Q. Can I use a video patchbay for S/PDIF signals?

By Hugh Robjohns

Is it possible to use a video patchbay, such as the one Studiospares sell, as an S/PDIF patchbay? The Studiospares one has 75Ω BNC connectors, so obviously appropriate 75Ω cables with BNC-to-phono connectors and/or adaptors would be required. From what I can see, the available S/PDIF patchbays, especially the programmable sort, are prohibitively expensive for the requirements I have.

Gavin

Mosses & Mitchell Musa.

Technical Editor Hugh Robjohns replies: Yes, it is perfectly possible to use passive video patchbays for routing S/PDIF (or unbalanced AES) digital audio signals — indeed, that is precisely what many professional installations actually do. As you say, the standard characteristic impedance for video systems is 75Ω, the same as that required for S/PDIF digital audio signals.

If you anticipate a lot of plugging and unplugging of patch cords, then its best to use a patchbay with video connectors designed for the job, such as the Mosses & Mitchell MUSA design. A BNC patchbay can be used just as reliably, but is rather more awkward to use. I'd stay clear of phono jackfields though, as these tend to become unreliable relatively quickly.

If possible, use BNC-to-phono cables (those designed for video applications will be ideal) rather than phono-phono cables with BNC adaptors, as the boundary between the adaptor and phono cable may well make the signal less reliable.

If you only need a few inputs and outputs, it may prove more cost-effective to use something like Canford Audio's modular patch panel system and install the required BNC sockets for the digital audio circuits, possibly along with other connectors to suit other applications (analogue audio, microphones, AES, MIDI and so on).

Published March 2005

Korg New at NAMM 2020 - XE20

Q. Is there a better balanced-to-unbalanced cabling solution?

By Hugh Robjohns

The line inputs on my computer interface are all balanced on TRS sockets, but the outputs of my hardware synths are all unbalanced on mono TS sockets. I've been connecting them with ordinary unbalanced guitar leads up until now, but some of them suffer from what I think is ground-loop hum, and I wondered if there was a better way? I thought about using DI boxes, but I don't have enough mic-level inputs.

Peter Bradley, via email

SOS Technical Editor Hugh Robjohns replies: DI boxes would certainly cure the ground-loop hum problem, because they generally employ a transformer to balance the signal, with a ground-lift switch to isolate the grounds between the synth and interface. However, it seems a bit silly taking a line-level signal from a synth and knocking it down to mic level, only to re-amplify it again inside the interface, and as you don't have enough mic inputs we can discount that option anyway! A better option would be to use line-level transformer isolation boxes. These use a transformer again to balance the signal and isolate the source and destination grounds but, like DI boxes, the good ones are quite expensive.Unless you're handy with a soldering iron — and have a lot of patience! — SOS's custom-made pseudo-balanced cables could be the best solution for connecting unbalanced equipment to balanced inputs.

Thankfully, there is a much cheaper and more convenient alternative, which takes advantage of the fact that a balanced input is also a 'differential' input. A differential input looks for a signal applied between the 'hot' and 'cold' sides of the connection, and unlike an unbalanced connection, the cable screen (which is grounded) plays no part in transferring the wanted audio — it's only there to trap unwanted external interference. We can use this differential input idea to our advantage by wiring a cable in a slightly non-standard way, to trick the balanced input into accepting an unbalanced signal, while also avoiding a ground loop.

The non-standard wiring is actually very simple. The signal side of the unbalanced output (quarter-inch plug tip) is connected to the 'hot' side of the balanced input (quarter-inch plug tip), while the 'trick' part is to connect the ground side of the unbalanced output (quarter-inch plug sleeve) to the 'cold' side of the balanced input (quarter-inch ring). In that way the balanced input 'sees' the wanted signal between its hot and cold inputs, as it should, but the unbalanced output's ground isn't connected directly to the balanced input's ground any more, so there can't be a ground loop!

