School Circuit Project

[Alienminds]

I'm doing a statistics project in school and my Idea is to build several different circuits all designed to accomplish the same task in different ways. Then I want to compare the circuits to see differences in resistance, efficiency, etc. Any ideas of types of circuits I could build? I was thinking about power converters, but I'm not sure. I don't know too much about them.

 

[Nigel Goodwin]

I'm doing a statistics project in school and my Idea is to build several different circuits all designed to accomplish the same task in different ways. Then I want to compare the circuits to see differences in resistance, efficiency, etc. Any ideas of types of circuits I could build? I was thinking about power converters, but I'm not sure. I don't know too much about them.

How about a simple DC power supply, fed from a 12V battery (car battery?) and giving out 5V DC. You could build a simple 7805 design, a discrete analogue design, and also switch-mode designs, both analogue and discrete. You could also build bipolar and MOSFET designs, to show the differences between them.

 

[stevez]

Tell us about the "statistics" if you would please. That might help us guide you. You might take a simpler approach and sample resistors, meter accuracy, capacitor values, etc.

 

[Alienminds]

Well, its for my IB Math Studies project. Basically, I have to get numerous numerical data that I can use to do various statistical stuff with (Most of which we haven't gotten to yet in class). The Stuff we've done so far are Bar Graphs, Time Plots, Time Series, Histograms (Several types of histograms), Stem and Leaf plots, and more. I can't really do a survey, because then the only info I can use are counts of yes's and no's to different questions.

 

[monttyle]

Well, its for my IB Math Studies project. Basically, I have to get numerous numerical data that I can use to do various statistical stuff with (Most of which we haven't gotten to yet in class). The Stuff we've done so far are Bar Graphs, Time Plots, Time Series, Histograms (Several types of histograms), Stem and Leaf plots, and more. I can't really do a survey, because then the only info I can use are counts of yes's and no's to different questions.

I agree with stevez, measuring component values would be the simplest way to go. Get a hundred 5% 1K resistors - really cheap, you can get 200 for 5 bucks from digikey(part no 1.0KQBK-ND) - and measure their resistances with a multimeter.

100 resistors would be a large enough population to be statistically useful, they'll all be slightly different, and their distribution should fit the bell curve. You'll be able to see what the manufacturer means by 5%.

 

[Alienminds]

That sounds like it would be a good idea, but I'd still like to try doing the different types of circuits. I don't know. On one hand, I'd like to get a good grade on the project and I think the thing with the resistors would allow me to show higher level statistic methods. On the other hand, this is a perfect opportunity for me to learn more about different electronic circuits (And to have an excuse to tinker around with them). Maybe I could try to incorporate both?