Car Audio Explained, Part 11: Bass in a Car: Cabin Gain, Front Bass, Enclosures and Subwoofer Placement (Andy Wehmeyer's Lessons)

This is Part 11 of 19 in the series Andy Wehmeyer on Car Audio, Explained for Beginners. Start here · Download the full guide as a PDF

Unofficial guide. Andy Wehmeyer's points are paraphrased from his public forum posts, blog, tuning guides, videos and podcasts, with a link to each source; the plain-English explanations were written for this guide. Not written, reviewed or endorsed by Andy Wehmeyer or Audiofrog.

4. Bass, power and factory systems

Andy Wehmeyer treats bass as the easy part of car audio and bass that sounds like it comes from the front as the hard part: it depends on the car's cabin gain, a correct sub-to-midbass crossover and freedom from rattles and distortion. On power and gain he argues that music needs far less power than people buy, and that digital clipping, not too little power, is what kills speakers. On factory systems he says the OEM processing was built by experts, and that you should measure it and work with it before ripping it out.

What this section is about: This section covers three topics. The first is bass: the deep notes played by the subwoofer (a large speaker for the lowest notes) and the midbass speakers (usually in the doors). The second is amplifier power and gain settings. The third is factory-installed ("OEM") systems. In short, he says making bass in a car is easy, but bass that seems to come from in front of you is hard. On power, he says music needs far less power than people buy, and that digital clipping (waveforms chopped off inside the digital processing), not too little power, is what kills speakers. And he says factory systems were built by experts and should be measured and worked with before anyone rips them out.

4.1 How bass behaves in a car: cabin gain and the transfer function

What this part is about: This part explains how a car's cabin adds a lot of deep bass on its own (called "cabin gain"), how to measure that boost, and what it means when you choose a subwoofer box and tune it.

4.1 #1 · The car boosts bass more and more as frequency falls; that boost belongs to the car, not the woofer.

Cabin gain (the car's transfer function: in-car response divided by the woofer's outdoor response) keeps rising below 20 Hz, usually broken by a cancellation between 50 and 120 Hz. The in-car peak is where the box's roll-off outruns that gain, not a "car resonance"; an 8" and a 15" with the same outdoor response measure the same in the car. [RARE] Sources: DIYMA 2012-01-31 · DIYMA 2008-12-06

In plain English: Frequency, in hertz (Hz), is how many times per second a sound wave vibrates; the lower the number, the deeper the note. About 20 Hz is near the bottom of human hearing. He calls the car's effect on bass its "transfer function": the woofer's response measured in the car, divided by its response measured outdoors. That difference is the cabin gain. He says it keeps rising as the pitch falls, even below 20 Hz, and is usually interrupted by a cancellation dip somewhere between 50 and 120 Hz. Many people believe the bass peak they measure in a car is a "car resonance". He says it isn't: it is simply where the box's natural loss of deep bass (its roll-off) starts to outrun the car's boost. Think of the car as a fixed tone control that is the same whatever woofer you put in. That's why, he says, an 8-inch and a 15-inch woofer with the same outdoor response measure the same in the car.

4.1 #2 · Every box type gets cabin gain; the car changes what you hear, not what the sub does.

Sealed boxes look flat in cars only because their 12 dB/oct roll-off matches the rise; vented boxes stay flat lower and then fall faster, so they show a peak. [RARE] Sources: DIYMA 2009-08-28 · DIYMA 2015-06-07 · DIYMA 2021-05-08

In plain English: A sealed box is airtight. A vented (ported) box has an opening, the port, that adds output around one low "tuning" frequency. Below its limit, every box loses output at some rate; for a sealed box that is 12 dB per octave (an octave is a halving or doubling of frequency, and a few dB is a clearly audible change). He says every box type gets the same cabin gain. A sealed box only looks flat in a car because its 12 dB/oct loss happens to match the car's rise, so the two cancel. A vented box stays flat down to a lower pitch and then drops faster, so the car's rise shows up as a peak. Many people believe one box type suits cabin gain better. His point is that the car changes what you hear, not what the sub itself does.

4.1 #3 · Where the sub sits barely matters below 40–50 Hz but shapes everything above 50 Hz.

In his tests and Harman's, moving a box around a car changed nothing below 40–50 Hz; above that, cancellations appear, and the first cabin mode can be predicted from the car's size. Overall level depends on car size (an MG vs a Suburban, not a Jetta vs a Passat). [RARE] Sources: DIYMA 2008-07-05 · DIYMA 2008-12-06

In plain English: He says that in his own tests and in Harman's (the big audio company where he worked), moving a sub box around a car changed nothing below 40–50 Hz. At those low pitches a sound wave is many feet long, longer than the cabin. Above about 50 Hz, placement matters: cancellations appear, where waves meet out of step. He adds that the first cabin "mode" (the first strong resonance of the cabin's air, which he says usually shows up as the first big dip above 50 Hz) can be predicted from the car's size. Overall bass level also depends on size, but only when cars differ a lot: a tiny MG sports car versus a big Suburban SUV, not a Jetta versus a Passat (two similar sedans).

4.1 #4 · Measure cabin gain with a ground-plane test.

Measure at the seat; then put the box outdoors on the ground, 20–30 ft from walls, with the mic on the ground 2 m away at the same drive level (equal to an anechoic 1 m test). The difference is the cabin gain; this works only if woofer and port share a baffle. Sources: DIYMA 2009-08-28 · DIYMA 2008-10-12

In plain English: This is his way to measure how much bass the car itself adds. Step one: measure the sub inside the car, with the microphone at the seat. Step two: take the same box outdoors and set it on the ground, 20–30 ft (about 6–9 m) from any walls. Put the mic on the ground 2 m (about 6.5 ft) away and drive the sub at the same level. He says this equals a 1 m test in an anechoic room (a special echo-free test room). Step three: the difference between the two measurements is the cabin gain. He notes one condition: it works only if the woofer and the port are on the same baffle (the same face of the box).

4.1 #5 · "Bass is the easiest thing to make in cars. Bass up front is not."

The real work is 60 Hz–1 kHz. The cabin is a leaky resonator: soft panels leak, the seat filters between cabin and trunk, and windows act as passive radiators; find resonances by recording a mic while you whack the glass and sheet metal. [RARE] Sources: DIYMA 2008-11-20 · DIYMA 2011-12-24

In plain English: The quote means that getting a lot of deep bass in a car is easy; making it sound like it comes from in front of you is not. He says the real work is in the 60 Hz–1 kHz range (midbass and lower midrange). He describes the cabin as a leaky resonator: a space that rings at certain pitches but also lets sound escape. Soft panels let bass leak out. The back seat acts as a filter between the cabin and the trunk. Windows act as passive radiators, panels that get pushed by the air pressure and give off sound of their own, a bit like a drum skin. To find where the car rings, record with a mic while you whack the glass and the sheet metal.

4.1 #6 · Don't design the box or pick the woofer to be the car's inverse; build the most efficient box you can fit, then EQ.

Matching a sub's F3 (−3 dB point) to the car is "a ridiculous crapshoot". FIR inversion can't cancel the car either: nulls can't be filled, big boosts lack power and excursion, and the fix holds at one head position. Sources: DIYMA 2008-06-24 · DIYMA 2013-01-30 · DIYMA 2013-02-04

In plain English: F3 is the frequency where a speaker's output has dropped 3 dB below normal. Some people try to pick a woofer or design a box whose bass loss exactly mirrors the car's bass boost, so the two cancel. He calls matching F3 to the car "a ridiculous crapshoot". His advice instead: build the most efficient box that fits, then use EQ. He adds that FIR inversion can't cancel the car either. (FIR is a kind of digital filter that can be set to the exact opposite of a measured response.) He gives three reasons: nulls (deep cancellation dips) can't be filled; big boosts need more power and cone movement (excursion) than you have; and the correction is right at only one head position.

4.1 #7 · No EQ available? Match the sealed box's −3 dB point to the car's +3 dB point.

