Audio tuning help - long read

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, Skimlinks, and others.


Oct 21, 2015
A few years back I put together some random notes I had kept on how to set up a stereo system and I posted on the Tacoma World forum I used to frequent. (Note: if I plagiarize without credit due, please forgive, it's been a while). Some ideas may be wrong headed, some may be my personal opinion, but I offer them for anyone trying to get really good sound but not knowing how to proceed (without throwing excessive amounts of money at it). I apologize for the length - and multiple posts (wouldn't all fit in one) - and do welcome updates, critiques, etc.

********* Begin Lee's treatise on stereo originally on Tacoma World *******************

One of my kids asked me about how I went about setting up a car stereo system. So, I pulled together some notes I've written to myself over the years (gathered from many sources that I failed to track and therefore sadly cannot credit - hope I am forgiven) - trying to describe in one place a "basics" course on how to accomplish audio tuning in the mobile environment.

Based on how big this document turned out to be, I may not be offering all that much of an easy solution to TW, but I'm posting in the hopes it helps one or two people anyway.

Disclaimer: I honestly have doubts about my skills to do this, especially given the knowledge I see from several of the members. I'm not a musician, I'm not an acoustic engineer, I'm not a sound professional, but I am an enthusiastic listener. I've learned some things in the 42 years since I installed a system in my 1st car, and I've been forced to discard things I "learned" that turned out not to be true – and who knows how many falsehoods I'm still harboring. I also lack experience in the Tacoma - what works, what doesn't in terms of specifics. Finally, I'm not even a particularly good writer - so please critique & comment.

I hope to cover the steps below in this thread. How each step is accomplished is somewhat gear dependent, and I'll do my best, but see the disclaimer paragraph above.
0. Preface
1. Eliminate system noise
2. Check speaker polarity/phase
3. Start with a clean signal
4. Setting Gains/smoothing frequency response
5. Equalization

0. Preface
As in any good project, a baseline/starting point has to be established, so:

a) There is no substitute for attendance at live concerts, preferably by a great musical ensemble in a great environment (what I mean here is NOT along the lines of attending the early Beatles concerts held in baseball stadiums). As trite as this might sound, you need to know what music is supposed to sound like. You are best served by hearing music in an un-amplified environment. That way you learn what an instrument should sound like, not what a sound engineer thought should come out of a loudspeaker.

b) You should find a well recorded CD from an artist of your choice and listen to it. This should not be done on the OEM stereo, nor with some MP3 rip-off. It should have decent dynamic range (not overly compressed) and preferably contain a wide variety of instruments in type and frequency range (including male & female vocals). Play it on a high-end home audio system. If you can't afford that (I can't), then get the best set of headphones you can lay your hands on. Listen to the system with the all the bells & whistles set for "flat" output. Listen over and over until you know the recording by heart - this will allow you to recognize "faults" in the system you will be installing/tweaking.

c) Now that you know what a system should sound like, use your budget wisely and from a total system perspective to acquire what you think sounds good - after all it's your money. I tend to spend my $$ roughly in thirds (headunit, speakers, amplifiers), but YMMV.

Finally, I sincerely hope this doesn't come across as pretentious on my part - I hope other knowledgeable forum members would be kind enough to jump in advance this discussion.

1. Eliminate System Noise
The goal here is to eliminate 100% of the noise - basically something that can’t be done - but it's a goal, so we do our best (and hope for the best). What is noise? In a vehicle audio system, there are many potential noise sources, I’ll try to explain a few (source noise, thermal noise, induced noise and ground loops) and concentrate on those that respond to treatment.

Source noise. Source noise comes from the source (duh!), and the only treatment is to use a different source. Source noise includes things like a weak FM signal (listen to another station), bit jitter in the production run of a CD (buy a different CD), etc. A real-world example would be the CD of David Crosby’s " If I Could Only Remember My Name". The original master tape (and vinyl record) contained a fair amount of hiss. When the recording was moved to CD, a decision was made to not attempt removal of the hiss. The only treatment is to use a different source but none is available so this noise we must live with to listen to this music.

