Don't have an account?
Reset your password
Fill in your registered email, and we will send you a link to reset your password.
- Mic University
KNOW THE BASICS ABOUT PHANTOM POWER
What is phantom power?
Why do microphones need power?
Is it standardized.
Is phantom power always 48 volts?
What if the voltage is less than 48 volts, how much current can you draw, must the p48 be switched off if i don't use it, what if my device does not have phantom, do miniature microphones also need phantom power.
How is it possible to carry both AC and DC?
Can P48 power anything else but microphones?
When was phantom power invented, can phantom power damage a microphone.
How to check the phantom power?
Does P48 circuitry have any effect on the input impedance?
How much current draw can you expect from a microphone.
DPA Pencil Microphones
DPA Microphones’ Recording Microphone Series contains a wide range of versatile pencil microphones that can be used for all of your high-end audio amplifying and recording needs. Known worldwide as an industry standard in the recording branch, these mics offer amazing sound in virtually any situation.
DPA Lavalier Microphones
Our miniature microphones are versatile workhorses that adapt to your needs. With eminent sound quality, as well as a large range of accessories and adapters available for all pro wireless systems, these mics are the obvious choice for any application requiring a big sound and a small footprint.
DPA Microphones’ ingenious adapter system is one of a kind in the industry. It protects your investment in fantastic sound for years to come. As your circumstances change, you can just switch your adapter, instead of incurring the cost of a completely new microphone. When you make the switch, you won't hear any loss or change in sound quality from the mic as the adapters are tailored to suit specific transmitter types.
2012 compact cardioid microphone.
Great close miking on any instrument on the live stage
The 2012 Compact Cardioid Microphone is an all-around generalist pencil mic that guarantees excellent results on any instrument on a live stage.
4055 Kick Drum Microphone
Designed to kick the bass to the next level
The 4055 takes the traditional concept of a kick drum mic and turns it on its head - no more pre-tailored sound - now you can take sound design back into your own hands.
2015 Wide Cardioid Microphone
Overhead mics for ambient pick up on the live stage
The 2015 Compact Wide Cardioid Microphone is for ambient pick up of a drum kit or other instrument groups.
Sign up for our newsletter
Be the first to hear about our new products, workshops, events, contests and more.
Find An Answer
What is phantom power, apr 14, 2021 • knowledge.
Phantom power is used with condenser microphones. It is called phantom power because there is no obvious external power supply for the condenser mic; the power supply is invisible and therefore a "phantom." The mixer sends voltage up the same wires that the audio is traveling down. Thus, the microphone is receiving the power remotely from the mixer. Phantom power is a DC voltage (usually 12-48 volts) used to power the electronics of a condenser microphone. For some (non-electret) condensers it may also be used to provide the polarizing voltage for the element itself. This voltage is supplied through the microphone cable by a mixer equipped with phantom power or by some type of in-line external source. The voltage is equal on Pin 2 and Pin 3 of a typical balanced, XLR-type connector. For a 48 volt phantom source, for example, Pin 2 is 48 VDC and Pin 3 is 48 VDC, both with respect to Pin 1 which is ground (shield). Because the voltage is exactly the same on Pin 2 and Pin 3, phantom power will have no effect on balanced dynamic microphones: no current will flow since there is no voltage difference across the output. In fact, phantom power supplies have current limiting which will prevent damage to a dynamic microphone even if it is shorted or miswired. In general, balanced dynamic microphones can be connected to phantom powered mixer inputs with no problem.
- Number of Views 57.47K
- Number of Views 25.32K
- Number of Views 2.16K
- Number of Views 9.72K
You are here
Phantom Power is a standardised professional method of providing power to the electronics of some types of microphones via a balanced XLR connection. The relevant standard was first conceived in the early 1970s, and is now recognised as IEC 61938:2018. Several variations of phantom power are detailed in the document.
Most professional systems operate with a nominal supply voltage of +48V DC, although there is an acceptable tolerance range of +/-4V (ie. 44 to 52V). This arrangement is described as the 'P48' format, and the phantom power supply is connected with the positive side going to each of the two balanced audio lines via individual 6k8 current-limiting resistors. The negative side is returned via the cable screen. In this configuration the maximum current available to the microphone is 10mA, providing up to 170mW of power. Most microphones draw around 4mA (or less) of current, but some models require more and a few need the full 10mA .