That gives flat lows, but he prefers vented plus a cut of the in-car peak (about 12 dB near 45 Hz in his example); short on space, use a high-Xmax woofer and lots of power. Sources: DIYMA 2008-05-25

In plain English: This is a fallback for a system with no EQ, from a 2008 post. If you choose a sealed box whose −3 dB point (where its output has dropped 3 dB) sits at the frequency where the car's gain has risen 3 dB, the two cancel and the lows come out flat. Even so, he prefers a vented box with EQ cutting the in-car peak, about 12 dB near 45 Hz in his example. He gives the reason: lower distortion and better dynamics, because power is left over for loud peaks. If you are short on space, he suggests a high-Xmax woofer (one whose cone can travel far while staying clean) with lots of power.

4.1 #8 · Cut the cabin-gain peak, but by less than the graph says.

For a boomy sub he cuts 6–10 dB rather than the 15 dB measured. A low-tuned box "lacks" 60–80 Hz only next to its 30–50 Hz peak: cut the peak and raise amp gain, which also lowers excursion and distortion. Sources: DIYMA 2015-10-09 · DIYMA 2011-12-26

In plain English: EQ lets you cut the peak that cabin gain creates. He says to cut it, but by less than the measurement suggests: for a boomy sub he cuts 6–10 dB rather than the 15 dB the graph shows. The source doesn't spell out exactly why he stops short. He also says that a box tuned low (one whose port works at a very low pitch) only seems to "lack" 60–80 Hz because it is being compared with its own 30–50 Hz peak. His fix: cut the peak and turn the amp gain up. That evens things out and also reduces cone movement (excursion) and distortion.

4.1 #9 · Open windows, sunroof or trunk turn the cabin from a leaky sealed box into a low-tuned vented one; EQ toward that sound instead of cutting holes.

Only very low bass rises. Tune with everything closed and save a second preset if you like the open sound; sealing the sub box to the cabin reduces the trunk-open boost but not the window-open one. [RARE] Sources: DIYMA 2008-08-25 · DIYMA 2013-08-25 · DIYMA 2018-03-23

In plain English: With everything closed, he describes the cabin as a leaky sealed box. Opening a window, the sunroof or the trunk turns it into a vented box tuned very low, with the opening acting like a port. Only the deepest bass rises. Some people cut holes in the car to get that sound. He says to EQ toward it instead. Tune with everything closed, and if you like the open-window sound, save it as a second preset. He adds that sealing the sub box to the cabin (so it doesn't play into an open trunk) reduces the extra bass when the trunk opens, but not when a window opens.

4.1 #10 · Convertibles get little cabin gain, so build a bandpass box and fire its port into the cabin.

A soft top is acoustically transparent; as a young installer he routinely did this. [RARE] Sources: DIYMA 2021-10-30

In plain English: A convertible's soft top is acoustically transparent: sound passes straight through it, so the cabin adds little bass. His answer is a bandpass box. In a bandpass box the woofer is fully enclosed and all the sound comes out through a port, so you can point that port into the cabin. He says that as a young installer he did this routinely.

4.1 #11 · Home-style auto-EQ logic fails in cars because of cabin gain.

Even a 6" door speaker can measure nearly flat below 20 Hz in a car, so the first Alpine Imprint (Audyssey-based), which put the sub crossover at the fronts' measured −3 dB point, always landed at its minimum. Sources: DIYMA 2016-01-25

In plain English: Auto-EQ systems measure the speakers and set things up automatically. Audyssey is a well-known room-correction system for home theaters. Home logic says: put the subwoofer crossover where the main speakers' bass fades, their −3 dB point. In a car, cabin gain props up the bass so much that even a 6-inch door speaker can measure nearly flat below 20 Hz. So the first Alpine Imprint (based on Audyssey), which used that home rule, always put the sub crossover at its lowest possible setting. His point: this home-style auto-EQ logic fails in cars because of cabin gain.

4.1 #12 · One mic position is enough for bass.

Five positions in a 7" circle around the head measured almost the same below 200 Hz; they diverge from 200 Hz to 1 kHz and by 25 dB or more above 1 kHz. Keep bass EQ Q below about 6. [RARE] Sources: Audiofrog blog Jan 2021

In plain English: When measuring, some people average many mic positions around the head. He measured five positions in a 7-inch (about 18 cm) circle around the head. Below 200 Hz they read almost the same; from 200 Hz to 1 kHz they started to differ; above 1 kHz they differed by 25 dB or more. So one mic position is enough for bass. He also says to keep bass EQ adjustments with a Q below about 6. (Q is how narrow an EQ adjustment is; a higher Q is narrower.)

4.1 #13 · Road noise: quieter tires were his best-ever midbass upgrade, and EQ cannot remove noise.

Noise is usually negligible above 200 Hz, but on pebble-paved I-10 (El Paso–Van Horn) it mattered up to 500 Hz, so he kept a preset for that road. EQ only lifts music above the noise; record the noise and look at it in Audacity. [RARE] Sources: DIYMA 2016-01-25 · DIYMA 2008-05-30

In plain English: He says road noise is usually negligible above 200 Hz, so it mostly drowns out the low notes, and quieter tires were his best-ever midbass upgrade. EQ cannot remove noise; it can only raise the music above it. Removing noise itself takes active noise cancellation, a separate technology. On one rough, pebble-paved stretch of the I-10 highway (El Paso to Van Horn, Texas), the noise mattered up to 500 Hz, so he kept a separate EQ preset just for that road. To see your own noise, record it and look at it in Audacity (a free audio editor).

4.2 Getting bass to sound like it comes from the front: the sub–midbass crossover, phase and delay

What this part is about: This part is about the hardest bass job in a car: making the bass sound as if it comes from the stage in front of you, not from the subwoofer in the trunk. It covers the crossover that hands the bass from the sub to the midbass (door) speakers, how their phase (the timing of their waves) must line up, and what delay can and can't do.

4.2 #1 · Front bass needs a correct sub–midbass crossover and zero rattles, not delay tricks or very low midbass.

Any rattle or high-frequency distortion from the sub gives its location away, and bass 20 dB louder than everything else can't sound like it's in front. From 2017 to the 2023 tuning PDF he gives two conditions: sub and midbass in phase at the crossover, and no distortion. [RARE] Sources: DIYMA 2013-01-03 · DIYMA 2017-08-17 · V2 PDF 2023

In plain English: The sub usually sits in the trunk, behind you. In a good system the bass still seems to come from the front. A crossover is the filter pair that hands the low notes to the sub and the next notes up to the midbass speakers. He says front bass doesn't come from delay tricks or from making the door speakers play very low. It needs a correct sub-to-midbass crossover and zero rattles. Deep bass on its own is hard to locate, but he says any rattle or higher-pitched distortion from the sub gives away where it is. Bass that is 20 dB louder than everything else (a very large difference) can't sound like it's in front either. From 2017 through his 2023 tuning PDF he names two conditions: the sub and midbass in phase at the crossover (their waves pushing together), and no distortion.

4.2 #2 · Distortion, not arrival time, is what makes bass localize to the trunk or a door.

Rear-sounding bass often improves when you turn it down, and door woofers driven near resonance become easy to find. Test it: add 2nd and 3rd harmonics to a low-passed track, remove the fundamental, and A/B it in the car. [RARE] Sources: DIYMA 2013-03-18 · DIYMA 2010-02-13

In plain English: To "localize" a sound is to tell where it comes from. Many people believe bass sounds like it's in the trunk because it arrives late. He says the cause is distortion: extra, higher-pitched tones the speaker adds that weren't in the music. A harmonic is one of those tones, at a whole-number multiple of the note: the 2nd harmonic of 40 Hz is 80 Hz, the 3rd is 120 Hz. He says rear-sounding bass often improves when you simply turn it down. Door woofers pushed down near their resonance (the low end of their range) also become easy to pinpoint. His test, step by step: take a track and keep only its bass (low-pass it); create 2nd and 3rd harmonics of that bass; remove the original bass notes (the "fundamental") so only the added distortion is left; mix that back into the track; then switch between the two versions (A/B) in the car and listen.

4.2 #3 · Kick-drum "impact" comes from the midrange (around the 3 kHz mid/tweeter crossover), not from 40 Hz.

