Searching \ for '[EE]: Temperature Sensors? How'bout lamps?' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/io/sensors.htm?key=temperature
Search entire site for: 'Temperature Sensors? How'bout lamps?'.

Exact match. Not showing close matches.
'[EE]: Temperature Sensors? How'bout lamps?'
2002\12\01@165608 by

-------    [OT]: => [EE]:   -------

I'm looking at a small incandescent  lamp 60V-50mA,
lying on my desk. (Even don't ask me what device I took
it from :-) When cold it shows 120ohm, when under 60V,
I guess, it's current should be 50mA, so resistance
60/0.05=1200ohm.
Assuming temperature under 60V is about 2000 C we
have (1200-120)/2000 = aprox 0.5 ohm per 1grad C. Or this
thing should double it's resistance at 250 C.
So needed current I = 5V / 250ohm = 20 mA.
Power dissipation  P = 5V * 20 mA = 0.1 W much less then
60V * 50 mA = 3W.

So voltage across the lamp should change from ~2.5V
when cold to 5V when oven is heated to 250 C.

Why not use this as "Temperature Sensor" ?
20 mA current source is not a big problem, I think.

Mike.
-------
Correct me if I'm wrong, please.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservmitvma.mit.edu with SET PICList DIGEST in the body

Mike Singer wrote:
{Quote hidden}

Suppose you insert a small lamp inside the oven and measure its filament
resistance, what can happens?

The lamp has vacuum or an dumb gas into the bulb to avoid oxigen to oxidize
and burn the filament. Small lamps normally have just vacuum, it is cheap
to produce that way.  With vaccum, the temperature will have some
difficulty to be transported from the oven internal radiation to the lamp's
filament.  I mean, the filament will receive just infra-red initially,
later on the thermal vibration can reach the filament from the lamp's body
and glass posts and wires holding the filament. It means, the lamp
resistence change will take some time to reflect the oven temperature, time
not fast enough to allow a good power control.

Then suppose we build a circuit to measure the actual oven thermal heating
elements electric current consume during heating.  A table could be built
doing lots of measurements using a thermometer and lots of patience, what
can happens?  The resistance of such element will reflect the temperature
of the element rubishing like a shy girl, not exactly the oven's
temperature.  I think that when the oven is at 200°C, the element red
probably will be around 1000°C, when the thermostat turns it off and stays
there, probably seconds before it turns on again its temperature should
still much higher than the oven, a guess would say around 300°C or more.
During this week I will try to measure it with the non-contact thermometer

If measuring the actual power consume of such thermal element, I guess it
will not help much, since the change in current will not (I guess) be much
noticeable from a warm to a hot oven, when you feel 100% power to an
element, it turns bright red, and this means very high temperature, for a
cold or hot oven, I may be wrong.

I remember seen a color table, meaning "color of the temperature". It was a
simple color bar printed on a thermal scale (like a ruller), from dark red,
going to shine red, redish yellow to bright yellow to bright white in 6 to
7 inches. You can guess DARK RED is how your car's brake pads and disk goes
after a strong break at the freeway when you spot a police officer with a
radar gun - and WHITE is simply hell's temperature of a star.  By simply
using this ruller in front of your eye and matching the color, you can say
what temperature the target is, of course, it should be at least dark red.
I never found this scale for sale, if someone has any tip, it will be
welcome.

You can relate this issue to a range's spiral heater, the pan temperature
could be around 300°F, but the spiral will be middle red, and it can
represent something around 1500°F. When the thermostat turns off the
heater, it probably stays at the pan temperature or a little bit higher,
but it happens since there is a direct physical contact between the parts.
If there was air in middle, as into the oven, I don't know what the heater
temperature could be.  Suppose the minimum it goes is exactly the oven's
temperature, then that's ok, but it requires to cools off to oven's
temperature before taking a valid resitance reading, and that time could be
a killing for any temp control process.

By other side, any other exposed thermal element, as a simply AWG32 (very
thin) nickel-chromium wire, or even a disconnected secondary heater element
could be used for such measurement.