Battery-powered equipment often uses the P12 format, which uses a 12V power supply with 680 Ohm feed resistors allowing up to 15mA and 100mW of power. A newer addition to the specification, called P12L (low-power), uses 3k3 resistors to give 4mA and just 22mW of power.
At the other end of the scale is the new SP48 variation which uses the 48V supply voltage but connects via 2.2K feed resistors allowing up to 22mA and 520mW of power.
Although rare, some manufacturers choose to provide non-standard supply voltages, such as 15 or 24V — something which seems prevalent on budget, compact, active PA loudspeakers. Some microphones are very tolerant of the supply voltage (many AKG mics can accept anything from 9-52V, for example), while others won't function correctly if the supply voltage falls too low.
Other microphone powering schemes are also available, including ' Plug-in Power ' for consumer electret and lavalier mics, and Tonader (or A-B) power for battery-powered legacy professional equipment.
- Plug-In Power
- Tonader Power
New forum posts
- Re: Electrik Bach sos_forumer > 26 Oct 2023, 00:35 Keyboards & Synthesis
- Alto TS415s super quiet with mixer? BigBrother > 26 Oct 2023, 00:01 Live Sound & Performance
- Re: Electrik Bach BillB > 25 Oct 2023, 23:21 Keyboards & Synthesis
- Re: Performance meter in Cakewalk The Paul meister > 25 Oct 2023, 22:37 Windows Music
- Re: Why does the version that is clipping and have less stereo width sound actually better? BJG145 > 25 Oct 2023, 22:34 Mixing, Mastering & Post Production
- Alto TS415s super quiet with mixer?
- Why does the version that is clipping and have less stereo width sound actually better?
- VSTBuzz: Up to 77% off 4 New Deals from Guareschi, Tom Wolfe, Handheld Sound & Nembrini Audio
- Transport bar in Cakewalk question
- Performance meter in Cakewalk
- Electrik Bach
- A rather strange MIDI issue
- Korg announce Modwave mkII
- Some Drums from Song Athletics
- Transparent Sounding Reverb Settings
Recently active forums
- Recording: Gear & Techniques
- Mixing, Mastering & Post Production
- New Products & Industry News
- Music Business
- Windows Music
- Apps & Other Computers/OS
- Guitar Technology
- Keyboards & Synthesis
- DIY Electronics & Studio Design
- Live Sound & Performance
- Music Theory, Songwriting & Composition
- User Reviews
- Remote Collaboration
Create an account
Phantom Power - What It Is And How It Works
Anyone involved with audio will know about phantom power, also sometimes known as P48, because the source voltage is (or is supposed to be) 48V DC. The term 'phantom' comes from the lack of any on-board power supply (such as a battery), and the power is delivered seemingly by magic. Of course there's no magic involved, just engineering. The P48 standard was pioneered by Neumann (a very well known German microphone manufacturer), and was developed during the 1960s. Rumour has it that Neumann was told by the Norwegian Broadcasting Corporation (NBC) that their new microphone at the time must use 48V phantom power [ 1 ] .
48V DC is very common, as it forms the power supply for the entire telephone system, and it was always provided by a large battery bank using 24 lead-acid cells in series. However, lest you imagine that phantom power was initially derived from the telephone battery bank, it wasn't. The telephone system operates with a negative supply, with the positive grounded. This was done to minimise corrosion should water get into the system (oxygen forms on the anode, and that would eat the wiring away in no time). Presumably the NBC had no such issues and used the more 'conventional' negative ground.
During the 1960s, great progress was made with 'solid-state' electronics, and providing a 48V DC supply in a mixing console or broadcast studio became simple and relatively inexpensive. P48 has been in constant use since it was introduced. It's now standardised in DIN 45596, and is used worldwide. There are alternatives as well, being P12 (12V DC) and P24 (24V DC), but the original is (IMO) still the best. Phantom power input circuits are always balanced, but the powered equipment may use unbalanced outputs, or just a simple impedance balancing arrangement. All work equally well in practice.