With a crossover at or below ~80 Hz the sub carries no attack, so by the precedence effect (the first arrival sets location) it can sit almost anywhere unless boosted hard. Low-pass a punchy track at 70–80 Hz on headphones and the impact disappears. [RARE] [Conflict: in 2021 he also told dealers that a club-goer's "impact" comes from correct midbass–sub phase.] Sources: DIYMA 2009-12-02 · DIYMA 2021-09-10 · CMA Connected Oct 2021 @46:57

In plain English: "Impact" is the punch or snap of a kick drum. Many people believe it comes from deep bass around 40 Hz. He says it comes from much higher, around 3 kHz, near the midrange-to-tweeter crossover. He says that without that region right you won't hear the snap of toms or a bass player's thumb slap. The "attack" is the sharp start of a note. With the sub crossed at or below about 80 Hz, the sub carries none of it. Your brain judges where a sound comes from by whatever arrives first (the "precedence effect"). Because the sub carries no attack, it can sit almost anywhere without giving itself away, unless it is boosted hard. His test: low-pass a punchy track at 70–80 Hz (keep only what's below) and listen on headphones; the impact disappears. The [Conflict] tag: in 2021 he also told dealers that a club-goer's "impact" comes from correct midbass-to-sub phase (3.13 #12). These are two different statements about where impact comes from, and the sources don't reconcile them.

4.2 #4 · Big sub/midbass overlap destroys the image; a sub low-pass under full-range mains is "a low-pass filter, not a crossover".

Without the matching high-pass, the sub goes out of phase with the mains about an octave above a 12 dB/oct cutoff (half an octave at 24 dB/oct). Harman's six-woofer optimizer (the basis of JBL BassQ) gave good response but "no image in any seat". [RARE] Sources: DIYMA 2009-11-14 · DIYMA 2013-04-27 · DIYMA 2022-02-10

In plain English: "Overlap" means the sub and the midbass both playing the same range of notes. A low-pass filter lets only the lows through; a high-pass filter lets only the highs through. A crossover is a matched pair of the two. Filters also shift phase, and a real crossover works because the two halves' shifts are designed to fit together. "Mains" are the main front speakers. He says a big overlap destroys the image (the sense of instruments placed on a stage). If the mains play full range and only the sub is filtered, that is "a low-pass filter, not a crossover". Without the matching high-pass, the sub goes out of phase with the mains about an octave above the cutoff with a 12 dB/oct filter, or half an octave above with 24 dB/oct. For example, with an 80 Hz cutoff at 12 dB/oct, that is around 160 Hz. He also describes a Harman optimizer that used six woofers (the basis of the JBL BassQ product): it gave good response, but "no image in any seat".

4.2 #5 · Integration is "95% EQ and 5% phase" and needs real parametric bands per channel.

With separate EQ on left midbass, right midbass and sub, he can blend at any crossover point or slope if nothing rattles; a 31-band graphic on a sub gives about 6 usable bands. Feed the sub from one left and one right DSP output so you can tune each whole side, sub included. [RARE] Sources: DIYMA 2013-04-27 · DIYMA 2015-05-04 · DIYMA 2021-08-22

In plain English: "Integration" means blending the sub and midbass into one seamless bass. A parametric EQ band lets you choose the exact frequency, the amount of boost or cut, and the width (Q). A graphic EQ has fixed sliders at set frequencies. He says blending is "95% EQ and 5% phase". With separate parametric EQ on the left midbass, the right midbass and the sub, he can blend at any crossover point or slope, as long as nothing rattles. A 31-band graphic EQ used on a sub gives only about 6 usable bands. (A 31-band EQ spreads its sliders across the whole audible range, so only a handful fall in the sub's range.) His wiring tip: feed the sub from one left and one right DSP output. That way you can tune each whole side, sub included.

4.2 #6 · Default crossover: sub low-pass 60 Hz and midbass high-pass 80 Hz, both LR4 (24 dB/oct Linkwitz-Riley), because you will turn the sub up.

Raising the sub 9–12 dB to sound natural (or riding a bass knob) moves the acoustic crossover up and out of phase. His rule: lower the sub low-pass half an octave per 12 dB of sub boost so the acoustic crossover stays at 80 Hz, with slopes EQ'd to target. [RARE] [Changed view: 2010–2015 default was ~80 Hz LR4 on both sides; 2020–2022 the 60/80 Hz stagger.] Sources: DIYMA 2010-02-14 · DIYMA 2020-12-30 · DIYMA 2021-08-21

In plain English: LR4 (a 4th-order Linkwitz-Riley filter) is a common crossover type that cuts 24 dB per octave and is designed so the two halves add up smoothly. The "acoustic crossover" is where the sub's and the midbass's actual sound levels cross in the car, which can differ from the number typed into the DSP. His default: sub low-pass at 60 Hz and midbass high-pass at 80 Hz, both LR4. The reason for the gap is that you will turn the sub up. Raising the sub 9–12 dB so the bass sounds natural (or riding a bass knob) pushes the real crossover point higher, where the two speakers are out of phase. His rule: for every 12 dB of sub boost, lower the sub's low-pass by half an octave, so that the acoustic crossover stays at 80 Hz. Half an octave below 80 Hz is about 57 Hz, close to his 60 Hz. This assumes the slopes have been EQ'd to the target. The [Changed view] tag: from 2010 to 2015 his default was about 80 Hz LR4 on both sides; from 2020 to 2022 it was the staggered 60/80 Hz setting.

4.2 #7 · Or add the bass lift with a shelf or a shared 80 Hz boost, not by delay.

His tuning guide uses a 12 dB/oct low shelf near 60 Hz, set to taste (2009: never above 160 Hz, on the midbass channels too). The 2023 PDF: with the sub low-pass half an octave below the midbass high-pass, one boost near 80 Hz on both sub and fronts puts the bass in front with no delay changes. [RARE] Sources: DIYMA 2009-11-14 · DIYMA 2021-03-26 · V2 PDF 2023

In plain English: "Bass lift" is the extra bass most listeners want on top of a neutral tune. A low shelf is an EQ setting that raises everything below a chosen frequency, like a step. He says to add the lift with EQ, not with delay. His tuning guide uses a 12 dB/oct low shelf near 60 Hz, set to taste. In 2009 he added: never set it above 160 Hz, and apply it to the midbass channels too. His 2023 PDF gives another way. With the sub's low-pass half an octave below the midbass's high-pass, one boost near 80 Hz on both the sub and the front speakers puts the bass in front, with no changes to the delays.

4.2 #8 · His 2008 integration recipe.

One amp on both woofers plus a parametric EQ (ideally separate L/R); measure the sub alone, low-passed "ridiculously high"; cross where the mids roll off (~80 Hz); set the sub about 9 dB above the mids/highs with a smooth slope down to ~160 Hz; flip sub polarity if needed; don't boost the midbass. [Era: 2008] Sources: DIYMA 2008-11-20

In plain English: This is an older, 2008 step-by-step recipe. Step 1: run both woofers from one amplifier, and add a parametric EQ, ideally one with separate left and right controls. Step 2: measure the sub on its own, with its low-pass set "ridiculously high". Step 3: cross over where the mids and highs naturally roll off, about 80 Hz. Step 4: set the sub about 9 dB louder than the mids and highs, with the level sloping smoothly down toward about 160 Hz. Step 5: flip the sub's polarity if needed (polarity means which way the cone moves; flipping it reverses that). Step 6: don't boost the midbass with EQ. The [Era: 2008] tag means this is how he described it then; later points in this part (such as #6 and #7) show his newer settings.

4.2 #9 · Sub/midbass phase only has to stay far from 180°, and each slope type has its own correct phase.

At 60 Hz a 1 ft distance error is ~19°, and even 60° apart sums louder than either alone. LR4 sums flat at 0°; LR2 at 180° (flip polarity); 2nd-order Butterworth at 0° gives a +3 dB hump; 3rd-order Butterworth sums flat at 90°/270° in either polarity, so it offers no polarity check. No speaker needs slopes steeper than 24 dB/oct. [RARE] Sources: DIYMA 2018-01-21 · DIYMA 2021-08-21 · DIYMA 2021-12-03

In plain English: Phase, in degrees, says how far one wave is shifted against another. At 0° their peaks line up and they add; at 180° one pushes while the other pulls, and they cancel. He says the sub and midbass only need to stay far from 180°. At 60 Hz a 1-foot (30 cm) distance error shifts them only about 19°, and even 60° apart they still add up louder than either alone. Crossover filters come in types (Linkwitz-Riley "LR", Butterworth) and orders (2nd order cuts 12 dB/oct, 3rd 18, 4th 24). Each type has its own correct phase. LR4 sums flat when the two speakers are at 0°. LR2 halves come out 180° apart, so you flip one speaker's polarity. A 2nd-order Butterworth at 0° gives a +3 dB hump at the crossover. A 3rd-order Butterworth sums flat at 90° or 270° in either polarity, so flipping polarity won't reveal a wiring mistake: it offers no polarity check. He adds that no speaker needs slopes steeper than 24 dB/oct.

