americanforest said:
I recently changed my major to Physics (from Computer Engineering) and love the subject. However, there is a problem. Even after doing tons of problems at home to study, there are questions on test that I just can't wrap my head around. The way I try to solve problems is to set them up without any numbers, only with variables, v for velocity etc and find a general equations. However, I often find myself in algebraic conundrums which take forever to solve (ie Quadratic Forumla). I'm actually quite good at math. I understand all the concepts like the back of my hand. I have friends who can do these problems without studying nearly as much as me. Am I just not cut out for Physics? What can I do to help my problem solving skills? It's very discouraging, having changed my major to something which inspires me and which I love, only to find that I may not be capable of doing it. Please help.
When I took my first university physics course we were allowed to use cheat sheets. I had to write down EVERYTHING. Let me give you an example.
In an ideal capacitor you can write the capacitance as:
C = \epsilon \frac{S}{d}
Now if I had a question that asked for the distance between two plates of a capacitor with a capacitance of C and a surface area of S, what is the distance". I wouldn't be able to 'see' C = \epsilon \frac{S}{d} and just write d = \epsilon \frac{S}{C}. Instead I would have to look at my 'cheat-sheet' and find the expression d = \epsilon \frac{S}{C}.
Or, let's say we had something like:
\frac{1}{\frac{a}{b}} it would take me a awhile to simplify it as \frac{b}{a}.
All of this just comes down to algebra. Once I had practice all of those hand waiving magic tricks made sense to me. Things used to feel like they just magically appeared out of thin air.
Just practice! If you like physics a lot, then just go through as many physics problems as you can, but keep an algebra book next to you.
Also, realize that there are times to solve symbolically and times not too. For example, the current through a NMOS can be expressed as:
i_D = \frac{1}{2} \mu_n C_{ox} \left( \frac{W}{L} \right)_n (v_{GS}-V_t)^2(1 +\lambda v_{DS})
Now let's say you are given values for \mu_n, \,\,\, C_{ox}, \,\,\, W_n, \,\,\, L_n, \,\,\, V_t, \,\,\, \lamda, \,\,\, v_{DS}
If you are asked to find v_{GS} you could yes solve this symbolically, but you would save some time if you did some multiplication and got a decimal equivalent. So pick and choose your times...