h a l f b a k e r y
Breakfast of runners-up.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
This is an electronic integrated circuit which measures its own temperature and
converts it to the ratio of voltage to current across it. The power to operate it is
drawn from the meter (multimeter in ohms range) used to measure it.
An ohm-meter measures resistance by applying a known voltage
half a volt) to the resistor and displaying the inverse of the resulting current as the
resistance. Or, forcing a constant current through it and displaying the resulting
voltage scaled as the resistance.
A multimeter is available in almost every technician's toolkit, and so this sort of a
temperature sensor would make easy and error free measurement of
Thermistors already exist, but their resistance is some convoluted function of
temperature, and require convoluted mathematical gymnastics to convert to a
LM34 Fahrenheit Temp Sensor
Suitable? [csea, Nov 18 2013]
Suitable? [csea, Nov 18 2013]
||Does your idea boil down to "a plug-in temperature
probe for a multimeter"? If so, it's baked - my
multimeter has one, although it uses a thermocouple
rather than a thermistor.
||//a plug-in temperature probe for a multimeter//
||Yes, but two-terminal and without batteries ie, entirely passive, like a
||My thermocouple-based one is exactly that - two-
terminal and passive. A thermistor would work too,
but it wouldn't be as precise.
||The multimeter can also give an internal temperature
reading (without the external probe connected),
which is presumably derived from an internal
thermistor. Good thermocouple thermometers need
such an internal reference anyway.
||The thermocouple based meter has a goodish bit of circuitry inside it -
an amplifier to boost the minuscule thermocouple voltage, cold
junction compensation, etc., and is therefore more complex (and
expensive) than an ordinary multimeter.
||My proposal is to bring all that complexity inside a purpose-built
sensor which can be used with any ordinary multimeter set to the
||Be careful with thermocouples where the cold junction
is potentially affected by the hot junction temperature.
In that respect, it was correct to choose something
else. But you should choose an RTD for your idea, I
think, and a thermocouple has a much wider range.
||The name of the circuit is a loop-powered transmitter,
where the loop is not 24V as normal, but only 0.5V.
This is a tough remit as, for example, a pn junction
requires more than that.
||That is why it is a halfbakery idea rather than a commercial product.
||An LM34 Fahrenheit sensor [link] is a three pin device
(V+, Tmp, GND) where Tmp is 10mV/deg F. A DVM
with a 4 terminal Ohms function [link2] should be able
to power the device, and display. Source current to
V+ ref GND, sense voltage between Tmp and GND.
||Thanks for the explanation; your probe has more smarts than I was imagining. In effect, the multimeter is just an external power supply and display.
||I wonder if it would be possible to create a MEMS thermometer that is basically an array of small bi-metalic switches that switch at different temperatures, connecting resistors in parallel as the temperature drops. This would create a completely passive two-terminal device who's resistance is the temperature in Fahrenheit. My guess is you'd have a few ohms of switch resistance, so you wouldn't be able to get accurate meaurements very close to 0.