4.2 #10 · A dip at 60–90 Hz is usually sub-vs-front phase or a cancellation, and EQ can't fill a cancellation.

Diagnose with sub and fronts crossed at 90 Hz and rears off; flip sub polarity if a big dip appears. A null near 70 Hz needs a sub move, or a crossover below it with the midbass matched. Sources: DIYMA 2008-11-18 · DIYMA 2016-04-03

In plain English: A dip is a range of notes that comes out too quiet. He says a dip between 60 and 90 Hz is usually the sub and the front speakers out of phase with each other, or a cancellation (sound waves meeting out of step). EQ can't fill a cancellation; boosting only makes both cancelling waves louder. His diagnosis: cross the sub and the fronts at 90 Hz, turn the rear speakers off, and measure. If a big dip appears, flip the sub's polarity. If there is a null (a deep dip) near 70 Hz, EQ won't help: move the sub, or set the crossover below that frequency and match the midbass.

4.2 #11 · Polarity is not delay, and "half a wavelength of delay plus a polarity flip" re-aligns only one frequency.

If the high-pass/low-pass phase relationship is wrong, delay fixes it only at the crossover point; that is the #1 reason bass won't come forward. A true half-wave path error at 80 Hz needs ~7 ft of path difference, so it is rare. [RARE] Sources: DIYMA 2016-01-22 · DIYMA 2021-08-19

In plain English: Polarity flips a wave upside down at every frequency. Delay shifts it in time, which shifts the phase by a different amount at each frequency. A wavelength is the length of one cycle of a sound wave; at 80 Hz it is about 14 ft (4.3 m). A common forum trick is to add half a wavelength of delay and then flip polarity. He says that re-aligns the speakers at only one frequency. If the crossover's high-pass and low-pass halves have the wrong phase relationship, delay can fix it only at the crossover point, and he calls that wrong relationship the #1 reason bass won't come forward. A real half-wavelength error caused by distance would need about 7 ft (about 2.1 m) of difference in path length at 80 Hz, so in a car it is rare.

4.2 #12 · Set sub delay with a tape measure; it compensates distance, not crossover lag.

At 80 Hz, 1 in of error is ~2° of phase, so being off by a lot hardly matters; the DSP can only add delay, so delay the nearer drivers to match the farthest. [Changed view: 2009–2010 "don't time-align subs" (none needed with a crossover ≤80 Hz); from 2015 delay everything to the sub; 2018–2023 PDFs and 2021 posts tape-measure the sub too.] Sources: DIYMA 2010-11-21 · DIYMA 2021-03-26 · V1 PDF 2018

In plain English: "Crossover lag" is the extra delay a crossover filter itself adds. He says to set the sub's delay with a tape measure, to make up for its distance only, not for crossover lag. At 80 Hz an error of 1 inch (2.5 cm) is only about 2° of phase, so even a sizeable measuring error hardly matters. A DSP can only add delay, never remove it, so you delay the nearer speakers to match the farthest one. The [Changed view] tag: in 2009–2010 he said not to time-align subs at all (none needed with a crossover at or below 80 Hz); from 2015 he said to delay everything to the sub; his 2018–2023 PDFs and 2021 posts tape-measure the sub along with the other speakers.

4.2 #13 · Impulse-response "auto delay" is useless for subs.

A low-passed channel's impulse peak moves with its high-frequency content, so the true arrival is where the trace first leaves zero (the zero crossing before the correct peak), and REW's own help says its sub delay estimate is inaccurate. The polarity-null delay test works only with even-order crossovers, matched levels and target slopes. [RARE] Sources: DIYMA 2016-03-25 · DIYMA 2021-09-02

In plain English: An impulse response is a measurement of how a speaker reacts to one sharp click; its graph shows a spike when the sound arrives. Some software, such as REW (a free measurement program), estimates delay automatically by finding that spike's peak. He says this is useless for subs. For a channel that plays only low notes, where the peak lands depends on how much high-frequency sound the channel has, so the peak is not the arrival time. The true arrival is where the trace first leaves zero, the zero crossing before the correct peak. He notes that REW's own help says its sub delay estimate is inaccurate. A common alternative is the "polarity-null" test: flip one speaker's polarity and adjust delay until the sound at the crossover cancels most deeply. He says it works only with even-order crossovers (such as LR2 or LR4), matched levels and slopes that match the target.

4.2 #14 · Group delay is a "nothingburger" in car bass.

It would matter only above ~50 ms, about one cycle at 20 Hz; people reject vented boxes over group delay, then add a 72 dB/oct sub low-pass that adds far more. A high-pass filter's group delay sits in its stop band, where the woofer isn't used. [RARE] Sources: DIYMA 2022-01-14 · DIYMA 2017-09-12

In plain English: Group delay is how much a filter or box delays one band of frequencies compared with others. Many people believe group delay makes vented boxes worse for sound quality. He calls it a "nothingburger" (a non-issue) in car bass. It would matter only above about 50 ms (fifty thousandths of a second), which is about one cycle of a 20 Hz wave. He points out that people reject vented boxes over group delay and then put a 72 dB/oct low-pass on the sub, which adds far more. And a high-pass filter's group delay sits in its "stop band", the range it cuts out, where the woofer isn't being used anyway.

4.2 #15 · All-pass filters on sub and midbass: a last resort.

An all-pass shifts phase over a band without changing level; a DSP "phase" slider (e.g. Helix) just moves an all-pass's center frequency. In 2015–16 a 2nd-order all-pass on the sub plus one half an octave higher on the near-side midbass centered bass in both seats of his car (aim ~90° or 270°, never 180°). [RARE] [Changed view: 2010–2013 never made them work and called all-pass "front bass" tricks one-seat fixes; 2021 prefers fixing the crossover, and if used, puts it on the high-pass side, where that driver's level is falling.] Sources: DIYMA 2013-03-16 · DIYMA 2015-10-19 · DIYMA 2021-09-26

In plain English: An all-pass filter shifts the phase (timing) of a band of frequencies without changing how loud they are. He says the "phase" slider on some DSPs, Helix for example, just moves an all-pass filter's center frequency. He calls all-pass filters on the sub and midbass a last resort. In 2015–16, in his own car, he used a 2nd-order all-pass on the sub plus another one, half an octave higher, on the nearer midbass. That centered the bass in both front seats. He aimed for a shift of about 90° or 270°, never 180°. The [Changed view] tag: in 2010–2013 he never made them work and called all-pass "front bass" tricks fixes for one seat only. In 2021 he preferred fixing the crossover itself. If an all-pass is used, he puts it on the high-pass side, where that speaker's level is already falling, so its effect on the combined sound is smaller.

4.2 #16 · Each front seat sees a different sub-to-midbass phase, even with a center channel.

It depends on sub location (he cites Harman patent US6876748); "pressure on the ears" can be a low-frequency phase problem, which he fixed in a CES demo car. Sources: DIYMA 2016-01-28 · DIYMA 2016-01-29

In plain English: A center channel is an extra speaker in the middle of the dash. Each front seat sits at different distances from the sub and from each midbass speaker, so the phase between sub and midbass is different in each seat. He says a center channel doesn't change that; it depends on where the sub sits. He cites a Harman patent on this, US6876748, about symmetrical stereo imaging in cars. He also says a feeling of "pressure on the ears" can come from a low-frequency phase problem; he fixed one in a demo car at CES (the big Consumer Electronics Show).

4.2 #17 · Keep bass below ~100 Hz out of the rear speakers.

