ROBOTECA FREE SAMPLE
Dashboard

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

LESSON · Numbers, Units & Estimation

Units, SI prefixes and dimensional analysis

Turn 1 45 min LESSON

ALearning Material

A measurement is a number together with a unit. Without its unit, a quantity becomes a bare number that can be combined with any other number, correctly or not, and the arithmetic gives no warning. With the unit carried through the working, the unit becomes a second, independent check on every line, and it catches errors that the numbers alone cannot reveal.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍The SI system builds every mechanical and electrical quantity from seven base units. Five of them appear throughout this course: the metre for length, the kilogram for mass, the second for time, the ampere for current and the kelvin for temperature. Every other unit is a combination of base units. A newton is , a joule is a newton metre, a watt is a joule per second, and a volt is a watt per ampere. These combinations follow from the physics, and they are what makes units usable as a check.

The check rests on dimensional homogeneity: an equation that describes the physical world must have the same dimensions on both sides. A length cannot equal a time, any more than 3 can equal "blue". The same rule applies within a sum: metres can be added to metres, never to seconds. Multiplication and division have no such restriction, because they build new dimensions, which is how a velocity in m/s arises from a length divided by a time.

A dimension and a unit are different things, and the name of the method refers to the first. Length is a dimension, and the metre, the inch and the mile are units of length. Two quantities can share a dimension while carrying different units, which is the situation a conversion resolves. Two quantities with different dimensions cannot be made comparable by any conversion. A formula therefore holds in any consistent set of units while a number does not: the physics constrains the dimensions, and the choice of unit is a convention added on top. For the same reason, a dimensional check can be run before any conversion, by asking only whether both sides describe the same kind of quantity.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Dimensional analysis is the procedure that applies this check. Each quantity is written as a number times a unit, the units are carried through the algebra as symbols, and they cancel in the same way as variables. If the units of the answer are wrong, the working is wrong, and the error is found before the number reaches a machine.

Unit conversion uses the same idea. A conversion factor is a fraction equal to 1, because its top and bottom are the same physical quantity written two ways. Since , both of these fractions equal 1:

Multiplying by 1 leaves a quantity unchanged, so either fraction can be applied at any point. The choice between them follows from the cancellation: the right fraction places the unit to be removed on the opposite side of the line from where it currently sits, so that it cancels. Converting 90 kilometres per hour into metres per second:

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍The kilometres cancel against kilometres and the hours against hours, leaving metres over seconds, which is the unit required. The cancellation decides whether to multiply or divide by 3.6, so no factor has to be remembered. A remembered factor can be applied in the wrong direction, while a cancellation cannot.

A pump lifts water at 45 litres per minute to a height of 12 m, and the hydraulic power is required in watts. Power is force times velocity, and the force here is the weight of the water being lifted. The calculation is simplest in base units from the start. Water has a density of about , and , so the mass flow is

Each unwanted unit has been cancelled by a factor equal to 1. The power, using , is:

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍The resulting unit, , is the watt: a joule is and a watt is a joule per second. Had the height been divided instead of multiplied, the answer would have carried , which is not a unit of power, and the error would have been visible without any knowledge of pumps.

SI prefixes are powers of ten with names, as defined in the previous lesson. The prefixes in daily use run from pico at to giga at , in steps of a thousand. Two details cause errors. The first is that the kilogram is the base unit of mass, not the gram, so a prefix on a mass appears doubled: a milligram is kg. The second is capitalisation, which carries meaning: M is mega and m is milli, a factor of apart, and K is not a prefix, since k is kilo and K is the kelvin.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Formulas & method.

  • Base units used constantly here. metre (m) length, kilogram (kg) mass, second (s) time, ampere (A) current, kelvin (K) temperature.
  • Derived units, as combinations. , , , , , and .
  • Homogeneity. Both sides of an equation, and every term in a sum, must carry the same units. Multiplication and division build new units and are unrestricted.
  • Conversion. A conversion factor is a fraction equal to 1. Choose the orientation that puts the unwanted unit on the opposite side of the line so it cancels. Chain as many as needed.
  • ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Procedure. Write each quantity as a number times a unit and carry the units through the algebra as symbols. Cancel them, then check the unit of the result before trusting the number.
  • Prefixes. p , n , u , m , k , M , G . Case matters: M is mega, m is milli.
  • The kilogram trap. The base unit of mass is the kilogram, so a milligram is kg, not kg.
  • Angles. The radian is dimensionless, being a length over a length, which is why angular speed in rad/s has units of and why comes out in watts with no stray factor.

