3.4 Systems in Three Variables

Linear equations in three variables and their graphs as planes, ordered triples as solutions, solving a three-equation system by eliminating one variable twice, the degenerate cases, and modelling a situation with three unknowns.

Subject: Algebra 2 · 68 slides · symbolic lesson

Open the interactive version of this deck

What this lesson covers

The lesson, slide by slide

1. Systems in Three Variables

Title

Algebra 2 · Lesson 3.4

The same elimination you already know, run one variable deeper

2. What this lesson gives you

Objectives

In Lesson 3.2 you took two equations in two unknowns and combined them so that one letter vanished. That left a single equation in a single letter, you solved it, and you put the value back to get the other one. Nothing about that method is about to be replaced.

What changes is the size of the room. Three unknowns need three facts, so you get three equations, and one combination is no longer enough. You will do the same elimination twice, on two different pairs, and land in exactly the two-variable problem you already know.

A warning about what actually goes wrong here. Almost nobody loses marks on this topic because the algebra is hard; each individual move is something you could do a year ago. People lose marks on bookkeeping, by losing track of which pair of equations they just combined and which letter they were supposed to be removing.

So the whole deck is organised around keeping that visible. Every equation gets a number, every combination says which pair it used, and every answer gets checked three times rather than once. Build those habits here and the rest of the chapter is comfortable.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 178 — Before, Now, Why and the Key Vocabulary

3. First, the two-variable version you already own

Warm-up

Before anything with three letters in it, run the move you will be repeating all lesson. Write your working down properly, one line at a time, because in a moment the same lines will be twice as long and twice as easy to muddle.

\[ \begin{aligned} 2x + y &= 5 \\ 3x - y &= 5 \end{aligned} \]

Discussion prompt

Solve this pair by elimination. Say which letter you removed, why adding the two equations removed it, and how you found the other letter afterwards.

Two things in that solution are worth naming before you go on, because both come back immediately. First, you did not choose x or y at random: you looked at the coefficients and noticed that y was already set up to cancel. Second, you finished by substituting back into an original equation, not into something you had derived.

Check the pair before moving on, and check it in both equations rather than the one you used. Two twos plus one is five, and three twos minus one is also five, so the pair is right.

Now hold that method in mind and picture the same question with a third letter in every equation. Adding one pair of equations can only ever kill one letter, so one addition leaves you with two letters still standing. That is the gap this lesson fills, and it fills it in the most obvious way available: do it again.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.2, p. 160 — the elimination method in two variables

4. Planes, triples, and what counts as a solution

Section

Part 1

5. A linear equation in three variables

Concept

Start with a single equation rather than a system, because the new object needs a name and a shape before it gets combined with anything.

linear equation in three variables — An equation of the form ax + by + cz = d, where a, b and c are not all zero. Every letter appears to the first power only, and none of them multiply each other.

\[ 2x + y - z = 5 \]

Notice what is not allowed. There is no x times y, no z squared and no letter under a root. That restriction is exactly what the word linear buys you, and it is the reason the graph turns out to be something as simple as a flat surface.

One equation like this has an enormous number of solutions. The triples (3, 0, 1) and (0, 5, 0) and (1, 4, 1) all satisfy it, and so do infinitely many others, because you may choose two of the three letters freely and the equation then forces the third.

That is the real reason three equations are needed. One equation leaves two free choices, two equations leave one, and only the third can pin every letter down to a single value. Counting free choices like that is a useful instinct for the whole of this chapter.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 178 — the definition of a linear equation in three variables

6. One equation, and far too many solutions

Intuition

It is worth seeing just how loose a single equation in three letters really is, because that looseness is the reason the lesson needs three of them.

\[ 2x + y - z = 5 \]

Pick any two of the three letters at random and the equation then forces the third. Choose x and y freely, and z has to be two x plus y minus five; nothing you choose can fail.

choose xchoose ythen z must becheck
3016 + 0 - 1 = 5
0500 + 5 - 0 = 5
1412 + 4 - 1 = 5
-21-8-4 + 1 + 8 = 5
10102520 + 10 - 25 = 5

Every row of that table is a solution of the equation, and the table could be extended for ever. So one equation does not narrow the possibilities down to anything you could call an answer; it only removes one degree of freedom out of three.

Now count what each extra equation buys you. Two equations leave one free choice, which is a whole line of triples, and only the third can remove the last freedom and leave a single point. Three unknowns, three facts, and not one fewer.

Hold that counting argument, because it comes back twice: once when you see that two equations can never produce a single triple, and once in Part 5, when a word problem gives you only two real facts about three unknown quantities.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 178 — the system of three equations and what a solution of it is

7. Reading the general form piece by piece

Notation

The general form is short, and every symbol in it earns its place. Take it apart now so that when you start multiplying whole equations by numbers you know exactly what you are multiplying.

Annotate

On: \( ax + by + cz = d \)

  • The three unknowns. They are the quantities you are hunting, and an answer has to give a value for all three, not just for the one that came out easiest.
  • The coefficients: fixed numbers already in the equation. The rule says they are not all zero, which simply means that at least one letter really is present. If all three were zero there would be no equation left to solve.
  • One or two of them may be zero, and that is common and useful. An equation like three x plus y equals four is a perfectly good equation in three variables whose z coefficient happens to be zero, and such an equation is a gift when you are eliminating z.
  • The constant term, alone on the right. When you multiply an equation by a number, this is the piece people forget, and forgetting it turns a true equation into a false one.

Pay particular attention to the third note. An equation with a missing letter is not a broken equation, it is a head start: the elimination you were about to perform has already been done for you.

The fourth note is the one that costs marks. Multiplying an equation by three means multiplying all four pieces, the three coefficients and the constant, because you are scaling a whole statement rather than editing part of one. There is a trap slide later built entirely on that slip.

Say the general form aloud once: a lots of x, plus b lots of y, plus c lots of z, makes d. Phrasing it as a sentence about quantities makes it much harder to treat the constant as decoration.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 178 — the general form and the Key Vocabulary

8. The graph of one equation is a flat plane

Picture it

An equation in two variables graphs as a line. Add a third variable and the graph becomes a flat surface, a plane, stretching out for ever in every direction. The activity before the lesson draws one by finding where it cuts each axis.

Figure (svg): Three-dimensional axes with the points four along x, three along y and two along z joined into a shaded triangle, the first-octant piece of the plane three x plus four y plus six z equals twelve.

The three intercepts come from setting two of the letters to zero at a time. Joining them shows the corner of the plane that sits in the first octant; the real plane runs on for ever past all three edges.

The recipe for the three intercepts is worth remembering because it is the fastest way to picture any plane. Set y and z to zero and the equation collapses to three x equals twelve, so the plane crosses the x-axis at 4. Set x and z to zero and four y equals twelve gives 3. Set x and y to zero and six z equals twelve gives 2.

The shaded triangle is only the piece of the plane in the first octant, the corner of space where all three coordinates are positive. The real plane keeps going past all three edges of that triangle, exactly as a line keeps going past the two intercepts you use to draw it.

Keep this picture in mind for one reason above all. When somebody says a system of three equations has no solution, what they mean is that these three flat surfaces have no point in common, and that is much easier to believe once you have seen how a single one of them sits.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 177 — the Investigating Algebra activity, graphing 3x + 4y + 6z = 12

9. An ordered triple is an address in space

Concept

A solution of a system in three variables is not a number and not a pair. It is an ordered triple, three numbers in a fixed order, and it names one point in space the way a pair of numbers names one point in a plane.

Figure (svg): A three-dimensional coordinate frame with x, y and z axes drawn from one origin, and a dashed path that goes four units along x, then three units parallel to y, then two units straight up to a marked point labelled the ordered triple 4, 3, 2.

Two numbers name a point in a plane; three numbers name a point in space. The order of the three is part of the address, so a triple is never just a set of values.

Read the picture as a set of directions. Walk four units along the x-axis, turn and walk three units in the y direction, then climb two units, and you are standing at the point the triple describes. Three coordinates, three separate moves, one place.

The word ordered is doing real work. The triples (4, 3, 2) and (2, 3, 4) are different points, because the first number is always the x coordinate whatever order you happened to discover the values in. A later trap slide shows how much that costs when it is forgotten.

solution of a system of three linear equations — An ordered triple whose coordinates make every one of the equations true. Not most of them, and not the ones you used to find it: all of them.

That definition is the whole of Part 1 in a sentence, and the word every is the part to hold on to. It is what forces the three separate checks that the next two slides are built around.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 178 — the definition of a solution of such a system

10. Three planes, three possible answers

Concept

Since each equation is a plane, a system of three equations is three flat surfaces sitting in the same space, and the question becomes geometric: what do they have in common? There are only three possible answers, and the book draws all three.

Figure (svg): Three panels, each showing three planes drawn as overlapping parallelograms. In the first the three cross at a single marked point; in the second all three contain one highlighted line; in the third two of them are parallel and never meet the same point as the crossing one.

Two planes either miss each other or cross in a line. Adding a third plane is what can cut that line down to a single point, leave it alone, or wipe the common ground out completely.

In the first picture the three planes tilt against each other so that only one point lies on all three, and that point is the one ordered triple that solves the system. This is the usual case and the one nearly every exercise you meet will be.

In the middle picture all three planes share a whole line, like three pages of an open book sharing the spine. Every point of that line satisfies all three equations at once, so there are infinitely many solutions. The same happens if two of the planes, or all three, are really the same plane written differently.

In the third picture there is no point on all three. Two of the planes are parallel and never touch, so no matter how the third one is tilted there is nothing for all three to share, and the system has no solution.

Nothing else is possible, and that is worth knowing in advance. A system of linear equations never has exactly two solutions, or exactly seven. Once you find a second solution you have found infinitely many, because two planes that share two points share the whole line through them.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 178 — the three diagrams of intersecting planes

11. How can three planes sit together?

Prediction

The activity closes by asking exactly this, and it is worth committing to an answer before reading the book's. Think about what two planes can do first, then about what adding a third can change.

Predict first

Three planes are graphed in the same coordinate system. Which list of possibilities is complete?

  • A single point, a whole line or plane of points, or no common point at all
  • A single point, or no common point at all
  • A single point, always
  • A single point, two points, or no common point at all

Correct: A single point, a whole line or plane of points, or no common point at all

Why: Two planes either miss each other entirely or cross in a line. A third plane can cut that line down to one point, contain the whole line, or avoid it. So the common ground is a point, a line or plane, or nothing, which is the three-way split the book draws and the three-way split the algebra will report.

The possibility people leave out is the middle one, because a unique answer feels like what an equation is for. It is worth seeing why exactly two points is impossible: if two different points lie on all three planes, then so does every point on the line joining them, since a plane containing two points contains the whole line through them.

The other tempting answer is the one that says the answer is always a single point. Two parallel planes kill that idea on their own, and you will meet a system whose first two equations are parallel planes within the next few slides.

Hold on to this list, because the algebra mirrors it exactly. Three possible geometries and three possible last lines of working, and by the end of Part 4 you will be able to say which is which without drawing anything.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 177 — Exercise 7 of the activity

12. Exercise 4: check the triple (-1, -2, 5) in all three equations

Worked example

Before solving anything, practise the definition. You are handed a triple and a system and asked whether the one solves the other, which is a question you answer purely by substituting.

\[ \begin{aligned} 4x - y + 3z &= 13 && \text{Equation 1} \\ x + y + z &= 2 && \text{Equation 2} \\ x + 3y - 2z &= -17 && \text{Equation 3} \end{aligned} \]

Work one equation at a time and write the substitution out with brackets round every negative number. Doing negatives in your head is where this kind of question goes wrong, and it is the cheapest error in the lesson to avoid.

