Welcome to No Limit Sound Productions

Company Founded

Our services include Sound Engineering, Audio Post-Production, System Upgrades and Equipment Consulting.
Our mission is to provide excellent quality and service to our customers. We do customized service.

Saturday, June 30, 2018

Q. Should I use my mixer's group outputs or its direct outs for recording?

By Mike Senior
Like other mixers, this Allen Heath GL2400-424 offers both direct outs on channels and group outs. But which should be used and when?

I recently started teaching music technology in a college and was asked to rebuild one of the studios. It uses a 32-channel mixing desk, patchbay and Alesis HD24 hard disk recorder to record to, as well as outboard gear. The desk has eight group busses arranged in four stereo pairs. There are 24 mono group output sockets, three per group buss, so that group 1 goes to outputs 1, 9 and 17, group 2 goes to 2, 10 and 18, and so on. The way it was set up previously was that these 24 group outputs were normalled through the patchbay to the 24 inputs on the HD24. The students were being taught that the signal should come into the desk and then be routed through the relevant group to get to the HD24. For instance, if your mic is plugged into channel 3 and you want to go to track 5, you have to route it to group 5-6, pan it hard left and bring up the channel fader and group fader. However, I changed it so that the direct outs of the first 24 channels are normalled through to the 24 inputs of the HD24, which seems to make more sense. One of the lecturers is kicking up a fuss, so my question is: which practice is most common in professional studios?

Thom Corah

Reviews Editor Mike Senior replies: You're both right after a fashion, but I'm afraid that I think the lecturer is probably more right in this case, as you appear to be using a group desk, rather than an in-line one. Your approach has two main limitations. Firstly, you can only route channel 1 on the desk to channel 1 on the recorder. This is admittedly less of a limitation with a digital recorder, where you can swap tracks digitally, but it's still quicker to do this from the desk than from the recorder.

The second (and more serious) limitation is that you can't record a mix of several channels to the same track on the recorder. Although 24 tracks is quite a lot to work with, you might need to submix a number of microphones to, say, a stereo pair of tracks — for example, when layering up a string quartet a few times to make a composite string sound for a pop production. Another problem is that you can't use the mixer's EQ on the way to the recorder, as direct outputs are often taken from before the EQ circuitry. Also, you couldn't bounce down a group of tracks through the desk in this way without sending them all to a group first, and then patching from the group output to a further channel. So you'll have more flexibility if you do things the lecturer's way.

One reason that you're not completely wrong is that you're implementing a kind of in-line methodology, treating the input stage up to the direct output as the input path and the rest of the channel as the monitor path. However, a group desk isn't really sufficiently well equipped to do this properly, most notably because there is no routing matrix between the input channels and the recorder inputs, as there would be on an SSL desk or similar. There's only one routing matrix per channel on a group desk, and that is situated after the channel fader. There's no real alternative, given the facilities, but to have separate channels for the input and monitor paths. In your case, as you have only 32 mixer channels, this means repatching for mixdown and monitoring purposes, I imagine, but I don't know all the details of your setup.

One situation where you can get away with using an in-line configuration on a group desk, exactly as you have, is where the recorder is actually a computer system. In this case, given the powerful processing facilities a computer offers, there's little advantage these days in pre-processing audio before it reaches the computer, so the lack of input EQ would not really be a problem. Also, there are comprehensive input routing and mixing facilities built into most modern audio-recording packages, so a hardware routing matrix would also be unnecessary. Perhaps you could justify your routing scheme as just being a little ahead of its time? You are simply anticipating the happy day when the college moves to a more flexible computerised system!

At the end of the day, which is the more appropriate arrangement depends on how many tracks you plan to record at one time. The group routing approach is more flexible when it comes to being able to do track bounces and partial submixes, and it is an important way of working to teach students. However, the down side is that you can record no more than eight (different) tracks at a time because there are only eight groups on your mixer.

Taking the direct outs approach allows up to 24 different tracks to be recorded at the same time and is ideal in areas designed purely for tracking, but you are then in for lots of replugging when it's time to mix. In any case, students should definitely be made aware of both techniques and configurations.
One possible solution that you could consider is using the patchbay to normal the group outputs to the recorder inputs, as before, but also send all of the desk's direct outs to patchbays on the row above, so that when you need to patch direct outs straight into recorder tracks it's just a case of plugging in some patch cords.

Published January 2006

Thursday, June 28, 2018

Q. What determines the CPU reading in Cubase SX?

By Martin Walker

I remain baffled by the CPU load in Cubase SX 2 (as shown in the VST Performance indicator). I'm particularly curious to know why in my larger projects the indicator shows a constant load (typically 80 percent or more) even when I'm not playing anything back! What exactly is the CPU doing when nothing is happening in the project? My projects typically have 15 to 25 audio tracks, five to 10 virtual-instrument tracks and a couple of MIDI tracks, with five or so group channels and maybe a couple of FX Channels. Some of the channels have an insert effect or two, typically a compressor or gate, and there's a couple of aux channels for send effects.

SOS Forum Post

PC music specialist Martin Walker replies: When Cubase isn't playing back, the CPU overhead is largely down to the plug-ins, all of which remain 'active' at all times. This is largely to ensure that reverb tails and the like continue smoothly to their end even once you stop the song, and it lets you treat incoming 'live' instruments and vocals with plug-ins before you actually start the recording process. However, this isn't the only design approach — for instance, Magix's Samplitude allows plug-ins to be allocated to individual parts in each track, which is not only liberating for the composer, but also means that they consume processing power only while that part is playing.

Freezing tracks, adjusting the buffer size and using single send effects instead of multiple inserts can all help reduce CPU overhead. 
Freezing tracks, adjusting the buffer size and using single send effects instead of multiple inserts can all help reduce CPU overhead.

Of all the plug-ins you'll be using frequently, reverbs are often the most CPU-intensive, so make sure you set these up in dedicated FX Channels and use the channel sends to add varying amounts of the same reverb to different tracks, rather than using them as individual insert effects on each track. You can do the same with delays and any other effects that you 'add' to the original sound — only those effects like EQ and distortion where the whole sound is treated need to be individually inserted into channels.

The other main CPU drain for any sequencer when a song isn't playing back comes from software synths that impose a fixed overhead depending on the chosen number of voices. These include synth designer packages such as NI's Reaktor and AAS's Tassman, where their free-form modular approach makes it very difficult to determine when each voice has finished sounding. However, fixed-architecture software synths are more likely to use what is called dynamic voice allocation. This only imposes a tiny fixed overhead for the synth's engine, plus some extra processing for each note, but only for as long as it's being played.

If you use a synth design package like Reaktor or Tassman, try reducing the maximum polyphony until you start to hear 'note-robbing' — notes dropping out because of insufficient polyphony — and then increase it to the next highest setting. This can sometimes drop the CPU demands considerably. Many software synths with dynamic voice allocation can also benefit from this tweak if they offer a similar voice 'capping' preference.