Technically, the balanced input loses its ability to reject electrostatic (RF) interference, because the impedances to ground from each input terminal are now different (unbalanced). However, with line-level signals connected with relatively short cables of under five metres or so, RF interference is unlikely to be a problem anyway, and in practice I've employed variations of this kind of 'bodge' interfacing for decades without any problems. Incidentally, this kind of 'bodge' interface is typically called a 'pseudo-balanced' connection.

The quickest, easiest and 'dirtiest' DIY approach would be to cut off the TS plugs from the interface end of your existing unbalanced cables, and solder on new TRS plugs for the balanced input. The signal core wire should be reconnected to the TRS tip terminal, and the screen to the ring terminal, leaving the TRS sleeve unconnected and isolated.

Although that solution will work, it has virtually no protection against RF interference, and a better solution can be obtained if you start with balanced cables and modify the unbalanced end instead. In this case, the hot signal wire is connected to the unbalanced tip, the cold signal wire to the unbalanced sleeve, and the balanced cable screen is left isolated and disconnected. In this way the cable screen will provide a useful degree of RF protection, although it isn't as good as it could be...

The optimum solution is to connect the screen at the unbalanced end too, because that maximises the RF screening. However, a direct connection will reinstate the ground loop, so that's not a good idea. Instead, the cable screen should be connected at the unbalanced end via a simple circuit that maintains a relatively high impedance to mains and audio frequencies to prevent the ground-loop hum, but a much lower impedance to RF to maintain effective RF screening. This simple circuit is nothing more than a 100Ω resistor in parallel with a 10nF capacitor, although these values are not particularly critical.

The practical problem with this approach lies in trying to squeeze a resistor and capacitor into the very limited amount of space inside a standard TS plug without everything shorting out. It's not impossible if you have decent soldering skills, small components and some patience, but for those who don't, help is at hand in the form of our very own custom-made SOS pseudo-balanced cables. These are available from the SOS shop to UK customers at £19.99 (or £16.99 to SOS subscribers). They come with either an XLR or a TRS connector at the balanced end. They are made to our own specifications and to very high standards by Pirahna Cables using Neutrik connectors and Pirahna Ultraflex cable. I've been using some recently to avoid the frustration of making up some new pseudo-balanced cables myself, and they work superbly well 

Published May 2013

Korg New at NAMM 2020 - Nu-Tekt

Q. Are high-end cables worth the money?

By Hugh Robjohns

In a shop recently, I was almost persuaded to buy some Mogami XLR cables. The sales guys said that there's an audible difference between using cheap XLRs and high-end ones, but we didn't have time to test their assertion. Is there any truth in this? I understand that they might last longer — but can they really sound better? My main use in this case would be for feeding active studio monitors.

Via SOS web site

SOS Technical Editor Hugh Robjohns replies: The short answer is no, they're very unlikely to sound any better. The longer answer is that there are complexities and subtleties involved that can, in specific circumstances, conspire to affect the sound when using different types of cables and connectors.

In essence, provided that the cable is appropriate in terms of its construction — particularly in having low capacitance and good screening — and the connectors used are of good quality and manufactured to meet the appropriate specifications, then there is no audible difference that I have ever detected reliably.Are branded audio cables better than unbranded ones? Probably not, but they might be worth it anyway! The problem is that some inferior cables allow interference to get in via the cable or connectors, or have some inappropriate properties such as high capacitance, or they are wired in a way which creates ground-loop problems, and in those cases a properly made cable may appear to work better, simply because the reality is that the inferior cable didn't work properly. However, you should be aware that some very expensive cables are not built properly either (and, amazingly enough, that's often on purpose!)

Mogami cables are generally well made and fit for purpose (so you should have perfectly decent cables there!) but so too are a great many other branded and unbranded cables. I tend to make all my own cables, but if I have to buy pre-made ones I always look for genuine Neutrik connectors — I reason that if someone is going to the trouble of fitting those, the chances are they know how to choose a decent cable too, and wire it up correctly!  