Substitute into Equation 1

Why: replace x by -1, y by -2 and z by 5, keeping each negative inside brackets.

\[ 4(-1) - (-2) + 3(5) = -4 + 2 + 15 = 13 \]

That is the first equation satisfied, and notice how the middle term behaved: subtracting a negative two added two. Nothing has been decided yet, though, because a triple only has to be unlucky in one equation to fail.

Substitute into Equation 2

Why: the coefficients are all 1, so this is simply adding the three coordinates.

\[ (-1) + (-2) + 5 = 2 \]

Check Equation 3, the last test

Why: if this one holds as well, the triple satisfies every equation and is genuinely a solution.

\[ (-1) + 3(-2) - 2(5) = -1 - 6 - 10 = -17 \]

All three come out right, so (-1, -2, 5) is a solution of the system. Geometrically you have just confirmed that this one point lies on all three planes at once.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — Exercise 4

13. Exercise 3: a triple that passes two tests and fails the third

Worked example

Now the same question with a different answer, and this is the one that shows why the definition insists on every equation. The triple below looks convincing for a long time.

\[ \begin{aligned} 2x - y + z &= -5 && \text{Equation 1} \\ 5x + 2y - 2z &= 19 && \text{Equation 2} \\ x - 3y + z &= -5 && \text{Equation 3} \end{aligned} \]

Substitute (1, 4, -3) into Equation 1

Why: two ones, minus four, plus minus three.

\[ 2(1) - 4 + (-3) = 2 - 4 - 3 = -5 \]

Substitute into Equation 2

Why: five plus eight plus six, since subtracting twice a negative three adds six.

\[ 5(1) + 2(4) - 2(-3) = 5 + 8 + 6 = 19 \]

Two equations satisfied in a row is exactly when people stop and write down that the triple is a solution. Resist it. So far all you know is that this point lies on two of the three planes, which puts it somewhere on the line those two share.

Check Equation 3, where it breaks

Why: one, minus three fours, plus minus three: the value comes out nowhere near the -5 the equation demands.

\[ (1) - 3(4) + (-3) = 1 - 12 - 3 = -14 \]

Minus fourteen is not minus five, so the triple fails Equation 3 and is not a solution of the system. One failure is enough; there is no partial credit in the definition.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — Exercise 3

14. On two of the planes, but not the third

Picture it

The last example deserves a picture, because the near-miss it describes is not a fluke. There is a whole line of triples that satisfy the first two equations, and almost none of them satisfy the third.

Figure (svg): Two planes drawn as parallelograms crossing in a highlighted line with a point marked on it, and a third plane above them that the dashed line from the point does not reach.

A triple that satisfies two of the equations sits somewhere on the line those two planes share. That is a whole line of near-misses, which is why the third check is the one that decides.

The green point sits on the highlighted line where the first two planes cross, so it passes both of those checks. The dashed arrow shows what the third check is asking: does this point also lie on the third plane? Here it does not, and so the triple is rejected.

Now you can see what a solution really is. It is the one place, if there is one, where that line pierces the third plane, and every other point of the line is a triple that gets two checks right and the third wrong.

The practical moral is short. Checking one equation, or even two, tells you almost nothing, so when a question asks whether a triple is a solution, budget for three substitutions and write all three down.

15. Only one of these triples solves the system

Sorting

Here is a system of my own, chosen so that the near-misses are especially tempting: every triple below has coordinates adding to 6, so every one of them passes the first equation.

\[ \begin{aligned} x + y + z &= 6 && \text{Equation 1} \\ 2x - y + z &= 3 && \text{Equation 2} \\ x + 2y - z &= 2 && \text{Equation 3} \end{aligned} \]

Sort into buckets

Each of these triples satisfies Equation 1. Sort them by whether they satisfy the other two as well.

Solves the whole system
(1, 2, 3)
Passes the first, fails a later one
(3, 2, 1); (2, 1, 3); (1, 3, 2); (2, 2, 2); (0, 2, 4)
yes
The triple with 1 for x, 2 for y and 3 for z gives 6, then 2 minus 2 plus 3 which is 3, then 1 plus 4 minus 3 which is 2. All three equations hold, so this is the solution.
no
Every one of these adds to 6, so Equation 1 accepts them all, but each of them breaks Equation 2. Reordering the coordinates changes the point entirely: 3, 2, 1 gives 6 minus 2 plus 1, which is 5 and not 3.

Four of the six are the same three numbers in a different order, which is the point of the exercise. Since the first equation only adds the coordinates up, it cannot tell those four apart, and the moment a coefficient other than 1 appears the order matters enormously.

If you found yourself testing every triple in every equation, that is exactly the right instinct, but there is a shortcut worth knowing: test the equation with the most varied coefficients first, because that is the one most likely to reject a wrong triple quickly.

This system comes back as the exit ticket at the end of the deck, where you will solve it properly instead of testing candidates, so it is worth remembering that its solution is 1 for x, 2 for y and 3 for z.

16. Eliminate one variable, twice

Section

Part 2

17. Lesson 3.2, run one variable deeper

Concept

Now the real work. The book's own sentence for it is that the elimination method you studied in Lesson 3.2 can be extended, and extended is the honest word: nothing is replaced, one step is added at the front.

Figure (svg): A row of three boxes reading three equations in three unknowns, two equations in two unknowns and one equation in one unknown, joined by forward arrows labelled with the eliminations, and a second row of arrows running backwards labelled with the substitutions.

Lesson 3.2 already took two equations in two unknowns down to one. The only new work here is the step that gets you into that picture in the first place.

Read the top row left to right. You begin with three equations in three unknowns. Combining a pair removes one letter and gives you one equation in two letters, and combining a second pair removes the same letter again and gives you a second equation in those same two letters. Two equations, two unknowns: you are standing exactly where Lesson 3.2 started.

From there the old method takes over. One more elimination brings you down to one equation in one letter, you solve it, and the bottom row of arrows walks the values back out: the first letter gives you the second, and both together give you the one you eliminated at the very beginning.

So the mental model is a ladder with three rungs, and the whole of this lesson is the rung that was not there before. If you can say out loud what happens at each arrow, you can already do the method; the rest of Part 2 is about not losing count while you do it.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the Elimination Method heading

18. Eliminate the same letter from two different pairs

Picture it

This is the single most important picture in the lesson. Everything that goes wrong here goes wrong because a student loses track of which pair of equations produced which new equation, so look at the shape of the process before doing any arithmetic.

Figure (svg): A flow diagram. Three boxes on the left hold Equation 1, Equation 2 and Equation 3. Arrows from the first and third meet in a box for New Equation A with no y, arrows from the second and third meet in a box for New Equation B with no y, and both new boxes feed a box holding two equations in x and z, which feeds a final box for putting y back.

Removing y from one pair gives one equation without y; doing it again with a different pair gives a second. Two equations, the same two letters left: that is a system you already know how to finish.

The two arrows into the upper box come from Equations 1 and 3, and the two into the lower box come from Equations 2 and 3. Equation 3 is used in both combinations, which is normal and fine; the constraint is not about which equations you reuse.

The constraint is the letter. Both boxes say no y, and they have to say the same thing. Remove y from one pair and z from the other and your two new equations will have different letters left in them, so they cannot be combined with each other and you have made the problem worse rather than smaller.

Notice also how much smaller the problem has become by the middle column. Three equations have become two, three letters have become two, and the whole of the right-hand side of the picture is work you have already practised. That is the trade this method makes: one careful bookkeeping step buys you a familiar problem.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the KEY CONCEPT box, Step 1

19. Why it has to be the same letter both times

Intuition

It is worth being able to argue this rather than just obeying it, because the reason also tells you what to do when a system looks awkward.

\[ \begin{aligned} \text{from pair 1 and 3:} \quad & 16x + 11z = -1 \\ \text{from pair 2 and 3:} \quad & -16x - 7z = -11 \end{aligned} \]

Both of those new equations mention x and z and nothing else, so putting them side by side is a genuine two-variable system and one more elimination finishes it. That is the good case, and it happened because y was the letter removed both times.

\[ \begin{aligned} \text{if you removed } y \text{ once:} \quad & 16x + 11z = -1 \\ \text{and } z \text{ the other time:} \quad & 8y - 7z = 29 \end{aligned} \]

Look at the second pair. The first equation talks about x and z, the second about y and z, and together they involve all three letters again. Two equations in three unknowns is not a system you can finish, so you have spent two eliminations and come out no further forward.

The useful consequence is about choosing. Since the letter has to be the same in both combinations, pick it by looking at all three equations at once and asking which letter is easiest to knock out everywhere, rather than by starting with the first pair and seeing what happens.

There is one more freedom worth naming. It does not matter which letter you choose, only that you are consistent, so a system can be solved several different ways and all of them give the same triple. The book makes the same point in its Another Way note beside Example 1.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the Another Way note beside Example 1

20. Before combining anything, choose the letter

Step zero

Here is the system Example 1 works with. Do not solve it yet. Look at the nine coefficients and decide which letter you would remove, and from which two pairs, before writing a single new line.

\[ \begin{aligned} 4x + 2y + 3z &= 1 && \text{Equation 1} \\ 2x - 3y + 5z &= -14 && \text{Equation 2} \\ 6x - y + 4z &= -1 && \text{Equation 3} \end{aligned} \]

Figure (svg): A triangle whose three corners are labelled Equation 1, Equation 2 and Equation 3. Each of the three sides is labelled as a pair, with two of the sides drawn thick to show the two pairs actually used.

Three equations offer exactly three pairs. You use two of them, and the one you skip is free; what is not free is using the same letter in both.

Discussion prompt

Which letter is cheapest to eliminate here, and which two pairs would you use? Say what feature of the coefficients made the choice easy.

The habit being built here is thirty seconds of looking before any writing. A coefficient of 1 or minus 1 anywhere in the system is a signpost, because every other coefficient of that letter can be cancelled by multiplying the easy equation by a whole number.

Compare that with choosing x instead. The x coefficients are 4, 2 and 6, and cancelling the 4 against the 6 needs multipliers of 3 and minus 2, or fractions. It is perfectly possible, and the answer would be identical, but you would be carrying larger numbers through every line.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1 and its Another Way note

21. What you are allowed to do to one equation

Notation

Elimination rests on two permissions, and both are worth stating carefully before you lean on them twice in a row. Take one equation of the system apart first.

Annotate

On: \( 6x - y + 4z = -1 \)

  • You may multiply an equation by any number that is not zero, provided you multiply all four pieces. Doubling this equation gives twelve x, minus two y, plus eight z, equals minus two. The same triples satisfy the new equation, because doubling both sides of a true statement keeps it true.
  • The coefficient of y here is minus 1, not nothing. Written in full the term is minus one lots of y, which is why multiplying the equation by 2 turns it into minus two y, and multiplying by minus 3 turns it into plus three y.
  • Minus one is part of the equation and gets multiplied like everything else. Doubling gives minus two. Leaving the right-hand side alone is the most common way this whole method collapses, because the new equation is then simply false.
  • You may add two equations of the system, left to left and right to right. Whatever triple satisfied both of the originals satisfies the sum, so nothing is lost; what you gain is that the letter whose terms cancelled has gone.

The second note is a quiet source of sign errors. A letter with no visible number in front of it has a coefficient of one, and a letter with only a minus sign in front of it has a coefficient of minus one, so both of them do change when the equation is scaled.

