Re: Looking for Temp and humidity sensors

[nobody@xxxxxxxxxxxx (Dave Houston)]

Sure. One could design a more functional Bobcat (e.g. SECU16i + temperature
sensors).

ADI's primary business is contract manufacturing of SMT boards for others.
If anyone is likely to be efficient, they are. However, Chinese factories
can do the same work in small quantities for much less. So, design what you
want and send it off to one of them to fabricate.

I designed a small battery powered generic wireless ADC module using a
protocol that is understood by the BX24-AHT. It costs less than $10 to make
and could retail for about $20 but will cost about $6K for the required FCC
ID tests.

The real problem is that *none* of the HA related companies have found much
of a market. I recall seeing something from Dan Boone about a year ago
before he left ADI that they had shipped 5000 Ocelots.

Anyone with the skills to roll their own can find inexpensive hardware. The
JK Microsystems PicoFlash & Pico-I/O can do a lot for about $175.

     http://www.jkmicro.com/products/picoflash.html
     http://www.jkmicro.com/products/expansionboards/pico-io.html

The $1.60 Insteon PLC chip *might* lead to plug-in sensors of the type you
want.

Of course, nobody will be able to walk about without causing disputions to
your readings. ;)

"Robert Green" <ROBERT_GREEN1963@xxxxxxxxx> wrote:

>"Dave Houston" wrote:
>
><stuff snipped>
>
>> The Bobcat temperature node includes a power supply, an RS485 interface,
>an
>> ADC chip (I think they use a PIC but it's been a while since I looked
>inside
>> one), an enclosure and an LM34 temperature sensor.
>>
>> The Bobcat humidity node includes all of the above but with a Humirel
>> humidity sensor replacing the LM34. Humidity sensors are not cheap.
>
>Too bad.  I see other topologies like Phil Anderson's that make me believe
>the Bobcats could either be made more functional or less expensive, overall,
>but I suppose for limited uses where the sensor needs to be bulletproof and
>highly reliable, the Bobcats make sense.  With 16 port Ethernet hubs going
>for less than $50 I would have thought we would have seen more things
>appearing more quickly for networking like remote temperature sensing and
>even electrical load control.
>
>
>> If you try to duplicate the design you will find that you really cannot
>> build them and sell them through distribution channels for much less than
>> what ADI charges.
>
><sigh>  And economies of scale never amount to much in small operations like
>ADI's so the price isn't likely to come down as manufacturing techniques
>improve.  I wonder if HA will ever see a time where component prices drop
>the way hard drive prices have for PC's?  X-10 seems to have come the
>closest in lowering prices through high volume.  It's probably what make it
>difficult for some people to jump to the next level - almost every other HA
>control technology is at least double the cost of X-10.  That sets up a
>mental expectation that it will also be twice as good.
>
>> If you already have some type of controller with ADC inputs and a power
>> supply and a little knowledge you can DIY and save (if you conveniently
>> ignore the cost of the controller). The cheap sensors need something like
>an
>> SECU16 or other controller.
>
>It doesn't seem that many people are interested in knowing the temps in each
>room in the house.  I'm probably just a data freak but it seems to me a home
>automation system HAS to know details like that to effectively conserve
>energy.
>
>> I think your plans are overkill but you should look at some of the
>one-wire
>> interfaces which allow you to network lots of relatively inexpensive
>sensors
>> (at least for temperature, RH is still not cheap). You still need some
>type
>> of controller or PC to make sense of the data.
>
>I just want something simple that can log the temperatures and RH's of the
>various sensors to an ASCII file that I can analyze later.  If I had the
>technosmarts to crack the RF protocol of the RatShack remote
>thermo/hygrometers (six units, three with RH, three without) and record that
>output to a text file, I'd be happy.  But in reality, I'd like to monitor
>three temps per room, at least (ceiling, floor and about 3' from the floor.
>I don't need RH from each room, but I'd like it from critical rooms like the
>bathrooms, the basement, the pantry, the attic and the kitchen.
>
>Why so many sensors?  I believe it will be the foundation of a truly "smart"
>home that can look for data anomalies and thus detect open windows, clogged
>furnace filters, improperly operating dampers and a whole host of other
>things.  A house is like a human being in that it tries to maintain a fairly
>constant internal environment.  To do that, you have to have lots of sensory
>input.