Last Updated: 9th Nov 2015

C3 Cheat Sheet
Chapter
1 Algebraic
Fractions

Usual types of questions

Almost always adding or subtracting
fractions.
Simplifying top heavy fraction using
algebraic division.
Simplifying fractions by first
factorising numerator and
denominator, where possible.

Tips
Factorise everything in each fraction first. e.g. If denominators
(2 + 1)( 3) and 2 9, common denominator will be
(2 + 1)( + 3)( 3)
Otherwise, just a case of practice!
If adding/subtracting a constant, turn into a fraction.
+1
+1 +2
1

+2
+2 +2
2
+1
Suppose we were asked to turn
into a mixed number. By
1
algebraic division we find the quotient is + 1 and the remainder
2. Remember we can express the result as () +

()
()

where ()

What can go ugly

Blindly multiplying the two denominators when
there might be a common factor, e.g. ( +
3)( 3) and ( + 3) should become
( + 3)( 3) rather than
( + 3)2 ( 3).
Classic sign errors when subtracting a fraction.
3
1
3( + 1) ( 2)

2 +1
( 2)( + 1)
Note use of brackets around ( 2) ensures -2
becomes +2.

is the quotient, () the remainder and () the divisor, just as

1
16 3 = 5 + where 5 was the quotient and 1 the remainder.
Thus

2 Functions

Specifying range or domain of

function.
Finding inverse of function.
Finding specific output of function,
e.g. (3) or (4)
Finding specific output using graph
only (without explicit definition of
function)
Finding composite function.
Sketching original function and
inverse function on same axis (i.e.
reflection in = )
Be able to find ( 2 ) or (2) for
example, when the function () is
known.
If () = 2 3 find all values of
for which () = 1 ()

www.drfrostmaths.com

2 +1
1

+1+

2
1

In general, to simplify fractions within fractions, multiply top and

bottom of the outer fraction by the denominator of the inner
fraction, e.g.:

+ 1 +

You should know and understand why 1 () =

I avoid getting domain and range mixed up by thinking ddrrr with
a silly voice. i.e. Domain first (possible inputs) followed by Range
(possible outputs)
Learn the domains and ranges of each of the common functions
1
( , sin(), , , ln , ( + )2 + , ). Be careful about > vs

Domains can be restricted for 2 main reasons:

o The denominator of a fraction cant be 0, so domain of
1
1
() =
is . Use symbol.
2+1
2
o You cant square root a negative number, so range of
() = is 0. Similarly you cant log 0 or negative
numbers so domain > 0 (notice its strict)
Range can be restricted for 3 main reasons:
o The domain was restricted. e.g. If () = 2 and domain is
set to 3 < 4, then range is 9 < () 16. Notice
strictness/non-strictness of bounds.
o Asymptotes. For reciprocals a division can never yield 0

Having a lack of care in domains/ranges with the

strictness/nonstrictness of the bound. For
() = , range is () > 0 not () 0.
Similarly range for quadratics are non-strict
because min/max point is included.
Putting the range of a function in terms of
instead of say the correct (). Similarly for the
range of an inverse, if the domain of the original
function was say > 2, then the range of the
inverse is 1 () > 2 (i.e. even though the
inequality is effectively the same, were
referring to the output of 1 now so need to
use 1 () rather than )
When finding (), accidentally doing first
and then .
If () = ln , then you should recognise that
( 2 ) = ln( 2 ), NOT (ln())2 , which would
suggest you dont quite fully understand how