His Logic 7 fix: rears high-passed at 250 Hz plus a separate 50–250 Hz band-pass with its own delay, level and EQ. Midbass arrays of 4–8 drivers evened bass across seats, but the more even, the worse the image; that is why BMW under-seat woofers and Mercedes "Front Bass" work. [RARE] Sources: DIYMA 2013-03-16 · DIYMA 2021-11-29

In plain English: He says rear speakers should get no bass below about 100 Hz. He adds one exception: when that rear bass is used to cancel some of the sub's sound in the rear seats, which are often much closer to the sub. Logic 7 is a Harman surround system that spreads stereo music across many speakers. His Logic 7 fix: the rear speakers get the ambience signal high-passed at 250 Hz, plus a separate 50–250 Hz band (a "band-pass", a slice between two frequencies) with its own delay, level and EQ. He also tried midbass "arrays" of 4, 6 and 8 drivers around the car. They evened out the bass across the seats, but the more even it got, the worse the image. He says that is why BMW's under-seat woofers and Mercedes' "Front Bass" work.

4.2 #18 · Multichannel midbass: steer only common content to a center midbass.

Below 50 Hz bass is nearly minimum-phase, so EQ fixes it; above that use left, center and right midbass, tune every location (sub included) to one target, then leave the sub gain alone. Sources: DIYMA 2009-11-14

In plain English: "Common content" is sound that is the same in the left and right channels, such as a centered bass guitar. He says only that shared part should be steered to a center midbass speaker. "Minimum-phase" is a technical term: in such a system, fixing the loudness with EQ fixes the timing too. He says car bass below 50 Hz is nearly minimum-phase, so EQ fixes it. Above that, use left, center and right midbass speakers, tune every location (the sub included) to one target curve, and then leave the sub's gain alone.

4.2 #19 · The front-bass illusion has a ceiling.

In his Volvo C30 demo car, tracks with bass mixed ~30 dB hotter than everything else still come from the back. Sources: OFCAP 38 @1:07:02

In plain English: Even done right, the illusion has limits. In his Volvo C30 demo car, tracks with bass mixed about 30 dB louder than everything else (a huge amount) still sound like the bass comes from the back. This fits his earlier point that bass far louder than the rest can't sound like it's in front (#1).

4.2 #20 · Check bass polarity by ear and with long test signals; cheap pulse testers fail on subs.

Fronts only, processing off, bass plus a centered vocal: a centered balance must give more bass than either end, or one midbass is reversed; never combine a polarity flip with delay. Pick the sub phase that gives the most midbass impact. Short (~9 ms) pulses miss subs (a 20 Hz cycle is 50 ms): use long positive pulses or a right-triangle wave; sub terminal markings are almost never wrong (3 of ~4,000). [RARE] Sources: DIYMA 2013-02-09 · Audiofrog blog Aug 2021 · DIYMA 2021-08-16

In plain English: Polarity is whether a speaker's cone moves outward when it should; reversed wires make it move backward. His checks, step by step. (1) By ear: play only the front speakers, with processing off, using a track with bass and a centered vocal. Slide the balance control from left to right. With the balance centered you should hear more bass than at either end; if not, one midbass is reversed. (2) Never combine a polarity flip with delay. (3) For the sub, pick the polarity that gives the most midbass impact. (4) Test signals: cheap pulse testers send short pulses, about 9 ms (nine thousandths of a second), but one cycle of a 20 Hz note lasts 50 ms, so they miss subs. Use a long series of positive pulses, or a right-triangle wave (one that jumps straight up and then ramps down). (5) The + and − markings on sub terminals are almost never wrong: in his count, 3 out of about 4,000.

4.3 Enclosures: vented vs sealed, sizing, ports

What this part is about: The box a subwoofer sits in. A sealed box is airtight; a vented (or ported) box has a tube or slot, the port, that lets air move in and out and adds bass of its own. He explains why he usually prefers vented boxes in cars, how to size boxes and ports, and which popular box ideas he thinks make little difference.

4.3 #1 · He prefers vented boxes in cars, tuned for minimum cone excursion, with the in-car peak EQ'd out.

Vented is more efficient across 30–50 Hz and, with a good port and one EQ band, has lower distortion and power demand and measures as flat as sealed. In a sealed box the sub must play below its in-box resonance (Fc), where the suspension distorts; "sloppy" vented bass only means a box tuned high for loudness. [RARE] Sources: DIYMA 2008-05-23 · DIYMA 2014-10-07 · DIYMA 2021-12-14

In plain English: A subwoofer ("sub") sits in either a sealed box or a vented box. Excursion is how far the cone moves back and forth; more excursion means more strain and more distortion (sound added that wasn't in the music). Many people believe sealed boxes are for sound quality and vented boxes are only for loudness. He disagrees. He prefers vented boxes in cars, tuned so the cone moves as little as possible, and he removes the bass bump the car then creates with EQ (an equalizer: a tone control that can cut or lift a narrow band of frequencies). His reasons: a vented box is more efficient across 30–50 Hz, and with a good port and one EQ band it has lower distortion, needs less power and measures as flat as a sealed box. In a sealed box, the sub has to play below Fc, the woofer's resonance in that box (the note where woofer and box naturally "bounce"), and there the suspension distorts. He says "sloppy" vented bass only comes from a box tuned high to be loud.

4.3 #2 · Tune the port to the woofer and box, not to the car.

Place the two impedance peaks at the passband edges so the port does most of the work, EQ the resulting peak, and check full-power excursion below Fb (the box tuning). Vented subs are almost always tuned below their sealed Fc; once a woofer is in a box there is no Fs, only Fc or Fb. [RARE] Sources: DIYMA 2011-10-05 · DIYMA 2021-03-05 · DIYMA 2021-05-09

In plain English: A vented box has a tuning frequency, Fb: the note at which the port resonates and does most of the work. Many people pick Fb to suit the car. He says tune the port to suit the woofer and the box instead. On an impedance graph (impedance is the electrical load the speaker presents to the amplifier), a vented box shows two peaks. He places them at the edges of the range the sub will play, with a wide trough between them, so the port does most of the work. In the car this gives a bass peak, which you cut with EQ. Then check an excursion plot at full power for frequencies below Fb, because below Fb the port stops holding the cone back. Fs is the bare woofer's resonance. He says that once the woofer is in a box there is no Fs any more, only Fc (sealed) or Fb (vented), and vented subs are almost always tuned below what their sealed Fc would be.

4.3 #3 · A good vented box has less excursion everywhere above Fb and is not loudest at Fb.

Choose volume and Fb so woofer plus port sum flat (they are ~90° apart at Fb: +3 dB, not +6); low tuning plus a boost near Fb is fine. Sealed's only edge: any sealed box for a given woofer sounds like any other. [RARE] Sources: DIYMA 2014-10-07 · DIYMA 2016-05-13

In plain English: Many people believe a vented box is built to be loudest at its tuning frequency (Fb). He says a good one is not, and that its cone moves less than it would in a sealed box at every frequency above Fb. You choose the box size and Fb so that the woofer's sound and the port's sound add up to a flat response. At Fb the two are about 90° apart (a quarter of a cycle out of step). Two equal sounds exactly in step add 6 dB; a quarter cycle apart they add only about 3 dB (dB, decibels, measure loudness differences). Tuning low and adding an EQ boost near Fb is fine, he says. The one advantage he grants sealed boxes: any sealed box you build for a given woofer will sound much like any other sealed box for it.

4.3 #4 · Keep the port and EQ the peak; but with no EQ at all, sealed is probably better.

Turning the gain down to tame a vented peak leaves a hole at the crossover; compare boxes level-matched. Without EQ, sealed wins if you have the amp power and voice coil. Sources: DIYMA 2016-05-03 · DIYMA 2016-05-04

In plain English: In a car, a vented box makes a big bass peak (around 40 Hz in his example). Some people tame it by turning the sub's gain (its level) down until the peak matches everything else. He says that leaves a hole around the crossover, the point where the sub hands over to the door speakers. Better: keep the port open and cut just the peak with EQ. And when you compare a sealed and a vented box, match their levels first, or the comparison misleads you. His exception: with no EQ at all, a sealed box is probably the better choice, provided you have enough amplifier power and enough voice coil (the wire coil that drives the cone, which must cope with the heat).

