The mainstay of workbench test equipment.
Multimeters today are meant to be so easy that you would be embarrassed if you couldn’t use one – right? Many apprentice electricians, technicians, and even instrument technicians are typically handed an expensive multimeter and expected to be able to use it the first day on the job, without so much as an introduction to instruments in general; but there is a lot more to using an instrument than turning it on and reading the numbers.
In a previous life teaching at TAFE, a lot of time was spent reinforcing one concept in particular: measuring the voltage across a component, or the current through a component.
With the multimeter set to an appropriate voltage range (i.e., the range with a maximum above the voltage you intend to measure) make sure the field of work is clear so you don’t get leads or hands tangled in machinery. Then be sure to place the negative lead first on the negative side of the component, followed by the positive lead on the positive side. Note the voltage reading and the polarity. Perhaps hold in that position while looking for changes in the value.
When done, remove the positive lead first, followed by the negative lead. This isn’t always essential, but is a good practice developed on safer DC voltages. At times, electricians can forget this basic safety practice, and place the active (red) wire first, leaving a healthy 240Vac on the unattached neutral (black) lead. The impedance of the meter provides some protection, but the voltage will be high enough to shock, which often result in spontaneous motion of both limbs and vocal chords.
Remember the voltage is always measured across a component, and always connect to the most grounded voltage first.
Current flows through a component so measuring current requires the meter to be placed in series with the component. With the power off, the component conductor must be disconnected and the instrument wired into the circuit; once again preferably on the lowest voltage side of the component.
Once a check is made that all is safe and as intended, the power is turned on and the reading taken. Then the power is turned off again and the conductor replaced directly onto the component.
The inconvenience of connecting the ammeter in series is why electricians rarely measure current this way. Typically they prefer a special instrument called a “tong tester” which places a magnetic ring core around a conductor and measures the current flowing through the conductor within the ring (i.e., the current inside the conductor).
Another name for the tong tester is a “clamp meter” because the magnetic ring is often split so it may be easily placed around a conductor without disconnecting the conductor, and then clamped usually by a spring action.
It is more common in electronics to measure the current of a complete circuit, usually to determine if the circuit is faulty, or to calculate the battery or power supply requirement of a circuit.
As mentioned in our previous issue, current measurements may also be achieved by measuring the voltage across a shunt resistor.
Resistors are best measured out of circuit, as they are often in series or parallel with other resistors or components. If you must measure a resistor within a circuit, remember it should never read above its labelled value; unless it or the circuit is faulty. If it does read less than its labelled value, then there may be other components in parallel with it.
Most resistance measurements are made on an “ohmmeter” which applies a voltage to the resistance, and simply measures the current through the resistor:
Old analog instruments had a needle that rotated clockwise for voltage and current, but anti-clockwise for resistance, as maximum resistance meant minimum current.
Digital instruments however, are simply set up to read the value in clear digits, and that seems easiest for most people.
Multimeters do not normally measure power. However, once you know how, it is very easy to do, particularly if you have two multimeters or a stable supply voltage.
For example, take a circuit connected to a battery.
Measure the voltage first to make sure of the state of charge. Car batteries can be from 11.5V to 14.2V and still be considered healthy and stable, for low powered circuits at least.
Then connect the multimeter in series with the load – disconnected from the battery of course – and make sure the multimeter is set to read the expected current. Many multimeters have a maximum scale of 10A. Also make sure you have the probes plugged into the correct sockets in the multimeter.
Energise the circuit and record the current. Multiply the current by the voltage, and you have the power, in the DC circuit.
Alternatively, for a single resistor, or a network of resistors, measure the total resistance across the terminals, without the battery connected. Then energise and measure the voltage across the circuit.
Power is then equal to V2/R.
Most of what has been said so far is true for DC or ac, but ac circuits begin to behave differently when inductors or capacitors are in the circuit. Both DC and ac can be challenging when active components, transistors, ICs and the like are in circuit, as the conditions may change when you move the probes from component to component.
For the most part, Ohm’s Law works as long as L and C are much smaller than R.
Other Common Functions
Taking the most basic multimeter I could find, actually I should say “lowest cost” as it proved not to be all that basic. The Jaycar QM1500 or Altronics Q1053B, for example, is all that is needed for a tabletop hobbyist.
