ROBOTECA FREE SAMPLE
Dashboard

A free lesson from Engineering Foundations (Maths, Physics & Chemistry): the whole module, nothing cut short.

LESSON · Algebra & Functions

Rearranging equations and solving for the unknown

Turn 1 40 min LESSON

ALearning Material

An equation states that two expressions have the same value. states that the number on the left equals the number on the right. Rearranging rewrites an equation so that a different letter stands alone on one side, while keeping the two sides equal. One rule governs every rearrangement: ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍apply the same operation to both sides. Two equal quantities remain equal when both are doubled, both are halved, or both have three added. A rearrangement repeats such operations until the required letter stands alone.

The letter being solved for has no special status in the equation. In , the equation states a relationship among the three letters, and it holds whichever two of them are known. A formula printed with on the left is printed that way because was the quantity of interest. Making the subject does not change the physics. It changes which question the same relationship is arranged to answer.

The practical method is to undo the operations in the reverse of the order in which they were applied. If an expression was built by taking , multiplying by 2 and then adding 3, then is recovered by subtracting 3 and then dividing by 2. The order is reversed for the same reason that clothes come off in the reverse of the order they went on.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍The second and fourth lines write the operation on both sides explicitly. This layout takes longer, but it is the safest habit until the steps become automatic. The most common error in rearranging is applying an operation to one side and not the other, and a layout that shows both sides makes that omission visible.

The same method works with letters in place of numbers. To make the subject of , note that the right side was built by multiplying by , so divide both sides by :

The method did not change when the numbers became letters. The operation to apply is chosen by looking at what was done to the required letter, not at how complicated the other side is.

In , the target sits in a denominator, so it is not multiplied by anything that can be divided away. To make the subject, multiply both sides by first, which lifts it out of the denominator, and only then divide:

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍This step occurs constantly: a letter in a denominator has to be multiplied out before anything else can be done to it. Dividing both sides by at the first step would bury it further, giving , which is true and of no use.

A resistor and a capacitor in series have a time constant , in seconds when is in ohms and in farads, a circuit that returns in the electronics course. A design needs a 10 ms time constant with a 4.7 kohm resistor, so the capacitor is:

Two checks are available at no cost. The first is the units: a second per ohm is a farad, which is the dimensional homogeneity rule of the units lesson applied to a rearrangement, and it exposes an inverted formula at once. The second is the behaviour: is ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍divided by , so a larger resistor needs a smaller capacitor for the same time constant. That is the expected behaviour of the circuit, and a rearrangement that predicts the opposite contains a sign or inversion error.

Three operations can change the set of solutions, and each needs care.

Dividing by something that might be zero. does not give : dividing both sides by is valid only when , and is a valid solution that the division deletes. The safe method is to collect everything on one side, , factor to , and read off both possibilities: either or . Division by a letter always removes a case, and that case is often important, since a zero current or a zero speed is a condition an engineer frequently needs to know about.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Squaring both sides. Squaring can introduce solutions that the original equation does not have. has no solution, but squaring gives , and . Squaring is not reversible, because two different numbers can have the same square, so it can turn an impossible statement into a possible one. After squaring, every answer must be substituted back into the original equation, and any answer that fails is discarded.

Taking a square root. The reverse operation loses a sign. has two solutions, and , and writing drops one of them. The correct step is , after which the physics decides: a length or a speed keeps the positive root, while a displacement or a voltage may take either sign. Both roots are produced, and one is discarded only for a stated physical reason.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍When the unknown appears on both sides, its terms are gathered first. becomes by subtracting and 8 from both sides, and only then does the undo-in-reverse method apply. The rule is to move every term containing the unknown to one side and everything else to the other, then undo.

The final check is to substitute the answer back into the original equation. A rearrangement cannot be checked by re-reading it, because a re-reading repeats any mistake made the first time. Substitution is an independent operation with a yes-or-no result. For a symbolic rearrangement, where no number is available to substitute, the equivalent check is to choose any convenient set of values, such as , , , and confirm that the original and the rearranged forms agree.

Formulas & method.