4.3 #5 · Put Fb near the bottom of the band and respect the power math.

That minimizes excursion where it is worst and removes the need for a subsonic. 3 dB of boost doubles power, 12 dB is 16×, and each halving of frequency needs 4× the excursion. Sources: DIYMA 2008-12-18 · DIYMA 2008-12-05

In plain English: Fb is the vented box's tuning frequency. He says to put it near the bottom of the range the sub has to play. At Fb the port does the work and the cone barely moves, so this keeps excursion low exactly where it would otherwise be worst, and it removes the need for a subsonic filter (a filter that blocks notes too low for the box). Then he spells out the cost of bass. Every 3 dB of boost needs twice the amplifier power, and 12 dB needs 16 times the power. Each time the frequency halves (one octave lower), the cone must move 4 times as far to make the same loudness. So in practice: put Fb near the bottom of the band, shape the response with EQ, and remember how fast boosts use up power and cone travel.

4.3 #6 · Small sealed box plus huge amp is "progress" in reverse, and a sub built for a big box fails in a tiny one.

Typical small-box subs (~85 dB/W/m) need ~2,000 W to match an old Cerwin-Vega LE12 vented on a Rockford Punch 150. Squeezed into a tiny box, a sub that needs a big one makes no bass, and power plus EQ drive it past its limits. His 2023 C30 runs two GB10D4 in a small sealed box, per his rule "little space, plenty of power: sealed plus EQ". [RARE] Sources: DIYMA 2011-10-07 · DIYMA 2021-02-03 · OFCAP 38 @1:10:05

In plain English: Many people believe a small sealed box with a huge amplifier is the modern, better way to get bass. He calls it "progress" in reverse. Sensitivity says how loud a speaker plays from 1 watt, measured 1 meter away; about 85 dB/W/m is low. He says typical small-box subs need about 2,000 W to make the bass that an old Cerwin-Vega LE12 made in a vented box driven by a Rockford Punch 150 (an older amplifier). And a sub that needs a big box makes no bass when squeezed into a tiny one; adding power and EQ to make up for it drives it past its limits. His own rule for little space and plenty of power is sealed plus EQ: his 2023 Volvo C30 demo car runs two Audiofrog GB10D4 subs in a small sealed box.

4.3 #7 · Small sealed boxes don't give "more cone control", and box Q barely matters when the sub is crossed below Fc.

A small box is a stiffer spring that raises resonance and Q. Crossed at 60 Hz, Qtc 0.707 vs 0.9 or 1 is immaterial (model the box with the low-pass applied); a too-small box shows up as weak lows and "one-note bass" near the crossover. [RARE] Sources: DIYMA 2012-07-28 · DIYMA 2021-02-03 · DIYMA 2022-01-13

In plain English: Many people believe a smaller sealed box gives "tighter" bass with "more cone control". He says it doesn't. The air in a small box acts like a stiffer spring, which raises the resonance and the Q. Q describes how much a system overshoots and rings at resonance (he contrasted a well-damped ripe avocado with a springy super ball). Qtc is the Q of a woofer in a sealed box. But when the sub is crossed over (filtered off) below Fc, as with his example of a 60 Hz crossover, whether Qtc is 0.707, 0.9 or 1 makes no real difference; to see this, model the box with the low-pass filter included. A box that is too small shows up as weak low bass and "one-note bass" near the crossover.

4.3 #8 · Box type decides the Xmax you need.

(Xmax: how far the cone moves before distortion becomes objectionable.) Sealed or IB at high power needs a long coil; vented can use a short one. Rules: Qts ≤0.4 with ≥10 mm linear Xmax suits sealed, vented woofers usually have Qts 0.4–0.7, ~7 mm "small-box" subs are usually better vented with a subsonic, and a 4.3 mm-Xmax sub "is a midrange". [RARE] Sources: DIYMA 2011-09-16 · DIYMA 2011-09-19 · DIYMA 2014-10-04

In plain English: Xmax is defined in the point itself. Qts is a number describing how well-controlled a woofer is at its resonance; lower means more controlled. He says the type of box decides how much Xmax you need. A sealed box or an infinite baffle (IB: the woofer on a board, with the trunk as its "box") at high power makes the cone travel far, so it needs a long voice coil. In a vented box the port does much of the work, so a short coil can do. His rules: Qts of 0.4 or less with at least 10 mm of linear Xmax suits sealed; vented woofers usually have Qts between 0.4 and 0.7; "small-box" subs with only about 7 mm of Xmax usually do better vented with a subsonic filter, even though they are marketed for small sealed boxes; and a sub with 4.3 mm of Xmax "is a midrange", not a subwoofer.

4.3 #9 · Thiele/Small numbers are a box recipe, not a quality score, and "SQ sub" differences are mostly box response.

Equal bass needs equal displacement (cone area × excursion). A bigger box only adds low bass: EQ it down and "tight" vs "boomy" disappear; heard differences without blind tests in one car come from elsewhere. Note Qts = (Qes × Qms)/(Qes + Qms). [RARE] Sources: DIYMA 2008-08-08 · DIYMA 2021-09-05 · DIYMA 2021-05-09

In plain English: Thiele/Small (T/S) parameters are the standard spec numbers of a woofer, such as Fs, Qts and Qes. Many people compare them as if they were a quality score. He says they are a recipe for choosing the right box, not a measure of how good a woofer is. To make the same bass, two subs must move the same amount of air (cone area times how far the cone travels). He says most of what people call "SQ sub" (sound-quality sub) differences is really the box's frequency response: a bigger box only adds low bass, and if you EQ that extra down, "tight" versus "boomy" disappears. Unless subs are compared blind in the same car, the differences people hear come from elsewhere in the system. The formula at the end is his correction of someone who had simply multiplied Qes and Qms.

4.3 #10 · What sub distortion is, and why power compression hardly matters in a car.

Odd-order "thud thud" comes mostly from the coil leaving the gap; even-order from asymmetry, above all cupped spiders. Subs alone run at and below resonance ("resonance IS distortion", though ~3% at Fs is fine); power compression only slightly steepens the low end of a sub low-passed at ~60 Hz. [RARE] Sources: DIYMA 2021-09-05 · DIYMA 2012-09-16

In plain English: Distortion is sound the speaker adds that wasn't in the signal. He sorts sub distortion into two kinds. Odd-order distortion, the "thud thud" sound, comes mostly from the voice coil moving out of the magnet gap, where the motor's force is. Even-order distortion comes from parts that aren't symmetrical, above all a cupped spider (the spider is the springy fabric ring that centers the cone). Subs are the only drivers that play at and below their resonance, and he says "resonance IS distortion", though about 3% at Fs is fine for a sub. Power compression is the loss of output as a voice coil heats up. He says it hardly matters in a car: for a sub low-passed at about 60 Hz it only makes the bottom end roll off slightly more steeply.

4.3 #11 · Set the subsonic filter about half an octave below Fb (IB needs one too).

Below Fb the box is "just a hole" and the cone unloads; on the A600.1D use 5 Hz when none is needed. Bass boost is costly: 15 dB at 45 Hz needs 32× the power. Sources: A-Series manual · DIYMA 2014-11-07

In plain English: A subsonic filter is a high-pass filter that keeps the very lowest notes away from the sub. Below Fb, he says, a vented box is "just a hole": the port no longer holds the cone back, so the cone "unloads" and can flap too far. Set the subsonic about half an octave below Fb (half an octave down means dividing the frequency by about 1.4). An infinite-baffle setup needs one too. The manual for Audiofrog's A600.1D amplifier says to set it to 5 Hz when no subsonic is needed. And bass boost is expensive: 15 dB of boost at 45 Hz needs 32 times the power.

4.3 #12 · Size ports for air velocity at your power, with big flares; area rules of thumb are wrong.