The last note is what makes the whole method legal. Adding two equations cannot invent a solution and cannot lose one, so the system you hold after the addition has exactly the same solutions as the system you started with. That is why the triple you eventually find is guaranteed to fit the originals.

Keep those four permissions in mind as a checklist. Multiply everything, respect an invisible one, carry the constant, and add matching sides. Every line in Example 1 is one of those four things.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the KEY CONCEPT box for a three-variable system

22. Example 1, Step 1: the first pair, Equations 1 and 3

Worked example

Now the arithmetic, with the choice already made: y is the letter to remove, and Equation 3 is the one to scale. Take the pair of Equations 1 and 3 first.

\[ \begin{aligned} 4x + 2y + 3z &= 1 && \text{Equation 1} \\ 6x - y + 4z &= -1 && \text{Equation 3} \end{aligned} \]

Multiply Equation 3 by 2

Why: because Equation 1 carries plus two y, so its partner needs minus two y for the pair to cancel. Every term doubles, the constant included.

\[ 2(6x - y + 4z) = 2(-1) \;\Longrightarrow\; 12x - 2y + 8z = -2 \]

Check the new line against the original before using it. Twelve is twice six, minus two is twice minus one, eight is twice four, and minus two is twice minus one. Four pieces, four doublings, and the equation still says the same thing about the same triples.

Add that to Equation 1

Why: the plus two y and the minus two y cancel, so the sum has no y in it at all.

\[ \begin{aligned} 4x + 2y + 3z &= 1 \\ 12x - 2y + 8z &= -2 \\ \hline 16x \quad\;\; + 11z &= -1 \end{aligned} \]

Label the result

Why: give it a name you can refer to, because a second one is coming and confusing the two is the classic failure here.

\[ 16x + 11z = -1 \qquad \text{New Equation A} \]

Stop and notice what has happened. One equation, two letters, produced from Equations 1 and 3. Half of the reduction is done, and the half that remains is the same move on a different pair.

Check the new equation on the answer

Why: the triple will turn out to be (2, 1, -3), and a new equation must accept it: sixteen twos is 32 and eleven times minus three is minus 33.

\[ 16(2) + 11(-3) = 32 - 33 = -1 \]

It does accept it, which is the guarantee from the last slide in action: combining equations cannot lose the solution, so anything you derive still holds at the answer.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Step 1, first combination

23. Example 1, Step 1: the second pair, Equations 2 and 3

Worked example

Same letter, different pair. This time combine Equations 2 and 3, and keep saying which pair you are using, out loud if it helps, because that sentence is the bookkeeping.

\[ \begin{aligned} 2x - 3y + 5z &= -14 && \text{Equation 2} \\ 6x - y + 4z &= -1 && \text{Equation 3} \end{aligned} \]

Multiply Equation 3 by -3

Why: Equation 2 carries minus three y, so its partner needs plus three y. Multiplying minus one y by minus 3 gives plus three y, which is exactly what is wanted.

\[ -3(6x - y + 4z) = -3(-1) \;\Longrightarrow\; -18x + 3y - 12z = 3 \]

Three sign changes in one line, so read it back carefully. Minus three times six is minus eighteen, minus three times minus one is plus three, minus three times four is minus twelve, and minus three times minus one is plus three.

Add that to Equation 2

Why: the minus three y and the plus three y cancel, leaving x and z only, the same two letters as New Equation A.

\[ \begin{aligned} 2x - 3y + 5z &= -14 \\ -18x + 3y - 12z &= 3 \\ \hline -16x \quad\;\; - 7z &= -11 \end{aligned} \]

Label the result

Why: New Equation B, and note that it mentions x and z exactly as New Equation A does.

\[ -16x - 7z = -11 \qquad \text{New Equation B} \]

The two new equations are now a two-variable system, which means the new part of this lesson is over. Everything from here on is Lesson 3.2 with the letter names changed.

Check this equation on the answer too

Why: minus sixteen twos is minus 32, and minus seven times minus three is plus 21.

\[ -16(2) - 7(-3) = -32 + 21 = -11 \]

Both new equations hold at (2, 1, -3), so no solution has been lost on the way down. Running that check on a derived equation is a cheap way to catch a sign slip before it poisons everything after it.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Step 1, second combination

24. Trap: a different letter from each pair

Trap

The trap

Eliminating y from Equations 1 and 3 gives a good first line:

\[ 16x + 11z = -1 \]

Then, because Equations 1 and 2 happen to look convenient, the same student eliminates z from that pair instead:

\[ 8y - 7z = 29 \]

Both lines are true. Neither multiplication was wrong, and no sign was dropped.

The fix

Read the two letters left in each line. The first has x and z, the second has y and z, so between them all three letters are still in play and two equations cannot settle three unknowns.

The fix is to go back and remove y from a second pair, so that both new equations mention the same two letters.

\[ -16x - 7z = -11 \]

Now x and z appear in both, and one more elimination finishes the job. Say which letter you are removing before each combination, and write it at the top of your page.

25. Which multiplier clears which pair?

Matching

Choosing the multiplier is a small piece of arithmetic that decides whether the next line is easy or ugly. For Example 1's system, match each combination to the number that makes a letter cancel.

\[ \begin{aligned} 4x + 2y + 3z &= 1 && \text{Eq 1} \\ 2x - 3y + 5z &= -14 && \text{Eq 2} \\ 6x - y + 4z &= -1 && \text{Eq 3} \end{aligned} \]

Match the pairs

  • a. Remove y using Equations 1 and 3
  • b. Remove y using Equations 2 and 3
  • c. Remove x using Equations 1 and 2
  • d. Remove x using Equations 2 and 3
  • w. multiply Equation 3 by 2
  • x. multiply Equation 3 by −3
  • y. multiply Equation 2 by −2
  • z. multiply Equation 2 by −3

Why: To cancel a term you need its exact opposite. Plus two y needs minus two y, so minus one y is multiplied by 2. Minus three y needs plus three y, so minus one y is multiplied by minus 3. For x, four x needs minus four x, so two x is multiplied by minus 2; and six x needs minus six x, so two x is multiplied by minus 3. In each case the multiplier is what turns one coefficient into the negative of the other.

There is a rule hiding in those four answers. To cancel a coefficient of p against a coefficient of q, multiply the second equation by minus p divided by q, and that division is only pleasant when q divides p. This is why a coefficient of 1 or minus 1 makes such a good target.

The last two matches are worth working through even though Example 1 does not use them, because they show the choice really is free. Removing x instead of y would give a different pair of new equations and the same final triple.

One practical note on signs. Every multiplier here is negative except the first, which tells you something true in general: unless two coefficients already have opposite signs, the multiplier that cancels them must be negative.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Step 1 and the Another Way note

26. Trap: multiplying only part of an equation

Trap

The trap

Doubling Equation 3, which is six x minus y plus four z equals minus one, a student writes:

\[ 12x - y + 8z = -2 \]

The minus y was left alone, because a lone minus sign does not look like a number to be multiplied.

Adding that to Equation 1 gives sixteen x plus one y plus eleven z equals minus one, and the y is still there.

The fix

An invisible coefficient is still a coefficient. Minus y means minus one lots of y, so doubling it gives minus two y, and only then does it cancel the plus two y in Equation 1.

\[ 12x - 2y + 8z = -2 \]

The same slip visits the constant. Writing twelve x minus two y plus eight z equals minus one leaves a false equation, because the left-hand side was doubled and the right-hand side was not.

Guard against both by counting pieces. This equation has four pieces, so after multiplying, four numbers should have changed. If fewer than four changed, something was skipped.

27. Solve the smaller system and work back

Section

Part 3

28. The reduced system is a picture you can draw

Concept

The two new equations mention only x and z, so they are ordinary lines and you can plot them. Doing that once is worth it, because it shows that the reduction really has brought you back to familiar ground.

Figure (svg): Two straight lines drawn on axes with x across and z up, one for sixteen x plus eleven z equals minus one and one for minus sixteen x minus seven z equals minus eleven, crossing at the point where x is two and z is minus three.

Once y is gone the two new equations are ordinary lines in two letters, and their crossing point is the x and z of the answer. This is the picture Lesson 3.2 taught you to read.

The axes are labelled x across and z up, which feels odd only until you remember that the names of the letters are irrelevant to the method. Two lines in two letters cross at one point, and that point gives the x and the z of the answer.

Read the crossing off the picture: x is 2 and z is minus 3. The algebra on the next slide will produce the same two numbers, and knowing in advance what they should be makes any slip obvious immediately.

The picture also previews Part 4 rather neatly. If the two new equations had turned out to be parallel lines they would never cross and the system would have no solution, and if they had turned out to be the same line they would cross everywhere. Those are the only other things two lines can do.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Step 2

29. Example 1, Steps 2 and 3: finish the system

Worked example

Two equations, two letters, and a lucky arrangement: the x terms are already exact opposites, so the next elimination needs no multiplier at all.

\[ \begin{aligned} 16x + 11z &= -1 && \text{New Equation A} \\ -16x - 7z &= -11 && \text{New Equation B} \end{aligned} \]

Add New Equations A and B

Why: sixteen x and minus sixteen x cancel, so only z survives on the left.

\[ \begin{aligned} 16x + 11z &= -1 \\ -16x - 7z &= -11 \\ \hline 4z &= -12 \end{aligned} \]

Solve for z

Why: divide both sides by 4.

\[ z = -3 \]

One letter down. Now put that value back into either new equation; both must give the same x, so the choice is only about which arithmetic you prefer. New Equation A has the friendlier numbers.

Substitute z = -3 into New Equation A

Why: eleven times minus three is minus thirty-three, so sixteen x must make up the difference.

\[ 16x + 11(-3) = -1 \;\Longrightarrow\; 16x - 33 = -1 \;\Longrightarrow\; 16x = 32 \]

Solve for x

Why: divide by 16.

\[ x = 2 \]

Two letters known, one to go, and here is the important instruction: go back to an original equation for it. The new equations have no y in them at all, so they cannot possibly tell you what y is.

Substitute into original Equation 3 and solve for y

Why: Equation 3 has the simplest y term in the system, so it is the cheapest place to finish.

\[ 6(2) - y + 4(-3) = -1 \;\Longrightarrow\; 12 - y - 12 = -1 \;\Longrightarrow\; -y = -1 \;\Longrightarrow\; y = 1 \]

Check the triple in Equation 1, the one not used at the end

Why: four twos plus two ones plus three times minus three: eight plus two minus nine.

\[ 4(2) + 2(1) + 3(-3) = 8 + 2 - 9 = 1 \]

That is the 1 Equation 1 demands, so the triple (2, 1, -3) survives a check in an equation that played no part in finding y. The next slide checks the other two as well, because one check is not the standard this lesson is setting.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Steps 2 and 3

30. Example 1: the triple checked in all three equations

Worked example

The book closes Example 1 with the instruction to check this solution in each of the original equations, and that instruction is worth taking literally. Here is what all three checks look like written out.

\[ (x,\, y,\, z) = (2,\, 1,\, -3) \]

Use the original equations, never the new ones you derived. A derived equation inherits any mistake you made while deriving it, so checking there could confirm a wrong answer with total confidence.

Substitute into Equation 1

Why: eight, plus two, minus nine.

\[ 4(2) + 2(1) + 3(-3) = 8 + 2 - 9 = 1 \;\checkmark \]

Substitute into Equation 2

Why: four, minus three, minus fifteen.

\[ 2(2) - 3(1) + 5(-3) = 4 - 3 - 15 = -14 \;\checkmark \]

Two down, and the third is the one most likely to catch an error, because Equation 3 is the equation that was used at the end to find y. If y went wrong, this is where it shows.

