Thermistor in self heating mode
Scott Seidman
mode? How do you figure out how much power you need to be putting through
it??
All the instrumentation texts describe this, but not at a real practical
level. I think I'm sneaking up on the current from the low side, but I
want to avoid smoking these things.
--
Scott
Reverse name to reply
John Popelish
What is your reason to be heating a thermistor? The search
for assistance depends a lot on your reason.
In the mean time, perhaps one of these general thermistor
application notes will help.
http://www.epcos.com/web/generator/Web/Sections/ProductCatalog/NonlinearResistors/NTCThermistors/PDF/PDF__Applicationnotes,property=Data__en.pdf;/PDF_Applicationnotes.pdf
http://www.epcos.com/inf/55/db/ptc_03/00310045.pdf
--
Regards,
John Popelish
Eeyore
Scott Seidman wrote:
The data sheet should help !
Graham
Phil Hobbs
Are you talking about a PTC or NTC thermistor? PTCs typically have a
snap characteristic, almost like a thermostat, where their resistance
increases about 3 orders of magnitude at some temperature between about
40C and 300C. That makes them useful as heaters as long as that's the
temperature you want, but they're not very settable, their accuracy is
quite poor, and the repeated thermal shock tends to be very hard on the
leads, if any. On the plus side, apart from the thermal shock issue,
they're almost completely insensitive to supply voltage variations.
That makes them good for stuff like keeping condensation off windows.
NTCs have a resistance that decreases about 3-4% per kelvin, which makes
them more settable but much more of a puzzle to drive.
Constant-current drive will always be stable, because an increase in
temperature will always result in a decrease in dissipation--P=I**2 R,
so dP/dT = I**2 R**2 dR/dT, which is always negative. On the other
hand, constant voltage drive will be unstable above some
current--P=V**2/R and dP/dT = -V**2/R**2 dR/dT, which is positive. Heat
loss Q-dot tends to be proportional to delta-T, as does the dissipation
in the thermistor, so when you crank up the voltage past the point where
dP/dT + d(Q-dot)/dT = 0, it'll run away.
Soooo, if I really had to build a temperature control loop with an NTC
thermistor in a self-heating mode, I'd start with a constant-current
drive and put a very slow P-I feedback loop around it. Figuring out
what the set point should be is left as an exercise for the reader. ;)
Good control results when P = Q-dot and dP/dT is large compared with the
thermal conductance G_th=d(Q-dot)/dT. (At short times, you also have to
contend with the fact that G_th isn't frequency-independent and that the
thermistor itself has some thermal mass.) You can't achieve both of
those things at the same time with a barefoot NTC unless your insulation
is really really amazing--the temperature coefficient is just too low.
Cheers,
Phil Hobbs
GregS
I like running in self heating a PTC, well I guess it does not matter. I like them in
water baths. When the water runs dry, it auto shuts down. You can get an excellant
self heating differential by dipping in water and comparing results with the air.
greg
Spehro Pefhany
<namdie...@mindspring.com> wrote:
What are you trying to do? Measuring fluid (liquid or gas) flow?
Constant temperature or variable?
The equations for Rx(T) and T(Rx) are quite straightforward (eg.
Steinhart-Hart.
Getting heat loss under your proposed conditions.. probably a lot less
so.
Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
sp...@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Scott Seidman
news:bphm34d065n2f5amc...@4ax.com:
> What are you trying to do? Measuring fluid (liquid or gas) flow?
> Constant temperature or variable?
At the moment, just trying to understand thermistor self-heating through
experience. No urgent practical use, but lets just call it measuring
liquid flow at constant temp. It's not going to be in a control loop. I
know its not the only, or best way to do it, but I'd like to understand the
method.
I have a feeling I'll just end up finding out why its not particularly easy
or reliable, but I'd like to know.
Spehro Pefhany
<namdie...@mindspring.com> wrote:
>Spehro Pefhany <spef...@interlogDOTyou.knowwhat> wrote in
>news:bphm34d065n2f5amc...@4ax.com:
>
>> What are you trying to do? Measuring fluid (liquid or gas) flow?