A choking port makes the box act sealed "with a nasty-sounding leak". More area needs more length; use big flares (a ¾" roundover barely helps), round beats slot, and a long 3" port on a 12" chokes, so go smaller woofer vented or seal and EQ. Audiofrog manuals give each sub a sealed box and a vented one (round flared or slot port). [RARE] Sources: DIYMA 2014-11-07 · DIYMA 2008-06-05 · Absolute Electronix 2023 @7:18

In plain English: The port is the tube or slot in a vented box. Many people size it with a rule of thumb based on port area and box volume. He says that's wrong: what matters is how fast the air moves through the port, which rises with the power you use (and depends on the cone size and motor). If the port is too small, friction "chokes" it at high power, and the box behaves like a sealed box "with a nasty-sounding leak". A port with more area must also be longer to keep the same tuning. Use big flares (rounded, trumpet-like ends); a ¾" (about 19 mm) roundover barely helps, and round ports beat slots. A long 3" (about 75 mm) port on a 12" sub chokes, so either use a smaller woofer in a vented box, or go sealed and use EQ. Audiofrog's manuals give each sub a sealed box and a vented one.

4.3 #13 · Port position, port direction and baffle shape rarely matter; rigidity does.

Port and cone would need ~16 ft of path difference to cancel at 35 Hz, so rear-firing ports are fine if the ends are clear. Brace the largest panels; a baffle would have to be ~5 ft across to matter at 100 Hz. Sources: DIYMA 2011-11-21 · DIYMA 2008-06-03

In plain English: Many people worry about where the port goes, which way it fires, or the shape of the front panel (the baffle). He says these rarely matter for subs. For the port's sound and the cone's sound to cancel at 35 Hz, their paths would have to differ by about 16 ft (about 5 m), so a port firing backward is fine as long as its ends aren't blocked. Likewise, a baffle would have to be about 5 ft (1.5 m) across to matter at 100 Hz. What does matter is stiffness: a panel that flexes makes its own sound, so brace the largest panels.

4.3 #14 · Model with voice-coil inductance (Le), then EQ at playing level.

A model without Le "isn't worth much" for multilayer car subs; build to a good model and you needn't measure the box. Large-signal shifts (a few Hz of Fb) are handled by EQ set at playing level; compare subs only after EQ'ing them to the same response. Sources: DIYMA 2008-12-06 · DIYMA 2015-05-29

In plain English: Box-design software predicts how a woofer will behave in a given box. He says the model must include voice-coil inductance (Le, an electrical property of the coil that affects how the woofer behaves as frequency rises). For car subs with multilayer coils, a model without Le "isn't worth much". If you build to a good model, you don't need to measure the box afterward. At real playing levels the woofer behaves a little differently (the tuning can shift by a few Hz); he handles that with EQ set at the level you actually listen at. And to compare two subs fairly, first EQ them to the same response.

4.3 #15 · Polyfill lowers Qtc but absorbs nothing at sub frequencies; a pole vent may exit outside the box.

Rear-wave reflections inside a sub box don't matter, and "deflex" pads add only panel mass. A 1–1.5" pole vent outside a sealed box changes little; route it into a small tube with inert material to keep noise out of the cabin. Sources: DIYMA 2015-11-15 · DIYMA 2022-01-04

In plain English: Polyfill is fluffy stuffing put inside sealed boxes. He says it lowers the box's Qtc but absorbs nothing at sub frequencies. Many people worry about sound bouncing around inside the box (the "rear wave"); he says those reflections don't matter in a sub box, and "deflex" pads only add a little mass to the panels. A pole vent is a hole through the middle of the woofer's magnet that lets air in and out behind the cone. He says a 1–1.5" (25–38 mm) pole vent that exits outside a sealed box changes little. To keep its noise out of the cabin, route it into a small tube with inert material.

4.3 #16 · Aperiodic membranes and "impedance-peak removal" are wasted money, except in special cases.

The resonance peak in impedance is just the driver's Q and is harmless. His one use: a membrane between an IB woofer and the listener to strip HF distortion. BMW under-seat boxes work as aperiodic boxes (frame airspace through small openings); high-Q drivers crammed in instead never sounded good to him. [RARE] Sources: DIYMA 2008-10-12 · DIYMA 2013-04-28 · DIYMA 2020-06-24

In plain English: An aperiodic membrane is a panel of cloth or foam that lets air leak slowly out of a box, damping the woofer's resonance. "Impedance-peak removal" means adding parts to flatten the bump in a speaker's electrical load at resonance. He says both are wasted money in almost every case: that peak simply reflects the driver's Q and does no harm. His one use for a membrane is between an infinite-baffle woofer and the listener, to strip away the woofer's high-frequency distortion. He adds that BMW's under-seat boxes work as aperiodic boxes, joining extra airspace in the car's frame through small openings. Setups that skip this and cram high-Q drivers in instead have never sounded good to him.

4.3 #17 · A down-firing box is an accidental bandpass, not "boundary loading"; linked boxes are unpredictable.

The space under the box and the gap out act as a front chamber that is nearly impossible to model (carpet losses, no clear port). Passive radiators are fine, but two boxes joined by an internal port shift tuning with power: use two sealed boxes plus EQ. [RARE] Sources: DIYMA 2021-10-29 · DIYMA 2013-05-21 · DIYMA 2013-04-28

In plain English: A down-firing box has its woofer facing the floor. Many people say this gives "boundary loading" (extra bass from sitting near a surface). He says it really makes an accidental bandpass box. A bandpass box is one where the woofer fires into a chamber and the sound leaves through an opening, which acts as a low-pass filter. Here, the space under the box and the gap the sound escapes through form that front chamber, and it is nearly impossible to model, because carpet absorbs energy and there is no clear port. Passive radiators (an unpowered cone used in place of a port) are fine. But two boxes joined by an internal port change their tuning as power changes; he says use two sealed boxes plus EQ instead.

4.3 #18 · Sub "break-in" is spider resin cracking.

It lowers Fs and Qts (sealed: Fc and Qtc drop; vented: more output above Fb); mids and tweeters crossed above resonance barely break in. Sources: DIYMA 2010-05-09 · DIYMA 2021-05-07

In plain English: "Break-in" is the idea that a new speaker changes as it is played in. He says that for subs it is real and has a simple cause: the resin that stiffens the spider (the springy ring that centers the cone) cracks with use. That lowers Fs and Qts. In a sealed box, Fc and Qtc drop; in a vented box you get more output above Fb. Midranges and tweeters, crossed over above their resonance, barely break in.

4.3 #19 · The manual box is the right box, not a "safe" one: Audiofrog designs woofers backwards from the car.

It starts with an in-car target averaged over ~30 cars, picks a convenient box, then designs the woofer for that box (modeled in LEAP); midband sensitivity specs don't predict in-car sub loudness. GB10: 17.6 mm Xmax plus ~8 mm more travel, so keep grilles ~1.25" from the cone. [RARE] Sources: Absolute Electronix 2023 @8:13 · DIYMA 2021-09-04 · DIYMA 2021-12-22

In plain English: Many people treat the box in a sub's manual as a cautious "safe" choice and build something else for more bass. He says the manual box is the right box, because Audiofrog (his company) designs its woofers backward from the car. It starts with an in-car target response averaged over about 30 cars, picks a convenient box, and then designs the woofer for that box, modeled in the LEAP software. He adds that the sensitivity figure in a spec sheet (measured in the midband) doesn't predict how loud a sub will be in a car. For the GB10, the cone can travel 17.6 mm (its Xmax) plus about 8 mm more, so keep grilles about 1.25" (32 mm) from the cone.

4.3 #20 · Audiofrog subs (product-specific).

GB10/GB12 are "dual impedance" (GB10D4: 8 or 2 Ω; D2: 4 or 1 Ω) with 3" copper coils; a GB D4 at 2 Ω on an A600.1D is "the right amount of power". GS8ND2: under 3" deep, 10 mm Xmax, sealed ~7 L, vented ~14 L or IB, plays to ~300 Hz, ~2 mm excursion at rated power crossed at 80 Hz. Sources: CMA 2021 @27:03 · GS8ND2 store listing · DIYMA 2020-06-24

In plain English: This point is about Audiofrog's own subs. GB10 and GB12 are "dual impedance": each has two voice coils that can be wired for either of two loads (the GB10D4 as 8 or 2 ohms, the D2 version as 4 or 1 ohm; ohms measure the electrical load). They use 3" copper voice coils. He says a GB D4 wired to 2 ohms on Audiofrog's A600.1D amplifier is "the right amount of power". The GS8ND2 is a very shallow woofer (under 3" deep) with 10 mm of Xmax. It works sealed in about 7 liters, vented in about 14 liters or as an infinite baffle, plays up to about 300 Hz, and moves only about 2 mm at its rated power when crossed at 80 Hz.