48V is not the only voltage supplied, and there are three variations. P48 is by far the most common, and is specified to provide +48V with a maximum power of 240mW (14mA short-circuit current). The other standards are less common, with 24V and 12V being 'sanctioned' variants (IEC 61938:2018). As always with standards, you have to pay to get the documentation.
The three 'official' standards are shown above. One problem is that with voltages below 48V, the resistance of the feed network is lower, making it harder for the electronics to drive the load. While more current is available at the lower voltages, this may not be useful, since it's generally expected that something designed to operate at (say) 12V should not be damaged (and should still work normally) with the more conventional and widespread 48V phantom supply. It should be pretty obvious that if something is designed to work only with P12 phantom power, it will be plugged into gear delivering P48 phantom power. If it doesn't work that's one thing, but you'd be quite rightly very annoyed if it killed the circuitry. Note that I added the 'Working Current', and it's not specified in the standards. It's also highly variable in practice, but the values shown are those that I'd recommend.
If a product is designed for P48, it may not work (or will work very poorly) if supplied with P24 or (worse still) P12. Ideally, standards are just that - standards . To have three different 'standards' that all use the same principle and the same connector is mad. If alternatives are offered, they should use a different connector unless the product can function normally with any of the supply voltages (and feed resistances). This is uncommon, but it is possible.
Any P48 system has to be considered as a whole. Each part relies on the others for its function, and the overview shown below covers the essential ingredients. Firstly, a power supply is required to produce the +48V DC used by the end devices, which are usually microphones, but may also be DI (direct injection) boxes allowing on-stage instruments to connect directly to the mixing console. The microphone preamp will invariably have adjustable gain so that the level of each signal can be set in relation to the others (this is one of the sound engineer's tasks).
The mic cable is self explanatory, but is also responsible for a great many faults (mic cables have a very hard life). Pin 1 of an XLR connector is always ground, and connects to the shield of the cable, and always at both ends. Pin 2 is considered 'hot', and with unbalanced (or impedance balanced) circuits, the signal appears on Pin 2. Pin 3 is 'cold', and in some cases it may be grounded by equipment. For 'true' balanced connections, Pin 3 carries a signal that is an inverted copy of that on Pin 2. The effective signal level is doubled by this technique.
Because Figure 1.1 is simplified to the extreme, each of the functions will be covered in detail below, with the exception of the mic preamp. There are countless variations, but one of the most important functions (the mic preamp protection circuit) is covered. Even with this, there are many variations, but the one shown is adapted from an ESP project and is known to work very well. It provides a level of protection that ensures mic preamps will survive most abuse (but not all - some faults can destroy everything in the signal path).
Because DC is provided to both signal conductors, and through equal-value resistors, the DC appears as a common-mode signal and it doesn't affect the audio. If the feed resistor tolerance isn't good enough, common-mode rejection is compromised, so hum and buzz (the most common noises injected into balanced cables) are not attenuated as well as they should be. The same applies to the microphone or DI box - it must also apply equal loading to both signal lines. Contrary to belief, the signal does not have to be present on both wires, and 'pseudo-balanced' electronics are common in microphones, even with high-end models. What matters is impedance, not the voltage. If the impedance for both signal lines is not the same, hum and other noises will be picked up by the cable.
Nothing is without limitations or compromises, and P48 is no different. It would not be sensible to allow external gear to draw as much current as it likes, as a faulty lead or equipment could cause serious damage. Some early mixers used centre-tapped transformers to provide the DC, but the centre-tap was never connected directly to the phantom power supply. Instead, it was supplied via a resistor to limit the current under fault conditions. Feeding the current to the centre-tap ensures that the transformer is not subjected to any DC magnetic field, as the two windings cancel the flux. If it were otherwise, the transformer core could saturate, causing gross distortion.
