View Full Version : EQ'ing a sound system


Gezzza
20-10-2003, 03:27 PM
Taken from Alto Mobile (http://www.altomobile.com) website :)

Provided By Kevin O'Byrne - Alto Mobile Tech Support


EQ-ing Your Sound System
---------------------------
Whether you are a sound-off competitor, a professional installer or just someone who likes their music, you are not going to be able to create a really great-sounding system unless you know how the music should really sound.

The ancient art of listening hasn't changed much over the years; the problem is, most people hear but don't listen. The easiest and probably the best way to learn what music sounds like at source is to visit a local jazz club or cafe-bar where a trio of piano, bass and drums is performing live. Get close to the group and listen. Take in the sound of the piano and the resonance of the double bass; listen to the bass drum, the snare and the cymbals. Now all you need to do is recreate that live sound in a car.

Of course a lot of the music we listen to nowadays – particularly dance music - is purely synthetic, created by music and drum synthesizers, samplers and vocoders. Even so, knowing what sounds right in natural instruments will take you a long way towards reproducing the sound that the recording engineer mixed at the studio, because all the music we make – even purely electronic music – is inspired by the natural sounds that we find pleasant and stimulating.

High Fidelity means faithful, accurate reproduction of the recorded work - those whose audio systems play booming bass and shrill high frequencies have obviously never really listened to music. How well the system reproduces the instruments in the various frequency bands is vital. But which instruments should you be listening to when auditioning sub-bass, midrange, mid-bass and high frequencies?

Auditioning tonal balance

You will not be able to successfully setup a sound system unless you have experience of how a particular instrument should sound and which instruments are good for revealing strengths and weaknesses in a sound system.

Electric bass guitar provides a good pointer to the quality of the mid-bass reproduction. A system which lacks mid-bass will make the guitar sound as if it is being played behind closed doors in the next room (poorly defined), while one that lacks sub-bass will make it sound light and thin.

The snare drum is a particularly good instrument for revealing tonal balance across quite a wide range of frequencies. Its main output is in the region of 400-500Hz, but if mid-bass is lacking then the drum will sound thin and boxy. It's great for revealing a midrange that is sitting back or too far forward in the mix, and it can sound more like a tambourine if the mid-highs and high frequencies are wrongly equalized (often in an attempt to give more presence to vocals).

You'll often find midrange dynamics on the snare badly suppressed if the speaker is being asked to work too hard in the bass or mid-bass region. On the other hand, it's possible to achieve striking dynamics on the snare by boosting some of the frequency bands, but this can leave piano sounding thin and brash.

Synthesized bass (usually played on a synthesizer keyboard) is unusual because it can contain very low sustained frequencies – sub-bass of around 40Hz or below - but at the same time produce strong output much higher up the frequency scale. Although not a natural instrument, synthesizers can provide a very good guide to the tonal balance of a system because they are capable of a very steady output level across an unnaturally wide frequency range, but they need to be auditioned with care. The nature of electronic instruments makes them very variable. With a synthesizer, the aim of the musician and recording engineer is often to create an interesting new sound, so the sound will be heavily processed. Although natural instruments like piano and saxophone are also likely to be processed in the recording studio (to make them slightly brighter or to add reverberation), they usually retain their distinctive sound and this makes them a more reliable and consistent reference.

Classic analog synthesizer tones are popular when testing SPL vehicles (car audio systems designed predominantly for sound pressure level competition) since they are capable of producing low fundamental frequencies, heavy output through the sub-bass region and amazing dynamics. But for sound quality vehicles it is generally advisable to use piano or organ, which also span a wide frequency range but provide a more reliable reference. By the way, some church organ music contains frequencies lower than most synthesizers can produce.

Bear in mind that a lot of today's popular music sounds as if real instruments are being played, when in fact the natural instrument sounds are coming from sampling keyboards. These can sound extremely convincing and can be used to audition and setup your system, but sometimes the musical scope and processing extremes of these instruments can mislead you. But whatever music you choose to use, the key is to listen to a bunch of tracks on as many top quality sound systems as you can. That way you get to know just how good it can sound. Then the trick is to get it sounding as good or better in your own vehicle.

Vocals (male and female), piano and saxophone are favourite sounds for judging midrange performance. Because most of us hear people talking every day (and due to the natural sensitivity of our ears across this particular frequency band), it is relatively easy to hear when the vocals are unnatural (though it takes a little more experience to know how to fix it!).

The saxophone sound is beautifully rich in harmonics, tenor sax in particular, and provides an excellent indication of the system's performance in the lower midrange through to upper midrange.

Cymbals and other metallic percussion instruments tend to be best for revealing the performance of tweeters, but brass and wind instruments can be good for this too.

Once you know on which instruments to concentrate when auditioning a sound system, it's a case of listening to the music tracks as often and as carefully as possible, using a high quality home hi-fi system or known high quality in-car system. Then try to make them sound that way in your vehicle using the Alto Mobile Control 8 or Control 16 dedicated equalizers, or the Drive 20, Drive 30 or UCS PRO digital crossovers which all have powerful EQ sections.

TIPS for achieving a good tonal balance
** Choose your test/setup discs carefully, bearing in mind the points mentioned above. If you are entering one of the autosound contests for sound quality, always setup your system for the contest using the disc the judges will use.

** If competing in a sound quality contest, the judging will usually be performed while the vehicle is parked and without the engine running, so make sure the system is setup for this. Specifically, check that the level of bass is not too high.

** Check speaker crossover points and amplifier gain settings, and subwoofer cabinet size, build quality and vent dimensions.