Even though I have access to several multimeters, the little Digitech meter is often the first one grabbed, especially if it’s going to be bounced around or made greasy. At around $10 each, these little multimeters are great when you need several readings running continuously for an experiment, as you can buy as many as you need. In fact, this simple little multimeter also has a diode test function, the ability to measure transistor gain, and duty cycle of a PWM power source.
Some other multimeters can measure temperature, capacitance, or frequency. More expensive instruments may measure even more uncommon, and sometimes specialised values. Automotive multimeters may include functions such as temperature, frequency, tacho (RPM), and dwell (the spark charge duration).
In fact, there are many specialised multimeters for a diverse range of applications.
Personally, a Star Trek Tricorder would be useful. The fictional Tricorder seems to measure whatever that episode requires with two or three tweaks and a blip or two!
Before we get too deep into instruments, we need to understand the most significant rating of any instrument, which is “category of use”. You may have seen instruments that carry the label Cat IV, Cat III, or possibly Cat II. What you will not have seen is Cat I on any instrument. This is because anything below Cat II is automatically Cat I, meaning it is uncategorised or not tested at all.
Therefore most hobbyists will never need more than a plain ordinary multimeter, as legally, they cannot work on “live ac power”.
In the simplest terms, anything that plugs into a wall socket, or generator or inverter, such as household electrical appliances requires Cat II instruments.
The house wiring requires Cat III, which is what electricians often use as their “ordinary” instrument, but Cat IV instruments must be used on switchboards and what electricians call “distribution circuits”. If an electrician only has one instrument, it should be a Cat IV.
Remember, if an instrument doesn’t have a category, it is not recognised as safe for any connection to ac power, and you shouldn’t use it on any circuit using 240Vac.
Your favourite multimeter may have a 1000V range, but it doesn’t have the safety features it requires to measure that voltage safely.
What Is The Difference?
The difference between the instrument categories is the level of protection the instrument has against an electric shock, or explosion.
When high power wires are shorted out, large currents can flow that can generate a very large quantity of heat. The current is called a “fault current”. A fault current rating of 5000A is considered a low fault current, which potentially generates 25 million joules of heat per second – 25 megawatts! You wouldn’t want to be around if a fault of that size occurs.
To avoid such accidents, instruments must be rated to safely handle a fault, without it getting to that size or duration. A 10A circuit breaker should allow 10A to pass through it, but break the circuit as more than 10A passes over time.
A fault current should be stopped within 5mS, but as circuit breakers would be a little too big, multimeters use special fuses called High Rupturing Capacity (HRC) fuses to stop the fault.
The potential fault current requires special consideration inside the meter, and also of the leads. Cat IV instruments have greater precautions against internal short circuits and have internal current limiting. In some early Cat IV instruments, the probes had special high fault current (HRC) fuses in the probes.
The second feature is insulation and isolation. The leads plug into the instruments with special shrouded plugs and mostly insulated tips on the probes. Only enough probe tip is left bare so the probe can make contact with the component being tested.
Both features have defined sizes and materials to ensure that you cannot accidentally contact the live parts, and metal parts cannot easily enter the probe sockets.
Note: Nothing should be pushed into the sockets except the probes that came with the instrument, or replacements made to suit the instrument design.
The meter cases also have a number of features at every opening or joint, to ensure that the user cannot come into contact with any live part, but also so that sweaty hands will not allow sweat to seep into the instrument and make a circuit with a live part. For example, the tracks on the printed circuit board must be a certain minimum distance from the edge of the PCB, and from the outside world, usually measured around several overlapping lips in the case halves.
Any damaged instrument is no longer Cat II or above unless it is repaired to it’s previous standard.
So when you see two instruments that both measure volts and amps, one for $10 and another for $700, they are, unsurprisingly, not the same quality or safety rating.
SO HOW DO METERS MEASURE?
“Movement” of a multimeter comes from a time when all instruments were based on a common analog meter. For example a suitable central movement might be a 100 microAmp FSD (full scale deflection), which simply means that the meter needle is deflected to full scale at 100uA.
A modern digital instrument may have a movement with a full scale of 200mV, but without movement or deflection, it is simply limited by it’s display and method of converting analog values to digital. Although the term “movement” sounds like an old analog instrument, techs today still use this term.
In effect, it is neither the voltage nor current of specific interest, but the impedance (resistance) of the movement.