Thermal Noise. Thermal noise wouldn’t be a problem if we were willing to live at absolute zero (last time I checked that was -459 degrees Fahrenheit so I think even my Canadian friends - that visit Florida in the winter - would think it too cold). Thermal noise sounds like the hiss that you hear between stations on FM radio and in simple terms exists due to the random movement of electrons caused by heat. Since many of us seem to prefer existence somewhat near +70 degrees F, all we can do is hope the manufacturer of our equipment exercised careful design, then we can apply a little attention to detail in hopes of minimizing this type of noise. That detail would include ensuring adequate heat ventilation of the headunit, amplifier, etc. As a side benefit, the equipment should have a longer life with proper ventilation. Remember, the goal is minimize, it can’t be eliminated.

Induced noise. Induced noise has numerous causes, many of them somewhat insidious. Sometimes it sneaks into your system through the power wiring, or it might sneak in via wires placed too closely together. Hey, it can even get in through the air, but the biggest problem occurs due to power wiring and signal cables.

We all know our trucks are battery powered and a battery is a pure DC source right? Wrong (at least in our case). Although the point can be argued (a little), our trucks are alternator powered, and the electrical system is a nasty environment for electronic equipment. The alternator outputs high-current pulsating DC into the battery, which then shows up everywhere else in the electrical system. When it becomes induced noise it sounds like a whine whose pitch is proportional to engine speed. Courtesy of Mr. Michael Faraday, we know that if you put two wires in close proximity to each other, and one of them is carrying an electric current, you generate electricity in the other by a) moving one of the wires relative to the other, or varying the electric field. Since the alternator (notice the name) puts out alternating current, we have a case of varying electric field inducing noise current in nearby wires. Imagine the “receiving” wire as the RCA cable from the head unit to the power amplifier. Describing how to diagnose this problem would be long and involves things like muting plugs (a follow-on post?). This problem usually exhibits itself as a one over R squared phenomenon (doubling distance results in 4-fold noise reduction, so rerouting the cables away from other wires is the traditional method of attack). I found this out the hard way on a 2002 MR2 Spyder when I initially ran RCA cables under the sill panel at the driver’s side (left) door and found bad, bad alternator whine. I fixed it by running them down the passenger side door (note: I retained the run of power, ground, and remote turn-on wires down the driver’s side). As an aside, I use a set of StreetWires RCA cables that have little directional arrows printed down the length of them. I know of no scientific reason why the electrons would like to go one way versus another down the wire, but I sleep a lot better knowing I ran them with the arrows pointing from the headunit to the amplifier. YMMV, feel free to be a rebel.

EMI. The other (major) source of induced noise is electromagnetic interference (EMI). In this case, the EMI functions as the transmitting station, the sound system’s wiring becomes an antenna, and the amplifiers within the system become the receiver(s). To each their own, but that station doesn’t play my kind of music. Some examples of EMI are ignition noise (like using solid core ignition wires vice suppression core), or even a loud buzz that you might hear if you drive near a television transmitter, electrical power station, under powerful fluorescent light systems, etc. The solution here is to encase the system in a Faraday cage or otherwise use technology smartly in an attempt to minimize this. That’s why the HU is in a metal case, signal cables are shielded, professionals use balanced line systems, etc.

Ground Loops
Ground loops are nasty things. I used to work in a lab that had about 50 rods driven deeply into the ground over about an acre outside that hooked up to the electrical power system and we still had ground loop problems inside with some sensitive lab equipment. They are caused by the non-zero resistance of the wire/vehicle structure used in interconnecting the equipment. Typically, ground loops are created by a system (like a headunit and amplifier) having multiple connections into the grounding system. In vehicles this problem is made worse by assuming that all points in the metal frame make for a good ground. They do work as an effective ground for things like cigarette lighters and tail lights, but not so good for audio. The alternator in the battery charging system makes things worse because the frequency of its AC output (that’s why it’s called an alternator, not a generator) is within the audible range. The low impedance involved (high available current means low impedance) make filtering out alternator noise even more difficult. So, these voltage drops occurring between the various grounds in the body usually have alternator noise riding on them, which gets into an audio system via a ground loop. Fortunately, the solution is easily described and not too hard to implement. Find a place on the body (preferably near the amplifier since it has the most electrons scurrying about) and bring the grounds of all the electronic devices (except the speakers) to that spot. Secure the collective ground(s) to the body & voila problem fixed. And by secure them to the body I don’t mean through layers of paint or other insulator. If you can’t find a place for good clean metal to metal body connection you may have to sand down to bare metal, drill and fit a bolt/nut- with washers on both sides, attaching the ground through the use of wiring lug.
2. Checking Polarity/Phase
I have to interrupt this part with an announcement. First break out your audio system manuals and trace out the wiring. What you’re looking for here is ensuring that (+) on the output side of one device goes to (+) on the input side of the other all down the chain. It kinda goes without saying this fixes any (-) side issues. While you’re at it, might as well check to make sure right side stuff (connections) stays with right side stuff, and left for left. For you guys with rear fill speakers, I’m not going to explain why I’d recommend you ditch them, as that’s another story (hint: even go to a concert with performers both in front and back - I haven't). And for those of you running something like a 5.1 setup why aren’t you giving me advice instead of reading this.