Anyone who has selected a buffer size for their audio interface that results in very low latency will also notice a hike in the CPU meter even before the song starts, simply due to the number of interrupts occurring — at 12ms latency the soundcard buffers need to be filled just 83 times a second, but at the 1.5ms this happens 667 times a second, so it's hardly surprising that the CPU ends up working harder. For proof, just lower your buffer size and watch the CPU meter rise — depending on your interface, the reading may more than double between 12 and 1.5ms. You'll also notice a lot more 'flickering' of the meter at lower latencies. If you've finished the recording process and no longer need low latency for playing parts into Cubase, increase it to at least 12ms.

Finally, if some of those audio or software synth tracks are finished, freeze them so that their plug-ins and voices no longer need to be calculated. Playing back frozen tracks will place some additional strain on your hard drive, but most musicians run out of processing power long before their hard drives start to struggle.

Published February 2006

Tuesday, June 26, 2018

Q. What is the difference between mono with one speaker and mono with two?

By Hugh Robjohns
I read recently that when top engineers check their mixes in mono, they don't just hit a mono switch, but instead route the mix through a single speaker to hear it in true mono. What's the difference between the two?

A single speaker in a sealed enclosure is the classic means of monitoring in mono. 
A single speaker in a sealed enclosure is the classic means of monitoring in mono.

SOS Forum Post

Technical Editor Hugh Robjohns replies: It's important to check the derived mono signal from a stereo mix to ensure that nothing unexpected or unacceptable will be heard by anyone listening in mono, as could be the case in poor FM radio reception areas, on portable radios, in clubs, on the Internet and so on. Mono compatibility, as it's called, is very important for commercial releases — the artist, producer and record company want the record to sound as good as possible in these less-than-ideal circumstances.

In addition to simply checking the finished product, mixing in mono — or regularly switching the monitoring to mono while mixing — is very useful and a good habit to get into. Summing to mono removes any misleading phasing between the left and right signals that can make a stereo mix sound artificially 'big'.

The crucial difference between auditioning the summed mono signal on a single speaker, as compared to a 'phantom' mono image between two speakers, relates to the perceived balance of the bass end of the frequency spectrum. When you listen to a mono signal on two speakers, you hear a false or 'phantom' image which seems to float midway between the speakers, but because both speakers are contributing to the sound, the impression is of a slightly over-inflated level of bass. Listening to mono via one speaker — the way everyone else will hear it — reveals the material in its true form!

Checking the derived mono is always best done in the monitoring section of the mixer or with a dedicated monitor controller. Although a mono signal can be derived in the output sections of a mixer (real or virtual), this is potentially dangerous — if you should forget to cancel the mono mixing, you'll end up with a very mono final mix. It does happen, believe me! Sadly, very few monitor controllers outside of broadcast desks and related equipment provide facilities to check mono on a single speaker. Most provide a phantom mono image, which is fine for checking imaging accuracy and phasing issues, but no good for checking the mono balance.

Published November 200

Saturday, June 23, 2018

Q. Should I opt for active or passive monitors?

By Hugh Robjohns
One advantage of passive monitors is that the two components of your monitoring system — the speakers and the amp — can be upgraded separately, allowing a more gradual and less expensive progression to better-quality gear. 
One advantage of passive monitors is that the two components of your monitoring system — the speakers and the amp — can be upgraded separately, allowing a more gradual and less expensive progression to better-quality gear.

I'm interested in buying a pair of Alesis Monitor 1 MkIIs. Should I buy the passive versions and a good amp or just go for the active versions, which cost £100 more? I've always thought that active monitors are a bit of a gimmick and don't give a good sound, but I have now been told that they will give the best sound, as there is no crossover. Can you help me?

SOS Forum Post

Technical Editor Hugh Robjohns replies: In the middle and upper parts of the monitor market there is no doubt that active models offer significant advantages over passive designs, such as optimised power amps for each driver, optimised driver-protection circuitry, short and direct connections between amps and drivers, more complex and precise line-level crossovers, and so on.

However, at the budget end of the market these advantages are somewhat clouded by the inherent problems of achieving a low sale price. Most notably, many models are saddled with poor-quality power amps and power supplies that have been built down to a price rather than built up to a standard. Obviously, I'm painting pictures with a very broad brush here — there are some good and some less good designs out there — but the generalisations are true.

Active speakers come in two forms: true 'active' monitors, which have a separate amplifier for each driver, and 'powered' monitors, which have a single amplifier built into the cabinet, feeding both drivers via a normal passive crossover. In examples of the latter, you often get a better amplifier because you are only paying for one amp and not two (or three, in the case of a true active three-way monitor), while retaining the advantages of having an integrated package with very short internal speaker cables and so on. In the case of a well designed two-way speaker, a passive crossover can deliver superb results, and there is often little, if any, quality advantage from employing a complex line-level active crossover instead.

However, one facility that's easy to implement in active designs with line-level crossovers is user-adjustable EQ tweaks. These can be helpful sometimes in matching the speaker to the room, but in inexperienced hands such facilities can often be more trouble than they are worth because they can be mis-set... and usually are!
Perhaps a more relevant argument against budget active speakers — for me, at least — is the difficulty of upgrading. When the time comes to move up to a higher standard of monitoring, you will have to change both the speaker and its integrated amps. This inherently means that upgrading has to jump in large financial steps. On the other hand, if you go down the passive route you can upgrade the speaker separately from the amp, and vice versa. That approach allows you to improve the quality of the complete system in several easier and more cost-effective stages.

For example, you could start off with the best passive monitors you can afford and a reasonable amp (possibly second-hand — there are plenty on the markets as people switch to the more 'fashionable' active monitors), then maybe upgrade the amp to something that will warrant a better speaker after a year or two, then upgrade the speaker, and so on.

For what it's worth, all my 'little speakers' are passive designs coupled to good quality amps, in some cases with the amps fixed to the back of the speaker to make a 'powered' unit. I have found this approach to provide the best-quality result whilst still being very cost-effective and flexible.

Published January 2006

Thursday, June 21, 2018

Q. Why is the signal louder when it is panned to the centre?

By Hugh Robjohns
Different mixers employ different panning laws.Different mixers employ different panning laws.

When I plug my guitar into my 16-track and send the same signal to two channels, if I pan both channels to the middle it sounds louder than if I pan one all the way left and one all the way right. Surely it should sound the same — if they are both in the middle, the signal is coming through both speakers, and if one is panned left and one right, it's still coming through both speakers. Can you explain what's going on?

SOS Forum Post

Technical Editor Hugh Robjohns replies: Panning laws vary between products, depending on whether they are designed to maintain constant voltage, constant power, or a compromise between the two. The compromise version is probably the most common these days, with pan pots designed to provide something like a 4.5dB attenuation when at the centre. Constant power gives 3dB of centre attenuation, while constant voltage gives 6dB.