Where this shows up.

  • ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Robotics. Torque in N m, angular velocity in rad/s and their product in watts. A Jacobian mixes metres and radians in one matrix, so its entries do not share a unit and the mixture has to be tracked deliberately.
  • Automation. Instrument ranges, flow totals and energy tallies are the most common places a factor of 60 or 3600 goes in upside down, because minutes and hours meet seconds there.
  • Electronics. The RC time constant comes out in seconds only because an ohm times a farad is a second, a dimensional check that can be run on any half-remembered circuit formula.
  • Digital design. Timing budgets mix nanoseconds, picoseconds and megahertz, and a period is the reciprocal of a frequency, so an error of in a prefix is an error of in a setup-time margin.

Why it exists. A number alone does not say what was measured, so it can be combined with any other number without objection. A unit makes the quantity self-describing, and it makes the algebra checkable: the units follow their own arithmetic, which must come out right independently of the numbers. Unit checking costs almost nothing, and it works even on a formula that is only half remembered.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Mental model. A unit behaves like a factor in an ordinary algebraic expression: metres cancel metres in the same way that cancels . A conversion factor is a form of the number 1, so inserting one is always valid, and the only decision is which way up to write it. The units form a second calculation that runs alongside the numbers.

Common misunderstandings.

  • "Units are bookkeeping you tidy up at the end." Units are a running check. Carried through the working, they catch an inverted factor at the line where it occurs. Added at the end, they only label an answer that has already been produced.
  • "A milligram is a thousandth of a gram, so prefixes on mass behave like the others." The base unit is the kilogram, so a milligram is kg. Mass kept in kilograms from the start avoids the problem.
  • "Radians are a unit like metres, so they must appear in the answer." The radian is a ratio of two lengths and is dimensionless, so gives watts directly. A speed in rpm must still be converted to rad/s, because rpm carries a factor of and a minute.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Connections. The prefixes are the engineering notation of the previous lesson with names attached, and the conversion factors are ratios, as defined in the ratios lesson, chosen to equal 1. Dimensional analysis is used as a check in every later topic: to verify a formula in the mechanics topic, to confirm an RC time constant in the electricity topic, and to keep the terms of a control equation consistent. The next lesson, on precision, adds the second part of a measurement: how many of its digits are meaningful.

BImmediate Active Recall

QUERY

What is dimensional homogeneity, and what does it forbid?

REVEAL
ANSWER

Both sides of a physical equation, and every term in a sum, must carry the same units. It forbids adding metres to seconds, and equating a length to a time. Multiplication and division are unrestricted, since they build new units, which is how m divided by s becomes a velocity.

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

Why is a conversion factor always legal to insert, and how do you decide which way up to write it?

REVEAL
ANSWER

Its top and bottom are the same physical quantity written two ways, so the fraction equals 1, and multiplying by 1 changes nothing. Choose the orientation that puts the unwanted unit on the opposite side of the line from where it sits, so it cancels. The cancellation decides the direction, so nothing has to be remembered.

Did you recall it?
QUERY

Convert 90 km/h to m/s using cancellation, not a remembered factor.

REVEAL
ANSWER

. Kilometres cancel kilometres, hours cancel hours, and metres over seconds is left.

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

A power calculation yields . Is that right?

REVEAL
ANSWER

Yes. A joule is and a watt is a joule per second, so is the watt. This is the check in use: had a height been divided instead of multiplied, the result would carry , which is not a unit of power, and the error appears without any knowledge of the physics.

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

What is the kilogram trap?

REVEAL
ANSWER

The base unit of mass is the kilogram, not the gram, so a prefix appears to double up: a milligram is kg, not kg. Working in kilograms from the start avoids the problem.

Did you recall it?
QUERY

Why does give watts with no extra factor, and where does a factor still hide?

REVEAL
ANSWER

The ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍radian is dimensionless, being a length over a length, so rad/s has units of and N m per second is a watt. The factor hides in rpm, which is not rad/s: converting needs radians per revolution and 60 seconds per minute.

Did you recall it?

CConceptual Questions

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

PROMPT

An engineer half remembers that the time constant of a resistor and capacitor is either or ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. Show how units settle it without looking anything up, and say what this tells you about using formulas you are unsure of.