Check Equation 3 as well

Why: twelve, minus one, minus twelve.

\[ 6(2) - 1 + 4(-3) = 12 - 1 - 12 = -1 \;\checkmark \]

Three equations, three ticks, so (2, 1, -3) is the solution and you can write it down without hedging. Three substitutions cost about a minute, and they are the difference between an answer and a guess.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the closing line of Example 1

31. The order you find the letters is not the order you write them

Concept

Trace the route the last two slides took and notice how little it resembles the shape of the answer. That mismatch is worth one slide on its own, because it is where a correct piece of algebra most often turns into a wrong final line.

Figure (svg): Four boxes joined by arrows: adding the two new equations gives z equals minus three, a new equation then gives x equals two, an original equation then gives y equals one, and the last box holds the ordered triple two, one, minus three.

You find the letters in whatever order the algebra allows, and here that order is z, then x, then y. The triple is still written x first, which is the single most-reported slip in this lesson.

The algebra handed you z first, because z was the letter left standing after the second elimination. Then x came out of a new equation, and only then did y appear, from an original equation. So the order of discovery was z, then x, then y.

The triple, though, is written x first, y second, z third, always. There is no rule that says the coordinates come out in order, and in this example not one of them did.

So build in a deliberate pause at the end. Write the three values with their letters beside them, as x equals 2, y equals 1, z equals minus 3, and only then assemble the triple. It takes five seconds and it removes the error entirely.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 181 — the Avoid Errors note beside Example 4

32. Trap: writing the triple in the order you solved it

Trap

The trap

Example 4 finds z first, then x, then y, and a student writes the answers down in the order they appeared:

\[ (12,\, 18,\, 30) \]

Every number in that line is correct. The three values found were 12, 18 and 30.

The book flags this exact triple with a cross, because as an ordered triple it names a completely different point.

The fix

Attach a letter to each value the moment you find it, then sort by letter at the end. Here x is 18, y is 30 and z is 12.

\[ (18,\, 30,\, 12) \]

The check catches it too, which is one more reason to run it. Eighteen plus thirty plus twelve is sixty as required, whereas twelve plus eighteen plus thirty is also sixty, so this first equation cannot tell them apart.

The money equation can. A thousand times 18 plus two hundred times 30 plus five hundred times 12 is thirty thousand, while the reversed triple gives twenty-one thousand six hundred. Check in every equation and the wrong order is exposed.

33. Complete the back-substitution

Fill the middle

Example 1's last step, with the middle taken out. You already know that x is 2 and z is minus 3, and Equation 3 is where y comes from. Fill in the two missing pieces.

\[ 6x - y + 4z = -1 \qquad \text{Equation 3} \]

Fill in the blanks

6(2) - y + 4(-3) = -1 \;\Longrightarrow\; -y = -1 \;\Longrightarrow\; y = 1

Why: Twelve minus twelve is zero, so the equation collapses to minus y equals minus one, and multiplying both sides by minus one gives y equal to 1. The near-miss here is stopping at minus y equals minus one and writing y equal to minus one, which is one sign short of finished.

The step that vanishes in your head is the last one. Minus y equals minus one is not an answer, because the question asked for y and not for minus y, so one more move is needed before the line is finished.

Notice how conveniently the numbers behaved: twelve and minus twelve cancelled exactly. That is not luck so much as choice, since Equation 3 was picked precisely because its y term was the simplest, and a simple y term usually means simple arithmetic around it.

If you had used Equation 1 instead you would get four twos plus two y plus three times minus three equals one, so two y equals two and y equals 1. Same answer by a slightly longer route, which is the reassurance that the choice of equation was free.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Step 3

34. Put the whole method in order

Ranking

Six actions, and the order of the middle four is the part that matters. Arrange them as you would actually carry them out on a fresh system.

Put in order

Order these steps for solving a system of three equations in three variables.

  1. Look at all nine coefficients and choose the letter to eliminate
  2. Combine one pair of equations to remove that letter
  3. Combine a different pair to remove the same letter
  4. Solve the two-variable system the two new equations make
  5. Substitute the two known values into an original equation for the third letter
  6. Check the ordered triple in all three original equations

Why: Choosing the letter comes first because it decides both of the combinations that follow. The two eliminations must remove the same letter, so they belong together, and only when both are done is there a two-variable system to solve. The third letter comes from an original equation because the new ones no longer contain it, and the check comes last because it needs the whole triple.

The step people are tempted to move is the choosing. It feels like a preliminary rather than a step, so it gets skipped and the first pair gets combined on instinct, and then the second combination has to work around whatever the first one happened to leave.

The step people are tempted to drop altogether is the last one. It is the only step that can tell you whether the other five went right, so of the six it is the worst possible one to skip when you are short of time.

Notice too that the substitution step says an original equation. Substituting into a new equation is not wrong exactly, but a new equation has no third letter in it, so it simply cannot answer the question being asked.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the KEY CONCEPT box, Steps 1 to 3

35. Guided Practice 1: choosing y again, with two different multipliers

Worked example

A fresh system from the book's checkpoint. Look first: the y coefficients are 1, minus 2 and 4, so Equation 1 is the easy one this time and it will be scaled twice.

\[ \begin{aligned} 3x + y - 2z &= 10 && \text{Equation 1} \\ 6x - 2y + z &= -2 && \text{Equation 2} \\ x + 4y + 3z &= 7 && \text{Equation 3} \end{aligned} \]

Multiply Equation 1 by 2 and add Equation 2

Why: Equation 2 carries minus two y, so its partner needs plus two y, and doubling the plus one y in Equation 1 supplies it.

\[ \begin{aligned} 6x + 2y - 4z &= 20 \\ 6x - 2y + z &= -2 \\ \hline 12x \quad\;\; - 3z &= 18 \end{aligned} \]

Divide that by 3 to keep the numbers small

Why: every term is a multiple of 3, and smaller numbers mean fewer chances to slip later.

\[ 4x - z = 6 \qquad \text{New Equation A} \]

Tidying a new equation is always allowed and usually worth it. Dividing an equation by 3 is just multiplying it by one third, which is one of the permissions from Part 2.

Multiply Equation 1 by -4 and add Equation 3

Why: Equation 3 carries plus four y, so its partner needs minus four y. Watch all four signs in the scaled line.

\[ \begin{aligned} -12x - 4y + 8z &= -40 \\ x + 4y + 3z &= 7 \\ \hline -11x \quad\;\; + 11z &= -33 \end{aligned} \]

Divide by 11

Why: again for tidiness, and now the second equation is as simple as the first.

\[ -x + z = -3 \qquad \text{New Equation B} \]

Add the two new equations and solve

Why: minus z and plus z cancel, leaving three x equal to three.

\[ (4x - z) + (-x + z) = 6 + (-3) \;\Longrightarrow\; 3x = 3 \;\Longrightarrow\; x = 1 \]

Back-substitute for z, then for y

Why: New Equation A gives z, and original Equation 1 gives y.

\[ 4(1) - z = 6 \;\Longrightarrow\; z = -2, \qquad 3(1) + y - 2(-2) = 10 \;\Longrightarrow\; y = 3 \]

Check (1, 3, -2) in all three equations

Why: three plus three plus four is 10; six minus six minus two is minus 2; one plus twelve minus six is 7.

\[ \begin{aligned} 3 + 3 + 4 &= 10 \;\checkmark \\ 6 - 6 - 2 &= -2 \;\checkmark \\ 1 + 12 - 6 &= 7 \;\checkmark \end{aligned} \]

All three hold, so the solution is (1, 3, -2). Notice that Equation 1 was the one scaled in both combinations here, where Example 1 used Equation 3 for both: the workhorse is whichever equation has the friendliest coefficient of your chosen letter.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 180 — Guided Practice 1 for Examples 1, 2 and 3

36. Exercise 9: when a letter can be cancelled by plain addition

Worked example

Sometimes the system hands you an elimination for free. Look at the y coefficients here before reading on: 1, 2 and minus 1.

\[ \begin{aligned} 3x + y + z &= 14 && \text{Equation 1} \\ -x + 2y - 3z &= -9 && \text{Equation 2} \\ 5x - y + 5z &= 30 && \text{Equation 3} \end{aligned} \]

Add Equations 1 and 3 with no multiplier at all

Why: plus one y and minus one y are already opposites, so they cancel on their own.

\[ \begin{aligned} 3x + y + z &= 14 \\ 5x - y + 5z &= 30 \\ \hline 8x \quad\;\; + 6z &= 44 \end{aligned} \]

Divide by 2

Why: all three numbers are even, so this costs nothing and makes the next lines shorter.

\[ 4x + 3z = 22 \qquad \text{New Equation A} \]

Multiply Equation 1 by -2 and add Equation 2

Why: Equation 2 carries plus two y, so the one y of Equation 1 is doubled and made negative.

\[ \begin{aligned} -6x - 2y - 2z &= -28 \\ -x + 2y - 3z &= -9 \\ \hline -7x \quad\;\; - 5z &= -37 \end{aligned} \]

Two equations in x and z again, but this time neither pair of coefficients cancels by itself, so the last elimination needs multipliers on both lines.

Eliminate z from the new pair

Why: multiply New Equation A by 5 and the second new equation by 3, so that the z terms become plus fifteen z and minus fifteen z.

\[ \begin{aligned} 20x + 15z &= 110 \\ -21x - 15z &= -111 \\ \hline -x &= -1 \end{aligned} \]

Solve, then back-substitute twice

Why: x is 1; New Equation A then gives z, and original Equation 1 gives y.

\[ x = 1, \qquad 4 + 3z = 22 \;\Longrightarrow\; z = 6, \qquad 3 + y + 6 = 14 \;\Longrightarrow\; y = 5 \]

Check (1, 5, 6) in all three equations

Why: three plus five plus six is 14; minus one plus ten minus eighteen is minus 9; five minus five plus thirty is 30.

\[ \begin{aligned} 3 + 5 + 6 &= 14 \;\checkmark \\ -1 + 10 - 18 &= -9 \;\checkmark \\ 5 - 5 + 30 &= 30 \;\checkmark \end{aligned} \]

The lesson from this one is to scan for a pair of coefficients that are already opposites. Finding one saves a whole multiplication and, more importantly, saves the sign errors that multiplication invites.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — Exercise 9

37. Exercise 12: an answer with a zero in it

Worked example

One more for practice, and this one ends somewhere students often distrust. Equation 1 has the only y coefficient of minus 1, so it is the natural equation to scale.

\[ \begin{aligned} 4x - y + 2z &= -18 && \text{Equation 1} \\ -2x + 2y + z &= 11 && \text{Equation 2} \\ 3x + 3y - 4z &= 44 && \text{Equation 3} \end{aligned} \]

Multiply Equation 1 by 2 and add Equation 2

Why: Equation 2 has plus two y, so the minus one y is doubled to minus two y.

\[ \begin{aligned} 8x - 2y + 4z &= -36 \\ -2x + 2y + z &= 11 \\ \hline 6x \quad\;\; + 5z &= -25 \end{aligned} \]

Multiply Equation 1 by 3 and add Equation 3

Why: Equation 3 has plus three y, so the minus one y is tripled to minus three y.

\[ \begin{aligned} 12x - 3y + 6z &= -54 \\ 3x + 3y - 4z &= 44 \\ \hline 15x \quad\;\; + 2z &= -10 \end{aligned} \]

Both new equations carry x and z, which is the signal that Step 1 was done properly. Now eliminate z, which needs a multiplier on each line because 5 and 2 share no factor.