Both instrument movements have a scale much less that the multimeter ranges they will have available, but never more – unless internal amplifiers are included in the instrument.
Multimeters all use the same movement for both voltage and current ranges, and indeed for all ranges. They also use resistors to add multiple ranges. Occasionally other components are used to create other ranges of interest.
For the voltmeter, resistors may be added in series with the meter movement to increase the range value. Panel meters cannot easily handle odd multiples for ranges, so decade ranges are easiest to manage. e.g. for the 200mV panel meter, 2V, 20V and 200V would be easiest to accommodate.
The panel meters often have an impedance of 10MΩ for that 200mV, or put another way, the panel meter impedance is 50MΩ per volt. The simplest way to increase the range is to add “multiplier resistors” in series with the panel meter. So to make a 2V meter from a 200mV panel meter, you could add 2 x 50MΩ – 10MΩ = 90MΩ in series with the panel meter, and the voltage will be as accurate as the resistance used. If 90MΩ is not available, 9 x 10MΩ can be used, but the problem grows as the range voltage increases.
For a 20V range, the multiplier would have to be 20 x 50MΩ -10MΩ = 990MΩ. You might notice that a 200V multiplier would require a really, really, large resistance with a really difficult value to source or fabricate! There must be another method, more appropriate to hobbyists.
The 10MΩ is a high impedance compared to older analog instruments. Analog metres were good for maybe 10kΩ per volt, so even 50kΩ per volt is an improvement. If a 10kΩ resistor is placed across the panel meter terminals, the panel meter will have an effective impedance of 10kΩ rather than 10MΩ. However, the multipliers will also be 1000 times less resistance. For the 2V range, 90kΩ is needed, which can be 2x180kΩ in parallel. For the 20V range 2 x 1.8MΩ in parallel (900kΩ) is added to the string of resistors already used. Then if needed, a 200V range only needs 2 x 18MΩ (9MΩ) but then 18MΩ is beginning to become difficult to acquire again.
Remember, unlike analog instruments where a different scale is used for different ranges, the digital panel meter will read from 0 to 200, and you can’t change that. Also purists will note that there will be a loss in accuracy, but for most in the hobby, the difference won't be noticeable.
Last month, in Fundamentals we looked at shunts, and that is exactly what a multimeter uses, but typically a commercial multimeter has two shunts. The larger of the two is usually 10 Amps if used, but the more common smaller shunt is 2 Amps. In order to provide more ranges, either more shunts can be used, or a voltage divider can be used. For the 2A range, using the 200mV panel meter, a shunt resistor can be calculated neglecting the 10MΩ meter impedance, and calculating R = V/I = 0.2/2 = 0.1 Ohm.
You can try 10 x 1 Ohm, 1% resistors in parallel, (Altronics R-7686, 0.6W 1% Metal Film) making a 0.1 Ohm , 6W resistor at statistically better than 1% tolerance. Including the 10MΩ meter, at 2A the panel meter reading will be V = IR = 2 Amps x 10MΩ // 0.1 Ohm.
In case your parallel formulae is a little rusty, assuming you know the symbol “//” as meaning “in parallel with”, use the common formula:
R1//R2 = 1/(1/R1 + 1/R2) = 1/(1/10E6 + 1/10E-1) = 0.09999999 etc. (i.e. ~0.1Ω), meaning the 10MΩ had far less effect than the tolerances of the resistors would have. The 200mV meter would, therefore, read 2 A x 0.1Ω = 200mV.
An Ohmmeter is either a series type or a parallel type. The device under test (DUT) – usually a resistor, conductor or such – may be supplied by a constant voltage, simply from a good battery or a regulated power supply, through a series ammeter which is calibrated in Ohms, kOhms, or MOhms, depending on the range of interest.
The series ammeter type is a common analog instrument type. The scales start at infinity and go through to zero on the right side of the instrument face. This is because R = V/I and the instrument is reading zero current as representing infinite resistance.
Digital instruments use a parallel circuit, which uses a constant current source in series with the DUT, and measures the voltage across the DUT. This is an easier circuit for a digital instrument to display as there is no math involved. I is constant so R is proportional to V. Once again, ranges will be decades of '2'. If the 200mV needs to read resistance values up to 200Ω, then by Ohm's law again, I = V/R = 200mV/200Ω = 1mA.