www.drfrostmaths.com

(unless numerator is 0), thus range of () =

is
+3
() 0. Use sketch! For exponential functions, output is
(strictly) positive: () > 0
o Min/max value of a quadratic (or any polynomial whose
highest power is even). Note bound is non-strict as
min/max value included.
Range of () = ( + )2 + is ()
For composite functions, if given say (), write as (()) then
substitute () with its definition so you have
( ). Youre less likely to go wrong.
Remember that domain is specified in terms of and the range in
terms of () (or () or otherwise).
Remember that the domain and range is swapped for the inverse
function.
If the domain is unrestricted, you should know to write ,
where the means is a member of and is the set of real
number. Similarly an unrestricted range is () .
If asked Why does the function not have an inverse, answer with
It is a many-to-one function.
If () = 2 3 find all values of for which () = 1 (). If
() = 1 (), this only occurs when the input is equal to the
output, i.e. () = . Thus solve = 2 3.
Sometimes you never actually have the explicit function, but are
provided with a graph. Suppose (3,5) and (5,0) were points on the
graph. Then (3) = 5, 1 (5) = 3, and (3) = ((3)) =
(5) = 0. For harder questions you must have a combination of a
sketch (say for ) and an explicitly given one (say ). If say
() = 3 then note that = 1 1 (3): wed first find 1 (3)
by using the line graph backwards, and sub this value into your
function for 1 , or set () equal to this value and solve.

functions work.

3 Exponential
and Log Functions

Be able to draw the graphs of =

ln and = .
Sketch more complicated
exponential graphs,
e.g. = 100 + 50
Solve equations involving and
(see right).
Solve equations which are quadratic
in terms of (see right).
Find the long term value of an
exponential graph.

www.drfrostmaths.com

The log function exists to provide an inverse of the exponential

function, e.g. if we use () = 2 then we could obtain original
value using 1 () = log 2 .
C2 laws of logs you are expected to know and use in C3/C4:
log + log = log()

log log = log ( )

log = log
Combine s first in the same way as you would in C2.
e.g. ln 3 + ln 2 = becomes ln 3 3 = which in turn becomes
3 3 =
For equations of the following form, just ln both sides first and use
laws of logs to split up the LHS:
2 = 5
ln 2 = ln 5
ln 2 + ln = ln 5
ln 2 + = ln 5
(ln 2 + 1) = ln 5
ln 5
=
ln 2 + 1
Some questions are quadratics in disguise: + 3 = 4
1
Since = , this suggest we multiply everything by , which

gives us: ( )2 + 3 = 4 . Rearranging:

( )2 4 + 3 = 0
Then we could make the substitution = and factorise, or just
factorise immediately to get ( 1)( 3) = 0
Thus = 1 = 3, thus = 0 = ln 3.
If youd managed to factorise an expression to say ( + 1) = 0,
remember that cant be 0 (as the range of exponential functions
is > 0)
Suppose the population is given by = 5000 + 1000 3 where
is the number of years. What is the initial population? What is the
long-term population?
If = 0, = 5000 + 1000 0 = 5000 + 1000 = 6000
1
As , 3 = 3 will tends towards 0. Thus = 5000

To draw the graph of = 5000 +

1000 3 , either do as a graph
transform from = (i.e. shift up
5000, ensuring you get -intercept
correct), or consider initial value and
long term value as above. Ensure you
draw asymptote!

The worst algebraic error you can make is going

from say ln() + ln(2) = to: + 2 = .
Applying to both sides doesnt apply it
individually to each thing in a sum. Its
equivalent to the misconception that
2 + 2 + . To undo a , you have to
isolate a single on one side of the equation
with nothing being added or subtracted from it
(as per the example on the left)
Misremembering C2 laws of logs.
Thinking 0 = 0 particularly in applied questions
(e.g. population growth).

4 Numerical
methods

Rearranging an equation into the

form = (), i.e. where
appears on both sides of the
equation, but appears in isolation
on one side.
Using a recurrence to get successive
approximations.
Justifying why a value is correct to a
given number of decimal places.
Justifying why a root lies in a given
range. Similarly, justifying why a
turning point lies in a given range
(see right).

