Home -> Magazines -> Issues -> Articles in this issue -> View
Studio Earthing Techniques (Part 2) | |
InterconnectArticle from Home & Studio Recording, October 1984 |
Studio earthing techniques: Part Two.
In this part, Ben Duncan reveals the role of balanced connections and ground compensation in overcoming hum loops, and shows how the various techniques figure in a typical studio set-up.
At mic level, balanced lines are used solely to keep induced hum and interference at bay; there's no ground loop connections to worry about - unless your microphone stands are grounded! But for equipment interconnection, balanced connections have another advantage, in that there's no direct interconnection between signal grounds.
Looking at Figure 1, you'll see that the cable screen is linked to the chassis at either end. If both chassis are tied to their respective signal grounds, and also to the mains earth, bingo! - a hum loop. Note that the screen wire here isn't needed to convey the signal; it's acting solely as a gutter, down which induced interference nasties flow. And a gutter doesn't need a drain pipe at each end, provided it's not blocked up with dead leaves ie. has a low resistance! Thus we can easily stop the hum loop by ditching the screen connection at one end. On an XLR connector, this means breaking the pin 1 connection; on a stereo jack, the sleeve contact.
We can do this at either end of the interconnection, but the break is usually made at the output end ie. at the source equipment, so that the screen wire's potential refers most closely to the differential input's reference point. This is necessary for best radio frequency (RF) interference rejection. Equally useful, it prevents the succeeding equipment from screeching off into the stratosphere if you should accidentally unplug the cable at the output end... whoops!
Standard cables are vital to sanity. Today, fewer tempers need be lost if all XLR users stick to universal pin-to-pin leads, with the screen tied to pin 1 (Figure 2). Cables so wired will work in all systems, balanced or unbalanced, whilst for US-UK equipment interconnection, you just shove a phase-change lead on one end. This gizmo features male and female line Cannons with pins 2 and 3 cross-linked. It's commonly made about 6 inches long, so it's visually distinct from ordinary pin-to-pin cables, and can be tucked safely away in a drawer when out of use.
And so back to breaking the pin 1 link: having established the advantage of standard pin-to-pin XLR leads, we'd better not complicate matters by introducing special leads with the pin 1 connection unhooked at one end. With this in mind, the pin 1 connection is invariably chopped off inside the source equipment. A few, civilised manufacturers anticipate this, and provide pin 1 isolation switches adjacent to line level output sockets, or oddly, inputs. For example, see our Yamaha amp review, HSR February 84. If you think about it, pin 1 isolation is inevitable at the inputs on a power amplifier, so it's good practice to turn down the gain before patching leads in these situations. In fact, it's good practice outright, and can save a lot of bent needles and blown speaker cones....
Looking now at the broad picture, fully balanced systems relinquish the need to worry about isolating the mains earth/chassis from the signal ground. If there's a hum loop at large, we can still lift the grounds if we wish (see last month), but it's often far quicker and easier to simply isolate pin 1.
Thanks to Ted Fletcher and other pro-audio doyens, we don't need to go to the lengths of a fully balanced system nowadays, at least in a relatively small studio set-up (cf. a broadcast installation with thousands of interconnects). The general idea is to simplify and lighten.
Looking back at Figure 1, you'll see that the source feeds the cable via a transformer. Multiply £12 - £20 (for a decent one) times a few dozen line interconnections, and you'll soon see why full balancing is the province of bigger and wealthier studios. 'Transformerless' balanced outputs do, of course, exist, but they're not that much cheaper, and have their own share of problems. For instance, simple balanced inputs of the 'electronic' or active variety show a different impedance on pins 2 and 3, and this can upset the balanced output circuitry.
The complete solution - well, almost - is to do away with balanced outputs altogether. Figure 4 shows an ordinary, unbalanced output, driving straight into a balanced input. This can be one with a transformer, or equally, an active ('electronic') input. The series of three interconnections emphasises the staircase grounding pattern, a form of ground compensation. With this system, it's not just optional, but essential, to connect the cable screen at the input end only, as shown. This is so even if there's no hum loop. Otherwise, you'll end up unbalancing the input, thereby throwing away the interference elimination properties of the balanced line at the same time. As before, this is usually achieved by chopping out the pin 1 connection on each chassis-mounted output socket.
The only proviso with this set-up is that differences between the ground potentials must be kept to a minimum. The error voltage shown between the grounds of processors 1 and 2 is seen by the input as a common mode signal. This means you can't hear it (because along with other interference, it's cancelled out), but it can still overload the input. All this is apt to be baffling - you'll wonder why any signal of whatever level you feed in comes out gruesomely distorted. This happens when the legitimate audio tries to ride on top of a high level common mode error signal.
To avoid any such error voltages from developing, all we need do is tie all the chassis grounds, and hence mains earth connections, to a common point. Ideally, this means connecting all the equipment to a common mains outlet, or a pair of adjacent outlets on the same circuit. In a small set-up, this is usually no problem. In a larger control room, there may be additional ring mains, or outlets may be wired individually back to the central fuse-box. The latter won't normally prove troublesome, but you should certainly stick to a single ring mains circuit for the complete system, even for equipment mounted in other rooms. If you're not familiar with modern permanent mains wiring practice, it's really best to consult a friendly electrician about any proposed modification to the mains earthing pattern between outlets. Rather this, than prejudice your personal safety and insurance.
The big joy of the unbalanced-to-balanced interconnection (Figure 4) is that it's readily applicable to existing unbalanced equipment, providing you're up to making simple DIY modifications. All we need do is add a differential input to any unbalanced connections which are causing hum loop hassles, though real enthusiasts will want to 'balance up' every input.