REVEAL MODEL ANSWER
MODEL ANSWER

The units of each combination settle it. An ohm is a volt per ampere, and a farad is a coulomb per volt, where a coulomb is an ampere second. So gives , and both the volts and the amperes cancel, leaving seconds. That is a time, so is the time constant. The other arrangement, , gives ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍, which is an ohm squared per second and not a time. The units have answered a question of memory without any reference to circuit theory. Dimensional analysis is therefore a partial check on any formula, and it costs nothing: it cannot show whether a dimensionless constant such as a factor of 2 or belongs, but it rejects every arrangement whose structure is wrong, which covers most misremembered formulas.

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

A totaliser is meant to accumulate litres from a flow meter reading in litres per minute, sampled every 100 ms. A programmer multiplies each reading by 0.1 and adds it to the total, and the total runs 60 times high. Diagnose it in units, and say why the error is so easy to ship.

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

The reading carries L/min and the sample interval is 0.1 s. Multiplying them gives , which is not litres: it is litres seconds per minute, and the minute has not been cancelled. Litres require the interval in minutes, so the multiplier is minute, not 0.1. Using 0.1 is therefore 60 times too large, which matches the observed error, and the missing factor is a conversion that was never written down. The error ships easily because 0.1 is the correct sample interval in seconds, so the line looks right to anyone who reads it as "interval times rate" without checking which time unit the rate uses. The defence is to carry the unit in the variable name, or in a comment at the point of use, and to derive the multiplier by cancellation.

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

Dimensional analysis will reject a wrong formula but cannot confirm a right one. Explain the limitation precisely, and say what it means for how you use the check.

REVEAL MODEL ANSWER
MODEL ANSWER

The check tests structure and cannot see dimensionless quantities. A factor with no units (a 2, a , an efficiency, a coefficient of friction) can be inserted, removed or inverted without changing the dimensions, so a formula that is out by a factor of 2 passes the check as cleanly as the correct one. The kinetic energy expression shows this: and are dimensionally identical, and only one is correct. The check is therefore a one-way filter. Failing it proves the working is wrong, which is useful because the check is cheap and catches the most frequent error, an inverted or omitted conversion. Passing it proves only structural consistency. In practice, units are a continuous guard during the working, paired with a second, independent check on the magnitude, usually an order-of-magnitude estimate or a known limit such as conservation of energy. The two checks together catch far more than either alone.

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

DPractice Problems

P1 (easy). Convert, showing the cancelling factors: (a) 2.5 hours into seconds; (b) 750 mm into metres; (c) 1500 rpm into rad/s; (d) 4.7 kilohms times 100 nanofarads, and state the unit of the result.

P2 (medium). A conveyor belt runs at 0.35 m/s and carries product at 12 kg per metre of belt. (a) What is the mass throughput in kg/s, and in tonnes per hour? (b) The motor driving it is 78% efficient and the belt needs 240 N of tractive force. What electrical power does the drive consume? (c) Over an 8 hour shift, how much energy is that in kWh?

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍P3 (harder). A hydraulic cylinder of 63 mm bore is fed at 14 litres per minute at a pressure of 160 bar, where 1 bar is Pa. (a) Find the piston area in square metres and the force it can exert. (b) Find the extension speed in m/s. (c) Compute the hydraulic power two independent ways, from pressure and flow, and from force and speed, and confirm they agree. (d) State what the agreement does and does not prove.

Solutionsclick to reveal

P1. (a) . (b) . (c) ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. (d) , and since an ohm times a farad is a second, the result is .

P1Compared to this solution - did you get it right?

P2. (a) Throughput is speed times loading: . The metres cancel, leaving kg/s. Converting: ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. (b) Mechanical power is force times speed: . Electrical input is that divided by the efficiency: . The input is larger than the output, which confirms the direction in which the efficiency was applied. (c) .

P2Compared to this solution - did you get it right?

P3. (a) Radius is 31.5 mm, so ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍. Pressure is Pa, and a pascal is a newton per square metre, so , about 49.9 kN. (b) Flow is . Speed is volumetric flow over area: , about 75 mm/s. (c) From pressure and flow: . The square metres cancel against two of the cubic metres, leaving N m/s, which is a watt. From force and speed: , agreeing to within rounding. (d) The agreement proves the two routes are dimensionally and algebraically consistent‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍, which they must be, since identically: the area cancels. It is therefore a real check on the arithmetic. It does not prove that the answer describes the real cylinder, because both routes share the same inputs and the same assumptions: no friction, no leakage, no pressure drop in the lines, and an exact bore. A check that uses the same assumptions twice cannot test those assumptions.

P3Compared to this solution - did you get it right?

E‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍Feynman Exercise

Explain to someone who has always treated units as a label to write after the number: what dimensional analysis actually is, why a conversion factor is allowed to be inserted anywhere, and why carrying units through a calculation catches mistakes that checking the numbers cannot. Write your explanation, then compare.