Note: Right now you’re probably thinking about skipping this step. I know you lazy types (cause I are one too), you’re planning on just twisting the balance knob, see if the sound moves in the correct direction, and call it a day. Well, I predict down the road you might be sorry. You’ll especially regret that as your system complexity increases, i.e., adding line out converters, equalizers, DSP devices, crossovers, sub-woofers, etc.

Another interruption before we begin any tweaking of the system: you need to consider speaker type and placement. Ideally, one would use a full spectrum single speaker system, it avoids all sorts of complicating issues; however, and unfortunately, no one has produced a practical speaker capable of playing all those frequencies (and in stereo). The Tacoma comes with a two-way stereo system as OEM - tweeter and midbass (or mid) for each side - so I’m going to assume a two-way system (but not OEM) is what we intend to use. OK, sorry, I‘ve already changed my mind; I’ll consider the addition of a sub-woofer too. Moving along smartly to placement, I will further assume we are using the OEM locations. If you’re ambitious enough to relocate the mids and tweeters then more power to you, remember to try them out in a variety of locations, even just laying them in place without mounting, you can test imaging, etc that way before investing in the fiberglass work and whatever else that will be required. So back to OEM locations...but first, “Danger Will Robinson! – Lee’s opinion based on his personal taste to follow. The sound could still be improved somewhat by making the baffle board for the mid sloped so the speaker is aimed slightly toward an imaginary dome light in the center of the cabin. I further recommend the tweeter be placed so each points toward the windshield -- imagine a line connecting each tweeter to a point mid way up the windshield and 2/3 of the way across from each other - the result would be the imaginary lines cross mid-windshield about mid-dash in the depth plane. However, pointing speakers at a windshield can introduce things like comb-filtering problems, etc, and of course this might be brand dependent, etc., so I give that up leaving it for your experiment, and turn to the other tweaks promised.

Part 2a - Polarity. You did take your door panels off to check (+) and (-) connections right? So, don’t put them back on yet. To make sure all your speakers have correct polarity, unhook the door mounted midbass speaker (but don’t lose track of which wire goes where). Have a friend look at the speaker, then using a 1.5 volt battery (AAA to D, it doesn’t matter), touch the positive terminal of the battery to the positive speaker terminal, then briefly (less than one second) touch the negative terminal of the battery to the negative speaker terminal. This isn’t as easy as I just made it out to be since the battery is likely longer than the distance between speaker terminals. I used the AA battery holder from an old clock and ran wires out to connectors for my speakers and this has to be tailored to your own setup. If the speaker moves outward, the polarity is correct. If the speaker pops in, the speaker is hooked up backwards. To fix this, simply reverse the wires when hooking the speaker back to the amplifier. Do the same for the other side. Let me emphasize something here, DO NOT run this test on a tweeter, you could fry the tiny voice coil. And if you hold battery power to the midbass speaker (or use a battery with more than 1.5V out) you can damage it as well. If you have the simplest of aftermarket system you can stop reading this part now unless you’re just curious. I define the simplest system as headunit, amplifier, and two full-range speakers (no tweeter, no crossover, etc.).