** To avoid buzzes and rattles, make sure everything is tightened down - door mechanisms, window regulators, door release handles. Dampen down any loose trim.
Use sound deadening material or a spray-on treatment to prevent the larger metal panels of the vehicle from resonating, which can cause peaks to occur at certain frequencies. Treat especially the door skin, rear doors/side panels and the sides of the trunk.

** The tonal balance will often change with volume level and this is indicative of variations in speaker sensitivity, a poorly tuned bass system (check the vent size) or simply the non-linear response of the chosen system components (especially the loudspeakers). If you have access to a real-time analyzer (RTA) then measure the frequency response at the different volume settings to help you to pinpoint the problem. Adjusting the speaker crossover points (to create an overlap or widen the gap between adjacent drive units) can help to overcome speaker sensitivity variations.

Gezzza
20-10-2003, 03:30 PM
cont.

EQ-ing Your Sound System
Chapter 1 - Introduction
This is the first chapter of a discussion which we will keep adding to as time permits. Let's start with some basics. Firstly, here is a diagram showing the relationship between the range of musical instruments and the frequencies they cover.

If you are fortunate enough to have access to a Real Time Analyzer (RTA) and quality microphone, and use pink noise to setup your system, needing to know the frequencies of musical instruments may seem unnecessary. But you should never simply go with what the RTA tells you - it gives you a great starting point for helping to pinpoint problems, but ultimately you need to use your ears and the music you generally listen to in order to fine tune your system. And to do that you need to know where to adjust the equalizer if the snare drum sounds boxy or the lead vocalist sounds as if she's singing from under a blanket.

And if you don't have an RTA, then you may have to rely on your ears and your music. But that doesn't necessarily put you at a disadvantage. I have heard plenty of systems set-up purely according to the print-out from an RTA that measure great but sound quite dreadful.

http://www.altomobile.com/assets/images2/musical_range.gif

Equalizers, CD test discs and Real Time Analyzers (RTAs) will generally provide control, test tones and read-out across a standard range of frequencies. Most commonly this will be at one-third octave steps. The following frequencies are the ISO standard 1/3 octave steps in Hz:

20, 25, 31.5, 40, 50, 63, 80, 100, 125, 160, 200, 250, 315, 400, 500, 630, 800, 1K, 1.25K, 1.6K, 2.0K, 2.5K, 3.15K, 4.0K, 5.0K, 6.3K, 8.0K, 10.0K, 12.5K, 16.0K and 20.0K.

You will see that these steps correspond with those in the diagram.


Chapter 2 - Measuring Systems
Sorry guys, I'm putting this stuff on EQ together in bits and pieces in no particular order, but I'm trying to construct some kind of logical progression in terms of chapters.

The easier way to measure the tonal balance inside a vehicle is to use a Real Time Analyzer (RTA).

A pink-noise signal is generally used as the input. This is a signal that gives a spectrum of noise which is linear (ie. identical) in its output level across the audible frequency band. This is usually injected into the audio system simply by using a test CD played from the vehicle's CD head unit. It is possible to plug a dedicated pink noise generator into the amp's input, but that would then side-step any non-linearity caused by the head unit.

The RTA reads the output level of the signal entering its measurement microphone, which is generally positioned on a tripod at head height in the driver's seat position. The RTA will analyze the level at each of the one-third octave frequency points covered in Chapter 1. This can then be read out on a display or printed out and used to pinpoint areas needing attention, such as dips or peaks in the response.

RTAs used to be (and most hardware versions still are) expensive, but there are ways around the cost. If you have a laptop computer, there is some very powerful software available that will help you turn it into a full-function RTA, though you will also need a good measurement microphone and probably an external sound module (the internal soundcards are generally not designed for measurement purposes and won't give linear results across a wide enough range).

There is also another solution, using a handheld SPL meter of the kind you might find at a Radio Shack store. In combination with a test CD that has one-third octave spot frequencies (that is, individual tones at each of the standard ISO frequencies), it is possible to use a simple SPL meter to plot the frequency curve of the inside of your car. This is effectively what an RTA does - except that an RTA measures all 30 bands in one go and presents the results to you. Nice and easy. With an SPL meter it's a touch more laborious, but with care the results can be surprisingly accurate.

Weighting
It is important to understand the 'weighting' settings you'll sometimes come across on a meter. Often you'll be given a choice of A-weighting or C-weighting.

A-weighting has a frequency characteristic which favours the frequencies falling in the 500Hz to 10KHz range, which is the range across which the human ear is more sensitive - so it effectively gives an SPL figure corresponding to how the human ear would hear it. But if you use the measurements taken with A-weighting switched in, this can mean that you over-compensate at the lower and upper end of the scale when you EQ the sound.

C-weighting causes the meter to respond in a linear way across a wider scale - about 30Hz to 10KHz. If you had a true 'LINEAR' setting on your meter, that would (at least in theory, depending on the quality of the mic etc.) give a linear test result across the entire audible band from 20-20K, which would be the ideal. But where the choice is A or C weighting, select C.

Here's a diagram comparing the different weighting curves:
http://www.altomobile.com/assets/images2/weighting_curves.gif


Provided By Kevin O'Byrne - Alto Mobile Tech Support

Gezzza
20-10-2003, 07:28 PM
and a bit more

http://www.btinternet.com/~gezzza/EQingSmall.jpg

larger version HERE (http://www.btinternet.com/~gezzza/EQing.jpg)

this came from www.ofsoundmind.com but i cant seem to find it on there site anymore:)