a farm.  And I'm guessing the kind of valves used in
farming irrigation systems are probably incompatible with garden-hose
systems, just because that would be too simple.  :)  But I'll let you
know how it goes.
--
Timothy J. Weber                 http://www.lightlink.com/tjweber
spam_OUTtjweberTakeThisOuTspamlightlink.com

2005\08\23@053204 by Chen Xiao Fan

face
flavicon
face
So you are in Southern California, right? I remember back in
Irvine, there are so many sprinklers and sometimes I got
watered as well. :)

Regards,
Xiaofan

{Original Message removed}

2005\08\23@074928 by olin piclist

face picon face
William Chops Westfield wrote:
> This is California where it doesn't rain in the summer, so you need
> to water everything.  the house came with 36 zones of sprinklers,
> controlled by three separate 12-zone controllers (which seems to
> be near "as big as they come.")  They don't coordinate.  Their
> configuration is baroque (probably limited by the 256bytes of
> ram in an 8051) and entirely volatile.  I can fit a schedule
> that lists every freaking minute of the year in a $10 flash
> card these days, and a used PC or palmtop costs less than the
> price of single 12-zone controller at the gardening store (even
> neglecting that they seem to multiply like coat hangers around here.)
> It's all STUPID.  I can do better than that.  It may take me 10
> years of spare time, but I'm sure I can do better :-)

Sounds like a good application for CAN.  At slow enough bit rate (which you
can easily tolerate) you can wire CAN in combinations of star and bus
configurations with maximum node to node wire length in the 100s of meters.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\08\24@204151 by William Chops Westfield

face picon face

On Aug 22, 2005, at 7:45 PM, Timothy J. Weber wrote:

>
> I'm supposed to be doing a wireless irrigation control system geared
> toward farmers in a few months.  The guy who wants it has a similar
> attitute - apparently farmers spend a HUGE amount of time and money on
> irrigation control, and what they do doesn't make a lot of sense.

> Wireless seems important for my case because running cables is a
> nightmare on a farm.

Isn't power a problem, even if control is wireless?  I guess solar
is more likely to work on farmland, but not if you have the sort
of power-hungry valves used in home-style irrigation.

Hmm.   Potentially, this would seem like a good application for
those ad-hoc low-power wireless networks I keep hearing about.
While each a valve site may be out of range of the master controller,
it is probably within range of at least one other valve site...

The most interesting thing I've noticed so far is that once you
notice that you can use a real computer for your master controller
without increasing the cost significantly, it opens up an awful lot
of possibilities.  I mean - megabytes of memory; gigabytes of file
system, graphics and GUIs, easily connected printers, remote access,
and lots of other stuff. (although I have worries
about the durability of the ancient desktop, laptop, or palmtop
platforms I intend to use, and environmentally "hardened" systems
stop being cheap.)

BillW

2005\08\24@212042 by Timothy J. Weber

face
flavicon
face
William Chops Westfield wrote:
>
>> Wireless seems important for my case because running cables is a
>> nightmare on a farm.
>
> Isn't power a problem, even if control is wireless?  I guess solar
> is more likely to work on farmland, but not if you have the sort
> of power-hungry valves used in home-style irrigation.

The valves my friend is currently using can apparently go for
essentially a year (growing season) on one 9V cell, at least for the
amount of watering he does now.  He's trying them with a few dumb timers
(which are apparently pretty expensive?!) this year, so we'll see if
that's true.

> Hmm.   Potentially, this would seem like a good application for
> those ad-hoc low-power wireless networks I keep hearing about.
> While each a valve site may be out of range of the master controller,
> it is probably within range of at least one other valve site...

Exactly - I'm planning to try ZigBee with a mesh topology.  The hype
claims "many months per battery for a typical node," and given that my
duty cycle doesn't need to be high from the application point of view (a
60-second delay after he asks for turn on/turn off is fine), that seems
feasible.

And yes, there's also the option of putting a solar charger on it.
Seems like if there's anyplace you can depend on good sun exposure, it
should be in a field!

> The most interesting thing I've noticed so far is that once you
> notice that you can use a real computer for your master controller
> without increasing the cost significantly, it opens up an awful lot
> of possibilities.  I mean - megabytes of memory; gigabytes of file
> system, graphics and GUIs, easily connected printers, remote access,
> and lots of other stuff.

Yeah!  This is one of the main motivations for the project - tracking
sensor data over time, annotated maps of the farm for easy reading,
remote access, altering behavior based on the weather report, etc.  Once
that sensor/actuator network is in place, the problem is reduced to
desktop UI, which is what I know best.  :)

I have no idea how generally-appealing this solution will be - i.e., how
typical my friend's setup is among farmers - but he claims that he and
many of the small farmers he knows spend HUGE amounts of person-time
controlling their irrigation, and much of it at night (at least in his
particular locale).  So the idea is, he checks his desktop now and then,
and it sounds an alarm if there's some really critical problem;
otherwise, he gets to sleep nights.  Seems like a good thing.
--
Timothy J. Weber                 http://www.lightlink.com/tjweber
.....tjweberKILLspamspam@spam@lightlink.com

2005\08\25@042537 by Alan B. Pearce

face picon face
>> I'm supposed to be doing a wireless irrigation control system geared
>> toward farmers in a few months.  The guy who wants it has a similar
>> attitute - apparently farmers spend a HUGE amount of time and money on
>> irrigation control, and what they do doesn't make a lot of sense.
>>
>> Wireless seems important for my case because running cables is a
>> nightmare on a farm.
>
>Isn't power a problem, even if control is wireless?  I guess solar
>is more likely to work on farmland, but not if you have the sort
>of power-hungry valves used in home-style irrigation.

Well I guess having some sort of battery to supply the peak current, trickle
charged of solar would solve that one,

But hand on, why use wireless? If it is an irrigation system, then there are
pipes wandering around the show, that he will not want to put his plough
through, so why not have power/signal cables in the same trench? Or if the
pipes have to come from a central point above ground to some spray unit in
the middle of a field, why not have the cable laced to the pipe?

2005\08\25@093800 by Timothy J. Weber

face
flavicon
face
Alan B. Pearce wrote:
>
> But hand on, why use wireless? If it is an irrigation system, then there are
> pipes wandering around the show, that he will not want to put his plough
> through, so why not have power/signal cables in the same trench? Or if the
> pipes have to come from a central point above ground to some spray unit in
> the middle of a field, why not have the cable laced to the pipe?

I think the main reason is that the pipes are already there and buried -
digging them up again would be another ~month's work.  And then if the
cable does get damaged somewhere, finding and fixing the problem might
require digging it up again.  I don't know that /all/ the pipe runs are
buried, but my recollection is that most are.  I'll confirm that, though
- we should make sure we're not just using wireless because it's cool.  :)

I read another farmer's account of a similar project where he decided to
use buried cables for simplicity and ended up regretting it - feeling
that wireless would actually have been simpler in the long run.  And
that was before these new low-power wireless protocols started coming
out (or before I was aware of them).

One other note on the project - another benefit could be increased
fault-tolerance.  The night before the last time I visited the farm,
some valve had broken or come detached or something, and so they'd been
pumping water at maximum pressure onto the ground (NOT near the crops)
for an hour or something.  Half-drained their reservoir pond and didn't
do the crops any good.  If there were a pressure sensor out at that
valve, sending readings back to home base, it could have detected that
and shut down the pump automatically.  Another feature that might be
well worth the cost.
--
Timothy J. Weber                 http://www.lightlink.com/tjweber
tjweberspamKILLspamlightlink.com

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...