As far I know, the analog DC voltmeter works by deflection of the moving coil in the magnetic field (utilizing the 'EM induction.')
How does the counterpart - AC voltmeter work? Can anyone explain how in spite of its alternating nature, the meter shows a precise measurement of AC voltage?
Can we convert an analog DC voltmeter to an AC voltmeter? For example, converting a 0-10V DC meter to measure 0-10V AC.
There are no convenient AC voltmeters; there are only DC voltmeters. Therefore, we are forced to convert the AC voltage to DC and then measure it with a DC voltmeter. We can do this by first rectifying the voltage and then smoothing (averaging) it. Let's see how through step-by-step experiments, revealing the problems and looking for ways to solve them.
Today, these electromechanical measuring instruments have fallen out of use, and we are examining them more out of curiosity and historical interest than for any practical purpose.
A characteristic feature of them is that their moving coil is inert and cannot oscillate at the frequency of the input AC voltage (50/60 Hz). This saves us from having to connect a low-pass filter; we only need a rectifier.
Another feature of these old-fashioned voltmeters is that they have a relatively low resistance (for example, 10 kΩ). We can set this in the parameters field of the voltmeter.
This can be simply a diode connected in series. If it is "ideal"...
simulate this circuit – Schematic created using CircuitLab
... exactly the positive half-waves will reach the voltmeter, and the pointer will deflect to the corresponding position. The simulator cannot show this mechanical averaging; we simply have to imagine it.
Diodes, however, have a forward voltage drop of around 0.7 V, which is subtracted from the input voltage.
If the latter is small, the error is significant.
We can compensate for this unwanted voltage drop by adding the same voltage to the input. For this purpose, we can connect the diode in the negative feedback loop of an op-amp follower.
The op-amp "lifts" its output voltage by as much as is lost in the diode. Thus, the "diode + op-amp" combination acts as an "ideal diode".
In essence, an analog voltmeter consists of an ammeter and a resistor in series. So, the best way is to control the ammeter directly.
For this purpose, we can connect the ammeter and the diode in the negative feedback loop of a non-inverting amplifier (voltage-to-current converter).
As above, the op-amp compensates for the voltage drop across the diode...
... and the current is exactly equal to Vin/R.
To reduce the ripple, let's make the rectification full-wave, for example, using a bridge rectifier circuit.
But now the voltage loss across two diodes in series is twice as large, and almost the entire input voltage is lost.
The op-amp once again helpfully assists us...
... by raising its output voltage by 1.5 V to compensate for the voltage drop across the diodes.
Modern digital AC voltmeters do not have the inertia of 19th-century electromechanical voltmeters. Therefore, they need a smoothing (averaging) filter (RC circuit) connected to the input.
There are two time constants represented by the two resistors (in series and in parallel to the capacitor). The first is for charging, and the second is for discharging. Depending on their values, you get either a peak or an average value detector.
In a peak detector, the charging time constant is minimal, and the discharging one is maximal.
Zero resistance in series, infinite resistance in parallel: If we connect the capacitor directly to the rectifier output...
... it will charge almost instantaneously to the peak of the positive half-wave...
... but will have nowhere to discharge (the voltmeter is 'ideal'). One solution is to force its discharge by short-circuiting it, but this is not convenient.
Zero resistance in series, finite resistance in parallel: Another (compromise) solution is to connect a resistor with a large resistance (or to set the voltmeter's resistance) in parallel with the capacitor, which will discharge it slowly.
In an average value detector, both have relatively large values.
R = 1 kΩ
R = 10 kΩ
These techniques can also be applied to the other diode circuits from above.
OP said in a comment:
What about this idea that - (after rectification) can we divide the peak voltage Vp by factor of 0.707 using voltage divider?
Your idea is feasible. By connecting a high-resistance voltage divider R1-R2 with a "gain" of 0.707 to the output of a peak detector, you can obtain an RMS detector (for sinusoidal AC voltage).
so, can we skip average detector (if not required) ?
Yes, you can. In my opinion, the average detector has the advantage of having fewer pulsations and that the resistors have lower resistance.
Not all, but some AC panel meters work in the same vein as "universal" motors - ones that operate on AC or DC voltage.
The idea is to replace the movement's permanent magnet with an electromagnet. In other words, you end up with two electromagnets - one stationary one. And the original "stator" one attached to the needle. This has the effect that the deflection of the needle is always unidirectional independent of input polarity. The two electromagnets are always oriented in repulsion since their B-fields always switch, in phase, with the applied voltage.
