The Student Room Group

Help urgent integration

Help… for this do I always have to put it as
ln|y| or can I just put lny
Because especially if the Q says something like

Integrate 1/x can I just write lnx and get the mark for aqa a level mathsIMG_2708.jpeg
And can I put c instead on lnk?
(edited 10 months ago)
It's always safer to put the modulus sign, as in AQA, sometimes they let it slide, and other times they mark you down for it. So just to be on the cautious side, put the modulus in it. Since you can't input a negative number into lnx.

Also, regarding the +c. You can most definitely put +c instead of ln(k). In fact, most of the time with differential equations, it's nicer to put it in the form y = (a function). So doing +c is easier than ln(k).

Hope this helped!
Reply 2
Original post by Bruce Wayne59875
It's always safer to put the modulus sign, as in AQA, sometimes they let it slide, and other times they mark you down for it. So just to be on the cautious side, put the modulus in it. Since you can't input a negative number into lnx.

Also, regarding the +c. You can most definitely put +c instead of ln(k). In fact, most of the time with differential equations, it's nicer to put it in the form y = (a function). So doing +c is easier than ln(k).

Hope this helped!


Omg thank you so much
Reply 3
Original post by Bruce Wayne59875
It's always safer to put the modulus sign, as in AQA, sometimes they let it slide, and other times they mark you down for it. So just to be on the cautious side, put the modulus in it. Since you can't input a negative number into lnx.

Also, regarding the +c. You can most definitely put +c instead of ln(k). In fact, most of the time with differential equations, it's nicer to put it in the form y = (a function). So doing +c is easier than ln(k).

Hope this helped!


So if I put it as

lny = (function) + c
it would be fine for aqa
Original post by Alevelhelp.1
So if I put it as

lny = (function) + c
it would be fine for aqa

Yes. Although most of the time, they'd probably ask you to either work out this constant, or re-write the expression as y = function.

See attached to understand how to get y = function of x.
(You don't have to convert it into this form if they don't ask for it, but this form is a lot more useful when modelling differential equations) :smile:.20230604_122950.jpg
(edited 10 months ago)
Reply 5
Original post by Bruce Wayne59875
Yes. Although most of the time, they'd probably ask you to either work out this constant, or re-write the expression as y = function.

See attached to understand how to get y = function of x.
(You don't have to convert it into this form if they don't ask for it, but this form is a lot more useful when modelling differential equations) :smile:.20230604_122950.jpg

Agree with your comment about the
y = A e^(kx)
being more useful etc and you mention about negative values (y) in the previous post, but for a differential equation
dy/dx = k
y(0) = A (could be positive or negative), then doing a definite integral and seperation of variables you get to the usual
ln(|y|)-ln(|A\) = kx
or
ln(|y|/|A|) = ln(y/A) = kx
or
y = A e^kx
the sign of y (and A) is taken into consideration when you do the division (as both the numerator ..and denominator are the same sign) so taking exponetials (inverse logs) is ok for both positive and negative y.

If you did indefinite with + c, then youd get to the usual
ln(|y|) = kx + ln(|A|)
If y and A are positive then doing what you suggest is ok. If y and A are negative, youd have to write
ln(-y) = kx + ln(-A)
as this is what the equation says and taking exponentials
-y = -A e^kx
and obviously the - sign can be removed on both sides. Strictly speaking you cant just assume the A=e^c step works out when things are negative.

Realise the above may not even be worth a mark, but if you see a negative initial condition, be careful with the reasoning and treating it as a definite integral from 0..x and A..y. is probably safer.
(edited 10 months ago)
Original post by mqb2766
Agree with your comment about the
y = A e^(kx)
being more useful etc and you mention about negative values (y) in the previous post, but for a differential equation
dy/dx = k
y(0) = A (could be positive or negative), then doing a definite integral and seperation of variables you get to the usual
ln(|y|)-ln(|A\) = kx
or
ln(|y|/|A|) = ln(y/A) = kx
or
y = A e^kx
the sign of y (and A) is taken into consideration when you do the division (as both the numerator ..and denominator are the same sign) so taking exponetials (inverse logs) is ok for both positive and negative y.

