< Back to A2 Question Papers

2015 Communications - worse even than 2003 Signals?

Posts: 19

Dear All

Control Systems was the fairest, most straightforward paper I've seen in years. My hopes were up. So I trolled down to the sports hall at the end of the Communications paper, curious to see my babies' smiling faces after another great exam.

All I saw were tears of desolation and despair. Comforting words fell on stony ground.

If you want to know how this paper would feel to your students then take the time to go through every question.

My personal suspicion is that the UMS correction will be massive.



Posts: 19

Dear Andy

It seems to me that this is set by a new setter. The questions are strange. They stick point by point to the spec (start at the beginning, work through), some of the questions are vague and very high level and there is so much writing needed.

As for individual questions here are my additional comments:

Q1 (a) (ii) Why two marks?

(b) why three marks?

Q2 (c) what's wrong with a normal diode detector. That's what we teach them. This circuit requires so much thought, especially as a treble cut filter is tested again in Q4.

(ii) I could write an essay with formulas for this. Where are the students supposed to stop?

Q3 (d) How do you answer this in words? What kind of monostable is it?

Q4 (a) your statement 'must be an A*' is light. Not enough room and no guidance AND the frequency is quite a way off 10kHz. I wish even my good students luck on that one.

(c) (i) Why no scale? was that really necessary?

(ii) Does it matter what voltage the demodulator outputs? What are the marks for?

Q5 (c) Again, this is so hard to answer in words. What are they looking for? Formulas?

Q6 (b) (ii) I have no idea what the answer is to this. Is this a regulation?

(b) (iii) Why three marks? Do they want a repeat of the schmitt trigger answer in 3 (c)?

Q7 (b) (ii) Surely the function of the filter is that it 'only transmits frequencies between 450kHz and 490kHz' like it says in the question.

(c) (i) I hope they would accept both aerial length or the impedance of the mixer.

8 (b) How can an ADC have this resolution AND this range? One of my students answered 7.3 bits and I sympathise.

(c) (i) I have never seen such a vague question set in an exam. Would it have been so difficult to name the 'parallel load' input or to make the whole thing 4-bit with an AND gate. It just makes it really hard to answer.

(c) (ii) The question depends on how you make your PISO. If you use the one in the book you have to add 1 only to the word length, but again, why words?

This is, of course, a personal view, but I hated this paper. It made me tired. I found myself skipping questions and had to come back to it after several hours. Clearly I don't like writing. Nor do most of my students.

best wishes



andy masters
andy ma ...
Posts: 18


I spoke with 6 students yesterday (mixture of grades) and they seemed fairly happy. I agree that the UMS correction should be large (as it usually is), but not to the 2003 Sig Proc level.

My initial thoughts on the questions (possible issues and harder questions in bold): 

Q1:     (a) (i) should be able to pick up some marks - but all 4?;

(ii) ok;

(iii) and (iv) rely too heavily on (ii);

(b) ok.

Q2:     (a) and (b) ok.

(c) (i) a range must be allowed;

(ii) tough.

Q3: Mostly ok.

Q4:     (a) must be A*;

(b) and (c) should be ok, but follow an A* question, so confidence could be affected by this;

Q5:     Words are always tricky marks for our students. Probably higher-end.

Q6:     Fairly set-piece.

Q7:     Fairly set-piece.

Q8:     (a) 'Function' is a word that'll discriminate the marks;

(b) higher-end, but manageable;

(c) (i) based on the textbook, but should be possible to work out;

(c) (ii) 3 marks as it's looking for 'start' and 'stop', but is fair.

I'm prepared to be wrong.

Regards, Andy