5 Transforming
graphs of
functions

Using an existing function = ()

to sketch = (||) and/or =
|()|
As at GCSE/C1, be able to sketch a
variety of functions by considering
the transformation involved, e.g.
1

= cos ( + )

Solve equations of the form

|()| = ()

Note that the specification explicitly

excludes multiple transformations
to the input of a function, i.e. you
will NOT see = ( + )

When asked to show that you can rearrange an equation to the

form = (), use the structure of the target equation to yield
clues about how to rearrange (since there are multiple ways of
isolating , some which will lead to dead ends)
Exploit the ANS key on your calculator to get successive
approximations. Put your 0 value into the value then immediately
press =. Then write an expression in terms of ANS and spam your =
key.
Remember the golden words change of sign when justifying why
a root is in a range.
If asked to justify why a max/min point is in the range, note that
the gradient goes from positive to negative (or vice versa) and thus
there is a change of sign.
Show that the solution to () = () lies in the range [3,4].
Rewrite as () () = 0, so that we can then evaluate (3)
(3) and (4) (4) and show theres a change in sign.

The way to remember (||) vs |()| is remember that changes

inside the function brackets affects the values and changes
outside affect the values. Thus in = (||), any negative is
made positive before being inputted into the function. Thus we
copy and reflect the graph from the right side of the -axis (and
discard anything that was on the left).
For curved graphs, at the point of reflection, the line should be
sharp (i.e. a sudden change in direction) not smooth.
For questions such as say Solve |3 2| = + 4 then for any
expression involving a |..|, have a separate equation where it is
positive or negative. i.e.

3 2 = + 4 = 3

3 + 2 = + 4 =

1
2

However you MUST check both solutions satisfy the original

equation:
|3(3) 2| = 7 3 + 4 = 7
1
7
1
7
|3 ( ) 2| =
+4=
2
2
2
2

Thus both solutions in this case are valid.

www.drfrostmaths.com

Using radians instead of degrees, or vice versa,

for recurrences involving trig functions. If your
values dont gradually converge (i.e. approach)
to a particular value when you spam the = key,
something probably went wrong.
Reaching a dead end when trying to rearrange
the equation to give a certain form.

Forgetting to add key coordinates to your

diagrams, e.g. intercepts with the axis, turning
points, etc.
Not checking your solutions are valid when
solving equations of the form |()| = () or
similar.
If considering |2 5| in solving an equation, it
would be a mistake to consider 2 5 and
2 + 5. You actually want 2 5 and 2 + 5.
i.e. Negate the whole expression, dont just
make each negative term positive!

6/7 Trigonometry

Questions mainly in two flavours:

Solvey questions.
Provey questions.
Either type may involve use of double
angle/angle sum formulae, or identities
1 + tan2 sec 2 or 1 + cot 2
2

Be able to express cos + sin in

the form ( ) or sin( )

To remember which way the tan and sec go in 1 + tan2 sec 2 ,

I remember the queen coming back from holiday, saying one is
tanned. Then slap a co on the front of the tan and sec to get the
second identity.
For proof questions, if you have a mixture of say 2 and , its
generally easiest to put everything in terms of first using double
angle formulae. E.g.:
cos 2 cos + sin 2 sin cos
Similarly for proof questions, when you have a mix of sin, cos, tan,
sec and the like, whenever youre stuck, just write EVERYTHING in
terms of sin and cos.
In proof questions, when you have fractions being
added/subtracted, combine into a single fraction. Things usually
end up cancelling.
Remember the 5 golden rules of trig angles, and you wont need
those silly CAST diagrams:
1. () = ( )
2. () = ( )
3. / repeat every 360.
4. repeats every 180.
5. () = ( ) (lack of knowledge of this one cost
students dearly in the June 2013 paper!)
As with C2 questions, when you have a combination of say tan2
and sec , then change the squared term using the identity, so that
you end up with a quadratic equation in terms of one trig function.
As per C2, if youre given a range for your solutions, then rewrite
the range as appropriate. E.g. If 0 < < and you had sin(2) =
1
, then 0 < 2 < 2. This ensures you dont lose solutions.
2

To solve cot(2) = 0, wed usually reciprocate both sides so that

1
cot(2) becomes tan(2), but we cant do !
0

We can see this would happen at asymptotes for tan, thus 2 =

www.drfrostmaths.com

Forgetting solutions in solvey questions.

