Ron Ford wrote:

> It really helped me to see the pictures that come with these listings.
> Temperature sensors look like cockroaches with eight legs.  The first pin
> is on the left as you read the name left to right, and it is symmetric
> about the axis that would be the length.
> 
> Do these things need to be mounted on a board where they are to measure the
> temperature, else how do pins interact with wires?

These devices are called SMD:

http://en.wikipedia.org/wiki/Surface-mount_technology

For prototyping or hobby it is possible to mount them as dead bugs:

http://www.geofex.com/Article_Folders/deadbug/deadbug.htm

Then you can solder the wires more easily.

But sounds like you don't know very much about electronics. Maybe you
should buy a small microcontroller evaluation kit and start with some
basics, like blinking LEDs. You can reuse the kit later for your fan
controller. And if you are not a programmer, maybe you should use something
more simple than an AVR, like this one:

http://www.rev-ed.co.uk/picaxe/

Otherwise you'll need some years until you know all the basics to implement
some C program on a microcontroller, develop a RS485 bus protocol and the
rest of the hardware.

-- 
Frank Buss, f...@frank-buss.de
http://www.frank-buss.de, http://www.it4-systems.de

Ron Ford wrote:
> CBFalconer posted:
>> Ron Ford wrote:
>>>
>> ... snip ...
>>>
>>> It really helped me to see the pictures that come with these
>>> listings. Temperature sensors look like cockroaches with eight
>>> legs. The first pin is on the left as you read the name left to
>>> right, and it is symmetric about the axis that would be the
>>> length.
>>>
>>> Do these things need to be mounted on a board where they are to
>>> measure the temperature, else how do pins interact with wires?
>>
>> Of course not.  All you have to do is tell the cockroach where to
>> sense the temperature.  This requires a clear knowledge of
>> cockroach language, together with clear enunciation.  Then you can
>> take proper advantage of the cockroaches extra-sensory perception.
> 
> This sounds like you're having a little fun with my inexperience. 
> Fair enough, I'm attempting to do something ambitious and am a
> certifiable neophyte.
> 
> The cockroaches here on the Rio Grand are commandos with *amazing*
> superhero abilities.  There are so many out on the streets at night
> that one can sensibly talk about a cockroach pressure around here. 
> They have a meeting with their buddies and then move out in pods. 
> I think they're smarter than your average Texans.

You're right, and maybe I should have softened it.  However your
'cockroach' characterization couldn't be missed.  Many thing can
sense temperature, including mercury thermometers, bimettallic
junctions, diode drops, your fingers, crickets, etc. etc.  In fact,
almost anything.  Even Texans.

-- 
 [mail]: Chuck F (cbfalconer at maineline dot net) 
 [page]: <http://cbfalconer.home.att.net>;
            Try the download section.

On Mon, 28 Jul 2008 08:53:25 -0400, Rich Webb posted:

> On Mon, 28 Jul 2008 00:15:28 -0600, Ron Ford <r...@nowhere.net> wrote:
> 
>>On Sun, 27 Jul 2008 03:59:53 +0200, Frank Buss posted:
>>
>>> Ron Ford wrote:
>>> 
>>>> This sounds like a piece of hardware.  Would it presumably be connected to
>>>> the microcontroller by a wire appropriate for a 1.5-5 volt signal?
>>> 
>>> Yes. You can use whatever you have, e.g. something like this one:
>>> 
>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=CKN1025-ND
>>> 
>>>> I would need to measure temperature next to the woodburner and at the other
>>>> end of the house, the part that is "downwind" of the fan.  If they're
>>>> inexpensive, I could have more.
>>> 
>>> Yes, they are inexpensive:
>>> 
>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=568-2051-1-ND
>>> 
>>> If you have long wires from the temperature sensor to the microcontroller,
>>> I would recommend a RS485 bus. Add a small microcontroller with UART
>>> support (about $1) and a RS485 driver, like this one:
>>> 
>>> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=497-2080-1-ND
>>> 
>>> at each temperature sensor. Then you can connect all temperature sensor
>>> boards in one line with one long 4-pole cable (2 for RS485, GND and supply
>>> voltage). Twisted cables for RS485 is a good idea. On each board you can
>>> add some jumpers to set a RS485 bus address.
>>
>>I'll assume that since no else objected amid a lot of opinions that this is
>>a good way to go.
> 
> There is (usually) no single best answer. The sensor noted above will
> offload a lot of the calculations at the cost of some complexity in its
> interface. Many other options, including
>  (1) volts/degree sensors
> <http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=TMP36GRTZ-REEL7CT-ND>;
>  (2) NTC thermistors
> <http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=495-2125-ND>;
>  (3) thermocouples
> <http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=TP-29-ND>;
>  (4) RTDs
> <http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=615-1041-ND>;
>  (5) thermopiles
> <http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=MLX90247-ESF-DSA-ND>;
>  and I'm sure several more that just fell out of my head this morning.
> 
> One solution driver will be the expected temperature ranges next to the
> woodburner and also at the other end of the house. A single sensor or
> sensor family may not be the best in both situations.


