Searching \ for '[EE]: Embedded Thermal sensing' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=embedded+thermal
Search entire site for: 'Embedded Thermal sensing'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Embedded Thermal sensing'
2002\05\10@040804 by Alan B. Pearce

face picon face
>What I'm looking at is getting the temps in cores and the cooling rates
when
>the structures are exposed to air movement. I think I'll actually need two
>types of measurements, one would be temps from the entire series of probes
>on a fairly rapid basis 10Hz for about 10-15 minutes worth of data and then
>another series where the time might be a few minutes between readings. This
>is the easy part.

>The hard part is making it accurate and getting it right before its
embedded
>into the structure.

I would seriously look at using the AD590 temp sensor from Analog Devices.
Use the ceramic package version, which is about 1mm thick, 3mm wide, and
about 12mm long (dimensions from memory, not data sheet). This should be
easily stuck to, or imbedded in whatever you are wishing to monitor. It has
very low mass, so tracks changes rapidly.

It is a two terminal device, which requires a minimum of about 5 volts
across it (max about 15V) to operate properly, and passes a current of
10uA/K so at 0C passes 273uA. Conversion to voltage to read with a micro is
easy, we use a 10k to ground, and because it is a current mode device can
easily be multiplexed with a single 10k load resistor at the A/D, resulting
in about 2.5 to 3 volts at room temp.

Being a current mode device long wires can be run to it without effecting
accuracy (providing sensor voltage requirement met, but should not be a
problem at this current). This should mean you can imbed only the sensors
and leave the monitoring electronics somewhere where it is accessible for
service and calibration. The only caveat I would recommend is a capacitor of
about 1nF across the leads at the sensor to stop RF and other EMC getting
into it.

Comes in various calibration tolerances, best being +/-1K, but can be
individually calibrated to better than that if need be. Depending on
accuracy you require, either single point or multiple point (for linearity)
may be required. AD have info in the datasheet and app notes about this.

This is how we commonly do temperature sensing for housekeeping monitoring
inside space instruments. For this use we just use the best tolerance and
don't bother individually calibrating them.

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


2002\05\13@034623 by Dave King

flavicon
face
I would seriously look at using the AD590 temp sensor from Analog Devices.
Use the ceramic package version, which is about 1mm thick, 3mm wide, and
about 12mm long (dimensions from memory, not data sheet). This should be
easily stuck to, or imbedded in whatever you are wishing to monitor. It has
very low mass, so tracks changes rapidly.

Thanks Alan

These look pretty good. I need the accurate range from 40° - 100°C
and need to have the probes be pretty neutral to the foam structure. I have
this nagging little voice warning me off about using a wire type probe so
this seems to fit the bill. Thanks for the heads up.

Dave

--
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


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