Electronically balanced microphone preamps don't use a transformer, and due to the cost of even 'average' mic transformers, most equipment uses a differential preamp, direct-coupled to the mic connector. The standard feed resistance is 6.81k, selected for two reasons. Firstly, specifying 6.81k demands that the resistors are 1% tolerance - you cannot buy 6.81k resistors with 5% tolerance. The other reason is pure compromise. Lower values would load the microphone, reducing its output level and increasing noise, and higher values would be unable to supply enough current to power any useful electronics.
Most equipment now just uses 6.8k resistors as they are readily available with 1% tolerance, something that was not the case in the 1960s or 70s. The two resistors are effectively in parallel for DC, so the total limiting resistance is 3.4k, which allows a total short-circuit current of 14mA. If the internal circuitry of the microphone (or other phantom powered gear) requires 10V for normal operation, then the maximum current available is 11mA. The available current for any operating voltage is easily calculated. A design current of up to 10mA is usually safe, and that gives the remote circuitry a maximum supply voltage of 14V.
The rather miserly current provided means that circuitry has to be low-power to ensure it can operate from the available voltage and current. This means that the designer has to be fairly clever to ensure minimal current drain along with good performance, sufficient to suit the application. High impedance circuits don't draw much current, but they are noisy due to resistor thermal noise and other noise from semiconductors etc. Low impedance circuits draw more current and are quieter, but they may not have enough voltage to handle high signal levels without clipping.
The available power with P48 equipment may seem to be limiting, but it's usually not a problem. You have to live with the minimal current available, and the remote circuitry doesn't have to drive low impedance loads. Most mic preamps have an input impedance of at least 3kΩ, and the extra loading by the 6.8k resistor (on each signal conductor) means that the overall impedance is high enough to be easily driven with relatively simple circuits. Contrary to belief in some circles, microphones should never be terminated with a value equal to their output impedance.
An often serious limitation is that circuitry using P48 phantom power cannot be ground isolated . This is of no consequence for microphones, as they are a 'floating' signal source (not electrically connected to anything else). With DI boxes and the like, there is ground continuity between the mixer and instrument amplifier, and this often leads to issues with hum. It's sometimes possible to reduce the hum to a low level by incorporating a 10Ω to 100Ω resistor in series with the shield at the remote end. This should be bypassed with a 100nF capacitor to ground RF (radio frequency) interference. If included, a switch should be added so the resistor can be shorted out if it's not needed.
There are several ESP projects that are designed to provide phantom power, with one of the most popular being Project 96 . The drawing below shows the general idea for a microphone input. Although it's possible to provide phantom power via a TRS (tip, ring, sleeve) ¼" stereo phone jack, this is not recommended. Many things (such as guitars) can be plugged into a ¼" (6.35mm) jack socket, most of which are not designed to handle any DC at all. While damage is unlikely in most cases, it is still possible, so P48 is normally only made available via a female XLR socket, designed to accept a balanced microphone lead.
Not all phantom supplies include the protection diodes, but I consider them to be absolutely essential. You'll note that my circuit uses zener diodes, which can handle very high instantaneous peak current, and clamp the worst-case voltage to around ±11V. If you are using equipment that's powered via USB (5V), it would be advisable to reduce the zener diode voltage to 3.9V to protect the mic preamp which (probably) runs from a 5V supply.
There are countless different ways to provide the 48V DC used to power the microphones (or other P48 equipment). The circuit shown in Project 96 is a well proven design, and a modified version is shown here. Many USB powered mic preamps for use with a PC use a small switchmode supply. A 12V to 48V version is described in Project 193 , which is capable of up to 100mA (roughly 10 microphones). USB versions are more limited, because they have to boost from 5V without exceeding the USB current limit. This usually means one or two mics at most, because a standard USB port can only supply 100mA unless it negotiates a higher current (up to 500mA) via software.
It's important to note that the performance of any regulator depends on the transformer. For example, if you use a voltage-doubler to provide the 'raw' DC, the transformer has to supply a minimum of twice the output VA. 48V at 200mA is 9.4 watts, so the transformer must be rated for at least 16VA, but there's a lot to be gained by using a 30VA transformer. The peak current is a great deal higher than you might expect, and that reduces the unregulated DC voltage. With the two regulator circuits shown next, the peak current may be as high as 1.7A with an output of 200mA. An under-powered transformer will cause the unregulated voltage to fall, and you may get ripple at the 48V output.