If you pan identical signals fully left and right, you have full-level signals in each output channel. However, if you pan the signal to the centre, the left and right outputs will be attenuated by (in the case of the common 'compromise' panning law) 4.5dB. But because you have panned both input channels to the centre, each output channel is receiving two lots of signal, each 4.5dB lower than the level of a single channel panned fully left or right. Since your two signals are identical, they will sum together and the level will rise by 6dB. So we go up 6dB from -4.5dB and find that each output channel is now carrying a summed mix of +1.5dB. Hence, each output channel is now carrying a signal that is 1.5dB higher than it was when you panned the channels individually left and right, so it will sound slightly louder.

For the record, if the mixing desk employed the constant power law, with 3dB central attenuation, the two channels panned centrally would produce an output of +3dB in each channel, while a desk with the constant voltage law would produce an output that was exactly the same level as the fully panned channels (in terms of signal voltage, at least).

The constant power panning law is used where you want a panned signal to remain at more or less the same perceived volume regardless of where you pan it. However, this panning law looks wrong on the desk meters, which only show a constant level if you use the constant voltage law! Hence the halfway-house compromise law, which tries to satisfy the demands of both situations reasonably well.

Published November 2005

Tuesday, June 19, 2018

Q. How do I hook up my reel-to-reel tape machine?

By Hugh Robjohns

I recently purchased a second-hand Tandberg reel-to-reel tape machine and I'm having difficulties connecting it to my external hi-fi. I was provided with a lead that has a five-pin socket at one end and phono leads at the other, which I plug into the 'analogue in' socket on my hi-fi. However, when I'm playing tapes the music only comes out of one channel. The back of the Tandberg has two of these five-pin sockets and also three other holes, marked 'p up', 'amp' and 'radio'. Can you tell me how I can get the sound coming from both speakers and not just one? Any help would be most appreciated by this novice reel-to-reel owner!

SOS Forum Post

Technical Editor Hugh Robjohns replies: There are several possibilities here. The most obvious one is that the DIN-phono lead you have is broken. DIN is the Deutsches Insitut für Normung, a German standards-setting organisation, and it specified a range of connectors using a similar body with between three and 14 pins. The three- and five-pin versions were used a lot on hi-fi equipment in the '60s and '70s, before the RCA 'phono' socket became the standard interface, and now the five-pin DIN is most commonly found on MIDI leads. If you have a test meter, check the connections between the phono plugs and DIN pins to see if the cable is faulty.
The 'standard' numbering scheme for DIN plugs. 
The 'standard' numbering scheme for DIN plugs.

For some bizarre reason, some manufacturers' implementation of the DIN wiring is exactly the opposite of others, so although I am giving the most common way of wiring them up, bear in mind that this is not always the case. The 5-pin DIN sockets were used to convey stereo unbalanced signals. The DIN pins on a male jack are numbered in the order 1, 4, 2, 5, 3, clockwise from right to left (see diagram). Normally, pins 1 and 4 were used for the left and right inputs, respectively, and 3 and 5 for left and right outputs, with the middle pin of the five (pin 2) serving as the common screen or earth connection for all four signals. If your DIN-phono lead only has two phono connectors on it, the centre pins of the two phonos will either go to 1 and 4, or 3 and 5 — a test meter will help you find out which.

The other possible explanations for why you're only getting output on one channel are broken electronics within the machine itself, or that you are trying to play a quarter-track tape on a half-track machine (or vice versa)...

You can check the latter by looking at the heads or making a test recording to a blank tape. A half-track head uses almost half the tape width for each channel, so you'll see the two head gaps occupying just under half the tape width, with only a small gap (guard band) between them. A quarter-track head uses slightly less than a quarter of the tape width for each track, and the two channels are separated by a quarter-track width, so the two head gaps are separated by the width of another head gap.

As for the 'p up', 'amp' and 'radio' sockets, this suggests that the machine has a built-in record selector and preamp. 'P Up' will be an RIAA phono pickup input, for example. 'Radio' is pretty self-explanatory, and 'Amp' is probably another line-level input — but it could possibly be an output intended to go to a preamp. It would be worth checking anyway!

Published September 2005

Saturday, June 16, 2018

Q. How can I permanently stop mains noise in my studio?

By Martin Walker
Systematically tracking down the source of mains hum may be tedious but necessary, and you'll only have to do it once.Systematically tracking down the source of mains hum may be tedious but necessary, and you'll only have to do it once.

Can you recommend products suitable for the European power grid that can be used to clean up the power signal and ground loops? I am experiencing both ground loops and a generally dodgy power signal. A lot of people recommend that I use some sort of UPS (Uninterruptible Power Supply), but I don't need the functionality they provide, and I would rather spend money on better power conditioning and filtering equipment. Your advice will be greatly appreciated!

Alexander van Rijn

PC music specialist Martin Walker replies: In my opinion it's only worth 'cleaning up the power signal' if it's dirty, and a huge number of background noise problems are caused not by mucky mains, but by audio wiring that results in ground loops. This is the source of lots of unwanted nasties that sneak into your audio signals, and removing them often requires no dedicated products at all. Problems range from straightforward 'hums' (which normally include various levels of the mains harmonics, such as 50Hz, 100Hz, 150Hz, and so on in the UK, or 60Hz and higher multiples in the US), to a wide range of scratches, ticking, buzzing and other digital gremlins that are often associated with computer activities such as graphic redraws, mouse movements, and hard-drive activity.

If you're experiencing any of these ground-loop problems, you won't solve them by installing a power conditioner or an Uninterruptible Power Supply, so before you even think of spending money on either of these options you should examine your basic wiring. Temporarily unplug all the audio cables from your setup, and if you've got gear bolted into a rack, it may also be worth disconnecting the mains cables of this other gear to rule out problems with several metal cases touching each other and causing yet more ground loops.
As tempting as it might seem, short cuts such as leaving the cables plugged in and just switching off the connected gear at the mains won't work, since the mains cables and any resulting ground loops will still be in place. Unplugging one cable can therefore make the background noises better or worse, depending on how this affects the remaining ground loops. Only by removing every audio cable and working through your studio item by item can you totally eradicate ground-loop problems.

You should now hopefully hear silence from your loudspeakers or headphones, apart from a little hiss and possibly a tiny amount of hum or buzz if you turn the amplifier right up and place your ears nearby (be very careful when doing this, since an unexpected signal at this point could damage your ears or blow up your speakers). If there's still more hum than you expect, it might be due to a nearby 'line-lump' power supply, in which case, you should move this as far as possible from audio cables, and at the very least try rotating it to find the 'quietest' position. If you're still unhappy with the levels of hum and noise from your amp/speakers you may need to get them checked out by a technician — remember that hum levels of both solid-state and valve amps can increase over time, due to deteriorating capacitors or valves.

Assuming all is well at this stage, turn down the speaker levels, connect your mixer to the amp, turn up and listen again (if you route all your gear directly to a multi-channel audio interface, this is your 'mixer'). You'll probably hear greater hiss levels from the combined contribution of all the input channels until you pull the master fader right down, but there still shouldn't be any obvious hum or other interference. If there is, it's generally because you've just created an earth loop — the amp/speakers are already earthed via their mains cable, and the mixer is earthed in exactly the same way, so when you connect the two with an audio cable its screen connection completes the loop, causing unwanted earth currents to flow.