Wait for this week news... :)

VV46NER

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservmitvma.mit.edu with SET PICList DIGEST in the body

Wagner Lipnharski wrote:

> Suppose you insert a small lamp inside the oven and measure its
filament
> resistance, what can happens?
>
> The lamp has vacuum or an dumb gas into the bulb to avoid oxigen to
oxidize
> and burn the filament. Small lamps normally have just vacuum, it is
cheap
> to produce that way.  With vaccum, the temperature will have some
> difficulty to be transported from the oven internal radiation to the
lamp's
> filament.  I mean, the filament will receive just infra-red initially,
> later on the thermal vibration can reach the filament from the lamp's
body
> and glass posts and wires holding the filament. It means, the lamp
> resistence change will take some time to reflect the oven temperature,
time
> not fast enough to allow a good power control.

I placed it in boiling water - resistance was 155 ohm, pulled it out - resistance dropped to 125 ohm in about 10 seconds. (120 ohm at 25C). At 200-250C this time will be much less, much less then oven's.    It's not a rocket science (Dale used this term recently), so automatic control theory should be implemented at a very basic level.

> Then suppose we build a circuit to measure the actual oven thermal
heating
> elements electric current consume during heating.  A table could be
built
> doing lots of measurements using a thermometer and lots of patience,
what
> can happens?  The resistance of such element will reflect the
temperature
> of the element rubishing like a shy girl, not exactly the oven's
> temperature.  I think that when the oven is at 200°C, the element red
> probably will be around 1000°C, when the thermostat turns it off and
stays
> there, probably seconds before it turns on again its temperature
should
> still much higher than the oven, a guess would say around 300°C or
more.
> During this week I will try to measure it with the non-contact
thermometer
> I should receive on Tuesday.

If you think that heating due to power dissipation of the lamp           P = 5V * 20 mA = 0.1 W  would  spoil measurements take two or more lamps, this will reduce current twice or more and power dissipation will be reduced four times or more on each lamp. You might work at lower voltages as well decreasing power dissipation quadratically. You may get a lamp with the bigger resistance.

> If measuring the actual power consume of such thermal element, I guess
it
> will not help much, since the change in current will not (I guess) be
much
> noticeable from a warm to a hot oven, when you feel 100% power to an
> element, it turns bright red, and this means very high temperature,
for a
> cold or hot oven, I may be wrong.

Lamp discussed works at very small current not at 100% power. You'll burn AVR with 60V input voltage.
Best regards.
Mike.

> I remember seen a color table, meaning "color of the temperature". It
was a
> simple color bar printed on a thermal scale (like a ruller), from dark
red,
> going to shine red, redish yellow to bright yellow to bright white in
6 to
> 7 inches. You can guess DARK RED is how your car's brake pads and disk
> goes
> after a strong break at the freeway when you spot a police officer
with a
> radar gun - and WHITE is simply hell's temperature of a star.  By
simply
> using this ruller in front of your eye and matching the color, you can
say
> what temperature the target is, of course, it should be at least dark
red.
> I never found this scale for sale, if someone has any tip, it will be
> welcome.
>
> You can relate this issue to a range's spiral heater, the pan
temperature
> could be around 300°F, but the spiral will be middle red, and it can
> represent something around 1500°F. When the thermostat turns off the
> heater, it probably stays at the pan temperature or a little bit
higher,
> but it happens since there is a direct physical contact between the
parts.
> If there was air in middle, as into the oven, I don't know what the
heater
> temperature could be.  Suppose the minimum it goes is exactly the
oven's
> temperature, then that's ok, but it requires to cools off to oven's
> temperature before taking a valid resitance reading, and that time
could be
> a killing for any temp control process.
>
> By other side, any other exposed thermal element, as a simply AWG32
(very
> thin) nickel-chromium wire, or even a disconnected secondary heater
element
> could be used for such measurement.
>
> Wait for this week news... :)
>
> VV46NER

I wrote earlier:

{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestmitvma.mit.edu

Mike Singer wrote:

> > (light globe) not fast enough to allow a good power control.
>
>    I placed it in boiling water - resistance was 155 ohm, pulled
> it out - resistance dropped to 125 ohm in about 10 seconds.
> (120 ohm at 25C). At 200-250C this time will be much less,
> much less then oven's.