A quicker way to do this is by using a commercially available polarity checker (around $100 & up), which BTW comes with a test CD. All you have to do is pop the CD in the head unit and hold the polarity tester in front of each speaker, watch the lights, follow the instructions to make any necessary corrections and you’re done. The big advantage here is that you can test for polarity of the system on all the speakers, including tweeters. I don’t own one, and have no direct brand experience to make a recommendation, so you’re on your own here. BTW, sometimes when speakers are not mounted close to each other (i.e., mids low on the doors and tweeters up higher - sound familiar to anyone), reversing the polarity between tweeters & mids makes the system sound better because it makes up for phase differences due to the distance. This effect may also be deliberately employed by a component speaker manufacturer to match up with the phase/slope of the passive crossover. You’ll just have to test and see which way sounds better to you, which BTW, is a perfect lead into

Part 2b Phase. Phase is an issue I wish I didn’t have to bring up. First, I don’t have enough engineering background to properly/technically treat the issue, and second, what little I do understand makes a full treatment so complicated it would have to be handled on a case-by case basis. Why? Well, a speaker’s output changes in phase with frequency and is brand/model dependent, phase changes with the presence of electronic components in the signal path (capacitors, inductors), phase changes according to how the speaker is mounted (Infinite baffle, sealed, ported, bandpass, etc), and phase changes according to the order of the crossovers (passive) used (6dB/octave, 12dB, 18dB, 24dB, each shifting phase by 90 degrees). There may be even more reasons for phase changes, these are just a few I know something about. But don’t give up yet, there is a method to address the issue without knowing all the answers. One advantage is gained by using the same speakers on each side so whatever phase changes are occurring they should be matched side to side (please don’t tell me have brand A speakers in one door with Brand B in the other, please).

Let’s assume you have the next step up in system complication, i.e., a component speaker system with separate midbass and tweeter, using a passive crossover between them to separate frequencies. At this point you need to be able to create a CD with pink noise and single frequency sine waves. Pick a point clearly well away from a crossover frequency (say 500Hz for the midbass, 5KHz for the tweeter). And burn the CD with about a minute of each frequency - mono BTW. Turn the volume to a comfortable level (no need to blast out the neighbors yet, but don’t worry that comes later so there’s still some fun to be had) and set the balance all the way right or left (doesn’t matter which for now). Get your head out there in the middle of the cabin and listen. Now bring the balance into the middle - did it get significantly louder, does the sound seem to come from the middle of your head - if yes, then you can move to the next frequency. If no, then try swapping (+)(-) leads on one of the speakers. Did this fix the problem - it should have. Basically what you’re looking for is maximum sound pressure level - which you will get with proper phase match. Remember, switch the lead on the speaker, not at the crossover or you will change the phase for the other end of the tweeter/mid speaker duo. I haven’t actually tested this, but if you have a Sound Pressure Level (SPL) meter, you could test for phase by trying out both ways of wiring and selecting whichever was louder - makes sense to me anyway, somebody please go try it and let me know.

Stepping up once more in complication, you just added a subwoofer. Now you need another frequency burned to a CD - this time it should be matched to your selected crossover frequency between mid and sub (I assume someplace between 50 and 125 HZ). Play this frequency with the sub off, then on, looking for the same sort of thing we did above, the sound should get significantly louder with a proper phase match. If not, this time swap the subwoofer leads. Note: your headunit may have a means of internally swapping phase for the sub (the woofer menu choice might be something like normal or reversed) which would undoubtedly be easier than pulling the woofer in/out.

If you step up again in complication, the usual next step involves bi-amplification and active crossovers. In which case, we’re really beginning to talk time alignment as an integral phase effect. I’m not going to address that in this thread, but suffice to say you’re going to need something that will produce precisely time synced signals and a means of graphically displaying same for a precise solution. The quick and dirty method is to simply measure path length differences and adjust for desired listening position (actually the easiest method would be to buy a modern processor such as the Alpine Imprint). This quickie method can be applied to a passive crossover system, but I have not had good luck in three different cars I tried so far. I (perhaps falsely) attribute the failure to phase issues inherent with the separation correction for time alignment/phase for the mid messes with the correction for the tweeter - hopefully you get the idea. However, if it works for you, then go for it.
3. Clean Signal
I already addressed (somewhat) noise so that isn’t what I mean by a clean signal. What I’m trying to set up here is the basics for a good sounding system.