If your amp has balanced inputs and your mixer/interface has balanced outputs, the cure is to connect the two via balanced audio cables (twin core plus screen). If not, you may be able to achieve the same results by disconnecting the screen of an unbalanced cable at one end (in the case of soldered cables you can do this inside the plug, normally at the destination end). Similarly, if the amp has a balanced input, but your mixer/interface only provides an unbalanced output, you can make up a pseudo-balanced cable, as I described in 'Computer Audio Problems' in SOS November 2004. Here, one end of the balanced cable is wired to a balanced jack or XLR as normal, while the other end is wired to an unbalanced jack with the screen disconnected or, preferably, connected via a resistor. These cost only a few pence more to make than unbalanced cables, yet provide an ideal solution for connecting any unbalanced source to a balanced destination. I've got such cables wired between all my hardware synths and mixer, and background noise levels are considerably reduced as a result.

By disconnecting the earth wire inside a mains plug, you are removing an essential electrical safety measure — never do it!By disconnecting the earth wire inside a mains plug, you are removing an essential electrical safety measure — never do it!

Occasionally the only way to cure a ground-loop problem is to install a line-level DI (Direct Injection) box between the mixer and amp, to 'galvanically separate' the two circuits, commonly by using a transformer to transfer the signal — the audio gets through perfectly, but there's no direct connection at all between the input and output cables inside the DI box. This is sometimes the only way to cure some laptop-related ground-loop problems, but in my experience, most others can be dealt with by cable modifications.

Once your mixer, amp, and speaker chain have an acceptably low level of background noise, plug each remaining item of gear into your mixer in turn and power it up, listening at each stage for unwanted noises. As soon as you hear any, you know you've either got a faulty piece of gear or a ground-loop problem, and can sort it out in exactly the same way as before. If it's rack gear, you may need to temporarily unbolt it from the rack to check that the problem isn't due to its case touching other earthed metalwork and creating a further ground loop (if it is, use nylon rack bolts or 'Humfrees' to isolate it). Alternatively, low-level circuitry such as mic preamps can pick up mains interference from the mains transformer inside a nearby rack unit. This systematic approach is the only way to deal with ground-loop problems. It may be tedious, but you only have to do it once, and the benefits can be enormous!

When you've got all your gear connected, and still have no hums or other nasties, then and only then is the time to consider adding a 'power conditioner' or UPS. A power conditioner will filter the mains signal to remove any radio-frequency interference plus any incoming spikes and other intermittent noises riding piggyback on the mains signal from the outside world. However, most modern electronic gear, including computers, already includes such filtering in its own power supplies, and in general, it's far better to suppress switch-related mains transients from distant devices such as refrigerators and central heating systems at source, as this will be far more effective.

If, after solving your ground-loop problems, you don't hear any other nasties then you probably don't need a power conditioner at all, but they can be very useful bolted into a rack for live use, to cope with unexpected 'incoming' problems due to stage lighting or grotty wiring in unfamiliar venues. However, if your mains power is 'generally dodgy' it may pay you to have an electrician check your house wiring, and contact the local electricity board to have your incoming mains checked for quality. If, for instance, you live in a remote rural location or close to an industrial estate, you may suffer from occasional but unavoidable interference problems that will benefit from a studio-based power conditioner, although I've never personally found the need for one (perhaps I've been lucky).

A UPS will, in addition, cope with 'brownouts' (occasional severe drop in mains voltage, generally for a few seconds only), plus the more severe 'blackouts' (complete loss of mains power), in exactly the same way as a laptop computer carries on running on battery power if you pull its mains plug. Even if you only use the UPS to power your desktop computer rather than the whole studio (generally a far cheaper approach), it can prove invaluable if you have paying clients in your studio, to avoid your computer rebooting in the middle of a session, and can give you a vital few extra minutes to save the current project before the UPS backup power runs out.

Published July 2005

Thursday, June 14, 2018

Q. What's wrong with my patchbay?

By Hugh Robjohns
Neutrik 48 Jack patch bay.
I have a Neutrik quarter-inch jack patchbay that I'm having problems with. The unit is only a few months old but already I seem to be suffering from poor connections, with signals being quiet or not coming through properly and becoming distorted. I have isolated my outboard and tried different cables to check if the problem lies with these, and all roads lead back to the patchbay. Are there any methods for cleaning the contacts in patchbays or fixing them? The patchbays I had before never had any of these problems and I was using them for over four years.

SOS Forum Post

Technical Editor Hugh Robjohns replies: It is very unusual for a patchbay to become unreliable in such a short time. I presume there is no obvious environmental problem such as excessive dust, damp or smoke? Dust, damp and smoke tend to work together, gathering on the socket contacts to form a sticky residue which acts as a high-resistance layer, giving the kind of problems you seem to be experiencing. It helps if you can make sure the faceplate of the patchbay is vertical in the rack, rather than horizontal or sloping, as this minimises the risk of dust falling into the sockets, and that the room is kept dehumidified and well ventilated.

Another related cause is dirty (tarnished) plugs. This used to be a real problem in professional studios using PO316 or bantam patch cords which employed brass plugs, but tends not to be an issue with the plated domestic quarter-inch plugs used in most home-studio rigs. Professional studios using brass patch plugs often use a mechanical burnisher to clean and polish the plugs, along with an aggressive cleaner for the sockets, but the equipment is designed to be cleaned in this way. The plated domestic plugs and sockets are often quite soft in comparison and will wear out very quickly if treated this way, so gentle hand cleaning with a mild metal polish or contact cleaner — Deoxit or Servisol, for example — might help. Don't get too enthusiastic though: excessive rubbing with an abrasive cleaner will quickly damage or even remove the plating, making your problems a whole lot worse! A quick wipe over with one of the gentle cleaners mentioned above every month or two should keep everything in good order, if surface contamination is the problem.

Another likely problem, probably the most likely, in fact, is that your patch cables are of a non-standard size. Some of the cheaper Chinese-made moulded patch cables are fitted with locally made plugs that are slightly undersized and don't conform to the correct quarter-inch specifications. Consequently, they sometimes don't make reliable contact with some types of socket. The solution here is obvious: try patching using good-quality leads (ideally with Neutrik jacks on the end), and see if that works any more reliably.

Published June 2005

Tuesday, June 12, 2018

Q. Is there any advantage to using two subwoofers?

By Hugh Robjohns
Because the EMES Black TV Active monitors are full-range speakers, when they're used with the Amber subwoofer, the crossover can be comfortably set at 80Hz rather than 120Hz. 
Because the EMES Black TV Active monitors are full-range speakers, when they're used with the Amber subwoofer, the crossover can be comfortably set at 80Hz rather than 120Hz.