Sounds usable. :o) How much quicker is it without
the glass?? You do know the solder on the globe
base contacts will melt... ;o)

If you are after a cheap and rugged sensor, what about
an exposed jug element, about \$2 from the supermarket.
You know the type, ceramic former wrapped in coils
of wire, two long brass connector rods, no soldered
connections and enough length of exposed wire to
give quick response (hopefully)...
-Roman

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestmitvma.mit.edu

Roman Black wrote:
> > > (light globe) not fast enough to allow a good power control.
> >
> >    I placed it in boiling water - resistance was 155 ohm, pulled
> > it out - resistance dropped to 125 ohm in about 10 seconds.
> > (120 ohm at 25C). At 200-250C this time will be much less,
> > much less then oven's.
>
> Sounds usable. :o) How much quicker is it without
> the glass?? You do know the solder on the globe
> base contacts will melt... ;o)
>
> If you are after a cheap and rugged sensor, what about
> an exposed jug element, about \$2 from the supermarket.
> You know the type, ceramic former wrapped in coils
> of wire, two long brass connector rods, no soldered
> connections and enough length of exposed wire to
> give quick response (hopefully)...
> -Roman

No solder, just small glass tube with two outgoing wires.
I didn't try it without glass, fast enough as it is to me.
Glass tube provides stability. Wolfram spiral is very thin.
I think, in non-mechanical sense it's more rugged then
exposed element.

Mike.

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestmitvma.mit.edu

On Sun, 1 Dec 2002, Mike Singer wrote:

*>   Why not use this as "Temperature Sensor" ?
*>20 mA current source is not a big problem, I think.

Because at 250C the lamp will relatively quickly plate the glass with
filament vapors unless you run it at a significantly lower current. All
you need to read is ohms, and you do not need 20mA. 1mA is enough. I have
made RTDs from small lamps before, they work better if one collapses the
glass so it touches the filament. This is done by holding the lamp into a
mild gas flame (one that does not soot) and then annealing the glass (SLOW
cooling - more than 10 minutes from 700C to 100C).

Peter

--

Peter L. Peres wrote:
> On Sun, 1 Dec 2002, Mike Singer wrote:
>
> *>   Why not use this as "Temperature Sensor" ?
> *>20 mA current source is not a big problem, I think.
>
> Because at 250C the lamp will relatively quickly plate the glass with
> filament vapors unless you run it at a significantly lower current.
> All you need to read is ohms, and you do not need 20mA. 1mA is
> enough. I have made RTDs from small lamps before, they work better if
> one collapses the glass so it touches the filament. This is done by
> holding the lamp into a mild gas flame (one that does not soot) and
> then annealing the glass (SLOW cooling - more than 10 minutes from
> 700C to 100C).
>
> Peter

... I will do some experiments with a thin Nicrhome wire, the AWG32 is 10
ohms / ft at 25°C.
perhaps a piece of FE4 (1 x 3 inches) all wrapped around with Nichrome, can
hold aprox 10 loops per inch, 30 loops, 60 inches, around 5ft or 50 ohms.
For SMD soldering, the FE4 will be perfect, since it will be in the middle
of other boards, and will reflect very close temp of other boards too.

VV46NER

--

Wagner Lipnharski wrote:

> Peter L. Peres wrote:
> > On Sun, 1 Dec 2002, Mike Singer wrote:
> >
> > *>   Why not use this as "Temperature Sensor" ?
> > *>20 mA current source is not a big problem, I think.
> >
> > Because at 250C the lamp will relatively quickly plate the glass
with
> > filament vapors unless you run it at a significantly lower current.
> > All you need to read is ohms, and you do not need 20mA. 1mA is
> > enough. I have made RTDs from small lamps before, they work better
if
{Quote hidden}

I suspect Nicrhome has too small temperature resistivity expansion index (I don’t know the term in English, correct me if possible).
Mike.

--

> Because at 250C the lamp will relatively quickly plate the glass with
> filament vapors unless you run it at a significantly lower current.

Huh?  Vaporize tungsten at 250C? (About 600K?  Aren't typical filament
temperatures during operation roughly indentical to the "color temperature"
of the light, about 3000K ?)  Below incandescent heats of the filament, I
wouldn't expect to see ANY filament vapors!

