weihfool wrote on Wed, 26 May 2004 16:47 |
There is a radio tower a few miles from where I'm setup and the interference isn't constant. It comes and goes. I don't have it at all with dynamic mics. It's only there with condensers (as you mentioned earlier). Is there ANYTHING I can do to combat this, barring moving my studio's location? It seems as if the radio transmitter is cranking up the power only at certain times of the week. Thanks again. -Sean |
josh wrote on Tue, 18 May 2004 12:44 |
The first part of the discussion is background. The type of electrical interference noise you hear in a recording studio is caused by one of two phenomena: magnetic fields and electric fields. Magnetic fields result from high currents flowing through a conductor loop which has some significant inductance. |
Quote: |
Similarly these fields are picked up by conductor loops with sigificant inductance and low impedance, such as a "ground loop" or a guitar pickup. |
Quote: |
Magnetic fields' intensity decays at a rate inverse to the cube of the distance, so doubling the distance from the emitter to the reciever results in 1/8 of the field intensity. |
Quote: |
You can think of this as an "inductive" transfer of energy, since magnetic fields couple from one "inductor" to another (such as the intended operation of a transformer). |
Quote: |
Transformers, inductors, magnetic transducers (such as a dynamic microphone, tape head or guitar pickup) utilize magnetic field coupling as their intended mode of operation. |
Quote: |
Electric fields result from high voltages with little or no current flowing in typically unterminated lines. |
Quote: |
Electric fields' intensity decays at a rate directly inverse to the distance, so doubling the distance from the emitter to the reciever only cuts the intensity by half. |
Quote: |
You can think of this as a "capacitive" coupling where air is the dielectric of the capacitor through which the current flows. |
Quote: |
Now a quick touch on Ohm's Law. Ohm's Law states that R = E/I where... various algebra can be performed, but basically you can see that impedance can be defined as the ratio of voltage to current. This is important in determining whether we are dealing with a magnetic field problem (low impedance), or an electric field problem (high impedance). |
josh wrote on Tue, 18 May 2004 13:02 |
So is it magnetic or electric field that's giving you grief? Do the math. First a constant. The impedance of air is 277 ohms. For electromagnetic energy purposes, below this threshold is "low impedance", and above this is "high impedance". A tuned antenna will operate at nearly exactly 277 ohms. |
Quote: |
So to find the source impedance of your problem 60 cycle field, you divide E by I, or divide 120 by 12 and you get 10 ohms. Since this is way below 277 ohms, you know for sure that your problem is magnetic coupling. In fact, if you look at 60-cycle hum, for it to be an electric problem, then you would have to have an impedance greater than 277 ohms. For this to be the case, the total power load on your AC power line would have to be less than about 52 watts. Just turning on a table lamp in most studios will exceed this and ensure that your 60 cycle hum problem is indeed magnetic and not electric. |
josh wrote on Tue, 18 May 2004 14:44 |
OK, so shielding in an audio cable... pin 1 of an XLR is, as j. hall stated, connected to the shield at least at one end, and perhaps at most at one end. This is classic shielding function. Let's look at it in two scenarios: 1. Microphone cable: In a microphone cable the shield is connected to ground and serves as the return for the phantom power supply. If there is no phantom power, the shield is typically electrically floating, or attached to the microphone body. This is in fact shielding the way it's intended to work. It will reduce RF interference of the electric field variety (you can hear WKRP coming in on the mic). |
Quote: |
There is no real opportunity to create a ground loop and pick up magnetic interference with this type of shield. |
Quote: |
2. XLR interconnect at +4dBu: In this case, the shield may be connected to ground at one or both ends and does not (intentionally) have current flowing. Ideally, in classic shielding design, the shield is connected only at the source end and is left hanging at the load. So inputs would float the shield, outputs ground it. This is a ground-loop prevention scheme. |
Quote: |
If everything (including the converters) are connected to common (earth) ground on pin 1, then we have a huge bundle of ground loops. |
Quote: |
The most effective way to reduce hum in the studio is as follows: 1. avoid ground loops by making sure there are not continuous ground connections between different pieces of gear |
Quote: |
2. control ground currents and loop area especially on high-powered devices. This will lower the amount of magnetic field intensity you have to isolate (treating the problem at its source). Reducing the length of a power cable going to a big power amp will reduce the circuit loop area and reduce the intensity of the magnetic field (hum) it produces. |
Quote: |
3. proximity. Locate your power supply circuits and high-power devices in one area, and locate your susceptible circuits in another area. Power distribution design can really affect how much "hum" you pick up, simply by locating wiring and circuits that share return currents in a way that isolates them physically from susceptible circuitry. For example, the way NOT to do this would be to route your XLR cables and power cables in the same cable raceway. |
ebeam wrote on Fri, 21 May 2004 16:20 |
So, I have the output of my DAW (Lynx Two) connected to my homemade volume control/switching box (basically a pot and some switches...) using XLR to TS with the shield floating on the TS side, pin 2 to tip and pin 3 to sleeve. |
Quote: |
So then for my outputs I decided to use the same twisted pair cable (actually star quad) to run out to my amp, TS to TS. |
Quote: |
When I first wired this I must have been half asleep because I left the shield floating on both ends and just used the twisted pair. This caused major hum and when I touched the cable ends or switches it went away indicating a grounding problem. |
Quote: |
Also, when I looked in the manual for my amp (Hafler P3000) they say the shield should be tied to pin 1 for both sides when using XLR to prevent 'instability'. Should I do this or just screw the extra conductor and just use one with the shield or what? |
weihfool wrote on Wed, 26 May 2004 11:47 |
Hey Josh, Thanks for being part of this thread. Very, very informative. I have a question and if this is something you were going to tackle later on, then I'm sorry for jumping the gun. I'm experiencing the electric interference you spoke about. There is a radio tower a few miles from where I'm setup and the interference isn't constant. It comes and goes. I don't have it at all with dynamic mics. It's only there with condensers (as you mentioned earlier). Is there ANYTHING I can do to combat this, barring moving my studio's location? It seems as if the radio transmitter is cranking up the power only at certain times of the week. Thanks again. -Sean |
gtphill wrote on Thu, 27 May 2004 01:41 |
Part of your problem is that you have unbalanced your setup by the use of T/S cabling. |
Quote: |
The extra cable capacitance of star quad is not worth it for your unbalanced interconnects. |
Quote: |
If the shield is left unterminated, is can be possible to create an internal amplifier feedback loop, where very high frequencies on the speaker output cables couple back into the signal inputs, resulting in ultrasonic oscillations. It is rare, but known to exist. Connecting the shield provides a shunt for most of this. |
Quote: |
Start by reading Rane Note 151: http://www.rane.com/note151.html Then read Rane note 110: http://www.rane.com/note110.html Then properly balance your setup utilizing 1:1 isolation transformers and proper cabling. Then go to AES.org and download Preprint 5747, "Common-Mode to Differential-Mode Conversion in Shielded Twisted-pair Cables (Shield-Current-Induced-Noise)" Finally, build the device in the posted figure and test all of you equipment to see if it is Pin 1 compliant. |