I have heard of having a subwoofer in your studio in combination with your monitors to reach the extra-low frequencies. However in the Bob Katz book [Mastering Audio reviewed SOS October 2003], there is a picture of his studio where he has two subwoofers, one below the right monitor and one below the left monitor. Sub frequencies in stereo? This was a first for me. I never thought of the low frequencies being in stereo. Usually you think of keyboards, voices, and guitars in stereo, but not bass frequencies. What are your thoughts on this?

SOS Forum Post

Technical Editor Hugh Robjohns replies: This is a very complex subject and we'll need to separate the real issues from the urban myths if we're to get to the bottom of it.
There is quite a lot of evidence to suggest that we cannot perceive what direction low-frequency sound is coming from, and hence a single subwoofer would appear to make sense in the context of a sub-equipped stereo or surround system, where full-range speakers aren't practical (or affordable).

However, the upper frequency limit of the sub (and thus the lower limit of the satellites) has to be set very carefully. A lot of home theatre systems, for example, use ludicrously small satellites and thus require the sub to operate well into the range of directional frequencies.

There are two 'proper' standards for crossovers in the home cinema world (for sensible-sized speakers, not mini satellites) — one is 120Hz and the other is 80Hz. Personally, I favour the latter, as I think it is possible to locate 120Hz fundamentals.

However, the major fly in this particular ointment is that distortion in low-frequency speakers is inherently quite high, especially in the case of ported cabinet designs. Distortion produces harmonics, and those harmonics, although low level, are in the directional frequency range and hence the location of the sub becomes very obvious.

The result is that unless the sub is located close to the centre of the frontal sound stage, low-frequency content will tend to produce harmonics which will pull the stereo image towards the location of the sub. And you can't normally place the sub close to the centre of the frontal sound stage, because that will tend to excite the most pronounced room modes and produce a very lumpy and uneven bass response.

So, one way around this problem is to ensure that the crossover point between the satellites and subwoofer is as low as possible (80Hz for example), even though that means that reasonably-sized satellite speakers are needed (not usually a problem in music studios, but not common in the cheap home theatre systems), and that the subwoofer has to have extremely low distortion figures.
Subwoofers which employ a closed cabinet design, like the Dynaudio BM9S, can offer a more precise low-frequency response, but at the expense of efficiency. 
Subwoofers which employ a closed cabinet design, like the Dynaudio BM9S, can offer a more precise low-frequency response, but at the expense of efficiency.

Using two subwoofers is an alternative way of tackling the problem. By driving the room from two points instead of one, two different sets of room modes are excited, which can result in a smoother overall low-frequency balance. Also, any tendency for the harmonics produced by distortion to skew the image can be balanced out and thus effectively nulled, though I'm ignoring the unwanted masking effect that such harmonics would have. We could also go on to discuss the issues of actually trying to match the on-axis and off-axis energy responses of the satellites and subs, which makes the use of satellite and subwoofer systems seem even less attractive to me.

Of course, there are some systems that do work very well, and there are undeniable practical and fiscal advantages to 2.1 (or 2.2) setups in certain situations, but they all take a huge amount of careful setting up, both in calibrating and positioning the subwoofer(s). We've touched on these issues in these pages and elsewhere in Sound On Sound on a number of occasions — have a look at Mallory Nicholls' article on subwoofers from SOS July 2002 (www.soundonsound.com/sos/jul02/articles/subwoofers.asp) and the Studio SOS feature in SOS April 2003 (www.soundonsound.com/sos/apr03/articles/studiosos0403.asp). Ideally, I'd always prefer a full-range stereo system, but the constraints of space and budget mean that for most people this is rarely practical in a domestic situation.

Published May 200

Saturday, June 9, 2018

Q. Should I buy a vintage analogue synth or a modern modelling synth?

By Steve Howell
Q Should I buy a vintage analogue synth or a modern modelling synth?
I want to buy a 'knobby' synth because I am fed up of setting up sounds with a data wheel. I also want a very analogue sound. I am thinking of buying a modelled synth but, at the same time, I would really like a genuine vintage synth to get a 'real' analogue sound (and because they look so cool!). Any pointers would be appreciated.

Ben Slater

SOS contributor Steve Howell replies: As was pointed out in Sounding Off in SOS February 2004 (www.soundonsound.com/sos/feb05/articles/soundingoff.htm), analogue synths are not without their pitfalls. Firstly, assuming you can actually find a good example of the synth you favour, they can be costly to buy but, more importantly, they can also be costly to maintain.

When buying a vintage synth, you should check for noisy pots and switches. Whilst these can often be fixed with a squirt of an appropriate contact cleaner, replacing them can be expensive, especially if the pots are surface-mounted to the PCB and/or the switches aren't now available. You might think that noisy pots aren't really a problem, but a large part of the appeal of a knobby synth is the ability to tweak controls during a performance — there's nothing worse than your ripping solo being spoilt by the intrusive sound of crackles!

You must also check out the keyboard. Often, the keyboard mechanism on these old synths is very simple and it is all too easy for the contacts to break (or become bent or twisted so that they don't make contact). You can sometimes fix these yourself if you're handy with a soldering iron, but getting them repaired or replaced by a specialist is likely to set you back a few bob! And what about MIDI? Most old synths don't have it, although they can usually be triggered by control-voltage (CV) and Gate signals. So if you want to integrate the vintage synth into an existing sequencing setup, you're going to have to seek out an example that has a MIDI retrofit, or budget for some kind of MIDI-to-CV converter.

Korg's analogue-modelling MS2000B synth is more flexible than the vintage MS20 (top) and still offers plenty of 'tweakability'... but is it as desirable?Korg's analogue-modelling MS2000B synth is more flexible than the vintage MS20 (top) and still offers plenty of 'tweakability'... but is it as desirable?

Then there's the sound-generating circuitry itself. By definition, it's going to be old, and components may be failing, leading to tuning and other instabilities as well as noisy outputs — I once tried an ARP Axxe that sounded as though someone was frying bacon in the background! Not only are these problems potentially costly to repair but it could well be that some components are simply not available any more, especially if the manufacturer used any integrated circuits that are now in short supply, or worse, custom components.

Of course, if you buy the synth from a reputable dealer who specialises in vintage synths, a lot of these issues can be avoided, as the stuff they sell will invariably be refurbished (or at least serviced prior to sale) and will often carry some form of warranty. You will pay a bit more for that peace of mind, understandably, but it can be worth it.

You should also listen carefully to anything you are thinking about buying — or even do a blindfold test — and ask yourself, "Does it actually sound good?". Do not allow yourself to be deluded by the attractive retro looks or the allure of owning a genuine analogue. Due to component tolerances (and failing components), not every analogue synth sounds good (or even the same as another identical model). And just because it has a Moog badge on it (or whatever), don't consider that a guarantee of 'fatness', 'warmth' or any other adjectives that are applied with dewy-eyed nostalgia to anything vintage.

Q Should I buy a vintage analogue synth or a modern modelling synth?If you have to have a true analogue synth, the Voyager by Bob Moog (above) might be expensive, but you won't find a MIDI-equipped original Minimoog (top) in pristine condition and perfect working order for less money, if at all. 
If you have to have a true analogue synth, the Voyager by Bob Moog (above) might be expensive, but you won't find a MIDI-equipped original Minimoog (top) in pristine condition and perfect working order for less money, if at all.