  • The one rule. If then for any operation applied to both sides, provided that operation is defined and reversible.
  • ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Inverse pairs. Addition and subtraction, multiplication and division, power and root , exponential and logarithm. Undo in the reverse of the order applied.
  • Unknown in a denominator. Multiply both sides by it first: .
  • Unknown on both sides. Collect all its terms on one side first, then undo.
  • Factoring out. , which is how to solve without dividing by a letter.
  • Zero-product rule. If then or . This rule makes factoring a solution method, not only a tidying step.
  • Square roots. . Keep both, then discard on physical grounds.
  • Squaring. Not reversible, so it may create false solutions. Substitute every answer back.
  • ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍The check. Substitute into the original equation, or, for a symbolic result, into both forms with arbitrary numbers.

Where this shows up.

  • Robotics. Every kinematic relation is used in both directions. Forward kinematics gives a position from joint angles, and inverse kinematics is the same relation rearranged. The from a square root is why an arm usually has an elbow-up and an elbow-down solution for the same target.
  • Automation. Instrument scaling is rearranged constantly: a 4-20 mA span is converted to engineering units one way and back the other, and a flow computation is solved for the differential pressure that a given flow will produce. A square-root extraction on a flow meter is where the lost sign and the impossible root both appear in a real loop.
  • Electronics. Ohm's law and the power relations are one relationship arranged four ways, and choosing a resistor, a supply voltage or a power rating each needs a different subject. Component selection is nearly always a rearrangement, as the capacitor above was.
  • ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Digital design. Timing closure is a rearranged inequality: a required setup time is a constraint on clock period, and finding the maximum frequency a path supports means solving that constraint for the period instead of checking it.

Why it exists. A relationship in physics or engineering is a single fact, but the question asked of it changes with the job. The same is a measurement when two of the three quantities are known, a design equation when a component is being chosen, and a diagnosis when a reading is wrong. Rearranging lets one established relationship serve all of these purposes without three separate formulas, and one relationship remembered correctly is far safer than three arrangements remembered approximately.

Mental model. An equation is a balance with equal weight on both pans. Any change made to one pan must be made to the other, or the balance tips and the equation no longer holds. Solving removes weight from the pan holding the unknown, one reversible operation at a time, until that pan holds only the unknown, and everything removed is accounted for on the other side.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Common misunderstandings.

  • "I can cancel anything that appears on both sides." Cancelling is division, and division by a letter is valid only when that letter cannot be zero. In , the cancellation deletes the solution . A term that is added, not multiplied, cannot be cancelled at all: in the values do not cancel, because the fraction bar divides the whole numerator by the whole denominator.
  • "A square root gives one answer." The square-root symbol denotes the non-negative root by convention, but the equation has two solutions. Solving a physical problem means producing both and then discarding one for a stated reason.
  • "Rearranging is just moving things across the equals sign and flipping the sign." That shortcut describes how the valid operations look on the page in simple cases. It is not a rule, and it fails when the unknown is in a denominator, inside a power or on both sides. Applying an operation to both sides always works, while the shortcut works only in the cases it was learned from.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Connections. The rule here puts the substitution and notation of the previous lesson to work, and the units check on a rearranged formula is dimensional homogeneity from the units lesson. Solving by factoring with the zero-product rule is the method that the quadratics lesson generalises. The from a square root returns in inverse kinematics and in every quadratic that follows. Logarithms, later in this topic, are the inverse operation that reaches an unknown inside an exponent, which none of the tools here can do.

BImmediate Active Recall

QUERY

State the single rule that all rearranging rests on, and the one condition on it.

REVEAL
ANSWER

Apply the same operation to both sides. The operation must be defined and reversible for the values involved, which is the condition that fails when dividing by something that may be zero or taking a square root without keeping both signs.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Did you recall it?
QUERY

You need to make the subject of . What is the first move, and why is dividing by wrong?

REVEAL
ANSWER

Multiply both sides by , giving , then divide by . Dividing by buries it further: it gives , which is true and of no use. A letter in a denominator must be multiplied out before anything else can reach it.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Did you recall it?
QUERY

Why is dividing both sides of by a mistake, and what is the safe method?

