Mc16, Mg15 - Implicit and parametric differentiation
A relation can reveal a gradient without being rearranged into . Mc16 develops implicit differentiation; Mg15 finds gradients from two parameter rates. Use first derivatives to match points, construct tangents and normals, and distinguish a genuine tangent from a failed derivative ratio.
Differentiating a relation with a changing y
What is the tangent to at ? You could isolate , but that immediately raises a branch question. Differentiate the relation itself, keeping the square intact.
You already know
The chain rule differentiates an outer function at its changing input. The product rule gives , and a tangent at a point uses the derivative evaluated there.
Along a differentiable branch, changes with , so the inner expression has derivative .
Implicit differentiation
Implicit differentiation finds the derivative of a relation between coordinates without first isolating one coordinate. Treat as a local function of : , , and for radian input.
The chain and product rules are not in the formula sheet: learn them. Keep the result as an equation until all terms have been collected.
Worked example 1: a square of a coordinate sum
Question. Show that has derivative , where .
Plan. Differentiate the outer square, include the full inner derivative, then isolate .
Why this step? The derivative of is not just one: is changing too.
Answer. The stated derivative follows for .
Check. At , the relation gives and the slope is .
Change only the coefficient on the right, from two to four.
Worked example 2: the same outer structure
Question. Differentiate .
Plan. Retain the same chain-rule left side and update the right-hand derivative.
Why this step? Only the right-hand rate changes; the outer square still multiplies the whole inner rate.
Answer. , where .
Check. The point satisfies the relation and gives slope zero.
Expanded relations need the product rule too. For ,
where the denominator is nonzero. Likewise gives . A logarithmic relation , , gives , hence ; its slope at is one.
Watch out
Wrong: differentiates to . Right: . The missing inner factor is the changing coordinate's contribution.
Change the right-hand coefficient again.
Using implicit gradients and checking special tangents
Two points with the same -coordinate can belong to different branches. A derivative containing is useful precisely because it keeps that branch information.
Worked example 3: the upper branch at the opening point
Question. Determine the tangent and normal to at .
Plan. Check the point, evaluate the implicit derivative, then use a negative reciprocal for the normal.
Why this step? The chosen -coordinate selects a branch; alone does not.
Answer. Tangent ; normal .
Check. Both pass through , and their gradient product is minus one.
Change only the selected branch point at .
Worked example 4: the lower branch at the same input
Question. Determine the tangent and normal at on the same relation.
Plan. Use the new value of before finding the line equations.
Why this step? A different branch changes the local gradient even though the horizontal coordinate is unchanged.
Answer. Tangent ; normal .
Check. The explicit lower branch also has derivative , giving at two.
The older mixed relation illustrates the same point: at , tangent gradient and normal gradient ; at , the gradients are and . The normal lines are and .
Special directions need their own treatment. For , the point has tangent , normal . At , differentiating in the reverse direction gives , establishing tangent and normal . On , a horizontal tangent would require , contradicting the curve, so there is none.
Exam tip
Bracket the two product-rule terms when differentiating . For a given point, show its substituted gradient before giving the line. A zero denominator alone does not establish a vertical tangent: a expression needs the curve inspected.
Use the same relation at a point where the tangent becomes horizontal.
Finding a gradient from two parameter rates
A parameter supplies both coordinates. To obtain change in per unit change in , divide their changes per parameter unit. Eliminating the parameter is optional.
Parametric first derivative
For differentiable coordinate functions , where ,
This follows by rearranging .
Not in the formula sheet: learn it.
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account