I had lusted after an ARP Odyssey since the time I tried one as a teenager in Rod Argent's Keyboards back in the mid-'70s, and when one was offered to me many, many years later for a very silly price, I bought it on spec without checking it out first — bad move! When it arrived, it looked gorgeous — a prime example of a white-faced original, with all its sliders intact — but it was a totally underwhelming example of the instrument, and not at all what I had been remembering so fondly. I guess what I am saying is, don't buy an old synth wearing rose-tinted spectacles. If you do, you may well be in for a disappointment!

Modern, modelled synths are often a much better bet as a long-term investment. To all intents and purposes, and perhaps contentiously, they sound equally as good as the majority of vintage synths, if not better in some respects. They are inherently more flexible, are usually polyphonic, and are often more versatile, with sound-shaping facilities that the originals could only have dreamt of. They are also usually multitimbral, come with effects to polish the sound built in, and may have sophisticated (and often programmable) arpeggiators. They might not sound exactly like a vintage Moog, ARP, or Roland, but they're pretty close, and (unless you're very unlucky) won't spend much time being serviced.

I guess the only slight downside to these modern, modelled synths is that whilst many have plenty of knobs, they don't always have a control or switch for every parameter, unlike original analogue synths. Often, the less frequently used parameters on the modelled versions are accessed via an LCD and menus. However, it's perfectly possible to create very vibrant and convincing analogue synth sounds without ever having to delve into the more obscure aspects of the synth's programmability.

No-one has a greater respect for old synths than I do — after all, they paved the way to the technology we enjoy today. But just because a synth is old and carries a badge doesn't make it good. Witness the Polymoog — what a weak-sounding, unreliable crock! There are some great old synths out there if you can find a good example of one that satisfies your requirements and budget, but don't dismiss the more recent modelled hardware synths.

If you're still in the market for analogue, check out Gordon Reid's guide to buying a vintage keyboard from SOS September 1994 — see www.soundonsound.com/sos/1994_articles/sep94/vintagesynths.html. And for a more detailed idea of some of the things that can go wrong with vintage gear, check out the two-part feature on equipment servicing that appeared in SOS March and April 1996 (see www.soundonsound.com/sos/1996_articles/mar96/servicing.html and www.soundonsound.com/sos/1996_articles/apr96/servicing2.html).

Published May 2005

Thursday, June 7, 2018

Q. What kind of ear plugs should I wear at gigs?

By Hugh Robjohns
Some generic attenuating ear plugs manufactured by Sensorcom. 
Some generic attenuating ear plugs manufactured by Sensorcom.

I've been coming home from gigs recently with my ears ringing and I'm worried about damaging my hearing. I think it's definitely time to invest in some kind of (preferably unobtrusive) ear protection, but what kind of ear plugs should I be looking at? I still want to be able to hear what's going on but keep my ears out of danger at the same time. I guess I can't wear earplugs when I'm actually performing, but at least I can reduce the chances of permanent damage when I'm watching the other bands. What's your advice?

Patrick Bailey

Technical Editor Hugh Robjohns replies: Hearing damage is directly related to both sound level and length of exposure. So, even if you don't want to wear ear plugs when you're performing, consider wearing them when you're rehearsing, as well as at gigs — it has been suggested that musicians often do more damage to their ears during the many hours of rehearsal than in the comparatively short time they spend on stage.

I would recommend investigating the options for good-quality ear plugs that reduce the overall level of sound but maintain an even spectral balance so that you can still hear everything clearly, although the overall level is reduced. Disposable solid-foam ear plugs won't give you this even balance and will adversely affect your enjoyment of the music. You can often find suitable generic ear plugs in the good musical instrument and equipment retailers, sold as 'musicians' earplugs', and available in different strengths (amounts of attenuation). Obviously, the greater the number of dBs of attenuation, the better overall protection they offer.

However, for a really comfortable and long-lasting solution, I would recommend making an appointment with a good audiologist who will be able to take ear moulds and make earplugs to your precise specifications that will be comfortable to wear for long periods and easy to clean and look after. Custom-made earplugs will cost more, but considering that hearing damage is irreversible, if you value your ears the cost should be irrelevant!

More information and advice is available from the RNID (www.rnid.org.uk). The web site of their ongoing 'Don't Lose The Music' campaign (www.dontlosethemusic.com) is aimed specifically at musicians, DJs, clubbers and concert-goers and is linked with two hearing protection specialists — Advanced Communication Solutions, or ACS for short (www.hearingprotection.co.uk), and Sensorcom (www.sensorcom.com) — who can produce custom-fitted ear plugs.
Some custom-moulded ear plugs, manufactured by Sensorcom. 
Some custom-moulded ear plugs, manufactured by Sensorcom

Published June 2005

Tuesday, June 5, 2018

Choosing Monitors

By Hugh Robjohns
Choosing MonitorsPhoto: Mike Cameron & Mark Ewing

There are many decisions to be made when choosing a monitoring system. Infinite baffle, reflex, or transmission line? Active, powered, or passive? Bi-wired or bi-amped? We help you find the answers you need.

Choosing a monitoring system can be a difficult and confusing task, not least because of the enormous number of models and designs on offer. For a start, there are three basic classes of monitoring loudspeaker: infinite baffle (sealed box), reflex (ported), and the less common transmission line. Some monitors use a single wide-band driver, but most are two-way or three-way, while others use four or more drivers. There are also systems which require a separate subwoofer. And finally, three different amplifier arrangements are widespread: passive, powered, and active, along with bi-wiring options. So let's have a look at the pros and cons of each of these designs.

Ports are used in the majority of project studio monitors, primarily because they help boost the output level at low frequencies, such as the M-Audio BX5. 
Ports are used in the majority of project studio monitors, primarily because they help boost the output level at low frequencies, such as the M-Audio BX5.Photo: Mike Cameron & Mark Ewing

Infinite-baffle Designs

The simplest kind of cabinet construction used in studio monitors is the infinite baffle, or sealed cabinet. Theoretically, an infinitely large baffle will divide the sound coming off the front of the loudspeaker driver from the opposite-polarity sound coming off the rear, but both sides of the loudspeaker cone are working into the same infinitely large volume of air and thus are loaded identically. Of course, such a concept is not workable in practice, and the closest we can come to the ideal is to build a large sealed box and place the loudspeaker in the front baffle, hoping that the sound coming off the rear of the speaker cone will be absorbed within the box.

Samson Resolv 80A. 
Samson Resolv 80A.Photo: Mike Cameron & Mark Ewing

Sadly, it's not quite that easy. Clearly, the rear of the cone is working against a much smaller volume of air than the front of the cone, and that volume of air is fixed. Consequently the loudspeaker cone feels a different degree of resistance when moving inwards than it does when moving outwards, which affects the distortion characteristics of the system as a whole. Internal resonances and standing waves can also be created within the cabinet, despite the use of lots of absorbent material, and this can produce various audible colorations in the sound.