REVEAL
ANSWER

It deletes the solution , which is often the important case. The safe method is to collect and factor: , so , and the zero-product rule gives either or .

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Did you recall it?
QUERY

Squaring both sides and taking a square root each cause a different fault. Name both.

REVEAL
ANSWER

Taking a root loses a solution: has , and writing keeps only one. Squaring introduces solutions: has none, yet squaring gives . A root therefore needs , and after squaring every answer must be substituted back into the original.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Did you recall it?
QUERY

A time constant is . You need 10 ms with a 4.7 kohm resistor. Give the capacitor and the two free checks.

REVEAL
ANSWER

. The unit check: a second per ohm is a farad. The behaviour check: is divided by , so a larger resistor needs a smaller capacitor, which is the expected behaviour.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Did you recall it?
QUERY

Why is re-reading a rearrangement a poor check, and what should you do instead?

REVEAL
ANSWER

Re-reading repeats any mistake made the first time. Substituting the answer into the original equation is an independent operation with a yes-or-no result. For a symbolic rearrangement, put arbitrary numbers into both the original and the rearranged form and confirm that they agree.

Did you recall it?

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍CConceptual Questions

Answer in your own words first, then compare with the model answer.

PROMPT

A colleague says rearranging is "moving a term across the equals sign and changing its sign, or moving a multiplier across and making it a divisor". Say what is right about that description and where it breaks.

REVEAL MODEL ANSWER
MODEL ANSWER

The description is accurate for how the valid operations look on the page in the simplest cases. Subtracting the same term from both sides leaves that term looking as though it crossed the sign and changed sign, and dividing both sides by a factor leaves it looking as though it crossed and became a divisor. As a summary of appearances it is correct, and it is faster than writing both sides out, which is why experienced engineers use it. It breaks because it is a rule about positions on the page, not about operations, so it gives no guidance once the page looks different from the examples it was learned from. It says nothing about an unknown in a denominator, where the required step is a multiplication that does not resemble crossing anything. It says nothing about an unknown on both sides, which must be gathered before it can be undone. It gives no warning that dividing by a letter can delete a solution, because the shortcut has no concept of a value being zero. It cannot express what to do with a square root or an exponential, where the inverse is an operation, not a repositioning. The shortcut also hides its own misuse: a wrong application still produces something that looks like the pattern, so the error stays invisible until the answer is used. Applying an operation to both sides always works, can always be checked, and reduces to the shortcut once the full steps are no longer needed.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Compared to the model answer - did you get it?
PROMPT

An arm's forward kinematics gives a fingertip position from its joint angles, and inverse kinematics is that relation rearranged. Explain why the inverse usually has two answers while the forward direction has exactly one, using what this lesson says about square roots.

REVEAL MODEL ANSWER
MODEL ANSWER

The forward direction is an evaluation: the joint angles are substituted into an expression and one position comes out, because evaluating an expression gives a single result. The inverse direction is solving, and solving undoes operations that may not be one-to-one. The geometry of a two-link arm produces a relation in which a joint angle appears inside a cosine, and recovering the angle involves a square root of a quantity built from the link lengths and the target position. The square root produces the two answers: the equation has two solutions, and physically they are the elbow bending one way or the other to put the fingertip in the same place. Both are valid, so neither is discarded as unphysical, and a real arm can usually adopt either. This differs from the cases in this lesson where the negative root is discarded because a length cannot be negative. Here the choice is made outside the algebra: by which configuration avoids an obstacle, which one is reachable without passing through a singularity, or which one is closer to the arm's current position. In general, a forward relation can map several inputs to one output, and inversion cannot recover the information that this mapping discards, so it returns a set of candidates and something else must choose between them.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Compared to the model answer - did you get it?
PROMPT

You rearrange a formula, substitute numbers, and get an answer that is out by a factor of about 1000. What does the size of the error tell you about where to look, and what check would have caught it before the arithmetic?