A better quality nstrument would use a 'Constant Current' circuit to supply 1mA to the test leads, but a simple method uses a series resistance calculated to provide 1mA at the required voltage drop. If the power supply to the panel meter is regulated at 5VDC, then the range resistor needs to be R = (Vs-Vm)/I = (5 – 0.2)/0.001 = 4.8kΩ.
While many are possible, one example may be of interest. The “diode tester”, which is also useful for testing transistor junctions, can be achieved by using a constant current source as for the ohmmeter, but as the DUT is a diode, the voltage will be the actual forward or reverse voltage. Presumably the reverse voltage will never be reached and so should read zero, except for a very small leakage current too small for the panel meter to detect. The forward voltage will be around 0.6V so the voltage range should be 2V.
Other ranges simply require a similar procedure to adapt the 200mV scale to measure the value of interest, depending on the DUT. We will leave you to find other ranges to experiment with. Have fun but be safe!
Accuracy is a measure of the uncertainty of a measurement. If you expect that a measurement will always be exactly what the number should be, then you correctly assume perfect accuracy. However, perfect is not always possible, or to be more accurate (!), accuracy depends on how close we look. For example, if a person throws 100 darts at a dartboard, and hits it every time, then accuracy would be 100%. But if the person was expecting to hit the bullseye, the accuracy may be only 3% (as no more than that number can fit on the target).
Maybe we’ll remove each dart after it is thrown, and the person may have a good accuracy, but is unlikely to hit the bullseye every time. How far the darts miss the bullseye would be known as the “variance”, which is also given as two values called “offset” and “gain”.
If the person throws more darts to the left than to the right, there is an offset error, but if the darts are all over the place, but the average is in the centre then there is no offset error.
The darts should center about the bullseye in a “Bell curve” but that’s getting a little deeper than we need at this point!
If an instrument makes continuous readings, as a digital instrument does, then every reading should give the same result for the same value being read. An instrument is precise if every time it reads a given value it shows the same result. For example, if a voltmeter continuously reports a battery to have a voltage of 1.5V, assuming the voltage is not changing then the reading has precision. But if the actual voltage is 1.49V, then the precision is still good, but the accuracy may not be as ideal as expected.
The value displayed on the 200mV panel meter can only be from 0 to 200, with no decimal places, so the resolution can be no better than 1mV, or 1/200th of the display range. An eight-digit digital meter can display a value up to 99,999,999, so if the voltage range was 100V, the resolution would be 0.000,001 or 1 microVolt (note: very few instruments have such a small resolution).
Another way to think of resolution would be having a voltmeter in a car that has a resolution of 1V. If it keeps changing between 12V and 13V then what it the likely voltage? The answer is somewhere between 12V and 14V.
The term sensitivity is used to describe the smallest variation in the monitored value that can be detected and, therefore, cause a change in a reading. Unfortunately sometimes the term “sensitivity” is used for what should be called “burden”.
Most instruments require some energy to actually measure, in order to make a reading. The panel meter has an impedance of 10MΩs and a full scale of 200mV, therefore the burden of this instrument is I = V/R = 0.2/10E6 = 0.00000002A or 20 nanoAmps, which is very good. A small burden means a more accurate reading as taking current from a circuit alters the true circuit, and therefore a higher burden means a greater error. Burden is a result of impedance so a high impedance, for a parallel connected meter, is a low burden and therefore a more accurate reading. Ammeters and other series instruments should, by the same logic, have a very low impedance, for a low burden.
Burden and loading are essentially the same effect with different names. Loading is the electronic guys way of complaining about a low impedance multimeter.
Back when analogue instruments had an impedance of 10kOhms, and often times much worse, people working on low voltages and low current loads; electronics techs mostly, found that their Ohm's Law didn't always work. In fact, sometimes the technology they were working on was losing voltage all over the place and they couldn't explain why, unless they understood loading.
Take three 10MOhm resistors in series across a 10volt supply. Not now, but please do try this at home!
Now to simulate a 10kOhm multimeter, wrap the legs of a 10kOhm resistor around your meter probes, and measure the voltage across each of the 10MOhm resistors.
Add the three voltages together and you get 10 volts right? Well of course you should, but where did the rest go? That is the effect of Burden or Loading.