The purpose, is I bought a silent enclosure for a hard drive. So the hard drive is sealed tight in this air tight box. The box has a little slit for wires to come out. And I will place the temperature probe`s wires through there and take a reading from the temp probe`s display.
The reading end of the temperature probe - the end in the case - should be like a pad, to get some surface area on the hard drive.
Does anybody have any suggestions of what type of temperature probe to get(what it is called) and/or where I can buy it?
I tried maplin(well known uk electronics place), they said they do not have anything appropriate.
Actually I bought the bits for mine from Maplin ! You want a bead thermistor. Mine was 5K. Then just use the Ohms range of a multimeter. Calibrate it with boiling water and Ice.
I called their tech support. He asked what temperature I am going up to. I said less than 100 degrees C. He asked how I am reading the temp, I said I do not have anything to connect it up to. He recommended SA1-T (temp probe) $60 (is a pack of 5 "unfortunately") HH501AT (as a reader). $75
yes, you are right But a)I would like to see the temperature all the time.. Whether in the BIOS, in DOS, e.t.c. b)I would like to be sure I have the right temp. Speedfan displays different temps, Temp1, Temp2. One may be cpu temp, another computer case temp. I think it estimates which is which, and other software, like "notebook hardware control" might estimate differently. I know the higher one is prob or definitely CPU. It may do the samea game with HDD temp too, there may be uncertainty, it may confuse one temp with another temp. And of course the temps read in can be unreliable. I would rather measure it myself. I have heard of one case where a CPU temp was given wrongly, and a BIOS update fixed it. c)I am relying on the measure more, it is not such a gimmick for me. I am putting a hard drive in a silent enclosure, (2 infact) so I want to be more careful about the temperature.
So for many reasons, I would like to measure it manually. I am very aware of the computer side of things, speedfan windows software, e.t.c.
I am not an electronics person, but I just had a chat with maplin technical support.. And he made sense. I just want to verify it with you guys and see if you have any comments.
He said make a table. Ohms and Water Temperature Heat the water and take the temperature and measure the resistance. (do you have to boil the water first and then cool it?) Do I drop the end of bead thermister in the water? (and presumably solder/tie the ends of it to a multimeter measuring resistance)
He said once the table has been written up. I remove the boiling water. And just measure the resistance of the hard drive with the thermister and multimeter. Then get the temp with my table.
I may not do this.. or i may try a few solutions for the curiousity of it. I am looking at alternative solutions at the moment. But since I have an interest in getting it done, I am quite interested in ready made solutions that cost less than =A315!
You should insulate the bead from contact with the water=20 during this test. But you can probably get a table of=20 resistance versus temperature for almost any thermistor you=20 can buy that is as accurate as a the one you would construct=20 by this water method. You shouldn't have to pay much more=20 than 1 Euro for a thermistor bead that has a couple degree=20 tolerance.
I assumed that you had some knowledge ! I bought a cheap £3.00 multimeter from the local supermarket and the bead thermistor from Maplin. I don't remember how much I paid for it but it was not dear, less than £2.00.
The thermistor measured almost exactly 5000 ohms at room temp. I soldered two wires to the legs of the bead. (The wire in my case was actually a length of co-ax out of a dead monitor video cable) Then I just dipped the whole end into a tin of yaught varnish and let it drip dry. The other end I fitted banana plugs to fit the multimeter.
To calibrate just dip the bead into a boiling kettle and note the reading. Mine dropped down to about 100 ohms. Then do the same with a cup of water and crushed ice. Mine goes up to about 12,000 ohms.
Definitely not a linear change. After that I used a glass thermometer and a glass of water to get readings every 5 degrees.
There are three good solutions, and two odd ones, to this problem.
(1) thermocouple. One can buy precisely formulated wire for various kinds of thermocouple (type K aka chromel-alumel is typical), and use voltmeter and any known temperature reference, or a 'cold junction compensator' circuit. The voltmeter/compensator pair is available as a standard kind of meter or module (Fluke model
51 is a familiar item in laboratories). Rugged, high temperature capable, excellent for furnace applications.
(2) thermistor. Calibration is by chart from the manufacturer, or formula, or (cheaper) do-it-yourself. Platinum resistors (aka RTDs) are state-of-the-art for low temperature work.
(3) diode equation. A $0.04 transistor with collector-base connected is a very precise thermometer, only needs a single measurement at a known temperature to fill out the full calibration chart from the known formula. ICs that report temperature usually have such a diode built-in, and ICs that convert the temperature to a simple voltage output (LM60 is typical) are available.
If you can solder wires, any of those three will make a probe that can work for you, (3) is probably the best option.
The two odd solutions, are
(4) aquarium thermometers (thin strips with colorchange scales). You can fish one into the narrow 'slot' and pull it out for reading.
(5) indicating paints/crayons. I recall 'tempilstick' was one brand. The pigment has a permanent color change if it ever hits the critical temperature, so you can have a record of the lifetime max temperature.
If the this is a new modern hard drive it will have built into its controls a function called S.M.A.R.T. (I can't for the life of me remember what that actually stands for)
If this is available on your hard drive then there are programs that can read and display, among other things, the actual temperature of the hard drive.
The program I use is called "Disk Checkup" by Passmark Software
formatting link
This is free for personal use, and only $15 US for a single user commercial license.
Some of the other items this program will read from the SMART data is the number of sectors that have been relocated due to potential failures and the error rates for the drive. Very handy for early predictions of drive failure.
There is another way. Buy an LM73 temperature sensor (Farnell order code 131-2598) or similar.
This is a pre-calibrated silicon temperature sensor with analogue to digital converter and i2c/smbus interface. Some motherboards have pin headers to allow access to the i2c/smbus which is used for things like cpu temperature monitoring and reading the proms on the dram modules. The Via M10000 single board computer for example has i2c headers.
If you want several sensors in the PC, then each sensor can be set to a different address on the bus. There is also a variant with a different address block to allow even more to be connected simultaneously.
If the motherboard does not provide easy access, then some careful soldering to the leads of the prom on a dram module will give access to the smbus.
If you are running linux, then look up lm-sensors to find out how to install i2c/smbus drivers and access the sensor. Windows might take a bit more work.
Use an ordinary diode or transistor junction, and measure on the Ohms scale of the DMM (which gives constant current). I recommend ignoring the "diode equation" and doing a simple
2-point calibration. The response is exceptionally linear and should work over a range similar to the LM34. The upper limit is the melting point of the junction. The lower limit is alleged to be a few degrees K... better than the LM34 (which is limited by its other active circuitry), not that it matters for this application. What *does* matter is that you can get some fairly tiny diodes and transistors. Perhaps not as small as thermistors or thermocouples, but small enough.
Best regards,
Bob Masta DAQARTA v3.50 Data AcQuisition And Real-Time Analysis
formatting link
Scope, Spectrum, Spectrogram, FREE Signal Generator Science with your sound card!
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.