Sound damping. Fixing vibrations through sound damping material has a very easy solution. Take the truck apart and layer on multiple sheets of sound damping material everywhere. Unfortunately that could add a few hundred pounds and cost a mini fortune. So, find the song in your collection with the most bass content and play it loud. Note any panel or part vibrating bad enough to make noise as that will definitely have to be addressed. Contrary to advertising (and IMHO faked test reports) all sound damping materials (Dynamat, Raamat, BrownBread, Second Skin, Stinger, the list is endless) damp sound through use of a viscous-elastic substance with the damping proportional only to the weight of the material. Unless you live in a year-round cool climate and or have a light color paint, I recommend you find a butyl-based material vice the traditional asphalt base. This would include, but not be limited to Dynamat Extreme (not their base product), Second Skin, Raamat, & others, just make sure it’s butyl based so it won’t sag/fall off in heat (trust me on this one, I drove a black MR2 and at work had to park out in the Florida sun every day). Cover anything that was vibrating with some damping material. I would also recommend a couple of pieces about 2 or 3 inches wide be placed down the length of the door on the outer skin (from the inside of course), parallel to the crash bars running down the middle. Put damping sheets on the back side of the metal inside skin around where the speaker will sit. Do the same for the outer side of the inner skin (where the speaker baffle gets screwed down. Finally, find yourself some closed cell foam product (ensolite is one) or other type of sound absorption product (deflex pad - or, Dynamat - DynaXorb, etc.) and glue up a sheet slightly bigger than the mid directly behind it to the inside of the outer door skin. Make sure your speaker and baffle plate are well sealed against the door frame. I prefer not to use MDF for a baffle, but lots of people have had success. If you use MDF and live in a wet/humid climate (like Florida) I suggest you seal it up with something like the resin used for fiberglass. An alternative product to wood/MDF is Corian or similar. I got mine free by dumpster diving outside a kitchen countertop shop (the price is right; however, those with more delicate natures can ask at the shop the for scraps and pay a few bucks so the guys can get a beer at lunch).

Before I get any farther along, how’s your wiring and cables - are they properly sized? Is your system properly fused (within 18” of the battery)? Are the speakers properly mounted and housed? These topics are a real bag of worms and could fill up several posts all by themselves. Given the space limitations in the Tacoma and the difficulty of implementing an Infinite Baffle subwoofer arrangement, I hope you have opted for a sealed subwoofer enclosure. Anyway, like I said that’s a whole different thread.

Now set all your sources and processors "flat". If you have it, turn the loudness off. Defeat (turn off) all bass, mid and treble controls on the headunit (basically set them to the middle of their range). If you have one, set all the equalizer bands to 0dB. Defeat any sound processing such as decompression routines (used to “restore” lost MP3 info), any sort of reverb capability, and automatic sound level mechanism. Balance and fader controls should also be in the middle of their range.

4. Gains/Smoothing
Gains. Now we begin to determine the maximum parameters using a clean sound. I’m going to assume your amplifier does not have a power rating that significantly exceeds the rating for the speakers. And by rating I mean the RMS rating for both the speakers and amp - not the somewhat bogus Max Power rating frequently advertised. If the amp does exceed the speaker rating, then the process gets a lot harder and I can help some, but it will have to be on a case-by-case basis. Maybe someone else on the forum has a good method for handling this, because mine is not. Anyway, back to the process in which the power ratings are compatible. Set the gains on all the amps and processors to a low power position. Have a subwoofer? Then you’ll need a CD with a sine wave tone near the middle of its operating frequency. For example, my sub has a low pass filter at 90 Hz, so I used a 50Hz tone. You will also need a tone well away from any crossover frequency for the mid speaker. Mine runs from 90Hz up to about 3KHz, but I picked 500Hz to ensure I didn’t have power bleeding over to the tweeters. I’d suggest recording about a minute of each tone.

If you own or have access to an oscilloscope you are way ahead of the game. Hook it up to the speaker outputs of the amplifier. Start with your headunit and play one of the tones turning up the volume until the signal begins to clip, graphically it will look like the top of the sine curve is getting flattened. If you have a separately driven subwoofer, you will have to run this test twice, once with the sub, again with the mid frequency. The test is run twice because some headunit OEMs do not have the same voltage level output on the sub channels as they do for the front/stereo channels - older Alpines were notorious for this. Note the lower position found, because you should plan to never turn the volume control past this point. Now turn your attention to the amplifier gain control and do the same thing. If you have a subwoofer you will have two settings, one for the sub channel, and the other for the midbass/tweeter channels. Same idea, i.e., note where the signal begins to clip.