Finally, the bass response of this kind of cabinet is relatively limited compared to that of other arrangements, for a given cabinet size, the low-frequency roll-off starting at a relatively high frequency. On the plus side, though, the phase response is very smooth, with relatively little phase shift, and the slope is also quite shallow, averaging 6dB/octave. Indeed, because of the shallow slope, even small infinite-baffle speakers can produce audible bass at surprisingly low frequencies.

In lower-cost monitors (such as the Event TR5 and Behringer B2030A Truth shown below), the main side-effect of this design is a smearing of low-end transients which makes it difficult to judge the balance of bass instruments. However, the problems of ported cabinet design can be overcome, and more expensive models such as the Earthworks Sigma 6.2, Fujitsu Ten Eclipse TD512, Tannoy Ellipse 10 IDP, and ADAM S3A are able to achieve professional performance. 
In lower-cost monitors (such as the Event TR5 and Behringer B2030A Truth shown below), the main side-effect of this design is a smearing of low-end transients which makes it difficult to judge the balance of bass instruments.Photo: Mike Cameron & Mark Ewing 

For many, the infinite-baffle design is the most highly regarded and least compromised solution to loudspeaker monitoring. It is also interesting to note that the most widely used mixing references — the Auratone and the Yamaha NS10 — are both infinite-baffle designs. One of the most revered high-quality infinite-baffle designs was the infamous LS3/5A — a BBC in-house design dating back to the early '70s.

A couple of more modern and high-tech examples of the infinite-baffle loudspeaker are the K+H O 300D monitor, most AVI monitors, and the smaller ATC monitors. These speakers demonstrate the characteristically smooth, natural-sounding bottom end and associated mid-range clarity of the closed-box design very well. To many, what the infinite-baffle approach lacks in raw low-end volume, it more than makes up for in quality and transparency.

Reflex Cabinets

The most common cabinet design is the reflex or ported cabinet, which makes deliberate use of the resonance of the cabinet to take advantage of the sound coming off the rear of the loudspeaker cone. Instead of being completely sealed, the cabinet has a hole in it through which the internal sound can escape and contribute to the overall sound in the listening environment.

Behringer B2030A Truth. 
Behringer B2030A Truth.Photo: Mike Cameron & Mark Ewing

The vent may be located on the front baffle, it may be on the rear, and it may take the form of one or more round holes or slots. Most usually, the vent is connected to a tube extending back into the cabinet, the diameter and length of which are carefully calculated to achieve the required frequency response. Across a specific frequency range determined by the various parameters of the port opening, the sound from the rear of the loudspeaker cone is allowed to resonate through this port, emerging in the same polarity as the frontal sound to bolster the low-frequency response of the system as a whole.

Fujitsu Ten Eclipse TD512. 
Fujitsu Ten Eclipse TD512.Photo: Mike Cameron & Mark Ewing 

The advantage of this approach is that it allows a much greater acoustic output at lower frequencies than the infinite-baffle design — you get a far more impressive bass response and overall volume level for the size of the box. However, there are a few disadvantages, one being that any resonant system smears transient signals over time. This can most clearly be seen on the waterfall response charts beloved of hi-fi magazine reviews, where one or more long resonant tails can usually be seen at low frequencies.

In monitoring terms, this inherent time-smearing and resonant behaviour can obscure small dynamic changes in the signal being auditioned, and may also reduce the transparency of the mid-range. In practical terms, a poorly designed reflex system can make it extremely hard to judge the relative levels of bass instruments properly, because their energy is stretched over time.

The problems of ported cabinet design can be overcome, and more expensive models such as the Earthworks Sigma 6.2, Fujitsu Ten Eclipse TD512, Tannoy Ellipse 10 IDP (pictured here), and ADAM S3A are able to achieve professional performance. 
The problems of ported cabinet design can be overcome, and more expensive models such as the Earthworks Sigma 6.2, Fujitsu Ten Eclipse TD512, Tannoy Ellipse 10 IDP (pictured here), and ADAM S3A are able to achieve professional performance.Photo: Mike Cameron & Mark Ewing

 Earthworks Sigma 6.2. 
Earthworks Sigma 6.2.Photo: Mike Cameron & Mark Ewing

Another issue is the frequency- and phase-response characteristics of the port resonance. While the low-frequency roll-off point can be extended to a significantly lower frequency using a reflex design than with an equivalently sized infinite-baffle cabinet, the slope is far steeper, and the phase shifts far greater. Thus the level of bass output is greater down to the roll-off point, but then falls away much quicker, and a reflex cabinet is likely to reproduce very low frequencies at a far lower level than an infinite-baffle speaker. The inherently large phase shifts of this design also reduce (or at least affect) the naturalness of the bass end — although not everyone appears to be sensitive to this aspect of sound reproduction.

ADAM S3A.Photo: Mike Cameron & Mark Ewing

The extent and impact of these inherent disadvantages depends enormously on the competence of the reflex cabinet's design and what the designer was trying to achieve. There are many excellent reflex designs around, including the larger ATC monitors, all the Genelec models, various Dynaudios, Mackies, and Tannoys, and many others.

The Mackie monitors are an interesting sub-class of reflex design, though, because the port is covered by a passive radiator — in essence an unpowered speaker cone that reacts to the sound pressure inside the cabinet. This is a more complex arrangement again, sharing some characteristics with both infinite baffle and reflex designs — although it falls most comfortably into the latter camp.

Transmission-line Systems

The third type of cabinet is the transmission line, and at the present time there is only really one commercial monitoring manufacturer using this approach — PMC in the UK. Like the passive radiator approach, the transmission line in some ways presents a combination of both infinite baffle and reflex characteristics, arguably offering the best aspects of both worlds. Essentially, a transmission-line speaker places the driver cone near to or at the end of a long large-diameter tube which is very heavily damped with absorbent material. To make the cabinet practical, the tube is generally folded several times internally, allowing a line length of several metres to be enclosed within even fairly compact cabinets.
Across most of the low and middle frequency range the transmission line is so well damped that all of the sound energy from the rear of the driver cone is completely absorbed and none of it reaches the outside of the cabinet. In that regard, it operates like a true infinite-baffle design — none of the rear sound reaches the listener. At very low frequencies, though, the line absorption becomes less effective and some very low-frequency sound reaches the end of the transmission line, much like the sound leaving a ported speaker. This allows a near flat response which extends down to at least an octave below any similarly sized reflex cabinet. One other advantage is that the overall frequency response varies very little with monitoring volume — the balance stays more or less constant regardless of listening level — which I personally find very useful.

How Many Ways?