Figure 3a shows a simple but effective DIY differential input circuit, easily built up on Veroboard. Incidentally, it's for line interconnections only - mic inputs are a different kettle of fish (See HSR August 84 p.58). If you'd rather avoid modifying your equipment, and finding suitable powering points therein, I recommend you mount the circuit in a diecast box, placing it close to the associated input (Figure 3b). Remote powering can be +/-12V from the HSR rackmount unit, or similar. This concept is, of course, the basis of the various active isolation or 'ground compensation' boxes available from pro-audio accessory manufacturers.
The voltage across the ends of any conductor is proportional to the current flowing and the resistance of the conductor. So high currents correlate with high resistances and/or high voltages. Here, we're not talking about hundreds of volts, but the few hundred millivolts that arise across a cable having a few hundred milliohms resistance. These sort of levels across line connections are quite enough to cause a deafening hum. Given that some resistance is inevitable (short of dunking all our cables in liquid helium!), interaction between currents can be kept at bay by separating out common connections as much as possible.
If this sounds contradictory, let's pause to reflect that the concept of common connections is really saying 'All these wires share a common reference'. By definition, a reference is dead solid, unwavering, a deck plate, so it shouldn't have any appreciable resistance. Once we start using wires as part of the reference, their resistance and inductance makes certain that the various currents all get jumbled up. Figure 5a shows a series grounding technique that's disastrous for just this reason. Preventing this additive interaction is what the famous star earth connection is all about (Figure 5b). Here, the wires are kept to themselves until the last moment, where they reach the central ground, which should feature a very low impedance. The star earth (No Star Wars today, Mr President) is an invaluable 'first principles' technique whenever you're landed with sticky grounding problems. It's especially useful if the symptom is one of HF (high frequency) instability and oscillation, rather than a simple hum loop alone.
The DI box is the everyday remedy to the hums that come when musicians plug their amps directly into the mixing desk. Low cost DIs feature a passive attenuator plus transformer, and may not be ideal for interconnecting at line level. Upmarket types, however, with either active electronics plus a transformer, or the wholly active variety (a variant on the ground compensation theme in Figure 4) can be usefully brought in to sort out hum loops within the control room interconnections.
As a permanent solution though, this option ties up useful gear, and may limit ultimate performance.
In any studio, it's useful to develop a house technique for quickly tracking down the cause of a hum. Simple tools include a 'U/G' lead - a mains lead with the earth wire disconnected. This should only be used for temporary checks as it's technically UNSAFE. Placing a U/G lead in series with various equipment's mains supply will tell you whether the hum can be tied down to a specific unit, and if so, which unit. But let the investigator beware of hum caused by several mains earths simultaneously!
Other tests involve unplugging input and output cables, and working through the mixing desk's mute buttons and faders in a logical sequence. When the hum stops, the lead you are holding is one of the two responsible. You should also check that hums don't disappear when the mains power is switched off in a specific unit, but the 13 Amp plug is left in place. In this event, the hum is nothing to do with ground loops at all - it's there by induction. The cure is to move the offending unit elsewhere.
Short of a full logical analysis (that's testing all 100 or so studio cables in turn, methodically against the other 99, a long task!) hum loops are found by intuitive thinking and associated shortcuts. These you can only achieve with practice. After a while, you'll be able to pinpoint the guilty source by unplugging half a dozen leads and plugs. This adds new meaning to the phrase 'green fingers'. Good luck.
Symptom | Problem | Solution |
---|---|---|
(1) Loud buzz on channel 6 fader. | Synth has bonded ground, and the mains earth is on a different ring mains. | Synth modification not feasible, so add a DI box. Or try using an extension mains cable, back to the mains outlet. |
(2) Quiet but edgy background buzz at all times. | Monitor amp's chassis is bonded. | Isolate signal ground from amplifier chassis, or add a balanced input, and break pin 1 connection. |
(3) Hums when DBX fed into mixer insert return. | Insert return is unbalanced and DBX has a bonded chassis, and is also also rack-mounted. | Modify DBX, or use aux return, which is balanced, or a channel input, (assumed to be balanced). |
(4) Desk distorts, oscillates and buzzes when DI and Mic tie lines are in use. | Ground wires are paralleled. Mic current is 1000 times smaller than DI feed current flow. | Separate out the ground/screen wires in the patchbay. Keep high and low level interconnections totally separate. Use separate patchbay if necessary. |
This is the last part in this series. The first article in this series is:
Studio Earthing Techniques
(HSR Sep 84)
All parts in this series:
Part 1 | Part 2 (Viewing)
Digital Sampler/Delay (Part 1) |
Polarity Unveiled |
Sourcing The Bottle - Valves |
Analogue Equipment Design - for Rock 'n' Roll (Part 1) |
Workbench - Sounding Out |
Lab Notes: The Polyphonic Synthesizer |
Assaulted Batteries |
Understanding Electronics - How To Make Music Projects |
Data To Music Conversion - A Low Cost Peripheral Circuit For Your Computer |
Technically Speaking |
Adding an Independent Tracking Output to the 4780 Sequencer |
How It Works - Drum Machine |
Browse by Topic:
Topic:
Series:
Part 1 | Part 2 (Viewing)
Feature by Ben Duncan
Previous article in this issue:
Next article in this issue:
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!
New issues that have been donated or scanned for us this month.
All donations and support are gratefully appreciated - thank you.
Do you have any of these magazine issues?
If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!