If you don’t have an oscilloscope, then the easiest method I know involves using your ears. Starting with the headunit again (and performing the tests twice if you have a sub), turn up the volume until you hear the tone change slightly in frequency. What’s happening here is as the signal begins to clip (forming a square wave) it generates higher frequency components in multiple of the base frequency (harmonics) and this higher tone you hear is a dead giveaway for clipping. Back it off until the tone drops back down and note the volume level. Do it for the sub and mid, as above, and note the position of the gain. You should also plan to never turn the gain beyond this point for a clean signal (note: if your amp is at or near the rating of the speakers, going beyond this point also introduces a serious risk of frying the voice coil, something usually not covered under warranty).

Smoothing. If you have a subwoofer, it’s time to break out the CD with the frequency at your crossover point. I’d suggest a mid volume point on the headunit and you want to turn off the sub (either through the headunit directly or via minimizing the amp gain control. Note the sound volume playing through the mid. Now switch over to the sub with the midbass speakers muted and adjust the gain so the volume is equal, It would be a help here if you had a cheap sound pressure level meter such as sold by Radio Shack. Make sure as you tweak amp gains that neither side (sub or mid) exceeds the place you already determined was a clean maximum.

At this point you have a noise free system (or as close as possible), one with correct polarity and optimized phase relationships, gains matched, and a clean signal/response. Your truck should be sounding pretty good and it’s time to deal with the topic of equalization, ranging from simple tone controls to parametric and graphic equalizers, and finally ending with digital signal processing modules. I know this has been a boat load of stuff presented before I even get to a bass/treble control, but please keep in mind equalization is NOT the way to compensate for installation flaws.

5. Equalization (EQ)
Here’s the problem: For all intents and purposes, when dealing with a closed vehicle, you are inside the speaker enclosure. Your position is fairly static - the driver can’t really move to a “sweet spot” like in the home environment. Reflecting surfaces and absorptive surfaces are where they are, unless you’re willing to do major surgery to the interior. The ambient noise level is fairly high, and the frequency content of the noise is not flat like it tends to be in a home. The choice of mounting locations and the available space is limited. And last, but not least, the vehicle was designed as a means of transportation first, and stereo heaven last (if at all). Bottom line, as an audio environment, the vehicle leaves a lot to be desired. Aside from the limited space, the OEM locations in door panels, woofer in the rear are way less than optimum. Also, when you consider the interaction of the speakers with the interior, peaks in the response curve occur when a reflecting surface occurs at odd fractions of a wavelength. Finally, peaks occur at the resonance of the interior. Although I’m not treating the issue, those of you with enough budget to employ active crossovers can try changing the crossover frequency and slope before implementing tweaks in equalization. The method of attack then ranges from as primitive as bass/treble controls up through graphic equalizers and ending with full-up digital signal processing (which includes parametric equalizers).

The solution. Essentially what we have to do here is to measure the frequency response of the installed system, then make the necessary corrections. The net result is a sound system having smooth frequency response, with carefully applied shaping to overcome the limitations of the environment. The best option is to use a professional Real Time Analyzer (RTA) and tweak away. If you have access to such a device I’m going to assume you don’t need my help. The second best option is to get a SPL meter (Radio Shack sells them for less than $60) and create a test CD producing tones (containing both pink noise and sine waves at least every 1/3 octave - actually if you don’t have the computer/RTA software you don’t need the pink noise, and if you do have the computer/RTA software you don’t need the sine waves, but it couldn’t hurt to burn them both to a CD either). The third tier down (and paradoxically it’s simultaneously the worst and best solution) is simply to use your ears.

Assuming the Radio Shack SPL meter (at least the mdoel I own), but no computer to run RTA software, you can record the sound pressure level readings from the meter at each frequency, then draw a response curve the same way an RTA does. In this case, you will be using your eyes to note the SPL values and recording them so you can determine where you need to apply equalization. Set your meter to slow "C" weighing and add the following values to each measurement: 20Hz add +7.5, 25Hz add +5, 31.5Hz add +3, 40Hz add +2.5, 50Hz add +1.5, 63Hz add +1.5, 80Hz add +1.5, 100Hz add +2, 125Hz add +0.5, 160Hz add -0.5, 200Hz add -0.5, 250Hz add +0.5, 315Hz add -0.5, 400Hz add +0, 500Hz add -0.5, 630Hz add +0, 800Hz add +0, 1kHz add +0, 1.25kHz add +0, 1.6kHz add -0.5, 2kHz add -1.5, 2.5kHz add -1.5, 3.15kHz add -1.5, 4kHz add -2, 5kHz add -2, 6.3kHz add -2, 8kHz add -2, 10kHz add -1, 12.5kHz add +0.5, 16kHz add +0, 20kHz add +1.