The regulator shown can supply over 200mA easily, and is simple to build. It needs a 25V AC input, which is fed to a voltage-doubler (D1, D2 C1, C2). A bridge rectifier would be a bit better (and dispenses with the voltage doubler), but then you need a 50V AC transformer winding. Otherwise 'odd' voltages are not a problem with a mixing console, because that will have a dedicated power supply that can provide all voltages needed at whatever current the circuits demand. Most mixing consoles will probably not be able to use phantom power on all channels at once, but that's rarely a problem. The circuit shown is easily modified to provide more current if it's needed. The regulator circuit needs an input voltage of at least 56V DC to ensure reliable operation and low noise. The output noise will usually be less than 50µV RMS (primarily residual 100/ 120Hz hum). It can be reduced by increasing the value of C4, but that shouldn't be necessary.
The transistors shown are examples, and virtually any device with similar specifications can be substituted. You might need to change the value of C6 (220pF) if the circuit shows signs of instability (radio-frequency oscillation). A larger value reduces the transient response and might allow more high-frequency noise to get through. An added LC (inductor-capacitor) filter can be added if you wish, but it should not be necessary.
IC regulators are available that can handle the voltage (standard 3-terminal regulators such as the LM317 cannot!). While using an IC may be superficially simpler than the discrete design shown, there's always the problem of sourcing the parts, and being able to find a replacement in 10 years time if the IC fails. A discrete circuit can always be repaired, especially if it uses common parts throughout. Many modern products are not made to be serviced, so if (when) they fail, the only option is to replace the entire unit. A disadvantage of the IC approach (at least with this particular type) is that it requires an output current of at least 15mA to ensure regulation, and that's why the 3.3k resistor (R3) has to be rated for 1W.
The TL783 IC is rated for up to 125V input, with an output current of up to 700mA. Interestingly, the datasheet doesn't mention the maximum power dissipation, but it would probably be unwise to exceed 20W, even with a good heatsink. The output voltage is determined by ...
V Out ≅ 1.27 × ( R3 / R2 + 1 ) Which works out to be ... V Out ≅ 1.27 × ( 3.3k / 82 +1 ) ≅ 52.4V Just outside the P48 specifications)
The output voltage will usually be different from the calculated value because the internal voltage reference (nominally 1.25V, but shown as 1.27V in the datasheet) can vary between 1.2V to 1.3V. This means that the output voltage will actually be somewhere between 49.5V and 53.6V with the values shown. This won't cause any problems in normal use. In a most unusual state of affairs, there is no mention of the minimum input-output differential in the datasheet, but it would appear that it should be greater than 25V (meaning an unregulated supply of at least 75V). With a 25V input-output differential, dissipation with 700mA output will be 17.5W, which will be alright with a good heatsink.
Small switchmode power supplies are often used to boost the voltage from (say) 12V to 48V. These usually need a lot of filtering, because the supplies themselves are noisy. The noise is outside the audio band (typically 50kHz or more), but it can still interfere with the audio by producing intermodulation artifacts. Without exception, USB audio interfaces use a switchmode booster, but they have to boost from only 5V. Most will require at least 500mA to be available from the host USB port, or it's not possible to get enough current for the P48 feed and the internal circuitry.
It's to be expected that most P48 supplies made today will use a switchmode converter (aka switchmode power supply or SMPS). There are countless ICs that do everything - the regulation and main switching MOSFET are all part of the IC itself, requiring a minimum of external parts. For DIY, getting a suitable inductor may be difficult, but otherwise it's a good solution provided the output is well filtered. Miniature DC-DC SMPS modules are available from a number of manufacturers, but if you expect to boost 5V to 48V the input current of a 1W converter will be about 250mA for 21mA output (only just enough for two P48 supplies). Issues with replacement parts are also a consideration, and the ICs available today may not be compatible with newer versions. This means that if the supply fails after 5 years it may need to be completely re-engineered.
