In PCB design I every time feel fear in my power trace included high impedance section or not? I've shocked on about 6k resistance putted on +48v of phantom power. I've heard this large resistor is just exist to prevent short circuit or misswiring disaster. What if voltage source be short-circuit/over-heat protected? Then:

  1. The only reasonable problem caused by low impedance supply is to kill audio signal, in other word it imposes high load to MIC. Then why not replace resistor with inductor?

  2. If phantom power can supply MIC and it's preamps well then including resistanced in power supply line must not be a catastroph, but really can prevent catastroph (ie. IC breaking and becomming short circuit). Then is there any rule to include resistor in our circuit design without any performance loss of supplied circuits?

  3. Is it inevitable that by using phantom power mixer input impedance have to degrade to 3k? Why this shoud be just CM load and not the diff load, if it is?

  • 1
    \$\begingroup\$ It's really 3.4K, as the 6.8K on each leg is in parallel. You also forgot the other reason it's high : it provides a 13.6K load to the microphone, in parallel with whatever load the mic amp input stage provides. \$\endgroup\$ Jul 3 at 16:13
  • 1
    \$\begingroup\$ So are you asking why both XLR wires have 6k8 to 48V, and can it be 0 ohms if supply is short circuit protected? \$\endgroup\$
    – Justme
    Jul 3 at 16:37
  • 1
    \$\begingroup\$ IEC 61938 is the current standard that applies to phantom power of microphones. The 6.81k resistors are specified in there. The specified source voltage is 48V +/- 4 V or 44-52V. These two specs set the nominal short-circuit current to be 48 / (6.81k || 6.81k) = 14.1 mA. Though it will vary a few mA higher or lower, because of the resistor and voltage tolerance range. In real circuits I've seen, the design uses 0.1% tolerance resistors, and the 48V is well-regulated to within 1%, typically by a LM317HV regulator IC. \$\endgroup\$
    – Rich S
    Jul 3 at 19:23
  • \$\begingroup\$ @Justme Partially yes, and it can't be 0 ohm since in this way phantom will kill the audio signal. And this is a trade-of we are forced to do it because of lots of limitation by using phantom power. My ques rises if phantom power can completely power MIC and preamp stages, using high resistance in series with voltage supplies. Resistive voltage supply must not be a catastrophe, maybe we can eliminate the effect of resistor by adding capacitors or ... but does such rule exist to design our circuits without fear and just follow the rule to easily add high voltage source resistance in our design. \$\endgroup\$ Jul 4 at 7:09
  • \$\begingroup\$ @user_1818839, Yes, then it's so bad since we must forgot high input impedance, in case of switching phantom power on. But why not using inductor instead? \$\endgroup\$ Jul 4 at 7:18

2 Answers 2


There are typically many mics that can be powered from the same supply - could be 20, 30 or more. If one cable has a short, only that mic will be affected, because of the resistors. The others will function as normal.

The +48V source is always protected against excess current anyway (at least in designs that I have seen, and it would be a bad oversight for it not to be so) but this will not stop all mics being affected by one bad cable, if the resistors were not there.

The current used by the FET preamp in a typical condenser mic is tiny, also there is decoupling within the preamp. These resistors have no effect in practice. This technique is well established and has been in use in professional systems for over 50 years.

(As an aside, it's also common in analogue mixers to use +/-18V for the main audio supply, typically a big brute that can supply many amps, and then use small - about 10R - "fusing" resistors on each module, which will burn and give an awful smell if one module develops a short across the rails.)

  • \$\begingroup\$ Upvote for valuable experince sharing. What about +/-18 volt? deriving it from +48 on MIC side? or in mixer side put - to cold and + to hot? and gnd to shield? \$\endgroup\$ Jul 4 at 8:41
  • 1
    \$\begingroup\$ no, you will usually have separate +/-18V and +48V, all coming from linear regulators. do a google search on (for instance) "Soundcraft PSU Schematic" and you will find lots of examples. \$\endgroup\$
    – danmcb
    Jul 4 at 9:25
  • 1
    \$\begingroup\$ here is a typical one : content.invisioncic.com/m315360/… --- note the use of two linear regulators for +/-17V (one is used upside down, so its + terminal goes to 0V) and a separate high voltage regulator for +48V. \$\endgroup\$
    – danmcb
    Jul 4 at 9:28
  • \$\begingroup\$ I think I've missunderstood, +/-18V for the main audio supply was not about the supply transferred to MIC, it imply main supply in mixer, doesn't it? \$\endgroup\$ Jul 4 at 10:28
  • \$\begingroup\$ yes, indeed. That is used to power the many internal opamps which need +/- balanced supplies, and maybe other circuits. Does not generally go outside the mixer. \$\endgroup\$
    – danmcb
    Jul 4 at 12:38

The choice of phantom power resistor depends on the load and supply voltage. It serves two purposes;

  1. low pass filter into preamp supply capacitor
  2. current limiter for transient shorts.

The voltage drop across the resistor, I would expect to be about 10% of the supply. It is not critical.

I might suggest not soldering while power is on, yet it is safe for fingers to touch low voltage < 50V.

  • \$\begingroup\$ your meaning from load is that: load applied to the microphone? Doesn't it better to use inductor instead? \$\endgroup\$ Jul 4 at 8:42
  • \$\begingroup\$ Low R Series LC can be resonant, so RC is better \$\endgroup\$ Jul 4 at 21:04
  • \$\begingroup\$ Then all LC filters and all LC tank included circuit, except for oscillators must be prohibited since maybe they resonate, doesn't it? \$\endgroup\$ Jul 9 at 10:23
  • \$\begingroup\$ Not true.......read again \$\endgroup\$ Jul 9 at 11:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.