The Student Room Group

Scroll to see replies

Hello everyone, I wanted to ask a question about STEP I 2005 Q4 last part of b)
It is given that tan3Ф= 3t-t^3/1-3t^2 (where t=tanФ) and that tan3Ф=11/2. Solutions suggest what I intended to do which is make a cubic equation but how am I suppose to find that one of the factors is (2t-1)? I understand the trial and error part but it is usually enough to try -2,-1,0,1,2. Should I attempt to put in 1/2 from now on or is there something I do not see?
Original post by Zacken
You've got plenty of time to do it! :biggrin:

Are you year 12? :smile:


I wish man, in year 13. But I ain't as good as you Cambridge lot, only doing STEP 1 for Warwick which I have an offer for.:wink: But the plan is if I get a high 1 or a S this year in STEP 1- might ditch Warwick take a gap and apply to Cambridge ...
Original post by -Gifted-
I wish man, in year 13. But I ain't as good as you Cambridge lot, only doing STEP 1 for Warwick which I have an offer for.:wink: But the plan is if I get a high 1 or a S this year in STEP 1- might ditch Warwick take a gap and apply to Cambridge ...


If you're this good at STEP after just starting... you're twice as good as me. :tongue:
Original post by 130398
Hello everyone, I wanted to ask a question about STEP I 2005 Q4 last part of b)
It is given that tan3Ф= 3t-t^3/1-3t^2 (where t=tanФ) and that tan3Ф=11/2. Solutions suggest what I intended to do which is make a cubic equation but how am I suppose to find that one of the factors is (2t-1)? I understand the trial and error part but it is usually enough to try -2,-1,0,1,2. Should I attempt to put in 1/2 from now on or is there something I do not see?


Look up the rational roots theorem.

In either case 1/2, -1/2 is a fairly common trial.
Original post by Zacken
If you're this good at STEP after just starting... you're twice as good as me. :tongue:


I doubt it man, it probably just seems like it because I am only doing the questions that I like the look of, I am a lost man with geometry/vectors and combinations trust me. BTW quick question do you know when I was changing the limits in the second part I actually got -3 and -2 as well. Can i always just choose the positive limits if I ever make another similar substitution ?
Original post by -Gifted-
I doubt it man, it probably just seems like it because I am only doing the questions that I like the look of, I am a lost man with geometry/vectors and combinations trust me. BTW quick question do you know when I was changing the limits in the second part I actually got -3 and -2 as well. Can i always just choose the positive limits if I ever make another similar substitution ?


As in, you got u2=4,9u^2 = 4, 9? Yeah, just take the positive root.
Original post by Zacken
As in, you got u2=4,9u^2 = 4, 9? Yeah, just take the positive root.


Oh yup thats what I meant :smile:, and will it always work ? Because I thought it needed to be an even function for this to work ?
Original post by -Gifted-
Oh yup thats what I meant :smile:, and will it always work ? Because I thought it needed to be an even function for this to work ?


This is probably wrong (somebody correct me), but I think that what you're *actually* doing is:

u=ex+1>0u = \sqrt{e^x + 1} > 0 and then squaring both sides, so uu is gonna be positive.
Original post by 130398
Hello everyone, I wanted to ask a question about STEP I 2005 Q4 last part of b)
It is given that tan3Ф= 3t-t^3/1-3t^2 (where t=tanФ) and that tan3Ф=11/2. Solutions suggest what I intended to do which is make a cubic equation but how am I suppose to find that one of the factors is (2t-1)? I understand the trial and error part but it is usually enough to try -2,-1,0,1,2. Should I attempt to put in 1/2 from now on or is there something I do not see?


Well i noticed that a term was 11/2 so it could not be an integer and then i just guessed it and got it. I remember this very strongly. Someone i know when we used to do prep together could jt get this part and he ended up getting 110/120 this year on II so don't worry too much :smile:


Posted from TSR Mobile
Alright this solution took me exactly an hour and like I feel like I have a wishy washy answer so someone please give me a mark and perhaps provide a more rigorous reasoning for what I did ? Thanks again ...
Original post by Zacken
This is probably wrong (somebody correct me), but I think that what you're *actually* doing is:

u=ex+1>0u = \sqrt{e^x + 1} > 0 and then squaring both sides, so uu is gonna be positive.


