implicit differentiation tan(xy)=x+y
Hey Tyler c: \[\Large\rm \tan(xy)=x+y\]So where we gettin stuck at?
im trying to find the slope of the line tangent at (0,0)
but i dont know where to begin exactly
We're going to have a derivative function that depends on BOTH x and y.\[\Large\rm y'(x,y)\]We'll evaluate that derivative function at (0,0) and this will give us the slope, \(\Large\rm m\), of the line tangent to the curve at that point. \(\Large\rm y'(0,0)=m\)
So we need to take a derivative, then to try solve for y'.
so we do tan(xy)-x=y?
the find y prime?
No, you don't NEED TO move the x. We're going to also get a y' popping out of the tangent, so it won't be as nice and neat as you would like.
\[\Large\rm \frac{d}{dx}\tan(xy)\]So what would be your first instinct for this? Do you remember the derivative of tangent?
sec^2(u) u=xy?
Ok good. Turns out we need to also apply our chain rule though:\[\Large\rm \frac{d}{dx}\tan(xy)=\sec^2(xy)\cdot(xy)'\]
We have a `product` of x and y, how should we deal with that? :O
thats where im getting stuck
Notice the word I used... `product`. What rule lets us take the derivative of a `product`? Comeonnnn you know this :3
ok, so x'y+y'x
\[\Large\rm (xy)'=(x)'y+x(y)'\]Ok good. So start by taking the derivative of x with respect to x. x' = ?
1
Good, that's just your normal power rule.\[\Large\rm (xy)'=(1)y+x(y)'\]
y+x?
The derivative of y, with respect to x... this is what we're calling our dy/dx. So we have,\[\Large\rm \frac{d}{dx}\tan(xy)=\sec^2(xy)\cdot(y+xy')\]
We don't get 1 when we take the derivative of y.
That's the part that's confusing? :o
ya
Lemme just try to give an example real quick. Technically, we're ALWAYS applying the chain rule, but when your variable matches the derivative operator, it doesn't give us anything significant. \[\Large\rm \frac{d}{dx}(x)^2=2(x)^1\frac{d}{dx}x\]Here I'm applying the power rule, then I'm applying the chain rule, multiplying by the derivative of the inner function. We could write it like this: \(\Large\rm =2(x)^1\dfrac{dx}{dx}\) But this dx/dx term has no significance. It basically is saying, "as x changes an immeasurably small amount, how much does x change?" \(\Large\rm =2x\)
But with y, this extra chain is really important :O\[\Large\rm \frac{d}{dx}(y)^2=2(y)^1\frac{d}{dx}y\]\[\Large\rm =2y\frac{dy}{dx}\]
I like to just remember it this way (even though this isn't very rigorous), Whenever you take the derivative of something and it contains a variable that doesn't match your derivative operator, you're getting a prime that pops out. So whenever we take the derivative of \(\Large\rm y\) with respect to any other variable besides y, we're going to get a \(\Large\rm y'\) popping out. This will become very important later in the semester when you deal with related rates.
Too confusing still? :c
you still get 2x and 2y for both though?
No :) You get 2x for the first and you get 2y y' for the other one.
We COULD write 2x x' for the first one... but again, the derivative of x with respect to x is just 1.
So for the right side of our equation we get,\[\Large\rm (x+y)'=x'+y'=1+y'\]The y' doesn't simplify in the same way the x' does.
so the derivative of y is just y'
derivative of x is 1
(y+y')sec^2(xy)=y'
?
Yes, since our derivative is being taken with respect to x. Here's another couple of examples to maybe help nail the point. \(\Large\rm (x^3)'=3x^2\) \(\Large\rm (y^3)'=3y^2y'\)
What happened to your x derivatives? They disappeared :O
y+y'*sec^2(xy)=1+y'
On the left side we got:\[\Large\rm \sec^2(xy)\cdot(y+xy')\]And on the right side we got:\[\Large\rm 1+y'\]
\[\Large\rm \sec^2(xy)\cdot(y+xy')=1+y'\]To solve for y' from here is a little tricky. Requires a few algebra steps.
subtract and divide everything, press ctrl alt delete, etc etc
We'll distribute the sec^2 to each term in the brackets.\[\Large\rm y \sec^2(xy) + x \sec^2(xy) y'=1+y'\](We usually write derivative terms on the right like this, hopefully that's not too confusing.
Yah, some of that stuff :3
Then get your derivative terms on the same side, using subtraction. And the other stuff to the other side.\[\Large\rm y \sec^2(xy)-1=y'-x \sec^2(xy)y'\]
Factor out your y',\[\Large\rm y \sec^2(xy)-1=y'\left[1-x \sec^2(xy)\right]\]
And then to solve for y', simply divide, yah?
That gives you your derivative function y'. Then they want to know what the value of y' is when x=0, y=0.
And now that I think about it..... You could have plugged in x=0, y=0 immediately after taking your derivative. Things may have simplified on their own.
-1?
Mmm yah that sounds right! :)\[\Large\rm y'(0,0)=-1\]
dang..
Bro bro bro. Chain rule is, by far, the hardest of the derivative rules to master. Really spend some time with it! :O This y' shenanigans will getcha!
alright
Here is what the graph looks like, just in case that helps. https://www.desmos.com/calculator/sxg4x1lhf6 It's kind of crazy looking... but you can clearly see that the slope at the origin is -1.
weird, those y's are confusing tho
still need to figure out derivatives "with respect to". its the "with respect to" part
im wondering about e^y now
\[\Large\rm \left(e^x\right)'=e^x~x'=e^x(1)\]
\[\Large\rm (e^y)'=e^y~y'\]And you would stop right there :) yah?
ya?
just tried doing this one on my own cos(y^2)+x=e^y i got y'=1/(e^y+2ysin(y^2))
Mmm yes, looks good! :)
Join our real-time social learning platform and learn together with your friends!