The circuit shown above is taken from Project 193 - Obtaining a +48 Phantom Supply From 12V. This circuit has been fully tested, and everything you need to know is in the project article. It can power up to ten P48 microphones easily, but additional filtering is recommended to ensure low noise. This is shown in the project article, and it's easily built on Veroboard.
A few brave souls have used a Cockcroft-Walton voltage multiplier (see Rectifiers, Section 7 ) to obtain the P48 supply, but this is (IMO) a rather poor way to obtain the required voltage. Even with a full 12V peak-peak input, you need a 7-stage multiplier just to reach just 42V when loaded (two microphones). You need a 9-stage multiplier to obtain 48V with a total load of 20mA, and a zener regulator is needed to keep the voltage stable. Voltage multipliers work well with very low current, but aren't suitable for more than a couple of milliamps. There's no doubt that this arrangement can be made to work, but the input current is fairly high from the switching circuit, requiring at least 250mA from a 12V supply. A small switchmode boost supply such as that shown in Figure 3.4 is a much better alternative, and can supply the same load with an input current of less than 100mA.
At the microphone (or DI box) end, there are many different ways to get the required DC and power the circuitry. I can only provide a few examples, because each design will be different. It's probably fair to say that there are as many different implementations as there are manufacturers, and I don't intend to even try to cover them all. There are two primary challenges at the signal source end of the cable, being able to extract the DC supply for the electronics, and superimposing the audio onto the DC supply. Neither is difficult, but it requires some ingenuity. The mic preamp in each circuit is simplified, and has internal DC blocking and overvoltage protection as shown in Figure 3.1.
In each case here, I've assumed an electret mic capsule, but 'true' capacitor (aka condenser) mics are also phantom powered. They generally use a very simple switching boost converter to provide the capsule polarising voltage - typically from 24V up to around 100V DC. Some other mics use an RF oscillator and demodulator to detect the change of capacitance in the capsule. These techniques aren't possible or necessary with an electret capsule.
The Figure 4.1 circuit is fully balanced, and the emitter followers (Q2, Q3) are used to buffer the signal and modulate the phantom supply. The supply is taken from the collectors of these transistors. It's up to the mic preamp's interface circuit to supply P48V and isolate the preamp's inputs from the phantom supply. The circuit draws surprisingly little current, and its operating voltage is around 19V. In some cases, this will be regulated with a zener diode for particularly sensitive applications. It's shown with an electret mic capsule, but the signal source can be any type (e.g. guitar, bass, keyboard). If it's not built as a microphone, R12 would be omitted. The circuit can then be used as a DI (direct injection) box, allowing instruments to connect directly to the mixing desk.
The second circuit is adapted from an ESP project ( Project 93, Recording and Measurement Microphone ). Unlike the previous circuit, this circuit uses impedance balancing, with the impedance set by the two 100Ω resistors and their series 100µF capacitors. From the perspective of a mic preamp, this is almost identical to a 'true' balanced circuit, unlikely as that may seem at first. The DC is derived from the two 2.2k resistors (R10, R11), and the circuit is designed to be able to drive the low resistance without overload.
Note that these are intended as examples only - the idea is to show two different ways to utilise the phantom power to derive a power supply for the electronics. Providing P48V is simplicity itself, needing only a power supply and a pair of resistors. Obtaining DC for the remote-end circuitry is another matter altogether, as the three examples shown indicate. Those shown are by no means the only circuits used, but they are representative of the techniques that can be used.
If the remote circuit uses opamps, their outputs need to be protected from possible damage if a 'hot' cable (with P48V turned on) is plugged into the unit (D1 and D2). The cable has capacitance, and without a load it charges to the full 48V, and there is little or no current limiting. The transient impulse is very short, but can easily cause damage, forcing the opamp's output to +48V before it has a supply voltage. All P48V DC 'take-off' circuits take some time before the supply is available to the circuit. It's usually only a few milliseconds, but sensitive circuitry can be damaged in microseconds!