The simplest & nominally best way to perform the analysis is to use pink noise and a real-time analyzer (RTA). The RTA looks at the audio spectrum in a series of constant-bandwidth chunks and displays the result on a number of bar graph displays. Each vertical row of the display represents the level of signal strength in that portion of the spectrum. Typical RTAs measure the audio spectrum in 1, 1/2 or 1/3 octave bandwidths. Here’s what to do:
1. Play a CD with calibrated pink noise.
2. Connect the microphone or SPL meter to the input of the device you will use for RTA (typically a sound card on a laptop).
3. Adjust the output of the amplifier so that the average sound level in the space to be measured is at least 6dB above the ambient (background) noise level.
4. Place the microphone at the listening position.
5. Observe the display of the RTA. The RTA’s display represents the overall frequency response. Note: the response is that found at the microphone position, and the result can vary widely based on the position selected. As a result it’s a good idea to make measurements at several points within the space, averaging the results.

Now if you use an equalizer to provide a frequency response curve that is the inverse of what you’re seeing on the RTA (or your graph created using the SPL meter), you’ll have flat frequency response. Now add bass boost to suit personal taste. Remember that you can always change the amount or placement of the boost to accommodate your taste.
As simple as this sounds, there are pitfalls. The actual equalization process is not quite as simple as just using the equalizer to make things level and smooth. Here is a relatively simple method to get you going:
1. Start with the highest peak in the overall response and reduce it to approximately that of its neighbors. You may find you need to use several adjacent controls to control the overall response. For those of you with parametric equalizers, you may also need to tune the Q factor.
2. Repeat step 1 until the overall response is as flat as you can achieve - it’s a virtual certainty that you can NOT get it totally flat.
3. Apply any corrections needed for personal taste when you’re actually out on the road and dealing with wind/engine noise, etc. You may need a rather substantial boost in the low frequencies to overcome real-world driving conditions noise. One danger is in trying to boost a particular frequency band to overcome a dip in the frequency response. From an amplifier power standpoint, this can lead to a damaged speaker. Lucky for us, our ears don’t hear dips as well as they hear peaks, so use moderation in going after the dips. If you do decide to go after a dip in the frequency response, use the least amount of boost necessary to do the job. A 3dB boost means that your amplifier must deliver twice the power at those frequencies. You should go back and check the amp at this frequency to ensure you do not drive the amplifier into clipping with your settings.

If you can't get either the RTA or the SPL meter, you will just have to rely on your ears. Hopefully you read my preface where I stressed attendance at live performances and knowing your test music CD inside & out - now you know why. Keep tweaking and measuring until you are happy with the results.

One thing to keep in mind, a perfectly flat sound curve will not necessarily sound great. You will ultimately have to use your ears to fine tune the system. Do this through the use of different types of music. Even if you don't like to listen to jazz or classical music, they are a great resource to set systems. Remember that at this point you are NOT listening to music; you are listening to your system. Music should appear to come from the front of the vehicle. The singers/band should seem to be up and in front of you. Classical music is very good for this because of all the different instruments that are used, covering pretty much the entire audio spectrum. The system should be completely transparent. The whole purpose of the system is to give you the illusion that the music is coming from a live band. You might be asking, Ok, you’ve just made the system flat and now you tell me to make it un-flat...what gives? The whole point of using electronic instruments to make the measurement is to remove the variable of human hearing. It’s a fact that our long term memory for sound isn’t very good. It’s easy to get used to listening to something, even if it doesn’t sound particularly good. Add in the fact that our ear response is variable from day to day, and you end up with the ear as an un-reliable test instrument. On the other hand, once we’ve got the system to a known reference point, our ears are exactly what’s needed to make that final finishing touch.

Users who are viewing this thread

Top Bottom