REVEAL MODEL ANSWER
MODEL ANSWER

A factor near 1000 is a power of ten, and a power of ten is almost never an algebra mistake. A rearranging error produces wrong structure, which shows up as an error factor that changes with the inputs, a wrong sign, or wrong behaviour when an input changes. A clean factor of 1000 that stays the same for any inputs points to a ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍prefix: a kilohm treated as an ohm, a millisecond as a second, a microfarad as a farad. The place to look is therefore the substitution, not the rearrangement, and specifically the point where a prefixed value entered the arithmetic. Carrying units through the substitution, as the units lesson requires, would have caught it earlier. Had the value been written as instead of as 4.7, the exponent would have been in the expression instead of in the writer's head, and the answer would have carried its own magnitude. An order-of-magnitude estimate before the calculation is a second check: a microfarad-scale capacitor with a kilohm-scale resistor gives a few milliseconds, so an answer of 10 seconds stands out at once. Algebra errors and arithmetic errors have different signatures, and the signature shows which half of the work to re-examine, which is much faster than repeating both.

Compared to the model answer - did you get it?

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍DPractice Problems

P1 (easy). Make the named letter the subject, showing the operation applied to both sides at each step: (a) in ; (b) in ; (c) in ; (d) in .

P2 (medium). Solve each, stating every solution and saying which if any you would discard and on what grounds: (a) ; (b) where is a speed in m/s; (c) ; (d) ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍.

P3 (harder). A voltage divider gives . (a) Make the subject. (b) You have a 12.0 V supply and need 5.00 V out, with kohm. Find and check by substituting back. (c) Show what the rearranged formula predicts as approaches , and say what that tells you about the divider as a way of making a supply rail.

Solutionsclick to reveal

P1. (a) Add 7 to both sides: . Divide both sides by 3: ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. Check: . (b) Divide both sides by : . (c) Multiply both sides by 2: . Divide both sides by : . (d) is in a denominator, so multiply both sides by to lift it out: . Then divide both sides by : . Reading it: a higher frequency or a larger resistor needs a smaller capacitor, which is the expected behaviour of a filter.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍P1Compared to this solution - did you get it right?

P2. (a) The unknown is on both sides, so gather first. Subtract : . Subtract 4: . Divide by 4: . (b) . Both are algebraically correct, and here is a speed, which is a magnitude and cannot be negative, so keep and discard on the physical ground that a speed is non-negative. Had been a velocity along an axis, both values would stand, describing the two directions. (c) Do ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍not divide by . Collect: , so , so is the only solution. Dividing by would have given , a false statement, and would have hidden the real answer. (d) Square both sides: , so , which factors as , giving or . Now substitute both back into the original, because squaring can introduce solutions. For : and , so it holds. For : while ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍, so it fails. The only solution is , and is an artefact of the squaring.

P2Compared to this solution - did you get it right?

P3. (a) Multiply both sides by : ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. Expand: . The unknown is now on both sides, so gather: . Divide: . (b) . Substituting back: , which confirms it. (c) As , the denominator , so . The rearranged form shows what the original hides: an output close to the input requires an far larger than . That is a warning about using a divider as a supply rail. A large means a high source impedance, so any current drawn by the load changes the output, and the divider holds its stated voltage only when the load draws almost nothing. Near the top of its range the division ratio also becomes very sensitive to component tolerance. A divider is a ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍reference, not a supply, and this limit in the algebra shows why.

P3Compared to this solution - did you get it right?

EFeynman Exercise

Explain to somebody who can do arithmetic but has never rearranged a formula: what a rearrangement actually is, why the same relationship can be written several ways without any of them being more true than the others, and what the two operations are that can lose or invent an answer. Write your explanation, then compare.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍REVEAL MODEL ANSWER
MODEL ANSWER