Most DI Boxes are battery powered, because this allows the ground connection to be broken (commonly called 'ground lift') to prevent hum caused by circulating ground currents. This isn't always convenient, and several DI boxes have the option of battery or phantom power. See Project 35 for a couple of other examples. One is completely passive, and uses a transformer. This is always a good solution, but decent transformers are expensive, and the input impedance is usually fairly low making them unsuitable for use with an instrument with no amplifier.
A phantom powered DI (direct injection) box has several limitations, with the common ground connection being the biggest problem. In the design shown above, the 10Ω resistor (R12, in parallel with C6) may be enough to prevent hum, but it also may not. While it's shown as optional, in most cases it will be necessary. If you wish, you can add a switch to short out the network if the unit proves to be hum-free.
The current drain is quite low, and the zener has been increased to 24V to allow an input level of up to 2V RMS. It can take a bit more, but distortion will become a problem with anything over 2.2V RMS. Input impedance is 25k, which is too low for direct connection to a guitar, but is fine for the line output from an amplifier or keyboard. The overall gain of the circuit is 1.9, and you can use a level pot at the input if higher input levels are expected.
It is possible to use batteries for phantom power. Five standard 9V batteries in series gives a voltage of 45V (nominal), which is within the P48 specifications. New batteries will measure about 10V each, giving 50V which is also within the allowable limits. A phantom powered mic will typically draw a maximum of 10mA in total, so a series string of five should last for at least 25 hours of continuous use, more if the mic draws less current. 'Standard' 9V alkaline batteries have a capacity of around 580mA/h, so with a 10mA discharge that can be worked out to be 58 hours operation. However, the battery voltage will be down to about 7.5V (each) if fully discharged, so you can't operate for as long as you might have thought.
Of course, you can use more batteries and then regulate the voltage, but the regulator would need to be very low power. Something like the Figure 3.3 IC based circuit is completely unsuitable, because it draws 15mA by itself. By comparison, the Figure 3.2 circuit draws around 6mA (no load), which is better, but too high if you were using batteries. It can easily be redesigned to use far less current, but that's outside the scope of this article.
Typically, a battery operated regulator would use 8 × 9V batteries (72V nominal, 60V with 7.5V [end of life] for each battery). Ideally, it would draw no more than 1mA or so with no load. Extreme filtering isn't needed because batteries are fairly quiet (they are not noise free), but a filter is easily added. Battery operation is normally a last resort, and if required it will end up being much cheaper to use a rechargeable Li-Ion battery pack and a switchmode boost converter. The initial cost is higher, but the saving on 9V batteries will add up fairly quickly if battery power is needed regularly.
There's another 'phantom' powering scheme that was developed in the 1960s, called 'T-Power' (aka Tonader, Tonaderspeisung, A-B powering, or parallel powering) [ 3 ] . This is completely incompatible with P48 powering, and dynamic mics are likely to be damaged if inadvertently used with T-power. The source is 12V DC, and the DC is provided between the two signal leads. The general scheme is shown below, but since T-Powering is considered obsolete I won't cover it in great detail.
In many respects, this method is something of a dog's dinner. A miswired cable could reverse the polarity of the DC, and because the DC is between the two signal lines, it may damage dynamic or ribbon microphones. It can only supply 33mA into a short-circuit, but that can still be enough to cause problems with sensitive microphones. None of this is helped by the fact that some T-Power mic connectors were wired in reverse to suit older Nagra tape recorders which used positive earth/ ground (almost certainly using germanium transistors).
A typical T-Powered microphone may still work (more-or-less) normally if the cable shield is open-circuit, potentially leading to a recording that is found to be unusable when played back in a studio. The ability to keep working with an open shield is 'admirable' in a way, but it can't be considered desirable .
Mics used with computer sound-cards do not use phantom power in any traditional sense. The tip of the 3.175mm (¹/ 8 ") TRS (tip, ring & sleeve) mini-jack plug is the signal from the electret mic capsule, and the ring is connected to the 5V supply by a resistor to power the capsule itself. The tip and ring are usually shorted at the microphone end. This is the most basic of all connections, and it powers only the mic capsule. There are no other electronics involved in almost all cases.