Multiply the first new equation by 2 and the second by 5

Why: the z terms become plus ten z and plus ten z, so this time you subtract rather than add.

\[ \begin{aligned} 12x + 10z &= -50 \\ 75x + 10z &= -50 \\ \hline 63x &= 0 \end{aligned} \]

Solve for x

Why: sixty-three times something is zero, and only zero does that.

\[ x = 0 \]

A coordinate of zero is a perfectly ordinary answer and not a sign that something has gone wrong. It means the point sits on the plane where x is zero, which is one wall of the coordinate frame, and points do sit on walls.

Back-substitute for z, then for y

Why: the first new equation gives z, and original Equation 1 gives y.

\[ 6(0) + 5z = -25 \;\Longrightarrow\; z = -5, \qquad -y + 2(-5) = -18 \;\Longrightarrow\; y = 8 \]

Check (0, 8, -5) in all three equations

Why: zero minus eight minus ten is minus 18; zero plus sixteen minus five is 11; zero plus twenty-four plus twenty is 44.

\[ \begin{aligned} 0 - 8 - 10 &= -18 \;\checkmark \\ 0 + 16 - 5 &= 11 \;\checkmark \\ 0 + 24 + 20 &= 44 \;\checkmark \end{aligned} \]

All three hold, so the triple is (0, 8, -5). Write the zero in: dropping it and writing a pair instead is the same error as writing the coordinates in the wrong order, and it is just as easy to make.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — Exercise 12

38. Two variables against three, side by side

Comparison

Before the degenerate cases, line up the old method against the new one and fill in the gaps. Seeing how few rows actually differ is the point of the table.

Comparison matrix

questiontwo variables (Lesson 3.2)three variables (this lesson)
equations you start with23
what one equation graphs asa linea plane
pairs you combine in the first step12
what is left after the first stepone equation in one lettertwo equations in two letters
substitutions on the way back12
checks a finished answer needs23
sign of no solution0 = a nonzero number0 = a nonzero number

The last row is the one to notice, because it does not change at all. A false equation means no solution whether you are working with two letters or three, and in Part 4 you will see the same line appear in a three-variable system.

The rows that do change all change in the same direction: one more equation, one more combination, one more substitution, one more check. Nothing about the method is new, there is simply one more of everything.

The graph row is the one genuine novelty. A line in a plane becomes a plane in space, and that is why the possible outcomes have a richer geometry even though the algebra reports them the same way.

39. When there is no answer, or far too many

Section

Part 4

40. A false equation means no solution

Concept

Everything so far has ended with a triple, but the method is honest enough to tell you when no triple exists. It does so by producing a line that is simply not true.

\[ 0 = -5 \]

There is nothing mysterious about that line. Every letter cancelled at once, which leaves a statement about numbers alone, and if that statement is false then no triple can possibly have satisfied the equations you combined to get it.

Remember why: combining equations never loses a solution. So if the original system had a solution, that solution would have to satisfy everything you derive from it, including the line zero equals minus five. Since nothing satisfies that, the original system had no solution either.

Geometrically you are in the third of the three pictures. Two of the planes are parallel, or the three of them are arranged so that no point is common to all, and the algebra has just detected it.

Practically, the moment you see a false equation you stop. There is no triple to find, and the answer to the question is the sentence the system has no solution.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the KEY CONCEPT box, the false equation case

41. Example 2: a system with no solution

Worked example

This one gives itself away early, and the book chooses it to show that you may not have to do much work before the verdict arrives.

\[ \begin{aligned} x + y + z &= 3 && \text{Equation 1} \\ 4x + 4y + 4z &= 7 && \text{Equation 2} \\ 3x - y + 2z &= 5 && \text{Equation 3} \end{aligned} \]

Look at the first two equations before choosing a letter. The left-hand side of Equation 2 is exactly four times the left-hand side of Equation 1, which is a warning worth acting on, so start with that pair.

Multiply Equation 1 by -4

Why: so that its left-hand side becomes the exact negative of Equation 2's, and every term cancels rather than just one.

\[ -4(x + y + z) = -4(3) \;\Longrightarrow\; -4x - 4y - 4z = -12 \]

Add that to Equation 2

Why: all three letters cancel at once, which is more than you asked for and is the whole story.

\[ \begin{aligned} -4x - 4y - 4z &= -12 \\ 4x + 4y + 4z &= 7 \\ \hline 0 &= -5 \end{aligned} \]

Zero is not minus five, so the two equations you combined cannot both be true for any triple. Equation 3 never even came into it, and it does not need to: two contradictory facts are enough to rule out an answer.

Check what the two equations are really saying

Why: Equation 1 says the three coordinates add to 3, so four times that sum must be 12; Equation 2 insists the same four times that sum is 7.

\[ 4(x + y + z) = 4(3) = 12 \quad \text{but Equation 2 says } 7 \]

Twelve and seven cannot both be right, so the system has no solution. That is the whole verdict, and it is worth phrasing the contradiction in words like this, because it shows the algebra was reporting something real rather than misbehaving.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 180 — Example 2

42. What a false equation looks like when you draw it

Concept

Two equations whose left-hand sides are proportional but whose constants disagree are worth picturing once, in two letters, where you can actually see them.

Figure (svg): Two parallel straight lines on axes, one for two x plus two y equals eight and one for two x plus two y equals fourteen, with a dashed vertical segment between them showing a constant gap of three.

The same left-hand side cannot equal two different numbers at once. Drawn, that is two parallel lines with a gap that never closes, and in algebra it is the line that reads zero equals something.

Both lines have the same left-hand side, two x plus two y, and different right-hand sides. Same slope, different intercept, so they run alongside each other for ever with a gap that never closes. There is no point on both, which is precisely no solution.

Now subtract one equation from the other and you get zero equals six, which is the false line from the last slide wearing different numbers. The picture and the algebra are the same fact told two ways.

In three letters the same thing happens with planes rather than lines. In Example 2, Equation 1 and Equation 2 describe parallel planes, one of them where the coordinates add to 3 and the other where four times that sum is 7, which would need the coordinates to add to one and three quarters instead.

So the useful habit is a quick scan before you start. If one equation's left-hand side is a multiple of another's, check whether the constants agree by the same multiple, because the answer to that decides the whole question before any elimination is done.

43. Guided Practice 2: spotting it before eliminating

Worked example

The book's second checkpoint hides the same situation slightly better. Scan the three left-hand sides before you choose a letter to eliminate.

\[ \begin{aligned} x + y - z &= 2 && \text{Equation 1} \\ 2x + 2y - 2z &= 6 && \text{Equation 2} \\ 5x + y - 3z &= 8 && \text{Equation 3} \end{aligned} \]

Equation 2 is Equation 1 with every coefficient doubled. If the system were consistent the constant would be doubled too, so it would have to read 4, and instead it reads 6. That is the whole problem, and the elimination below is only confirmation.

Multiply Equation 1 by -2

Why: to make its left-hand side the exact negative of Equation 2's.

\[ -2x - 2y + 2z = -4 \]

Add Equation 2

Why: every letter cancels and a false statement is left.

\[ \begin{aligned} -2x - 2y + 2z &= -4 \\ 2x + 2y - 2z &= 6 \\ \hline 0 &= 2 \end{aligned} \]

Zero is not two, so there is no solution, and Equation 3 is irrelevant to that verdict. Two of the three facts already contradict each other, and a third fact cannot rescue a contradiction.

Check by reading the two equations in words

Why: the first says a certain combination equals 2, and the second says twice that same combination equals 6, which would need the combination to equal 3.

\[ x + y - z = 2 \quad \text{and} \quad x + y - z = 3 \]

One quantity cannot be both 2 and 3, so the system has no solution. Writing the contradiction out in this form is a good way to be sure you have not simply made an arithmetic slip somewhere.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 180 — Guided Practice 2

44. An identity means infinitely many

Concept

The other unusual ending is the opposite of the last one. Instead of a statement that is never true, you get a statement that is always true.

\[ 0 = 0 \]

Nothing is wrong when that appears. It means the second of your two new equations told you nothing the first had not already said, so the two lines you were about to cross are really the same line, and the system leaves a whole family of triples satisfying it.

Geometrically this is the middle picture from Part 1: the three planes share a whole line, or two of them are the same plane. Every point of that shared line is a solution, and a line has infinitely many points.

The extra work in this case is describing the family, and that is what most of the marks are for. Saying infinitely many is only half an answer; the question is which triples, and the answer comes out as a pattern with one letter still free to choose.

So the two odd endings are easy to tell apart. A false line means stop and say there is no solution; a true but empty line means keep going and describe the family.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the KEY CONCEPT box, the identity case

45. Example 3: infinitely many, and how to describe them

Worked example

This system looks ordinary and the elimination starts ordinarily, which is why it is such a useful example: nothing warns you in advance.

\[ \begin{aligned} x + y + z &= 4 && \text{Equation 1} \\ x + y - z &= 4 && \text{Equation 2} \\ 3x + 3y + z &= 12 && \text{Equation 3} \end{aligned} \]

Add Equations 1 and 2 to remove z

Why: plus z and minus z are already opposites, so no multiplier is needed.

\[ \begin{aligned} x + y + z &= 4 \\ x + y - z &= 4 \\ \hline 2x + 2y \;\; &= 8 \end{aligned} \]

Add Equations 2 and 3 to remove z again

Why: minus z and plus z cancel, and this is the second pair, so the same letter has gone twice.

\[ \begin{aligned} x + y - z &= 4 \\ 3x + 3y + z &= 12 \\ \hline 4x + 4y \;\; &= 16 \end{aligned} \]

Combine the two new equations

Why: multiply the first by -2 so that both x terms and both y terms cancel.

\[ \begin{aligned} -4x - 4y &= -16 \\ 4x + 4y &= 16 \\ \hline 0 &= 0 \end{aligned} \]

An identity, so the system has infinitely many solutions. The two new equations were the same statement written twice, which you can see directly: dividing four x plus four y equals sixteen by 2 gives two x plus two y equals eight.

Describe the family

Why: divide the first new equation by 2 to get x plus y equals 4, make y the subject, then put that back into an original equation to pin z down.

\[ x + y = 4 \;\Longrightarrow\; y = -x + 4, \qquad x + (-x + 4) + z = 4 \;\Longrightarrow\; z = 0 \]

Figure (svg): A straight line drawn on axes with x across and y up, with four marked points on it at minus one and five, zero and four, two and two, and four and zero, labelled as the triples whose third coordinate is zero.

The answer is not a point but a whole line of points, all of them lying flat at height zero. Picking any x hands you the matching y, and every one of those triples satisfies all three equations.

So z is forced to be zero while x stays free, and y follows from x. Every triple of the form x, then 4 minus x, then 0 solves the system, which is the line drawn above sitting flat at height zero.