Tbh if what you said is right it would sense :smile: BTW not gonna annoy you too much but whenever you have time could you look at ^ , and provide a mark ?
Original post by -Gifted-
Alright this solution took me exactly an hour and like I feel like I have a wishy washy answer so someone please give me a mark and perhaps provide a more rigorous reasoning for what I did ? Thanks again ...


You got the integrals wrong.

It should have been π2αxsinxπ2dx+αππ2xsinxdx\displaystyle \int_{\frac{\pi}{2}}^{\alpha} x \sin x - \frac{\pi}{2} \, \mathrm{d}x + \int_{\alpha}^{\pi} \frac{\pi}{2} - x\sin x \, \mathrm{d}x

Since you can see that cosecx2xπ\cosec x \leq \frac{2x}{\pi} in the interval [π2,α]\left[\frac{\pi}{2}, \alpha\right] This is the same as saying that xsinxπ2x \sin x \geq \frac{\pi }{2} by inverting both sides. (Since if one thing is bigger than the other, then 1/big thing will be smaller than 1/small thing).

Then you can do the same thing in the interval [alpha, pi].

That'll lose you an accuracy mark and 2 method marks.

Same mistake in part(ii), in the interval [0, ln 2] we have that ex11=2ex\displaystyle ||e^x - 1|-1| = 2-e^x Since for in the interval [0, ln 2] you're gonna get |e^(some number less than ln 2) - 1| - 1 = |some number less than 2 - 1| - 1 = some number less than 1 - 1 = some number less than 0. And hence negative. So the modulus signs equals effectively negates the function. Although note that e^x - 1 will be positive in that region, so the modulus signs around that simply disappear. If anything needs clarification, please ask. :smile:

I'd say you get 13/20.
(edited 8 years ago)
Original post by -Gifted-
Tbh if what you said is right it would sense :smile: BTW not gonna annoy you too much but whenever you have time could you look at ^ , and provide a mark ?


I'm a bite of a lifeless human being right now, with jan exams coming up - so feel free to tag me whenever you want a solution marked. I honestly don't mind, it gives me a much needed break. :smile:
Original post by Zacken
You got the integrals wrong.

It should have been π2αxsinxπ2dx+αππ2xsinxdx\displaystyle \int_{\frac{\pi}{2}}^{\alpha} x \sin x - \frac{\pi}{2} \, \mathrm{d}x + \int_{\alpha}^{\pi} \frac{\pi}{2} - x\sin x \, \mathrm{d}x

Since you can see that cosecx2xπ\cosec x \leq \frac{2x}{\pi} in the interval [π2,α]\left[\frac{\pi}{2}, \alpha\right] This is the same as saying that xsinxπ2x \sin x \geq \frac{\pi }{2} by inverting both sides. (Since if one thing is bigger than the other, then 1/big thing will be smaller than 1/small thing).

Then you can do the same thing in the interval [alpha, pi].

That'll lose you an accuracy mark and 2 method marks.

Same mistake in part(ii), in the interval [0, ln 2] we have that ex11=2ex\displaystyle ||e^x - 1|-1| = 2-e^x Since for in the interval [0, ln 2] you're gonna get |e^(some number less than ln 2) - 1| - 1 = |some number less than 2 - 1| - 1 = some number less than 1 - 1 = some number less than 0. And hence negative. So the modulus signs equals effectively negates the function. Although note that e^x - 1 will be positive in that region, so the modulus signs around that simply disappear. If anything needs clarification, please ask. :smile:

I'd say you get 13/20.