Published June 2014

Patch of the Week 43: Nu:Tekt NTS-1 + SQ-1 Sequencer

Choosing Guitar Cables

By Paul White

At one time, you'd buy a guitar cable and if the plugs didn't fall off or short out, it was a good cable, right? Now we have esoteric guitar cables that cost as much as some guitars, and claim all kinds of tonal benefits — but can cables really make a difference, or is it all just smoke and mirrors? If you use a guitar or bass fitted with active pickups you can sit down now, because your choice of cable will make very little difference, due to the very low output impedance of active circuitry. However, if (like the majority) you use conventional passive pickups, then yes, cables do make a difference.

The electrical resistance of a guitar cable is insignificantly tiny compared with the impedance of the pickups and controls, so that won't affect your tone a great deal, but cable capacitance is another matter altogether. A capacitor is formed whenever two electrical conductors are in placed in close proximity, and for a given spacing, the larger the surface area of the conductors, the larger the capacitance. The core and screen conductors of typical guitar cable may produce a capacitance of around 30 picofarads per foot (or 100 picofarads per metre), so it follows that the longer the cable, the more capacitance you have hanging on the output of your guitar.

Putting a capacitor across an audio circuit creates a low‑pass filter, and given this, you might expect a long cable (or one with a higher capacitance than usual) to kill your guitar's high end — and this is exactly the argument used by those companies making esoteric guitar cable. However, there's actually far more to getting a great guitar sound than simply choosing a low‑capacitance cable.

Firstly, before anyone gave cable a second thought, guitarists were making great‑sounding records using standard guitar leads, so if you're after imitating their sounds, there's little point in seeking out an ultra‑low-capacitance cable: that would make your guitar sound brighter than theirs did. After all, the entire history of guitar sound is built on technological imperfections. Today we could make a guitar pickup with a 20Hz to 20kHz response, we could build perfect amplifiers with no audible distortion and we could connect those to state‑of‑the‑art speaker systems that cover the whole audio spectrum. But we'd hate it! It might be fine for quasi‑acoustic rhythm playing, but it just ain't rock & roll!

Another interesting and important fact about guitar cable capacitance is that it doesn't only act as a low‑pass filter. An electric guitar's pickups are made from coils of wire, which makes them highly inductive. Wire a capacitor across an inductor and you get a tuned circuit, rather like a mixing desk's mid EQ set to boost. In the case of a typical guitar, the tuned circuit is fairly well damped, because of the resistive parts in the volume and tone circuits and the resistance of the pickup coil itself, but you can still end up with a 1‑2dB hump in the response. If you choose low‑capacitance cable, the tuned circuit will resonate at a higher frequency, whereas a high‑capacitance cable will push the pickup resonance downwards. Either way, the tonality of the pickups will change.

This knowledge can be helpful, because if your instrument lacks sparkle, choosing a low-capacitance cable could improve matters significantly. What's more, you can check the outcome before spending your money by making up a very short conventional cable (a couple of feet should do the trick) and seeing how your tone changes compared with your standard cable. On the other hand, if your guitar sounds thin and lacking in punch, a standard high‑capacitance cable might actually make it sound better.

A further consideration, taking all this into account, is the behaviour of radio systems designed for guitar. Many of these ignore the cable capacitance issue, or simply put a low‑pass filter in the receiver, but that won't affect the resonant frequency of the pickups like a real cable does. If you use a radio system and find that the tone seems thinner, why not try wiring a capacitor of 200‑400pf across the jack plug at one end of the short cable that connects the guitar to the transmitter and see if that helps?

Summing up then, cables really do make a difference with guitars that have passive, magnetic pickups. However, there's no simple answer as to what type of cable will work best with your instrument because, just like the amplifier and speaker, the cable is part of your sound. Paul White 

Published November 2009