Check the description on two members of the family

Why: take x equal to 2, giving (2, 2, 0), and x equal to minus 1, giving (-1, 5, 0), and test both in Equation 3.

\[ \begin{aligned} 3(2) + 3(2) + 0 &= 12 \;\checkmark \\ 3(-1) + 3(5) + 0 &= 12 \;\checkmark \end{aligned} \]

Both work, and so would any other choice of x, which is what infinitely many means. Checking two different members is the right check here: a single triple would only show that one solution exists, not that the whole family does.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 180 — Example 3

46. Guided Practice 3: the same ending, different family

Worked example

The book's third checkpoint has the same shape as Example 3, and working it through is the fastest way to see that the description step is a routine rather than a trick.

\[ \begin{aligned} x + y + z &= 3 && \text{Equation 1} \\ x + y - z &= 3 && \text{Equation 2} \\ 2x + 2y + z &= 6 && \text{Equation 3} \end{aligned} \]

Add Equations 1 and 2

Why: the z terms cancel with no multiplier.

\[ 2x + 2y = 6 \qquad \text{New Equation A} \]

Add Equations 2 and 3

Why: minus z and plus z cancel, so z is gone from this pair as well.

\[ 3x + 3y = 9 \qquad \text{New Equation B} \]

Before combining them, look at the two new equations. Dividing the first by 2 gives x plus y equals 3, and dividing the second by 3 gives x plus y equals 3 as well. They are the same equation, which already tells you what is coming.

Combine the new equations to confirm

Why: multiply New Equation A by 3 and New Equation B by -2, then add.

\[ \begin{aligned} 6x + 6y &= 18 \\ -6x - 6y &= -18 \\ \hline 0 &= 0 \end{aligned} \]

Describe the family

Why: from x plus y equals 3 make y the subject, then substitute into Equation 1 to find z.

\[ y = -x + 3, \qquad x + (-x + 3) + z = 3 \;\Longrightarrow\; z = 0 \]

So every triple of the form x, then 3 minus x, then 0 is a solution. The pattern is the same as Example 3's with a 3 in place of the 4, which is worth noticing: the structure of the family comes from the shape of the equations, not from the particular numbers.

Check two members in all three equations

Why: try (0, 3, 0) and (5, -2, 0).

\[ \begin{aligned} 0 + 3 + 0 = 3, \quad 0 + 3 - 0 &= 3, \quad 0 + 6 + 0 = 6 \;\checkmark \\ 5 - 2 + 0 = 3, \quad 5 - 2 - 0 &= 3, \quad 10 - 4 + 0 = 6 \;\checkmark \end{aligned} \]

Both triples pass all three equations, including one with a negative coordinate, which shows the family really is unrestricted in x rather than only working for small positive values.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 180 — Guided Practice 3

47. Exercise 13: infinitely many, with the third letter moving too

Worked example

Both families so far had z pinned at zero, which can leave the impression that the free letter is always x and the trapped one always z. This exercise breaks that impression.

\[ \begin{aligned} 5x + y - z &= 6 && \text{Equation 1} \\ x + y + z &= 2 && \text{Equation 2} \\ 3x + y &= 4 && \text{Equation 3} \end{aligned} \]

Equation 3 has no z in it at all, which is a free elimination: it is already an equation in x and y only, so you need just one more of those.

Add Equations 1 and 2 to remove z

Why: minus z and plus z cancel with no multiplier needed.

\[ \begin{aligned} 5x + y - z &= 6 \\ x + y + z &= 2 \\ \hline 6x + 2y \;\; &= 8 \end{aligned} \]

Divide by 2 and compare with Equation 3

Why: the tidied new equation is three x plus y equals 4, which is Equation 3 word for word.

\[ 3x + y = 4 \qquad \text{same as Equation 3} \]

Combine them to see the identity

Why: subtracting one from the other leaves nothing on either side.

\[ (3x + y) - (3x + y) = 4 - 4 \;\Longrightarrow\; 0 = 0 \]

Infinitely many solutions again, so the job is to describe them. This time take x as the free letter and work y and then z out of it.

Describe the family

Why: make y the subject of three x plus y equals 4, then use Equation 2 for z.

\[ y = -3x + 4, \qquad z = 2 - x - y = 2 - x - (-3x + 4) = 2x - 2 \]

So the family is x, then 4 minus three x, then two x minus 2, and here both y and z move as x moves. The shape of the answer is still a line in space, but this one climbs rather than lying flat.

Check two members in all three equations

Why: x equal to 1 gives (1, 1, 0), and x equal to 2 gives (2, -2, 2).

\[ \begin{aligned} 5 + 1 - 0 = 6, \quad 1 + 1 + 0 &= 2, \quad 3 + 1 = 4 \;\checkmark \\ 10 - 2 - 2 = 6, \quad 2 - 2 + 2 &= 2, \quad 6 - 2 = 4 \;\checkmark \end{aligned} \]

Both hold, so the description is right. Notice the second member has a negative y and a positive z, so nothing about the family is restricted to tidy small numbers.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — Exercise 13

48. Slide the coefficient until the crossing disappears

Tweak it

Suppose the first equation of a two-variable system could be tuned. The curve below plots the gap between the two lines at each value of x, so the lines meet exactly where the curve crosses zero.

Parameter explorer

Slide the coefficient and watch where the curve crosses zero, which is where the two lines meet. Find the one setting at which the curve never crosses zero at all, and say what that setting means for the system.

\[ \text{gap}(x) = 3 - (a + 2)x \qquad \text{with } a = {a} \]

  • a — from -4 to 3: coefficient of x in the first line

Read the curve as a distance rather than as a graph of anything. It tells you how far apart the two lines are at each x, so the crossing point of the curve is the crossing point of the lines, and a curve that stays above zero is two lines that never touch.

The one setting that matters is minus 2, where the curve flattens into a horizontal line 3 above the axis. At that setting the two lines have the same slope and a permanent gap of 3, so there is no solution, and this is exactly what the algebra reports as zero equals 3.

The tilt of the curve carries information too. Near minus 2 the curve is almost flat, so the crossing point runs far off to the side, and the solution becomes enormous. A system whose lines are nearly parallel has an answer, but a very large and very sensitive one.

Back in three letters, this is the last elimination of the lesson. After both new equations are formed you are always looking at two lines, and the slider is the whole of Part 4 in one picture: they cross once, or never, or lie on top of each other.

49. Read the last line and give the verdict

Discrimination

Six things you might have written at some point in a solution. Sort each by what it tells you about the system, and say to yourself which of the three pictures from Part 1 you would be in.

Sort into buckets

Sort each line you might reach by what it says about the number of solutions.

Exactly one solution
4z = −12; x = 2 and z = −3
No solution
0 = −5; 0x + 0y = 7
Infinitely many solutions
0 = 0; both new equations tidy to 3x + y = 4
one
A line that gives a definite value for a letter, or a pair of definite values, means the elimination has pinned the answer down, so there is one triple and back-substitution will finish it.
none
Both of these say that zero equals something that is not zero, once written plainly and once with the letters still visible but carrying coefficients of zero. No triple can make either true, so the system has no solution.
many
An identity, and two new equations that turn out to be the same equation, are the same situation: the second combination added nothing, so a whole family of triples survives and the family needs describing.

The item worth a second look is the one that still shows the letters while their coefficients are zero. It is the same statement as the plain false line, just not yet tidied up, and recognising it early saves you from trying to divide by zero to find a value for x.

Notice how much these verdicts have in common with Lesson 3.2. The same two odd endings appeared there, for the same reasons, and the only new thing is that the solution family can now be a line in space rather than a line in a plane.

One habit to take from this slide: when a line surprises you, tidy it before you interpret it. Collect the letters, collect the numbers, and then decide which of the three verdicts you are looking at.

OpenStax Algebra and Trigonometry 2e, §11.2 Systems of Linear Equations: Three Variables §11.2 — the same three outcomes, with the same algebraic signatures

50. Find the error: a sign lost while solving for z

Error analysis

The substitution method works here just as well as elimination, and the book's error-analysis exercise is about the first move of it. The system below is mine, chosen so the numbers come out cleanly; the error is the book's.

\[ \begin{aligned} 2x + y - 2z &= -4 && \text{Equation 1} \\ 3x + 2y + z &= 11 && \text{Equation 2} \\ x - y + z &= 3 && \text{Equation 3} \end{aligned} \]

Annotate

On: \( z = 11 + 3x + 2y \)

  • Equation 2 has the friendliest z term in the system, a plain plus z, so making z the subject of it is a sensible plan. The plan is right; only the execution slips.
  • To free z you must take three x and two y away from both sides, not add them. Equation 2 says three x plus two y plus z equals 11, so z is 11 minus three x minus two y. Both signs were flipped the wrong way.
  • Substituting the wrong expression into Equation 1 gives minus four x minus three y equals 18, while the correct expression gives eight x plus five y equals 18. Same constant, completely different left-hand side, which is what makes the slip so hard to spot afterwards.
  • Write z equal to 11 minus three x minus two y, then substitute that into Equations 1 and 3 to get eight x plus five y equals 18 and two x plus three y equals 8.

\[ z = 11 - 3x - 2y \]

Finish the corrected version to see that it behaves. Multiplying two x plus three y equals 8 by 4 gives eight x plus twelve y equals 32, and subtracting the other equation leaves seven y equal to 14, so y is 2, then x is 1 and z is 11 minus 3 minus 4, which is 4.

Check the triple (1, 2, 4) in all three equations: two plus two minus eight is minus 4, three plus four plus four is 11, and one minus two plus four is 3. Every one holds, so the corrected route lands where it should.

The general guard is to expand your substitution back out. If z equals 11 plus three x plus two y, then three x plus two y plus z would be eleven plus six x plus four y, which is not what Equation 2 says. Twenty seconds of checking kills the error at birth.

It is also worth noticing that substitution and elimination are doing the same work by different bookkeeping. Substituting for z removes z from the other two equations, which is exactly what the two eliminations of Part 2 achieved.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — Exercise 22

51. Can two equations ever settle the question?

Counterexample

The method says to use two pairs and then check all three, so it is fair to ask whether the third equation is always needed. Argue it both ways before reading on.

Discussion prompt

Give one system where two of the three equations settle the answer completely, and explain why a third equation cannot change that verdict. Then say why two equations can never be enough to produce a single triple.

So the asymmetry is the interesting part. No solution can be decided by two equations, because a contradiction is permanent, but one solution never can, because two equations always leave a free choice behind.

This is also the cleanest reason for the rule that three unknowns need three facts. Each equation can remove at most one degree of freedom, so with two equations at least one choice always remains open.

It cuts the other way too, and Part 5 will use it. If a real-world problem gives you only two genuine facts about three unknown quantities, no algebra will produce a single answer, and the honest reply is to say which family of answers fits and what extra information would settle it.

52. Modelling with three unknowns

Section

Part 5

53. Three unknowns need three separate facts

Concept

Word problems in this lesson all have the same skeleton. Something is unknown in three ways, the problem states three facts, and each fact becomes an equation.

  1. Name the three unknowns with letters first, and write down what each letter counts or measures, including its units.
  2. Turn each fact into one equation, in the order the problem states them, so you can check them off.
  3. Count the equations before solving. Three unknowns and only two facts means no single answer exists.
  4. Solve by elimination or by substitution, whichever the equations invite; a fact of the form one letter equals an expression is an invitation to substitute.
  5. Check the triple against every stated fact, not just against the equations you wrote, because a mis-translated fact makes a wrong equation that your answer will satisfy perfectly.

The third item is the one that separates this from Lesson 3.2. Two facts about three unknowns leaves a whole family of possibilities, and a question written that way is asking for a description rather than a triple.

The last item is the one that separates a careful solution from a lucky one. Checking in your own equations only confirms that your arithmetic matched your translation; checking against the sentences of the problem is what confirms that the translation itself was right.

