How the current in a diode varies with temperature (-20 to +150)

[dualcore]

Homework Statement

"You are required to design a laboratory experiment to investigate how the current in a diode varies with temperature in the range of -20 to +150 Degree's C"

Homework Equations

N/A

The Attempt at a Solution

Perhaps using a freezer for the negative temp's and a oven/kiln for high end temp's.
Use a themocouple to measure the temp' (Using ICE - 0^C).


I wondered:

How to attach the termocouple to the diode, (if thermopaste - which one?)
Which diode? (Can withstand -20 up to +150)
What Volatage Power Supply? 3V?

How to ensure it was all accurate and reliable?


I have a rough idea of the circuit diagram.

Any help would be appreciated.

Thanks!

 

[marcusl]

Any thermal heatsink grease is fine. Suggest you wait at each temperature for the oven and its contents to reach equilibrium (when the thermocouple reads a stable temperature, it will be a few to 10 minutes after your oven says it reaches the set temperature).

Use a standard diode 1N914 or 1N4148, and run at low current (1 to 10 mA is ideal) so the diode doesn't produce internal heat that is missed by your thermocouple. That means using a current limiting resistor of around 4.3k for a 5V supply. Put the resistor outside the oven so its resistance doesn't change with temperature.

As for accuracy, take your time and be methodical. Use two meters, one to determine current from the voltage drop across your known resistance (this is called a four-wire measurement and is an accurate way to measure current) and another to measure the voltage across the diode.

Have fun!

 

[dualcore]

Thanks

Thanks for your reply!

However, perhaps i have misunderstood. I am simply measuring the current, therefore shouldn't an Ampmeter be fine? Why is there any need to measure resistance?

Also, how could i place the circuit so that the diode is in the oven, surely the wires would melt at 150 degrees?

Im confused now of how to put the circuit in/around the oven and which wires to use.

 

[marcusl]

Yes you can use an ammeter instead.

Insulation indeed requires attention, but wires are available with insulation rated to 150C. Look in Belden's catalog at hookup wire, for instance. The temp rating is listed for each type. Some magnet wires also have 150C insulation.

 

[Bubble Sheep]

I had thought about using thermocouples, but have no real idea about how to use one. After research on the internet, I thought a Type T one would be the ideal one, but how do I connect it in a circuit? Do I need a volt meter to measure the voltage through it?

Any help appreciated

 

[marcusl]

You are better off using a commercial temperature meter and sensor. Take a look at Fluke thermometers, for instance. Plug your Type T sensor into their model 50 and you are set. Omega is a company specializing in temperature measurement (warning, their catalog can be overwhelming).

 

[Ramead]

Hi, if we didnt want to use a thermocouple, what do u think would be the best way to measure the temperature?

Cheers

 

[marcusl]

Depends on the accuracy you want to achieve. The original poster wanted to ensure accuracy. On the simple but inaccurate end, use an oven thermometer from your supermarket. You'll probably get +/-10C accuracy or maybe as good as +/-5C, depending on the unit.

 

[cultist1947]

hi guys - in your opinion what would be the best method to reach the temperatures required. should i stick with the fridge/oven method? thanks

 

[Sully]

I also did this plan, I would advise that you do some research on water baths capable of working at those temperatures, (high temperature baths and cryo baths), then you will also need to find suitable fluids to fill them with. Obviously there are various components which need to be taken into consideration, but I think if you find the right baths, it is an easy solution to how the temperature would be controlled and varied. Just putting the idea out there.

 

[jackmaxi]

Just caught this stream. Why use a diode at all, rather than simply measuring current across circuit? Are there advantages (practical/theoretical) in measuring current directly or indirectly (i.e assessing resistance and using Ohm's Law to calculate I)? Or, if you're playing around with temperature, should you avoid using Ohm's Law?