I think I want 2 sensors near the woodburner.  I can put one in the chimney
I constructed where temps would presumably be near 2000 degrees.  This
would inform the "woodburner is burning" event.  Another would measure the
air temperature near it.  This won't eclipse 85 degrees if the fan is doing
its job.

Would any of the above sensors be better if it were mounted in the metal
trim that goes around the stovepipe as it penetrates my living room
ceiling?  It would be hot to the touch but not so hot as to violate
building codes as far temps near combustibles.  (Everything I've installed
or bought--the stove itself-- is E.P.A. and local building codes approved.)

I have an unnuanced idea of say, how a thermistor would operate.  How would
you know which part of the wire is changing in its resistance?

> 
> Also pay attention to accuracy as well as precision. The SE95 "Ultra
> High Accuracy" sensor provides 13 bits of precision (0.03 C) but only
> +/- 2 C of accuracy (depending on temperature range & supply voltage).
> That's probably a lot-to-lot max/min and individual chips can probably
> be calibrated more closely.
> 
>>It really helped me to see the pictures that come with these listings.
>>Temperature sensors look like cockroaches with eight legs.  The first pin
>>is on the left as you read the name left to right, and it is symmetric
>>about the axis that would be the length.
>>
>>Do these things need to be mounted on a board where they are to measure the
>>temperature, else how do pins interact with wires?
> 
> Yep. That's another solution driver. Other sensor types, e.g.,
> thermistors, can be mounted remotely at the end of relatively long wire
> runs with a (also relatively) negligible effect on the result; 100K ohms
> for the thermistor, versus say AWG 24 hook-up wire at about 25 ohms per
> 1000 feet.

Thanks, Rich.  There are so many options that it makes it more difficult to
close in on the very vanilla solution that I'm looking for.

I'll go with digikey.  I tried their chat service and am apprised that I
can call 1 800 DIGIKEY to talk with someone who will help me design this
(during business hrs).

When I get a parts list, I'll broach the topic again, with a more realistic
appraisal of the complexity.
-- 
When a new source of taxation is found it never means, in practice, that
the old source is abandoned. It merely means that the politicians have two
ways of milking the taxpayer where they had one before. 8
H. L. Mencken

On Wed, 30 Jul 2008 08:02:45 +1000, Mark L Pappin posted:

> Stef <s...@yahooI-N-V-A-L-I-D.com.invalid> writes:
> 
>> John Speth <j...@yahoo.com> wrote:
> 
> (Actually, Chuck wrote, in response to a post where John had not
> provided attributions, just like he did when he then replied to Chuck)
> 
>>>> Because you snipped attributions, I have no way of knowing who
>>>> wrote the material with >>> or >> leading the sentences.
>>>
>>> Does it really matter?
>>
>> Depending on your definition of "really matter" yes...no.
>>
>> But I sure like to know who wrote what, it makes conversations much
>> clearer.
> 
> Indeed.
> 
> Knowing who wrote what helps the reader decide how much to trust
> what's written - it's not just what you say, but what you've said in
> the past (and what others have said about you, and you about them yea
> back unto the N-th generation) that matters in a reputation-based
> system like Usenet.  When you do not supply attributions for something
> you quote then instead of serving as backup or counterpoint to what
> you're saying, it's the equivalent of "I heard somewhere that ...".
> 
> mlp

Mark,

Thanks for the heads-up to bring this little project to comp.arch.embedded.

My sysadmin buddy is bringing over a solder kit tomorrow along with a book
he recommends.

Bill Cunningham might be a nice guy, but I don't think he's ever gonna
learn C.  I skim your tutorials.  Maybe you could herd them together and
have a book idea to peddle to a publisher.  _My Life with Bill: a
Compendium of usenet C_
-- 
Unquestionably, there is progress. The average American now pays out twice
as much in taxes as he formerly got in wages. 1
H. L. Mencken

On Wed, 30 Jul 2008 09:52:24 -0700, Mike H posted:

> Ron Ford wrote:
> 
> - snip-
> 
>> I have no idea what to even google for to find out more.  Who said
>> "jeepers, mr. peabody," rocky the squirrel?
>> 
> 
> That would be Sherman the Boy.
> 
> Mr. Peabody was the time-traveling, talking dog
> (with glasses and a pipe)
> 
> same Rocky & Bullwinkle program tho.

Those were some of the first cartoons I saw, pre second grade.

I always remember the cartoons I was watching in second grade, because they
were routinely interupted by those boring people with their Watergate
hearings.

My favorites were always Looney Tunes, which I watched religiously on
staurday morning on one of the available 4 channels.  My girlfriend brought
over Looney tunes from her Netflix the other night, and I can assure you
that they are as classic as ever.  Chuck Jones is a genius.

As I was weatherproofing the chimney from the monsoon rains, I went out on
my lawn with rubber gloves and roof cement in my hands. Fifteen feet away
was a bird I hadn't seen before.  The ladies passing me on the sidewalk
assured me that it was a road runner.