Watch for facts that do not look like equations. A sentence such as there are as many radio advertisements as television and newspaper ones combined is a perfectly good third equation, and it is often the one that makes substitution the easier method.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 181 — Example 4, Steps 1 and 2

54. Example 4: thirty thousand dollars of advertising

Worked example

A marketing department has 30,000 dollars to spend. A television advertisement costs 1000 dollars, a radio one 200 dollars and a newspaper one 500 dollars. They want 60 advertisements a month, and as many radio ones as television and newspaper ones combined.

Name the unknowns

Why: let x be the number of television advertisements, y the number of radio ones and z the number of newspaper ones. Three quantities, three letters, all of them counts.

\[ x = \text{TV},\quad y = \text{radio},\quad z = \text{newspaper} \]

Turn the three facts into three equations

Why: the total count, the total money, and the rule relating radio to the other two.

\[ \begin{aligned} x + y + z &= 60 && \text{Equation 1} \\ 1000x + 200y + 500z &= 30\,000 && \text{Equation 2} \\ y &= x + z && \text{Equation 3} \end{aligned} \]

Equation 3 already has one letter alone on one side, which is the signal to substitute rather than eliminate. Replacing y by x plus z in the other two equations removes y from both in one move.

Substitute x plus z for y in Equation 1

Why: the count equation becomes an equation in x and z only.

\[ x + (x + z) + z = 60 \;\Longrightarrow\; 2x + 2z = 60 \]

Substitute x plus z for y in Equation 2

Why: the money equation becomes the second equation in x and z.

\[ 1000x + 200(x + z) + 500z = 30\,000 \;\Longrightarrow\; 1200x + 700z = 30\,000 \]

Eliminate x from the new pair

Why: multiply two x plus two z equals 60 by -600 so that the x terms cancel against 1200x.

\[ \begin{aligned} -1200x - 1200z &= -36\,000 \\ 1200x + 700z &= 30\,000 \\ \hline -500z &= -6000 \end{aligned} \]

Solve for z, then work back

Why: z is 12; the count equation gives x, and Equation 3 gives y.

\[ z = 12, \qquad 2x + 24 = 60 \;\Longrightarrow\; x = 18, \qquad y = 18 + 12 = 30 \]

Figure (svg): Two horizontal bars drawn to scale. The upper bar shows sixty advertisements split into eighteen television, thirty radio and twelve newspaper. The lower bar shows thirty thousand dollars split into eighteen thousand, six thousand and six thousand for the same three kinds.

Each bar is one of the three equations drawn to scale. The top bar shows the counts adding to sixty and the radio rule; the bottom bar shows the same three quantities weighted by their prices.

Before writing the triple, attach the letters: x is 18 television advertisements, y is 30 radio ones, z is 12 newspaper ones. The values came out in the order 12, 18, 30, and the triple is not written that way.

Check all three facts, not just the equations

Why: eighteen plus thirty plus twelve is sixty advertisements; the money comes to exactly thirty thousand dollars; and thirty radio advertisements is eighteen plus twelve.

\[ \begin{aligned} 18 + 30 + 12 &= 60 \;\checkmark \\ 1000(18) + 200(30) + 500(12) &= 18\,000 + 6000 + 6000 = 30\,000 \;\checkmark \\ 30 &= 18 + 12 \;\checkmark \end{aligned} \]

All three facts hold, so the department should run 18 television, 30 radio and 12 newspaper advertisements. Each check is a sentence from the problem rather than a line from your working, which is what makes it a real check.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 181 — Example 4

55. Guided Practice 4: the same problem on a smaller budget

Worked example

The checkpoint changes one number and nothing else: the budget drops to 25,000 dollars. This is where the effort of setting the problem up properly pays off, because almost none of it has to be redone.

\[ \begin{aligned} x + y + z &= 60 \\ 1000x + 200y + 500z &= 25\,000 \\ y &= x + z \end{aligned} \]

Substitute x plus z for y again

Why: the count equation is unchanged, and only the constant of the money equation differs.

\[ 2x + 2z = 60, \qquad 1200x + 700z = 25\,000 \]

Predict the direction before solving. Less money and the same sixty advertisements means the mix has to shift towards the cheaper kinds, so expect fewer television advertisements and more newspaper ones than last time.

Eliminate x

Why: multiply two x plus two z equals 60 by -600 exactly as before.

\[ \begin{aligned} -1200x - 1200z &= -36\,000 \\ 1200x + 700z &= 25\,000 \\ \hline -500z &= -11\,000 \end{aligned} \]

Solve for z, then work back

Why: z is 22; the count equation gives x, and the radio rule gives y.

\[ z = 22, \qquad 2x + 44 = 60 \;\Longrightarrow\; x = 8, \qquad y = 8 + 22 = 30 \]

The prediction held: television advertisements fell from 18 to 8 and newspaper ones rose from 12 to 22, while radio stayed at 30. Radio could not move, because the count equation and the radio rule together force it to be half of sixty whatever the budget is.

Check the triple against all three facts

Why: eight plus thirty plus twenty-two is sixty; the money is exactly twenty-five thousand; and thirty is eight plus twenty-two.

\[ \begin{aligned} 8 + 30 + 22 &= 60 \;\checkmark \\ 1000(8) + 200(30) + 500(22) &= 8000 + 6000 + 11\,000 = 25\,000 \;\checkmark \\ 30 &= 8 + 22 \;\checkmark \end{aligned} \]

So the answer is 8 television, 30 radio and 22 newspaper advertisements. That observation about radio being fixed at 30 is worth more than the answer itself, since it is a fact about the structure of the problem rather than about one budget.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 181 — Guided Practice 4

56. Which sentence became which equation?

Matching

Translation is where most modelling marks are lost, so practise it in isolation, with the algebra already done. Match each sentence of Example 4 to the equation it produced.

Match the pairs

  • a. The department wants 60 advertisements per month
  • b. The budget for advertising is 30,000 dollars
  • c. As many radio ads as television and newspaper combined
  • d. A television ad costs 1000 dollars each
  • w. x + y + z = 60
  • x. 1000x + 200y + 500z = 30,000
  • y. y = x + z
  • z. the term 1000x inside the money equation

Why: Counting advertisements gives a sum of the three counts. The budget gives a sum of three amounts of money, each one a price times a count. The comparison sentence gives one count equal to the sum of the other two. And a single price is not an equation on its own: it is one term inside the money equation, which is why the prices and the budget appear together in one line.

The last pairing is the one that matters most. A price on its own says nothing about how many advertisements there are, so it cannot be an equation; it becomes a coefficient in the equation about money.

Keeping the units straight is what keeps the two sum equations apart. One of them adds counts and equals a count, the other adds amounts of money and equals an amount of money, and a term like 1000x is dollars per advertisement times advertisements, which is dollars.

Comparison sentences are the ones to slow down on. As many radio as television and newspaper combined becomes y equals x plus z, and the common wrong version puts the sum on the wrong side, giving x equals y plus z, which says something quite different.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 181 — Example 4, Step 1 verbal models

57. Three points decide a parabola

Real world

Three-variable systems turn up in places that do not look like word problems at all. Here is one you will meet again in Chapter 4: finding the curve through three given points.

Figure (svg): A parabola opening upward drawn through three marked points at one and four, two and three, and three and six, with its equation labelled.

Three points, three unknown coefficients, and exactly one curve of this shape through all of them. Each point turns into one equation, which is why this is a three-variable system in disguise.

A quadratic function has the form y equal to a x squared plus b x plus c, so it hides three unknown numbers. Each point you are given is one fact about them, which means three points give exactly three equations, and this is why three points is the magic number for a parabola.

Discussion prompt

The points (1, 4), (2, 3) and (3, 6) lie on such a curve. Write the three equations that each point gives, then solve for a, b and c and check the curve passes through all three points.

The shortcut worth noticing is that subtracting equations, rather than adding them, was the natural move here. Each equation carried plus c, so the difference of any two has no c in it, and subtraction is simply adding minus one times one of them.

Look at what the picture confirms. The curve dips below all three points between the first two and climbs steeply after, so the middle point being lower than the outer two is exactly what a positive value of a should produce.

This is the same idea behind a great deal of real fitting work. Unknown coefficients in a formula, known observations, one equation per observation: the system gets larger but never stops being a system of the kind you have just solved.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — the modelling exercises following Example 4

58. When the answer is not a whole number

Edge cases

Every worked answer so far has been tidy, and modelling problems are not obliged to be. This one is about tickets, and it shows what to do when the algebra is right and the answer is still unusable.

\[ \begin{aligned} a + s + c &= 100 \\ 15a + 8s + 5c &= 1000 \\ s &= 2c \end{aligned} \]

Discussion prompt

A hundred tickets were sold for 1000 dollars in total, at 15 dollars for adults, 8 for students and 5 for children, and twice as many students as children. Solve the system and say what the result means for the original question.

Say the answer out loud as a sentence and the problem is obvious: about twenty point eight three children, forty-one point six seven students and thirty-seven point five adults. Counts of people are whole numbers, so this triple, though it satisfies every equation, cannot describe a real sale.

That is a genuinely useful outcome rather than a failure. The system has exactly one solution and it is not a whole-number one, which tells you that the three facts you were given cannot all be true together in the real situation.

The habit to take away is to check an answer against the kind of quantity it counts, not only against the equations. Negative lengths, fractional people and speeds faster than light all satisfy algebra perfectly happily.

If a question like this appears, the right reply names the contradiction: the three facts are inconsistent with whole numbers of tickets, and one of them, most likely the total takings, has been misreported.

59. Estimate before you solve

Estimation

A quick estimate is worth thirty seconds before any three-variable model, because it turns an arithmetic slip into something you notice. Try it on this one.

Predict first

Three friends' ages add to 60. The eldest is twice the youngest, and the middle one is 4 years older than the youngest. Which is closest to the youngest friend's age?

  • about 14
  • about 20
  • about 10
  • about 25

Correct: about 14

Why: If the three ages were equal they would each be 20, but the eldest is twice the youngest, so the youngest must sit well below 20 to make room. Calling the youngest y, the three ages are y, y plus 4 and two y, which add to four y plus 4. Setting that equal to 60 gives four y equal to 56, so the youngest is 14, the middle one 18 and the eldest 28, and those do add to 60.

The estimate came from one observation: the average age is 20, and since one friend is twice another, the ages cannot all sit at the average. The youngest has to be below it and the eldest above it, which rules out two of the four options immediately.

Then the algebra confirms it. Three unknown ages, three facts, and the last two facts each express one age in terms of the youngest, so substitution turns the whole thing into a single equation in one letter.

Check the triple against the sentences rather than the equations: 14, 18 and 28 add to 60, twenty-eight is twice fourteen, and eighteen is four more than fourteen. All three facts hold.

Worth asking what happens if one fact is dropped. Without the middle friend's rule you would have two facts about three ages, so a whole family of answers would fit, and the honest response would be to describe that family.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 182 — the modelling exercises with three unknown quantities

60. Putting it together

Section

Part 6

61. The method, and the three things it can tell you

Pattern

Everything in this lesson fits on one page. The procedure is fixed, and the only branching happens at the very end, when you read the last line you wrote.

  1. Number the equations 1, 2 and 3, and choose the letter to eliminate by looking at all nine coefficients at once. A coefficient of 1 or minus 1 is the signpost.
  2. Combine one pair to remove that letter, multiplying whole equations as needed, and label the result.
  3. Combine a different pair to remove the same letter, and label that result too.
  4. Solve the two-variable system the two new equations make, exactly as in Lesson 3.2.
  5. Substitute back into an original equation for the letter you removed, then write the triple in the order x, y, z.
  6. Check the triple in all three original equations, and for a word problem against all three stated facts.

