High School Experiment : Effect of Temperature on Strength of Magnet I am running an experiment for my school which investigates the effect of temperature on the magnetic field of a magnet, and I encountered a little problem. For the measurement of data, I have gotten a thermistor and a linear radiometric hall effect sensor and I will be using an Arduino for the data logger. The problem I am facing is that I don’t know how to measure the magnetic field strength and the temperature efficiently without losing a lot of heat. 
So far I thought of two methods, however, I am not sure if they will work. One way that I thought of was placing the magnetic in a test tube with water and then heating the water until its boiling point, then remove the head and connect the thermistor to the water and I measure the magnet through the test tube. The second method that I was thinking of was to simply place the magnet in boiling water and then transfer it with ceramic tip tweezers to a little set up with the thermistor and hall effect sensor already placed. 
In the first method I feel that the water would not have exactly the same temperature as the magnet at all times, and the glass/water medium will affect the magnetic field strength that I'll be measuring. In the second method I think a lot of heat will be lost through radiation during the transfer; thus the temperature will drop very quickly and the date wouldn’t be accurate. I’m not exactly sure how to heat it up and measure the temperature & magnetic field strength efficiently. I would really appreciate some suggestions/help.
Thank you very much!
P.S I am a High School student thus I do not have access to advance apparatus
 A: It doesn't really matter whether or not the glass/water medium affects the magnetic field strength as long as your measurements are consistent; always make measurements with it present, but for the record, some water and glass won't impede a magnetic field. Additionally, if the magnet is a thermally conductive material and is smallish, it can safely be assumed it's the same temp as the water.
If all you're looking to do is investigate effects of temperature on a magnet, then you don't need much precision. Record the water temperature the magnet is in and the magnetic field strength outside the water container from the magnet (don't forget to take a zero reading without the magnet present). So long as you don't go above the Curie temperature, that should be enough to show if the magnetic field is influenced by temperature. I also advise suspending the magnet in the water via a string (or something thermally insulating) to prevent it from contacting the sides of the tube.
A: I think the simplest solution would be to heat the magnet in a bunsen flame then wrap it in a small amount of fire-proof insulating material and clamp it a fixed distance from the Hall Effect Sensor. The magnet will slowly lose heat while maintaining a fairly uniform temperature. Fairly slow cooling allows the magnetic domains sufficient time to re-organise. Temperature and magnetic field readings can be logged at regular intervals of time (or preferably temperature) as the magnet cools. The temperature sensor should be attached before heating, so that you can avoid exceeding the Curie Point at which the magnetism is lost. 
This is similar to your 2nd method. The advantages compared with a water bath method are :


*

*You can take readings over a much larger temperature range, and get much closer to the Curie Temperature (iron $770^\circ C$, neodymium up to $400^\circ C$, nickel about $360^\circ C$ - but gadolinium only $19^\circ C$). The result is a larger range of variation of magnetic field strength, which may be small over the range $20-100^\circ C$. You should be able to verify that below the Curie Temperature the magnetization is proportional to $(T_C-T)^\beta$ where $\beta$ is a constant.

*It is (probably) easier to maintain a fixed distance and orientation between the magnet and the sensor, because the variation due to temperature may be swamped by variations due to movement.
The main difficulties of this method are :  


*

*The temperature range is probably too large for a thermistor. A thermocouple or resistance thermometer (RTD) is required.  

*How to attach the thermometer? The magnet will be too hot for standard solder, but you need a good thermal contact.   

*Too much insulation will make the experiment drag on for a long time (hours?), during which there may be significant drifts in the instruments, or the possibility of some accidental interference.

*Avoiding going above the Curie Temperature when heating the magnet.  

*Avoiding heating of the Hall Effect sensor (as mentioned by Sam Weir). Placing the sensor in the water bath ensures that you know its temperature, so that you can apply a temperature correction.  

*Safety when handling hot metal.

A: You didn't say what temperature range you want to make the measurement over. Going up to the Curie temperature of something like iron (Tc=1043 K) would be quite a challenge and potentially dangerous for a high school lab level experiment.
If your intent is simply to go up to a bit less than the boiling temperature of water, here is how I would do it:
Get a sous vide water oven (I have this sous vide model for cooking short ribs but there are cheaper ones on places like Amazon), fill it with water, and put the magnet and your temperature sensor in the water (you may have to protect the temperature sensor from the water by putting it in a plastic bag or coating it with silicone rubber cement, etc.). If you're measuring the magnetic field strength from a ferromagnet, you don't need to worry about water, plastic, glass, etc., affecting the magnetic field strength from the magnet. The susceptibilities of all those materials are so small that they have virtually no effect on the magnetic field as far as your measurements are concerned.
Now the question is how to place the magnetic sensor relative to the magnet? The simplest approach would seem to be to just place the magnetic sensor in the water oven bath, too, so that it is at a fixed distance from the magnet (and, again, making sure that the magnetic sensor is also protected from the water by a plastic bag, etc.). The problem with that seemingly direct approach is that your semiconductor-based magnetic sensor almost certainly has a large temperature dependence, so when you're recording data as a function of temperature you won't know how much the measured signal is changing due to the change in the magnetic field and how much the measured signal is changing due to the temperature change of the magnetic sensor itself. If your temperature sensor comes with calibration specs giving its temperature dependence, you could in principle separate out the two effects but it would be messy.
So I would suggest instead that you try to arrange the experiment so that the magnetic sensor is kept at a fixed temperature. You could, for example, place a thin sheet of Saran-wrap or other plastic film over the water bath (in place of the usual metal lid), and then mount the magnetic sensor above the water bath and plastic film but close enough to the magnet so that the sensor can measure the magnetic field but at the same time the temperature of the sensor is maintained fixed at room temperature. That way you can be sure that the signal changes that you see when the water bath temperature changes are really due to changes in magnetic field strength and not to any temperature changes in the magnetic sensor itself.