Single drivers can't really handle the entire audio spectrum at monitoring levels, so the vast majority of monitoring speakers employ two drivers — a low-frequency/mid-range woofer and a high-frequency tweeter. The former generally handles frequencies below about 2kHz and the latter everything above, the actual changeover point being called the crossover frequency.
Getting two drivers to match each other in terms of level, phase, and dispersion at the crossover point is far from trivial, and the on-axis frequency response of a loudspeaker is only one aspect of its performance that must be right. The relative phase through the crossover region is just as important, and the smoothness of the off-axis responses arguably more so — after all, most of the sound energy we hear in a room is reflected off-axis sound rather than direct sound. This is often what differentiates a really good monitor from a less good one.
All of the monitors in Mackie's HR series, including the HR626 shown here, use a variation on the reflex design, where a kind of passive speaker cone is fixed over the end of the port. This design retains some of the advantages of the infinite baffle, even though a port it used. 
All of the monitors in Mackie's HR series, including the HR626 shown here, use a variation on the reflex design, where a kind of passive speaker cone is fixed over the end of the port. This design retains some of the advantages of the infinite baffle, even though a port it used.Photo: Mike Cameron & Mark Ewing

Loudspeaker monitors have polar responses just like microphones or acoustic instruments. Some designers argue that a loudspeaker should have an omnidirectional polar response, and there are commercial designs built to do that — but in most typical studio situations a directional speaker works far better with typical acoustic treatment designs. At very low frequencies, speakers tend to radiate omnidirectionally because the wavelengths of low-frequency sound are generally far larger than the speaker cabinet. As the frequency rises, the cabinet starts to influence the dispersion of sound, and so the polar response starts to narrow into a more directional lobe. At higher frequencies still, the size of the driver itself starts to influence the dispersion, and the sound lobe reduces to something more like a beam.

Through the crossover region, the sound will be generated by both mid-range/woofer and tweeter, but given the relative size of the two drivers in relation to the wavelengths being produced, the woofer's polar response is likely to be very 'beamy', while the tweeter will have a much broader dispersion. Such a disparity in dispersion angles will cause a huge step in the off-axis frequency response, and consequently a very coloured off-axis sound. This is one reason why a speaker can sound very different when placed in a highly damped room than it does in a more lively, reflective room. It's only in the last twenty years or so that the importance of the off-axis sound and the careful matching of dispersion has been realised. So the width of the front baffle, the relative size of the drivers, and their crossover frequencies and filter responses are all chosen very carefully to optimise the response of the complete system.

Many systems these days employ waveguides around the tweeter to help control dispersion and sometimes to create different polar responses in the horizontal and vertical planes. This is usually to reduce early reflections from console and ceiling, and it's also one reason why turning a nearfield monitor on its side is not a good idea!

Designing a two-way speaker is hard enough, but most designers agree that a three-way system offers the best overall performance. Although there are two crossover regions to perfect, the disparity in size from woofer to mid-range driver to tweeter is much smaller, so the dispersion matching between adjacent drivers is easier. Each driver also has to operate over a much narrower frequency range, which enables each to deliver far better performance. Indeed, the improvement in the mid-range resolution and clarity of a good three-way system compared with a two-way system is very significant. Systems with additional drivers — four-way systems and systems with multiple tweeters, bass units, and so on — become a lot more complicated, and often the advantages are outweighed or at least balanced by the disadvantages.

Powered Models

The traditional way to build a loudspeaker is with a high-level crossover that accepts the full-bandwidth, high-power output of a power amplifier and splits that signal into two or more separate frequency bands to feed the appropriate drivers. However, the quality of such a 'passive crossover' can affect the sound dramatically, and there are limitations as to what can be achieved in terms of response shapes and phase alignment using passive filtering. Additionally, passive systems are inherently lossy in terms of power dissipation.

A passive loudspeaker is powered from a separate amplifier, typically installed some distance from the speaker and connected via a two-wire cable. Given the need to transfer power from the amp to the speaker and the relatively low impedance of the speaker itself, this cable has to have very low resistance and be able to carry large current pulses. Bell wire is not recommended, but any relatively substantial two-core cable will do. Two-core lawn-mower mains cable is ideal in most circumstances, and very cost-effective — far more so than the esoteric cables promoted in hi-fi shops.

Assuming good clean and tight connections and a competent amplifier, a passive speaker connected with respectable cable will perform very well. However, there are potential quality gains to made, if the speaker's innate resolution warrants it, by doing what is known as 'bi-wiring', where the tweeter and woofer are connected to the amplifier by separate cables. The passive crossover must be designed for bi-wired operation, and must provide separate pairs of terminals for each driver (normally linked with bars or brackets which must be removed for bi-wiring). This allows the amplifier to control the damping of each driver more effectively, as this bi-wiring separates the large sustained current flows to the bass driver from the smaller high-frequency signals. However, the crossover filter for each driver is still placed at the end of a long piece of connecting cable.

A related configuration is called 'bi-amping'. Here separate power amplifiers are used to drive the bass driver and tweeter. Each amp is fed with a 'Y'-cord so that it is amplifying the same signal through both channels, and is then connected to the relevant terminals of the speaker. The idea is to remove the interaction between bass and treble signals completely, but the practical disadvantages of this approach usually outweigh any performance gains.

Bolting the amplifier directly to the back of the speaker cabinet reduces the length of the speaker cable considerably, and thus also improves performance. The result of this approach is known as a 'powered speaker', but it is important to remember that the crossover circuitry is still passive.

Active Electronics

The final step is to remove the crossover function from the speaker and perform it at line level using active electronics. Speakers designed this way are known as active speakers, and potentially have a lot of advantages. Firstly, far more complex filter shapes and characteristics can be implemented with active electronics than with passive circuitry, and the need for large capacitors and inductors is removed. This facilitates better matching of drivers through the crossover region. Most active designs also incorporate room equalisation and tailored response facilities that can be very useful too.

By performing the crossover separation ahead of the amplification, it is necessary to power each driver with its own amplifier. This affords the second advantage, which is that the amplifiers' responses and power ratings can be tailored precisely to the speakers they are driving and driver protection systems can be easily built in. Most active systems intended for home studios are fully integrated solutions, so the cabling between the amplifier and driver is very short, although larger high-end active systems usually employ separate amplifiers with rackmounting active crossover units.

There are also some disadvantages with active designs, the most obvious one being that the number of amplifiers required has doubled or trebled compared to a passive design, and good amplifiers are inherently expensive. There are cost savings to be made in translating a passive crossover into an active one, but nothing like the amount needed to fund the kind of amplifier usually employed with good passive speakers.

In order to make active speakers for the budget end of the studio market, most manufacturers have to employ cost-effective 'chip amplifiers' — fully integrated designs — rather than traditional discrete circuits. Alternatively, many have chosen to use very efficient Class-D digital switching amplifiers, but in both cases the signal quality is often compromised in comparison with a good rackmounting amplifier. Whether these potential amplification losses are outweighed by the filtering and connection gains and careful system optimisation within any specific system is largely open to debate. However, overall I would say that, on pure resolution and transparency grounds, the losses generally outweigh the gains on most active speakers costing less than about £250 each in the UK, and true monitor resolution doesn't start to emerge until at least double that figure.

Published July 2005