Figure (svg): A flow diagram. One box at the top holds the act of combining the two new equations, and three arrows lead down to three boxes reading zero equals a nonzero number, zero equals zero, and a value for one letter, each with its verdict underneath.

Three possible last lines, three verdicts. Deciding which one you are looking at costs nothing, and it stops you hunting for a triple that does not exist.

The diagram covers the branch. A false line means no solution and you stop; an identity means infinitely many and you describe the family with one letter left free; anything else means a single triple and back-substitution finishes it.

Notice that every one of the three branches ends with something to write down. None of them is an error state, and none of them means you should start again from the top.

If you carry one sentence out of this lesson, make it this one: the same letter, from two different pairs. Every other difficulty here is arithmetic you already had.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — the KEY CONCEPT box in full

62. Which first move actually removes a letter?

Elimination

Four possible opening moves on the pair of Equations 1 and 2 below. Only one of them makes a letter disappear; work out what each of the others would leave behind before eliminating it.

\[ \begin{aligned} 4x + 2y + 3z &= 1 && \text{Equation 1} \\ 2x - 3y + 5z &= -14 && \text{Equation 2} \end{aligned} \]

Eliminate the wrong options

Which opening move removes a letter from this pair?

  • A. Multiply Equation 2 by −2, then add Equation 1
  • B. Add the two equations exactly as they stand
  • C. Multiply Equation 1 by 2, then add Equation 2
  • D. Multiply Equation 2 by 2, then add Equation 1

Survives elimination: A

Why: Doubling Equation 2 with a negative sign turns its two x into minus four x, which is the exact opposite of the four x in Equation 1, so the x terms cancel and the sum is eight y minus seven z equals 29. A multiplier only helps when it makes one coefficient the negative of the other; the same number without the minus sign makes the terms add instead.

The move that survives is the one whose multiplier is negative, and that is the pattern worth taking away: unless two coefficients already have opposite signs, the multiplier that cancels them must be negative.

Test the survivor on the answer to be sure. The system's solution is (2, 1, -3), and eight ones minus seven times minus three is eight plus twenty-one, which is 29, so the new equation does hold at the answer.

The two near-misses that differ only by a sign are the ones to study. Doubling Equation 2 and doubling it negatively look almost identical on the page and do opposite things to the x terms, which is why writing the scaled equation out in full before adding is worth the line it costs.

McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables §3.4, p. 179 — Example 1, Step 1

63. Check: what the two eliminations must have in common

Check

This is the bookkeeping question the whole of Part 2 was about. Read each option and ask what letters would be left in the two new equations.

Check your understanding

You remove z from Equations 1 and 2, and then remove y from Equations 2 and 3. What have you got?

  • A. Two equations that cannot be combined, because between them all three letters survive (correct)
  • B. A two-variable system ready to solve
  • C. A system with no solution
  • D. The solution, after one more addition

Answer: A

Why: The first new equation mentions x and y, and the second mentions x and z, so together they still involve all three letters. Two equations in three unknowns cannot pin anything down, so the two eliminations have to remove the same letter for the reduction to work.

Why B tempts people
A two-variable system needs both equations to mention the same two letters. These mention different pairs, so it is not one.
Why C tempts people
Nothing here says the system is inconsistent. The work has simply not reduced the problem, which is a bookkeeping failure rather than a verdict about solutions.
Why D tempts people
Adding two equations that carry different letters removes nothing, so there is no way to finish from here without going back and eliminating the same letter twice.

The wrong answer that tempts most people is the one claiming a two-variable system, because two equations with two letters each does sound like one. What matters is that it is the same two letters in both, not merely two letters in each.

If this happens to you in an exam, the repair is cheap. Keep whichever new equation you like, then go back to the original three and remove that same letter from a different pair, and you are back on track with one extra combination.

The habit that prevents it is written at the top of your page: the name of the letter you are removing. Look at it before every combination and the error becomes almost impossible.

Notice also that reusing Equation 2 in both pairs is not the problem. Using one equation twice is normal and Example 1 does it with Equation 3; the letter is the thing that must be held fixed.

64. Check: reading the last line of the working

Check

Three possible endings, one of them in front of you. Decide which of the three pictures from Part 1 this system is in before choosing.

Check your understanding

Combining the two new equations of a three-variable system gives 0 = 7. What does the system have?

  • A. No solution (correct)
  • B. Infinitely many solutions
  • C. Exactly one solution, with a zero coordinate
  • D. Not enough information to say

Answer: A

Why: Zero is not seven, so the line is false. Since combining equations never loses a solution, any triple satisfying the original system would have to satisfy this line too, and nothing does, so the system has no solution and the three planes have no common point.

Why B tempts people
Infinitely many solutions is signalled by an identity such as zero equals zero, a statement that is always true. A false statement says the opposite.
Why C tempts people
A coordinate of zero comes from a line such as sixty-three x equals zero, which has a perfectly good solution. Here no letter is left at all, and the statement that remains is untrue.
Why D tempts people
The line is enough on its own. Two of the original equations already contradict each other, and no third equation can undo a contradiction.

The distractor worth thinking about is the one about a zero coordinate, because a zero does turn up in genuine answers and Exercise 12 produced one. The difference is whether a letter is still present in the line: sixty-three x equals zero has a solution, and zero equals seven has none.

Mixing up the two odd endings is the other common slip. A false line is never true and means no solution; an identity is always true and means infinitely many, which is as far apart as two verdicts can be.

What you should actually write down here is a sentence, not a triple. The system has no solution, because two of its equations contradict each other, and there is nothing left to substitute back.

As a sanity check, try to invent a triple that satisfies both equations you combined. The impossibility of doing so is the same fact the false line reported, and feeling that impossibility is what makes the verdict believable.

65. Check: where the third letter has to come from

Check

You have reached the end of the reduction and hold two of the three values. One question remains: which equation do you go back to?

Check your understanding

You have eliminated y twice, solved the new system to get x = 2 and z = −3, and now want y. Which equation do you substitute into?

  • A. Any one of the three original equations (correct)
  • B. Either of the two new equations
  • C. The new equation with the smallest coefficients
  • D. The original equation you never used in a combination

Answer: A

Why: The new equations had y eliminated from them, so neither of them mentions y and neither can tell you what it is. Any original equation will do, and all three give the same value, so choose whichever has the simplest y term.

Why B tempts people
Both new equations were built precisely to have no y in them, so substituting x and z into either one gives a true statement with no y to solve for.
Why C tempts people
Small coefficients are pleasant but irrelevant here: the new equations contain no y at all, whatever size their numbers are.
Why D tempts people
That equation works, but so do the other two. There is no rule reserving an unused equation for this step; all three originals contain y and all three give the same answer.

The temptation to reach for a new equation is understandable, because those are the lines you have most recently written and they look like progress. Look at what is actually in them: only x and z, by construction.

Since any original equation works, choose on the arithmetic. In Example 1 the third equation has a y term of minus one, so it costs one subtraction and a sign flip, while the first has plus two y and costs a division as well.

There is a bonus in the fact that all three originals agree. Substituting into a second original equation is then a free check of the value of y, and it is the cheapest check available at that moment.

One last caution about this step: substitute both known values, not one. Putting x in and forgetting z leaves two letters in the equation and no way to finish, and it is a surprisingly easy slip when the numbers are untidy.

66. Say the whole method in your own words

Explain it to yourself

Put the deck down for a minute and write the method out as if for somebody who has just finished Lesson 3.2 and knows nothing about three variables. Write your version first, then compare.

Discussion prompt

In four or five sentences, explain how solving three equations in three unknowns reuses the two-variable method, why the two eliminations must remove the same letter, and why the answer gets checked three times.

If your explanation used the word new more often than the word same, write it again. Nothing in this lesson is new except the extra elimination; the word doing the work is same, as in the same letter, both times.

A good test of whether you have it: could you teach the warm-up pair from the start of this deck and Example 1 as a single lesson, with the second one as a sequel? If yes, you have the idea the deck was built to deliver.

Try explaining the third check without using the word plane as well, in pure algebra. Two equations leave one free choice, so they are satisfied by a whole family of triples, and the third equation is what picks one member of that family out.

67. Exit ticket

Exit ticket

One system, worked from the beginning, with every habit the lesson asked for. Choose your letter out loud, name the pairs, and check three times.

\[ \begin{aligned} x + y + z &= 6 && \text{Equation 1} \\ 2x - y + z &= 3 && \text{Equation 2} \\ x + 2y - z &= 2 && \text{Equation 3} \end{aligned} \]

Discussion prompt

Solve this system. Say which letter you eliminated and from which two pairs, give the triple in the right order, and check it in all three original equations.

The place this system invites a slip is the second pair. Equation 1 and Equation 3 both carry a positive y coefficient, so adding them doubles the y term instead of clearing it, and the multiplier that does the job has to be negative.

Notice how the values arrived: z first, then x, then y. Writing them in that order would give (3, 1, 2), which is a different point entirely, so attach the letters before assembling the triple.

The triple here is the one from the sorting slide in Part 1, so you have now seen the same answer arrived at two ways, by testing candidates and by solving. The second way is the one that works when nobody hands you candidates.

Finally, say the three checks out loud as sentences about the original equations. That is the habit this deck has been pushing hardest, and it is the one that turns a probably into a definitely.

68. What you can do now

Recap

One new step, bolted onto a method you already had. Three equations become two by removing the same letter from two different pairs, and from there it is Lesson 3.2 to the finish.

what you seewhat it meanswhat to write
a value for one letterthe planes meet at a pointback-substitute, then the triple in x, y, z order
0 = a nonzero numberno point lies on all three planesthe system has no solution
0 = 0the planes share a whole linedescribe the family, leaving one letter free
two new equations in different lettersthe eliminations removed different lettersgo back and remove the same letter twice

\[ \begin{aligned} \text{choose a letter} &\;\longrightarrow\; \text{two pairs} \;\longrightarrow\; \text{two new equations} \\ &\;\longrightarrow\; \text{solve} \;\longrightarrow\; \text{back-substitute} \;\longrightarrow\; \text{check three times} \end{aligned} \]

The first row of the table is the ordinary case and the one to be fluent in. The last row is not a verdict about the system at all but a note about your own working, and it is the single most common way a correct method produces nothing.

Two habits are worth carrying out of this lesson more than any formula. Write the name of the letter you are eliminating at the top of the page, and check every finished triple in all three original equations rather than the one that was easiest.

Lesson 3.5 takes the next step and drops the letters entirely, keeping only a grid of the coefficients. Every multiply-and-add you did here becomes a single operation on a row of that grid, and the reason it is worth learning is that the bookkeeping this lesson made you do by hand becomes automatic.

OpenStax Algebra and Trigonometry 2e, §11.2 Systems of Linear Equations: Three Variables §11.2 — the same method, with the three outcomes named the same way

Sources

  1. McDougal Littell Algebra 2 (Texas Edition), Ch. 3 Linear Systems and Matrices — Lesson 3.4 Solve Systems of Linear Equations in Three Variables — Larson, Boswell, Kanold & Stiff, McDougal Littell / Houghton Mifflin, 2007, pp. 178-181
  2. OpenStax Algebra and Trigonometry 2e, §11.2 Systems of Linear Equations: Three Variables
  3. OpenStax College Algebra 2e, §7.2 Systems of Linear Equations: Three Variables

Want this taught 1-on-1? Alexander tutors Algebra 2 — $55/session, free consultation.

Book on Wyzant · Text (657) 465-8108