If you did indefinite with + c, then youd get to the usual
ln(|y|) = kx + ln(|A|)
If y and A are positive then doing what you suggest is ok. If y and A are negative, youd have to write
ln(-y) = kx + ln(-A)
as this is what the equation says and taking exponentials
-y = -A e^kx
and obviously the - sign can be removed on both sides. Strictly speaking you cant just assume the A=e^c step works out when things are negative.

Realise the above may not even be worth a mark, but if you see a negative initial condition, be careful with the reasoning and treating it as a definite integral from 0..x and A..y. is probably safer.

Correct me if I'm misunderstanding your point, but isn't A just any constant? So it can refer to anything, regardless of its sign. For instance, if you had -e^c, can't you just call that entire term "A" instead of including the negative within the final expression?
Reply 7
Original post by Bruce Wayne59875
Correct me if I'm misunderstanding your point, but isn't A just any constant? So it can refer to anything, regardless of its sign. For instance, if you had -e^c, can't you just call that entire term "A" instead of including the negative within the final expression?


I’m so lost 😭 which one is correct my a level exam is on Tuesday 😢😭😭😭
Original post by Alevelhelp.1
I’m so lost 😭 which one is correct my a level exam is on Tuesday 😢😭😭😭


I wouldn't overcomplicate it. Just stick with putting a +c on the end. The reason why ln(k) is there is so when you re-write the expression to how I've shown it on my whiteboard, you can have the constant as a coefficient of the e rather than just adding it on at the end. But both ways are perfectly valid, unless they ask for your final answer to be in a specific form. Which in that case just fiddle around with the logs until it works.
Reply 9
Original post by Bruce Wayne59875
Correct me if I'm misunderstanding your point, but isn't A just any constant? So it can refer to anything, regardless of its sign. For instance, if you had -e^c, can't you just call that entire term "A" instead of including the negative within the final expression?

Not sure what you mean. Im trying to say that treating it as an indefinite up to
ln(|y|) = kx + c
is ok, and obviously if y is positive its straightforward. Similarly writing it as
y = A e^(kx)
is the "best", A represents y(0) so is effectively the constant of integration and this works for both positive and negative y and A.

If y and A are negative, then doing the indefinite gives
ln(|A|) = c
or |A| = e^c. As A is negative you have to remember to write it as
A = -e^c
if youve determined c in the first place. So you have to be careful when/where you remove the abs function.

To me, it seems fiddly and easier just to consider the definite integral from
0 to x and from A to y and it all pretty much works out as the abs gets removed when you do the division inside the log as y and A have the same sign.
(edited 10 months ago)
Original post by mqb2766
Not sure what you mean. Im trying to say that treating it as an indefinite up to
ln(|y|) = kx + c
is ok, and obviously if y is positive its straightforward. Similarly writing it as
y = A e^(kx)
is the "best", A represents y(0) so is effectively the constant of integration and this works for both positive and negative y and A.

If y and A are negative, then doing the indefinite gives
ln(|A|) = c
or |A| = e^c. As A is negative you have to remember to write it as
A = -e^c
if youve determined c in the first place. So you have to be careful when/where you remove the abs function.

To me, it seems fiddly and easier just to consider the definite integral from
0 to x and from A to y and it all pretty much works out as the abs gets removed when you do the division inside the log as y and A have the same sign.

Do you think AQA care about sign changes so long as you define a new constant? Because then, you don't have to worry about modulus or anything. 20230604_180816.jpg
(edited 10 months ago)
Reply 11
Original post by Bruce Wayne59875
Do you think AQA care about sign changes so long as you define a new constant? Because then, you don't have to worry about modulus or anything. 20230604_180816.jpg


Guys this is stressing me out 😭
Reply 12
Original post by Bruce Wayne59875
Do you think AQA care about sign changes so long as you define a new constant? Because then, you don't have to worry about modulus or anything. 20230604_180816.jpg


I said in #6 that all this is worth 0 or 1 marks, probably 0, and Id agree that AQA probably wouldnt care and it wouldnt be important if y and A are positive. The only rare time it would be important (1 or 2 marks) is if they design a question which has negative y and A and theyre specifically examining it. I also agree that
y = A e^(kx)
is the "best" way to finish the question and it works for both positive and negative A (and y). The only minor quibble I had with your working was the A = e^c which assumes positive which comes about from removing the abs between lines 1 and 2. The only real reason for mentioning it was just for a bit of understanding.
(edited 10 months ago)

Quick Reply

Latest