The wiring and operation of these very basic interfaces aren't covered here, as they are irrelevant to the topic. If you want to know more, I suggest a web search, or you can read the Electret Microphones - Powering & Uses article on the ESP website. This article also covers many of the topics here, but with less detail. Some computer mic sockets are stereo, with the tip being 'Left', the ring is 'Right' and the sleeve is ground. There is often no way to know for sure how your computer mic interface is wired without taking measurements. It might be in the manual, but I wouldn't count on it.
The mic input can often double as a 'line' level input, so you'll need to get into the software that controls the 'mic in' and 'line out' connections to set it up the way you need it to be. This also happens with many external interfaces - until you get the settings right in the sound controller, you won't get the results you need. This setup is outside the scope of this article. and is not covered here.
If you are designing your own phantom-feed mic preamp, you may be tempted to increase the available output current to suit a particular piece of gear you wish to power. In a word, "DON'T" . There is a risk that doing so may damage other gear, and it becomes non-standard. There are good reasons for keeping to the standard resistances and voltages, because custom 'solutions' are not compatible with commercially available (and ubiquitous) equipment.
Although it's generally accepted that dynamic microphones will function normally if P48 is applied (usually by accident), it should be disabled. There's always a remote possibility that leakage paths within the mic may cause nasty noises between the voicecoil and mic housing, and it serves no purpose. Some microphones are very sensitive to external voltages, particularly ribbon mics. If they use a transformer to step up the voltage they might be alright, but good practice demands that phantom power is used only when it's needed.
Operators and installers should make it a habit to ensure that mics are not connected with P48 turned on. When a mic is connected to a 'hot' mic lead, large transients can be created that place both the microphone and the mic preamp at risk [ 2 ] . A long cable with the conductors (and mic preamp coupling capacitors) charged to 48V can deliver a considerable peak current, which may be more than enough to damage the mic, mic preamp, or both.
Although you may see dissent elsewhere, if a P48 circuit shorts the two signal lines to ground, the worst-case current is 14mA (7mA for each signal conductor), which will cause the 6.8k resistors to dissipate less than 400mW. The resistors will get quite hot, and it might be sufficient to affect their tolerance, but there's little or no evidence to indicate that it's actually a problem. Normal dissipation when supplying 5mA on each signal lead (10mA total) dissipates 170mW in each resistor, and leaves 14V available for the remote electronics. This is generally considered 'optimum', and many lower-current mic capsule amplifiers use a zener diode to limit the internal working voltage.
While T-Powering is rare, you need to be aware of it, especially if working in the film industry. While most systems now will use P48, there's still a possibility that you'll come across it. One would hope that such mics would use a Tuchel/ DIN connector to ensure they couldn't be connected to P48, but many T-Powered mics used XLR, which is most unwise.
Phantom feed, and 48V in particular, has been with us since 1966. It initially replaced separate power supplies for microphones, and has proven itself to be invaluable for minimising stage and studio clutter. It provides a simple, safe and convenient way to power remote electronics. Microphones remain the most common P48 powered devices, but DI boxes and even piezo pickups for various instruments can be powered just as easily.
Although some manufacturers have dallied with P24 (and even fewer with P12), P48 remains the dominant phantom power scheme. Even manufacturers of USB microphone interfaces for PC sound recording have realised that if they are going to provide phantom power, it has to be 48V, because many microphones simply don't work with lower voltages. It requires little extra effort to provide the full 48V supply, and it means that compatibility with popular professional microphones is assured.
There are a few things that users need to adjust to (such as making all connections before turning on the P48 supply), but even if you forget, no harm will come to equipment that's been designed properly. Yes, you'll get very loud noises through the monitors if the fader happens to up, but that's a lesson quickly learned. P48 is so common now that few people involved with audio will be unaware of it, even if they don't know how it works.
This article is fairly comprehensive, and there's also a vast amount of additional information available on-line. However, finding it isn't always easy, and not all writers manage to get their facts straight. Finding info that's technically accurate can be a minefield, as anyone can write on-line articles, and not everyone gets it right. Fortunately, P48 has been around for long enough that most of the info you find will be fairly close to reality, but there are still some misconceptions and/ or errors.