REVEAL MODEL ANSWER
MODEL ANSWER

Units are part of the arithmetic. A metre or a second can be treated like an in algebra: carried along, multiplied, divided and cancelled. That is all dimensional analysis is. It gives a rule about adding: metres can be added to metres but never to seconds. The same holds for the two sides of any equation, because a length is not a time. Multiplying and dividing have no such restriction, and they are where new units come from: metres divided by seconds is a speed. Conversion factors follow from one observation. In the fraction sixty seconds over one minute, the top and the bottom are the same amount of time written two ways, so the fraction equals 1. Multiplying by 1 changes nothing, so the fraction can be inserted at any point in a calculation, and the only decision is which way up to write it. The unit to be removed goes on the opposite side of the line from where it currently is, so it cancels. For this reason there is no need to remember whether to multiply or divide by 3.6 when converting km/h to m/s: the factors are written so that the kilometres and the hours cancel, and the arithmetic gives the direction. Carrying units catches mistakes because it is a second calculation alongside the first, and it does not depend on knowing the physics. A multiplication done where a division belongs still gives an answer that looks like a number. The units give kilograms per second cubed, which is not a unit of power, and that shows the error at the line where it happened, before the number is used anywhere.

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

FError Analysis Framework

The mistake: applying a conversion factor the wrong way up, or omitting one entirely, usually where a rate quoted per minute or per hour meets a calculation working in seconds.

Why it happens: a factor applied in the wrong direction produces a plausible number. A totaliser running 60 times high still rises and falls with the flow, so it behaves correctly in every respect except magnitude, and the magnitude has no independent reference until a reconciliation is done weeks later. The factor is also usually written as a bare numeric literal, which hides its direction: 0.1 looks correct as a sample period whichever time unit the rate uses.

How to recognize it: an answer out by a suspiciously round factor, 60, 1000, 3600, or 9.81; totals that do not reconcile against a manual measurement or a delivery note; a result whose unit is not the unit you wanted; an efficiency applied so that output exceeds input; a formula that gives the right answer in one set of units and not another.

‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍How to correct it: carry units through every line instead of attaching them at the end, and check the unit of the result before reading the number. Derive each conversion by cancellation instead of recalling a factor, since a cancellation has no direction to get wrong. Convert to base units at the start of a calculation and back to convenient ones at the end, so the middle of the calculation has no prefixes. Pair the unit check with an order-of-magnitude estimate, because units cannot detect a missing factor of 2 and an estimate can.

GMini Challenge

A specification says a robot joint must deliver 50 N m at 30 rpm, and a supplier offers a motor rated 250 W with a 20:1 gearbox of 88% efficiency. Decide whether the combination meets the specification, using units as your check at every step, and then state which part of your answer a dimensional check could never have told you.

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

The joint's demand is worked out in watts, since the motor is rated in watts. Angular speed first: . Mechanical power at the joint is torque times angular speed, and because the radian is dimensionless the result is directly in watts: . Working back through the gearbox, the motor must supply , which is within the 250 W rating with about 29% headroom. The speed also works: the joint needs 30 rpm, so the motor turns at rpm, and the motor torque is . On paper, the combination meets the specification. The dimensional check cannot show most of what decides whether it works in practice. It cannot say whether 250 W is a continuous rating or a peak one, whether the motor can hold 2.84 N m at 600 rpm without exceeding its thermal limit, whether the duty cycle is continuous or intermittent, or whether the 88% figure holds at this load as well as at the gearbox's best point. It also cannot detect a missing term with the same dimensions: if the specification meant 50 N m as a peak during acceleration instead of a steady value, the real requirement includes an inertia term that has not been computed, and that term has the same dimensions as the one that was. Units confirmed the structure of every line, but they say nothing about whether the right quantities were compared.

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

Quiz Check

Answer without looking back.

QUIZAuto-graded check · feeds your mastery score
  1. A conversion factor may be inserted anywhere in a calculation because:

  2. Which of these is forbidden by dimensional homogeneity?

  3. A milligram expressed in the SI base unit of mass is:

  4. ‍​‌‌​​‌‌​​‌‌‌​​‌​​‌‌​​‌​‌​‌‌​​‌​‌​​‌​‌‌​‌​‌‌‌​​‌‌​‌‌​​​​‌​‌‌​‌‌​‌​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​‌​‌‍A calculation of power yields units of . This means:

  5. Dimensional analysis cannot detect:

This is a free sample

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