An equation says that two things have the same value. Because the two sides are equal, the same operation can be applied to both and they stay equal: double both, subtract five from both, divide both by three. That single rule is all of rearranging. It is used repeatedly, each step removing something attached to the required letter, until that letter stands alone on one side and everything else is on the other. The same relationship can be written several ways because no single arrangement is the relationship itself. The relationship is the fact that three quantities are tied together, and an arrangement only chooses which of them stands on the left, usually the one that needs to be found. Voltage equals current times resistance, resistance equals voltage over current, and current equals voltage over resistance are one fact written three ways, and a circuit obeys all three at once. Which one to write depends on what is known and what is wanted, not on which is true. Two operations can go wrong, in opposite directions. Taking a square root loses an answer, because two different numbers have the same square, so undoing a square has to offer both the positive and the negative value, and the physical situation then decides between them. Squaring both sides can add answers, because it can turn an impossible statement into a possible one, so every result found after squaring has to be tested in the original equation and discarded if it fails. A third trap is dividing both sides by a letter. If that letter could be zero, the division deletes zero as a possible answer, and zero is often the answer an engineer needs. The safe version collects everything on one side and factors, which keeps every possibility visible. The check is to put the answer back into the original equation and see whether the two sides come out equal. Re-reading the working repeats the same reading that produced any mistake, while substitution is a different operation and can catch what re-reading cannot.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Compared to the model answer - did you get it?

FError Analysis Framework

The mistake: applying an operation to one side only, and cancelling a term that is added instead of multiplied.

Why it happens: both errors come from treating rearranging as a visual rule about moving symbols instead of as an operation applied to a whole side. With the picture of symbols crossing the equals sign, there is no side to apply anything to, so a step that changes only one side does not look like an omission. The cancelling error has the same root: appears to contain a repeated symbol, and repeated factors cancel in fractions, but cancelling is division, and the fraction bar divides the whole numerator by the whole denominator, so nothing divides out.

How to recognize it: an answer whose units do not work out, which is the fastest detector and costs nothing; a result that behaves the wrong way when an input is changed, such as a capacitor that gets larger when the resistor does for a fixed time constant; an expression that has become simpler than the physics can support, which is the signature of an illegal cancellation; an answer that fails when substituted back into the original, which is the definitive test; a solution set with fewer members than the equation's highest power suggests.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍How to correct it: write the operation on both sides explicitly, as a separate line, until the habit is automatic, and treat any step where only one side changed as an error. Before cancelling anything in a fraction, factor the numerator and the denominator and confirm that the cancelling piece is a factor of each, not a term. Carry units through every substitution, so that a wrong arrangement shows up in the units. After solving, substitute back into the original equation, and after any squaring step treat that substitution as mandatory. Where a step divided by a letter, check what happens when that letter is zero, and if the result is physically meaningful, redo the step by factoring instead.

GMini Challenge

The lens equation in optics, which you will meet again in the waves topic, is . Make the subject, in a single fraction with no stacked fractions left. Then use your result to find the image distance for a 50 mm lens with an object 2.0 m away, and say what your formula predicts as approaches ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍.

REVEAL MODEL ANSWER
MODEL ANSWER

Every unknown is in a denominator, so the first step isolates the term containing , and the second inverts it. Subtract from both sides: . Combine the right side over a common denominator , which makes the inversion possible: . Both sides are now single fractions, so taking the reciprocal of both sides is valid and gives . The reciprocal step is valid only because each side is one fraction. Inverting term by term into is a standard error, and it is the same error as cancelling an added term, since the reciprocal of a sum is not the sum of the reciprocals. For the numbers, work in millimetres so that the prefixes cannot cause an error: mm and mm. Then ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. The image forms just beyond the focal distance, which is why a lens focused near infinity sits close to its focal length from the sensor, and why focusing on a distant subject moves the barrel very little. As , the denominator and : an object at the focal point produces no image at any finite distance, because the rays leave the lens parallel. The rearranged form states that limit directly while the original form hides it, as the voltage divider in P3 also showed. For , the denominator is negative, so is negative, which is the sign convention's way of reporting a virtual image on the same side as the object. The algebra continues to work past the point where the physical picture changes, and the sign of the result shows that the change has occurred.

Compared to the model answer - did you get it?

Quiz Check

Answer without looking back.

QUIZAuto-graded check · feeds your mastery score
  1. The single rule underlying all rearranging is:

  2. ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍To make the subject of , the first step is:

  3. Dividing both sides of by is unsafe because:

  4. After squaring both sides of an equation you must substitute every answer back into the original because:

  5. An answer that is wrong by a clean factor of 1000, unchanged whatever inputs you use, points at:

This is a free sample

Progress and the spaced-repetition reviews are part of the course. The full track continues from here.