>> Constant temperature or variable?
>
>
>At the moment, just trying to understand thermistor self-heating through
>experience. No urgent practical use, but lets just call it measuring
>liquid flow at constant temp. It's not going to be in a control loop.
Then the temperature will be at the mercy of the liquid.
>I
>know its not the only, or best way to do it, but I'd like to understand the
>method.
What are you trying to predict or design?
It's perhaps not a bad way to make a level switch. You should be able
to predict the steady-state temperature with and without the liquid
being present for a given current or series resistor (given a constant
and (let's say) equal temperature for the liquid and air). Then play
with that single variable (the current or resistance) and see what
limits you run afoul of, and then you can play with variations in
ambient temperature.
>I have a feeling I'll just end up finding out why its not particularly easy
>or reliable, but I'd like to know.
Best regards,
John Fields
<namdie...@mindspring.com> wrote:
>Spehro Pefhany <spef...@interlogDOTyou.knowwhat> wrote in
>news:bphm34d065n2f5amc...@4ax.com:
>
>> What are you trying to do? Measuring fluid (liquid or gas) flow?
>> Constant temperature or variable?
>
>
>At the moment, just trying to understand thermistor self-heating through
>experience. No urgent practical use, but lets just call it measuring
>liquid flow at constant temp. It's not going to be in a control loop. I
>know its not the only, or best way to do it, but I'd like to understand the
>method.
Jim Thompson
Or my Bachelor's Thesis...
"THERMISTORS AS BLOOD FLOW-RATE TRANSDUCERS",
MIT, June 1962... with cooperation of Peter Bent Brigham Hospital,
Harvard University Medical School
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
America: Land of the Free, Because of the Brave
John Ferrell
<namdie...@mindspring.com> wrote:
I have this in my "things to do file when I am smarter".
I attended a seminar at the EAA Convention in Oshkosh in 1982(?) where
there was a presentation by a scientist from Langley AFB who built an
aircraft autopilot based on the principal. Two thermistors were in a
bridge with an air jet (tiny) that drove an RC servo on a trim tab on
one aileron. There was a professor from Ohio State University
(Walt(?)) who had it working on his Thorpe.
Apparently it worked very well but Gyro's won out in that application.
John Ferrell W8CCW
Beware of the dopeler effect (pronounced dope-ler).
That's where bad ideas seem good if they come at you
fast enough.
Jim Thompson
<johnf...@sprintmail.com> wrote:
[snip]
>Beware of the dopeler effect (pronounced dope-ler).
>That's where bad ideas seem good if they come at you
>fast enough.
Stolen for personal use ;-)
Scott Seidman
news:53gr34h2vtkrvsgvj...@4ax.com:
> Or my Bachelor's Thesis...
>
> "THERMISTORS AS BLOOD FLOW-RATE TRANSDUCERS",
>
> MIT, June 1962... with cooperation of Peter Bent Brigham Hospital,
> Harvard University Medical School
Thanks. I'm trying to put together a classroom demo or lab exercise for a
sensors and instrumentation course. I had a nice thermodilution lab, but a
self heating setup just never came to be, so the class never saw one beyond
reading about it.
Jim Thompson
<namdie...@mindspring.com> wrote:
Fluid level sensors usually use NTC thermistors... in air their
resistance plunges... in the fluid, they cool and their resistance
rises.
Scott Seidman
> Fluid level sensors usually use NTC thermistors... in air their
> resistance plunges... in the fluid, they cool and their resistance
> rises.