With a meter using a d'Arsonval movement (such as you've shown in the photo) the incoming AC is internally rectified to DC since the movement uses a DC magnet to make a bias field. The panel meter you show is internally calibrated to account for the rectification effects. And since it's for a relatively high voltage the diode-drop error will be negligible at full scale.
Were you to have only a DC version of this meter on hand and needed to measure AC, you could drive it with a diode bridge and get a decent RMS reading. Assuming a 20k ohm/V (50uA) meter, that would look something like this (simulate it here):
There's a couple of drawbacks of this approach. The diode bridge inserts 2 diode drops, making low voltage performance very poor. Also, the diodes themselves need be be able to sustain the full reverse voltage: for a 300V RMS input they need to be 600V diodes. If you didn't care about low voltage accuracy it would be fine for monitoring line voltage, noting that the RMS voltage will read about 1.2V lower than the actual RMS value at full scale.
A VOM with a d'Arsonval meter uses an improved approach: a pair of signal (low forward voltage) diodes do the rectification before presenting the voltage to the meter movement. The diodes are typically germanium or Schottky with Vf about 0.3V. This improves low-range accuracy. And, only one diode is conducting at a time so only one diode drop occurs.
You can see exactly how the rectification is done in an analog VOM by examining the Simpson 260 meter schematic (you can find a bunch of them here.)
From here: https://www.simpson260.com/downloads/simpson_260-7_and_7m_user_manual.pdf
Here's a simplified sim of the Simpson 260 meter doing 0 ~ 2.5V AC range mode (simulate it here):
While not perfect, the Simpson 260's AC low-range performance is significantly better than the proposed bridge rectifier above. Because there is a high-value resistor in series with the input, the diode reverse voltage is limited, allowing more reasonable choices for the diodes.
For better AC accuracy there's an alternative meter movement that doesn't need diodes: the iron vane (sometimes called 'AC repulsion') meter. Iron vane meters use a solenoid coil that induces a field to two parallel plates: one fixed, and one moving and attached to the pointer. The applied field induces the same magnetic polarity in the plates, causing them to repel each other with a force proportional to applied current, regardless of the solenoid current direction. This allows iron vane meters to work on DC or AC without rectification.
Iron vane meters can be built to measure current directly (again, AC or DC) without a shunt, just by looping the wire path around the solenoid. This allows them to read current without inserting an IR drop.
More about meter types here: https://electricalacademia.com/instrumentation-and-measurements/basic-analog-meter-movement-darsonval-iron-vane-meter-movement/
Short answer: No, a DC voltmeter cannot directly measure AC voltage accurately, as it only measures the average or DC value of the signal. But with a few tricks, you can get some idea of AC voltage with it.
This can be done by adding a rectifier circuit. First a diode bridge for rectification and a capacitor to smooth (DC) the output signal. But as you mentioned, it can fluctuate. Especially if the capacitor used is of low capacity.
Important note: This method is accurate for measuring pure sine voltage because it has a direct relationship to the DC voltage. Vac=Vdc/1.4142
But in practice, AC waveforms are not pure sinusoids and are distorted due to the loads. One technique for accurately measuring these types of waveforms is to use True RMS.
Following are the contents written in the comments: how does the 1.11 factor apply?
This is done with two tricks:
There are two basic kinds, moving iron and moving coil. Both are (at their heart) current meters and have some series resistance to create a current from the input voltage.
From this catalog, we can see similar examples of both types:
The above one uses a silicon bridge rectifier and a moving coil (d'Arsonval) movement so there is some compression at the bottom of scale, which is of little consequence in the normal operating range. In this case the manufacturer has reflected that compression in the scale, however it is not a given that this is always done. Some makers just appear to ignore it.
Moving Coil Meter Centre cored, self shielding moving coil movement, using pivots, hairsprings and sprung jewels. Seven variations have been designed in movement ranges: all intermediate ranges are achieved by shunting the next lowest range. All DC voltmeters are 1000 ohms per volt, rectified product run at 900 ohms per volt, millivolt meters use the 5 milliamp movement.
The above are moving iron (iron vane) types. The kV scale shown would be scaled to use with a potential transformer since the maximum voltage at the meter terminals is 600VAC.
Moving Iron Meter Clapper type repulsion design using pivots, hairsprings and jewel movements. The bottom jewel is oil filled to provide damping while the top is sprung for resilience. All voltmeters are manufactured with external voltage dropper resistors to substantially reduce the self heating effects
The moving iron types are said to use "silicon" (sic) oil damping. Moving coil use eddy current damping.
In both cases what we have is either a series resistor + moving iron movement (mA AC) or a series resistor + bridge rectifier + moving coil movement (mA DC).
