Physics 2014 Entry Hopefuls!

Only say you want to do a PHD if its true. They expect a much more involved and informed student when looking at someone who wants to do a PHD. It can act against you if they don't think you're sincere about it.

Also I missed out on the Cambridge Headstart (i got Manchester which is still great) i applied in December (10A*'s 2A) and my friend who applied in sept (6A*'s 3A's B) got a place. So i emailed and they pretty much said they allocate places to the first people who qualify for the course standards. So apply as soon as applications open and your more likely to get a place.

I've applied for the European Space Camp too!
I would say definitely go for the summer school - they are generally very competitive to get into, so you are unlikely to get any more summer school offers. Also, you will learn so much! I went for a week long headstart summer school at UEA last summer (I'm currently an A2 student), and it was very useful when writing my personal statement and deciding exactly which course to apply for (Physical Natural Sciences, 'cause I don't really know which of Physics/ Chemistry/ Materials/ whatever I prefer!). I also applied for UNIQ, Cambridge Physical Sciences, and Sutton Trust summer schools, but didn't get in, and got rejected from my first two choices for the Headstart one too :'( My top tip for anyone is to apply or as many as possible, as they all select on different criteria (most were based upon background, e.g. being from an ethnic minority group, living in a rough area, having low income or getting free school meals, etc.) although they do also look at your GCSE results and predicted AS results and some (e.g. the Cambridge Physics one) require a sort of admissions test thing. And definitely accept your places on them!
To get work experience at a uni, first look through their physics department website - some will advertise work experience placements (although you're probably coming quite close to the deadlines now...). Also, email whoever's in charge of 'outreach' in their physics department with a short letter about who you are, what you currently study, and what your ambitions are (to study physics at a top uni like theirs (a bit of flattery never goes amiss...) and then do a PhD and work in research at somewhere like CERN).
Also, make sure you do loads of extra reading and go on any school trips offered to you to visit science labs, go to masterclasses, or listen to lectures on particle physics etc.

so long as they aren't A level text books. If they are its just you writing that you do your subject at school

Thanks for your advice I'll spend some time this afternoon browsing physics department websites to see if I can find something and I'll try to write some emails, even if it's quite late.

I'm also starting to plan which open days to go to this June and I'll probably visit Imperial. I actually can't decide between the Physics First Year Project open day or the Science and Engineering one. Which would be best?

What's a fiducial marker?

As for parallax error, as far as I know it's error introduced by not bringing your eye level to that of the measuring instrument (so the obvious fix would be to bring your eye level to that of the measuring instrument!).

Are you sure they don't ask for how to reduce errors? e.g. here's a question from June 2011, I have no idea how to come up with answers for these sorts of thing:

4. In Section A Part 1 you measured the diameter of a wire using a micrometer screw gauge.

(i). Suggest a possible source of random error in this measurement.
(ii.) Describe and explain a procedure that can be followed that may reduce the effect of the source of random error you identified in part (i).
(iii.) Suggest a procedure that can be followed that may reduce the effect of systematic error in the determination of the diameter.

I know that random errors are those which differ throughout the system (e.g. the wire may have different diameters at different points in its length) and systematic errors are basically zero errors or other errors that come up from not being able to get accurate readings. But how do we reduce either of them (ii and iii)?

Fiducial markers are things that you place at the fastest point of an oscillator (e.g., the equilibrium position of a pendulum) to improve the accuracy of you timings (if you are measuring, for example, time period).

You use a mirror to reduce parallax error. When the pointer (or whatever is causing the parallax) lines up with its image, then that means you are perfectly level and there is no parallax.

i) Wire is not a consistent diameter.
ii) repeat and average to improve reliability
iii) close the jaws and zero the gauge (if it is digital), otherwise close the jaws, make a note of the initial reading, and subtract this from all further readings.

Other ways to reduce error: measure more than one, then take an average (e.g., to measure the thickness of paper, you would measure ten or twenty sheets, then divide by ten or twenty). Use multiple oscillations to measure time period.

Ah thanks

So random error is possible reasons for inconsistency in your measurement; systematic errors refer to parallax error or zero error; reducing random errors is simply to take multiple measurements at various different points, remove anomalies and take the average (to reduce the effect of inconsistency); reducing systematic errors is simply to calibrate in order to remove zero error (or bring your line of sight to the level, perpendicular to the measuring instrument, e.g. using a mirror, to remove parallax error). Is there anything you would add to that?

I'm not too sure where fiducial markers fit in since I have in fact not heard of them before now but I'll be sure to keep them in mind just in case

Make sure you know how to quantitatively prove inverse proportion.

Hi again - what did you mean by this when you said it? How do you "prove" inverse proportion?

If x is inversely proportional to y, then xy=k where k is a constant. In an EMPA/ISA we had to show that this was true to within ±5% of the mean of k to prove inverse proportion.

So to do this we would just pick two other values of (x,y) pairs and then find the value of k they correspond to; then we just have to show that the magnitude of [100-((either of these k values/the mean k)*100)] is less than 5%. In fact this works for any graphical relationship not just inverse proportion (e.g. with a directly proportional line, you just work out the values for k using pairs of y=kx values). Am I right?

When working out percentage uncertainty if you have an angle to +- 3% what is the uncertainty of the sine of that angle?


Posted from TSR Mobile