>
Thermodilution is the method currently used to measure cardiac output,
which isn't necessarily easy to do. A catheter is inserted into the right
side of the heart, and a balloon-like thing is used to float a thermistor
into the pulmonary artery. A cold water bolus is injected into the heart,
and the integral of the temp change is proportional to flow
Spehro Pefhany
<To-Email-Use-Th...@My-Web-Site.com> wrote:
>On Wed, 28 May 2008 15:17:12 -0500, John Fields
><jfi...@austininstruments.com> wrote:
>
>>On 28 May 2008 16:04:16 GMT, Scott Seidman
>><namdie...@mindspring.com> wrote:
>>
>>>Spehro Pefhany <spef...@interlogDOTyou.knowwhat> wrote in
>>>news:bphm34d065n2f5amc...@4ax.com:
>>>
>>>> What are you trying to do? Measuring fluid (liquid or gas) flow?
>>>> Constant temperature or variable?
>>>
>>>
>>>At the moment, just trying to understand thermistor self-heating through
>>>experience. No urgent practical use, but lets just call it measuring
>>>liquid flow at constant temp. It's not going to be in a control loop. I
>>>know its not the only, or best way to do it, but I'd like to understand the
>>>method.
>>
>>---
>>http://www.google.com/search?hl=en&rls=GFRC%2CGFRC%3A2006-50%2CGFRC%3Aen&q=thermistor+flowmeter&btnG=Search
>>JF
>
>Or my Bachelor's Thesis...
>
>"THERMISTORS AS BLOOD FLOW-RATE TRANSDUCERS",
>
>MIT, June 1962... with cooperation of Peter Bent Brigham Hospital,
>Harvard University Medical School
>
> ...Jim Thompson
It's not on your web page? ;-)
Jim Thompson
<namdie...@mindspring.com> wrote:
My thesis was based on two thermistors on the end of a catheter, one
with current flow to heat it, one simply to measure blood temperature.
This was back in the days of Germanium transistors, so I built my own
chopper-stabilized amplifier ;-)
Jim Thompson
I'd have to scan it... it dates to the days of trypewriters, except I
used a flexowriter + PDP-8 + punched paper tape ;-)
Scott Seidman
> I'd have to scan it... it dates to the days of trypewriters, except I
> used a flexowriter + PDP-8 + punched paper tape ;-)
>
>
How many Ph.D candidates back in the day do you think got married just for
the cheap typing??
Jim Thompson
<namdie...@mindspring.com> wrote:
Quite a few ;-)
I was already married, had a typewriter, but decided the PDP-8 was
ideal for generating a correctable copy, and I could spit out an
additional copy at will. (I think I still have the punched paper tape
around here somewhere ;-)
Michael A. Terrell
Jim Thompson wrote:
>
> This was back in the days of Germanium transistors, so I built my own
> chopper-stabilized amplifier ;-)
In 1962 my 'Chopper' was a Rat Terrier and he was self stabilizing.
;-)
--
http://improve-usenet.org/index.html
If you have broadband, your ISP may have a NNTP news server included in
your account: http://www.usenettools.net/ISP.htm
Sporadic E is the Earth's aluminum foil beanie for the 'global warming'
sheep.
legg
<johnf...@sprintmail.com> wrote:
>On 28 May 2008 13:15:13 GMT, Scott Seidman
><namdie...@mindspring.com> wrote:
>
>>Can anyone point me to a GOOD ref for running a thermistor in self heating
>>mode? How do you figure out how much power you need to be putting through
>>it??
>>
>>All the instrumentation texts describe this, but not at a real practical
>>level. I think I'm sneaking up on the current from the low side, but I
>>want to avoid smoking these things.
>I have this in my "things to do file when I am smarter".
>
>I attended a seminar at the EAA Convention in Oshkosh in 1982(?) where
>there was a presentation by a scientist from Langley AFB who built an
>aircraft autopilot based on the principal. Two thermistors were in a
>bridge with an air jet (tiny) that drove an RC servo on a trim tab on
>one aileron. There was a professor from Ohio State University
>(Walt(?)) who had it working on his Thorpe.
>
>Apparently it worked very well but Gyro's won out in that application.
>
Must have been pretty low velocity application.
RL
Michael A. Terrell
Jim Thompson wrote:
>
> (I think I still have the punched paper tape
> around here somewhere ;-)
Send it to Jorge. He can use it to light his wood stove his summer.