As I finished my task, it crossed the road, a large yellow object in its
beak.
-- 
We must be willing to pay a price for freedom. 4
H. L. Mencken

On Wed, 30 Jul 2008 21:07:28 -0600, Ron Ford <r...@example.invalid>
wrote:

>I think I want 2 sensors near the woodburner.  I can put one in the chimney
>I constructed where temps would presumably be near 2000 degrees.  This
>would inform the "woodburner is burning" event.

Bearing in mind that there is rarely exactly one solution to a real
world problem:

The high-temp monitor sounds like a job for a type K thermocouple
married to something like an Analog Devices AD597. That handles much of
the dirty work and can also be configured to provide an "on/off" signal
which could be pretty easily interfaced to a microcontroller.

> Another would measure the
>air temperature near it.  This won't eclipse 85 degrees if the fan is doing
>its job.

That could be a bog-standard thermistor hooked up in a voltage divider
network to a microcontroller's A/D port.

>Would any of the above sensors be better if it were mounted in the metal
>trim that goes around the stovepipe as it penetrates my living room
>ceiling?

Probably not, since there would presumably be a considerable time lag.

>  It would be hot to the touch but not so hot as to violate
>building codes as far temps near combustibles.  (Everything I've installed
>or bought--the stove itself-- is E.P.A. and local building codes approved.)
>
>I have an unnuanced idea of say, how a thermistor would operate.  How would
>you know which part of the wire is changing in its resistance?

For a thermistor, it's not the wire (although the wires will exhibit a
small but negligible resistance change) but the magic stuff inside the
epoxy or glass bead. For a thermocouple, the signal is the voltage
generated by the thermal gradient along the wires; using different
metals for the two wires results in a difference in the voltage gradient
and a measurable voltage difference.

>Thanks, Rich.  There are so many options that it makes it more difficult to
>close in on the very vanilla solution that I'm looking for.
>
>I'll go with digikey.  I tried their chat service and am apprised that I
>can call 1 800 DIGIKEY to talk with someone who will help me design this
>(during business hrs).
>
>When I get a parts list, I'll broach the topic again, with a more realistic
>appraisal of the complexity.

Sounds like an excellent learning experience!  ;-)

-- 
Rich Webb     Norfolk, VA

On Wed, 30 Jul 2008 22:11:41 -0400, CBFalconer posted:

> Ron Ford wrote:
>> CBFalconer posted:
>>> Ron Ford wrote:
>>>>
>>> ... snip ...
>>>>
>>>> It really helped me to see the pictures that come with these
>>>> listings. Temperature sensors look like cockroaches with eight
>>>> legs. The first pin is on the left as you read the name left to
>>>> right, and it is symmetric about the axis that would be the
>>>> length.
>>>>
>>>> Do these things need to be mounted on a board where they are to
>>>> measure the temperature, else how do pins interact with wires?
>>>
>>> Of course not.  All you have to do is tell the cockroach where to
>>> sense the temperature.  This requires a clear knowledge of
>>> cockroach language, together with clear enunciation.  Then you can
>>> take proper advantage of the cockroaches extra-sensory perception.
>> 
>> This sounds like you're having a little fun with my inexperience. 
>> Fair enough, I'm attempting to do something ambitious and am a
>> certifiable neophyte.
>> 
>> The cockroaches here on the Rio Grand are commandos with *amazing*
>> superhero abilities.  There are so many out on the streets at night
>> that one can sensibly talk about a cockroach pressure around here. 
>> They have a meeting with their buddies and then move out in pods. 
>> I think they're smarter than your average Texans.
> 
> You're right, and maybe I should have softened it.  However your
> 'cockroach' characterization couldn't be missed.  Many thing can
> sense temperature, including mercury thermometers, bimettallic
> junctions, diode drops, your fingers, crickets, etc. etc.  In fact,
> almost anything.  Even Texans.

The acronym I learned last night was DIP: dual in-line package.  That's for
these devices with cockroach symmetry. 
-- 
We must be willing to pay a price for freedom. 4
H. L. Mencken

On Thu, 31 Jul 2008 04:00:15 +0200, Frank Buss posted:


> But sounds like you don't know very much about electronics. Maybe you
> should buy a small microcontroller evaluation kit and start with some
> basics, like blinking LEDs. You can reuse the kit later for your fan
> controller. And if you are not a programmer, maybe you should use something
> more simple than an AVR, like this one:
> 
> http://www.rev-ed.co.uk/picaxe/
> 
> Otherwise you'll need some years until you know all the basics to implement
> some C program on a microcontroller, develop a RS485 bus protocol and the
> rest of the hardware.

I bookmarked this page  (Es ist ja einfacher, englisch zu reden.)  It seems
like it has a lot of stuff for starters.  I've got quite a few items in my
"to buy" list for this project.  Yeah for my rebate check.  It improves
computer science for the hobbyist so much to have a proper budget for it.

-- 
When a new source of taxation is found it never means, in practice, that
the old source is abandoned. It merely means that the politicians have two
ways of milking the taxpayer where they had one before. 8
H. L. Mencken