BillW

--

> -----Original Message-----
> From: Mike Singer [SMTP:msingerPOLUOSTROV.NET]
> Sent: Tuesday, December 03, 2002 9:11 PM
> To:   PICLISTmitvma.mit.edu
> Subject:      Re: [EE]: Temperature Sensors? How'bout lamps?
>
>    I suspect Nicrhome has too small temperature resistivity
> expansion index (I don't know the term in English, correct
> me if possible).
>
>    Mike.
>
I think the term you are looking for is "temperature coefficient of
resistance".

Mike

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

Michael Rigby-Jones wrote:
> > From: Mike Singer [SMTP:msingerPOLUOSTROV.NET]
> > Sent: Tuesday, December 03, 2002 9:11 PM
> > To:   PICLISTmitvma.mit.edu
> > Subject:      Re: [EE]: Temperature Sensors? How'bout lamps?
> >
> >    I suspect Nicrhome has too small temperature resistivity
> > expansion index (I don't know the term in English, correct
> > me if possible).
> >
> >    Mike.
> >
> I think the term you are looking for is "temperature coefficient of
> resistance".
>
> Mike

Thank you.

Mike.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

On Tue, 3 Dec 2002, Wagner Lipnharski wrote:

*>... I will do some experiments with a thin Nicrhome wire, the AWG32 is 10
*>ohms / ft at 25°C.

Nichrome wire has a (very) low tempco. Try bare copper wire. Iron wire
works too (guitar string). Crimp or spot weld terminals. Having an
ohmmeter with low ohms reading (to 0.1 ohm resolution at least) should
help. You can calibrate it directly in degrees C using the room
temperature and the boiling point of water (put destilled water in the
oven on top of the sensor, in a small, thin walled cup, and note
relatively constant until the water boils off completely).

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

On Tue, 3 Dec 2002, William Chops Westfield wrote:

*>> Because at 250C the lamp will relatively quickly plate the glass with
*>> filament vapors unless you run it at a significantly lower current.
*>
*>Huh?  Vaporize tungsten at 250C? (About 600K?  Aren't typical filament
*>temperatures during operation roughly indentical to the "color temperature"
*>of the light, about 3000K ?)  Below incandescent heats of the filament, I
*>wouldn't expect to see ANY filament vapors!

There is some because of the vacuum and external heat combined with the
running filament (Mike wanted to run the lamp at 20mA). I have 'plated'
the inside of the glass bulb of lamps like this before. The combined
effects of vacuum, electrically heated filament and external heat can
cause the filament to reach a temperature where it will plate itself out
on the glass. The lamps I had 'plated' like this still worked but most of
the envelope was black/metallic on the inside. The temperature was
slightly higher than 250C.

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

Peter L. Peres wrote:
{Quote hidden}

I wonder this experiment;

A on lamp, filament at 1000°C, external temperature 25°C.
Then you put the lamp into a 250°C oven, what will be the filament
temperature? 1250°C? less? much more?

The point in question, is that the filament heat is radiating all around,
and being sucked by the 25°C environment.  That 1000°C is caused because
the external environment helps it to reach such temperature.

Probably turning that same lamp at -50°C environment, the lamp brightness
will be affected since the filament temperature will be different.  The
temp sucking will be much faster.

And then again, if the lamp is into 250°C, what happens when you apply
power?  the external temp is working like a blanked all over the lamp, the
internal temperature will build up very fast...

Just curious.

VV46NER

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

But won't the brightness just go down? I though most light bulbs blow
out when you first turn them on (cold filament) because of the low
resistance of the cold filament. This causes a big inrush current and
pops the already weakened filament. While I don't know if the 250C would
be enough, at some point wouldn't the environmental temperature cause
the resistance of the filament to change?

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.

Wagner Lipnharski wrote:
> And then again, if the lamp is into 2500C, what happens when you apply
> power?  the external temp is working like a blanked all over the lamp, the
> internal temperature will build up very fast...

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

On Wed, 4 Dec 2002, Wagner Lipnharski wrote:

*>And then again, if the lamp is into 250°C, what happens when you apply
*>power?  the external temp is working like a blanked all over the lamp, the
*>internal temperature will build up very fast...

Both bodies radiate and absorb heat at the same time. The problem lies
with the glass being transparent for short wavelength ir but not for long
wavelength, plus it may reflect some of it back inside. I did not bother
with the physics but the lamp does shine brighter and it does plate out.

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...