Forgetting the when say dealing with
tan2 = 2 (and thus losing solutions).
Dividing both sides of a equation by say sin ,
rather than moving everything to one side and
factorising (as you again would lose solutions).
Getting to a dead end in proof questions (as per
advice, combine any fractions into a single one,
and write everything in terms of sin and cos if
completely stuck).
Ive seen some students try to skip steps with
cos + sin questions, and end up making
errors particularly in working out . Ensure you
show the expansion of ( ) or
sin( ), compare coefficients, and divide
and cos the correct way to get tan .

8 - Differentiation

Use of one or more of product rule,

quotient rule, chain rule.
Differentiating , ln , sin , cos ,
tan .

1
=

Be able to simplify a more

complicated expression via
factorisation (see tips).

Dont use the quotient rule if the numerator is a constant its

simpler to re-express as a product:
=

3
= 3(2 5)4
(2 5)4

= 3(4)(2 5)5 2

To remember the signage of differentiating and integrating and

, I visualise sin above cos, and that differentiating is going
down and integrating going up. If you have to go the wrong way
(e.g. integrating sin, but this is at the top) then the sign changes.
For nested functions, we obviously use chain rule. I use the bla
method: Differentiating the outer function with respect to bla,
2
then times by bla differentiated. e.g. with +3 : e to the bla
2
differentiated is e to the bla (i.e. +3 ) and bla differentiated is
2.

If is expressed in terms of and you need to find , then

sometimes it is not convenient to make the subject first: instead

find and then take the reciprocal to get .

Generally

leads to an expression in terms of . Use trig identities

or otherwise to get back in terms of . For example:

www.drfrostmaths.com

= sin

= cos

1
=
cos
But we need sin to get in terms of . Since sin2 + cos 2 = 1:

1
1
=
=
1 sin2 1 2
When differentiating a trig function to some power, write first
putting the power outside a bracket, so that its clearer you should
be using chain rule:
= sin3 2 = (sin 2)3

= 3(sin 2)2 2 cos 2

You will often have to simplify a differentiated expression.

Remember that when factorising things with indices, factor out the
smallest power, and factor out any fraction using the lowest

Substituting into the quotient rule wrong. E.g.

Doing instead of the correct

Doing something horrid like this: ( 3 ) =

3 3

Or even worse: ( 3 ) = 3 31

In general, forgetting to use the chain rule. If

theres some nested linear expression, make
sure you always multiply by the coefficient of :
d
(cos(3)) = 3 sin 3
d
Getting the signs wrong when differentiating (or
integrate) and , particularly when the
chain rule is involved.
e.g. Incorrectly differentiating sin(2 ) to
cos(2 ) rather than cos(2 ) due to
chain rule.
1
1
( 1)3 , NOT
Note that
3 is equal to
2(1)

2( 1)3 !
If using the product rule with more complex
expressions (e.g. with considerable use of chain
rule, particularly when there are lots of
negatives floating about), you are advised to

work out and separately first rather than

write all in one go.

common multiple:
=

www.drfrostmaths.com

1
1)2

1
1
= 1 ( 1)2
6

6(
3
1
1 1
1
1
= ( ) ( 1)2 + (1) 2 ( 1)2
6
2
6
3
1
1
1
= 1 ( 1)2 2 ( 1)2
12
6
3
3
1 2
1
=
( 1)2 [ 2( 1)] = 2 ( 1)2 (2 3)
12
12

If the tangent is parallel to the -axis, then its gradient is infinite:

this would only happen because of a division by 0 in the original
1
equation. e.g. If =
, then the tangent is effectively the
+3
asymptote, which has equation = 3.