I am not sure if I quite get you, you know in the interval between pi/2 and alpha, its clear that 2x/pi is greater than cosec x. If 2x/pi corresponds to pi/2 and cosecx corresponds to xsinx, I need to do pi/2-xsinx don't I ? I get the second part of your comment though :smile: And, thats beautiful, expect me to throw a lot of solutions at you this coming week :biggrin::colondollar:
Original post by -Gifted-
I am not sure if I quite get you, you know in the interval between pi/2 and alpha, its clear that 2x/pi is greater than cosec x. If 2x/pi corresponds to pi/2 and cosecx corresponds to xsinx, I need to do pi/2-xsinx don't I ? I get the second part of your comment though :smile: And, thats beautiful, expect me to throw a lot of solutions at you this coming week :biggrin::colondollar:


I agree that in that interval you have 2x/pi > cosec x

So 2x/pi > 1/sin x ***************

so pi/(2x) < sin x so pi/2 < x sin x which is the same thing as x sin x > pi/2.

Hence x sin x- pi/2 > 0 in the interval [pi/2, alpha].

Moving from ********* to the second line, we take the reciprocal of both sides and flip the inequalities. If x >y then 1/x < 1/y.
Original post by Zacken
I agree that in that interval you have 2x/pi > cosec x

So 2x/pi > 1/sin x ***************

so pi/(2x) < sin x so pi/2 < x sin x which is the same thing as x sin x > pi/2.

Hence x sin x- pi/2 > 0 in the interval [pi/2, alpha].

Moving from ********* to the second line, we take the reciprocal of both sides and flip the inequalities. If x >y then 1/x < 1/y.

AHH man that actually makes I am getting so careless... BTW could you please offer some guidance on STEP 1 2006 question 2 ? What perplexes me is that if the length of the rope is 4a, which is larger than the largest length possible from a corner of (8^0.5)a which is the diagonal length, why can't the goat just graze the entire field because every part of the barn is accessible? I think I am misunderstanding the question ? Thanks :smile:
Original post by -Gifted-
AHH man that actually makes I am getting so careless... BTW could you please offer some guidance on STEP 1 2006 question 2 ? What perplexes me is that if the length of the rope is 4a, which is larger than the largest length possible from a corner of (8^0.5)a which is the diagonal length, why can't the goat just graze the entire field because every part of the barn is accessible? I think I am misunderstanding the question ? Thanks :smile:


I remember seeing somewhere that you lost 85% of the marks if you assume the goat is tethered to the corner.

Try letting the goat be tethered a distance x from a corner and then working through the algebra. :-)

Edit: Q3 on that paper is a beauty.
(edited 8 years ago)
Original post by Zacken
I remember seeing somewhere that you lost 85% of the marks if you assume the goat is tethered to the corner.

Try letting the goat be tethered a distance x from a corner and then working through the algebra. :-)

Edit: Q3 on that paper is a beauty.


Yh the solutions says you get few marks


Posted from TSR Mobile
(edited 8 years ago)
@Zacken
Anyways, that question was quite ugly tbh might attempt it later, but for the moment have done this problem but couldn't really get the last part so help would be nice. So could someone please comment? Preferably a mark out of 20 and feedback ? STEP 1 2006 Q6
Original post by -Gifted-
@Zacken
Anyways, that question was quite ugly tbh might attempt it later, but for the moment have done this problem but couldn't really get the last part so help would be nice. So could someone please comment? Preferably a mark out of 20 and feedback ? STEP 1 2006 Q6


First part looks ok.

Second part: in these situations you always need to ensure that the values you find satisfy all the conditions. You might lose a mark or two for not checking that 2 and 3 verify not just the 'main condition', but also the other two.

Third part: Solving the equations properly is a ballache. However, I believe the question only asks you to find a solution as opposed to all solutions, so spotting a solution, as you seem to have done, is absolutely fine. That is, provided that you verify that your solutions verify all the equations you generate (which you have again omitted). You'd probably lose a couple of marks here.

(NB. I haven't had time to go through the algebra, but it looks good.)

So I'd say 16-17 marks for this, good job! The moral of the story is that if you just need to find a solution to an equation, just 'spot' one and verify that it satisfies all the necessary conditions. If you liked this question it might be worth looking into 'Pell equations'; I can direct you to another question on this theme if you're interested.

Posted from TSR Mobile

Latest

Trending

Trending