4.3 #21 · His earlier designs, and an Xmax method conflict.

JBL GTi subs used opposed dual coils (Differential Drive); the 2012 MS-BassProSQ used a short coil and a flared port tuned for minimum excursion; he designed the Infinity Perfect 10 VQ (variable-Q inserts). Compare Xmax only within one method. [Conflict: 2008, JBL Xmax was one-way overhang "with no +10% fudge"; 2020, GB60's 9 mm was (coil − gap)/2 + 15%, vs Klippel ~4.7 mm.] Sources: DIYMA 2008-12-26 · Erin's Audio Corner Aug 2020 · DIYMA 2021-12-15

In plain English: This point covers subs he designed before Audiofrog, and a conflict about how Xmax is measured. At JBL, the GTi subs used two opposed voice coils (Differential Drive). The 2012 MS-BassProSQ used a short coil and a flared port tuned for minimum excursion. He also designed the Infinity Perfect 10 VQ, which came with inserts that change its Q ("variable Q"). Xmax can be calculated in different ways. The conflict, in simple words: in 2008 he described JBL's Xmax as the coil's one-way overhang "with no +10% fudge", that is, no added margin. In 2020, the GB60's 9 mm figure was worked out as (coil height minus gap height) divided by 2, plus 15%, while a Klippel measurement (a lab system that measures a driver's actual usable travel) gave about 4.7 mm. The point takes no side; its advice is simply to compare Xmax figures only when they were worked out the same way.

4.4 Infinite baffle (IB: woofer on a baffle, trunk as the "box")

What this part is about: Infinite baffle (IB) means mounting the woofer on a sealed board, usually the rear shelf, so the trunk serves as a very large box. This part covers how he installs IB subs, how he tunes them, which woofers suit them, and what happens when bass fires through the rear seat.

4.4 #1 · Fire IB woofers through the package tray, rebuild only the deck, and don't seal the seat back or trunk.

Seal the tray openings and kill rattles (MDF baffle, deadener, foam between trim and deck); seal the seat only if it's thin and unbacked. Never seal the trunk (cars vent cabin air through it); his own IB sub hangs on rubber isolation mounts. [RARE] Sources: DIYMA 2009-08-24 · DIYMA 2014-07-20 · DIYMA 2017-12-09

In plain English: The package tray is the shelf behind the rear seat, under the back window. In an IB install the woofers are mounted there and fire up into the cabin. Many people believe IB needs the whole trunk and seat back sealed airtight. He says rebuild only the deck: seal the openings in the tray and kill rattles, using an MDF board as the baffle, deadener (heavy sticky damping sheet) and foam between the trim and the deck. Seal the seat back only if it is thin and has nothing behind it. Never seal the trunk, because cars vent cabin air out through it. His own IB sub hangs on rubber isolation mounts, which he used to avoid rattles.

4.4 #2 · EQ the cabin gain out instead of re-engineering the trunk.

Trunk volume changes ~6 dB at 20 Hz while the car adds ~20 dB: cut much of the cabin gain, flatten to 150–200 Hz, low-pass to suit the midbass, set level with gain — faster than "a thousand trips to Home Depot". Dacron-lined trunks and holes to the outside are "ridiculous". [RARE] Sources: DIYMA 2013-05-11 · DIYMA 2013-05-05

In plain English: Cabin gain is the car's natural boost of low bass. Many IB installers spend a lot of effort modeling and changing the trunk as if it were a precise box. He says don't bother: changing the trunk's volume moves the bass by about 6 dB at 20 Hz, while the car itself adds about 20 dB. So use EQ instead: cut much of the cabin gain, flatten the response up to 150–200 Hz, set the low-pass to suit your midbass, and set the level with the amplifier's gain. He says that is faster than "a thousand trips to Home Depot". He calls lining the trunk with Dacron (polyester batting) or cutting holes to the outside "ridiculous".

4.4 #3 · IB woofers need Qts ≤~0.7, enough Xmax for the power, and a subsonic.

In a car, sealed plus cabin gain is about flat, vented peaks above Fb, and IB rises as frequency falls; IB handles less power but needs less. Low-Qts (~0.38) "small sealed" woofers in IB act like a vented box: cut the peak above ~36 Hz and raise the gain. Sources: DIYMA 2009-08-24 · DIYMA 2018-02-02 · DIYMA 2014-07-25

In plain English: His requirements for an IB woofer: a Qts of about 0.7 or less, enough Xmax for the power you will use, and a subsonic filter. He compares the three setups in a car: a sealed box plus cabin gain comes out about flat; a vented box peaks above its tuning (Fb); and IB rises steadily as frequency falls. IB handles less power, but it also needs less. Woofers sold for "small sealed" boxes with a low Qts (around 0.38) act like a vented box when used IB, so cut the peak above about 36 Hz with EQ and raise the amplifier gain.

4.4 #4 · IB in a BMW 3-series: cut the tray and use metal or plastic mesh, never grille cloth.

IB is "the easiest installation in the world" and removes the "peaky sealed box sound"; cloth can snap against the mesh. Sources: DIYMA 2012-09-03

In plain English: This is about BMW 3-series cars, where the sub can fire through the rear shelf. He calls IB "the easiest installation in the world" and says it removes the "peaky sealed box sound". Cut the tray open and cover the hole with metal or plastic mesh, never grille cloth, because the cloth can snap against the mesh as the air moves.

4.4 #5 · Firing through the rear seat is an acoustic low-pass filter.

It removes midbass, so add strong midbass up front; an 8" behind a 5" opening only low-passes far above an 80 Hz crossover. Sources: DIYMA 2009-08-24 · DIYMA 2012-08-12

In plain English: Some trunk or IB subs play into the car through the rear seat. He says the seat acts as an acoustic low-pass filter: deep bass gets through but midbass (the upper bass) is lost, so you need strong midbass speakers up front. A small opening is different: an 8" woofer behind a 5" hole is also a low-pass, but only at frequencies far above an 80 Hz crossover, so it doesn't matter (as long as it isn't noisy).

4.5 Sub location, number of subs, multiple-sub wiring

What this part is about: Where to put the subwoofer, how many to use and how to wire more than one. It also covers the "cross-car mode": a dip in the bass, usually near 70 Hz, caused by sound bouncing across the width of the car.

4.5 #1 · Trunk subs work; trunk midbass doesn't.

Sub bass reaches the cabin (you hear the next car's at a light), but midbass may not, and a small hole won't help: use a bandpass box firing through the opening, or real front midbass. Sources: DIYMA 2012-01-22 · DIYMA 2013-05-03

In plain English: Subs in the trunk work: low bass gets into the cabin, which is why you can hear the bass of the car next to you at a traffic light. Midbass (the upper bass, just above what a sub plays) may not get through, and cutting a small hole between trunk and cabin won't help much. If you want midbass from the trunk, use a bandpass box whose port fires through the opening; otherwise use real midbass speakers up front.

4.5 #2 · Behind the rear seat (mid-car) is the worst sub location, and a deep null at the seat there is placement, not crossover phase.

It is the easiest spot to design for, but his experience and a member's measurements rank it last. [RARE] Sources: DIYMA 2021-09-16

In plain English: Behind the rear seat, roughly the middle of the car, is the easiest spot to design a sub box for. He says it is the worst sub location, based on his experience and a forum member's measurements. A null is a deep, narrow drop in the response. He says a deep null at the seat with the sub there comes from where the sub and the listener sit, not from a phase problem (the sub and front speakers pushing and pulling out of step) at the crossover.

4.5 #3 · Confirm the cross-car mode with a tape measure.

Measure sub → mic, then sub → far side of the car → mic; a difference near 7 ft is the standard cross-car mode (the dip near 70 Hz). Move the mic to the other seat to see whether the dip follows; one dual-sub dip vanished when the subs were crossed at ~60 Hz. [RARE] Sources: DIYMA 2021-12-02 · DIYMA 2021-12-05

In plain English: A mode is a frequency at which sound bouncing between surfaces reinforces or cancels itself. The standard one across the car's width shows up as a dip near 70 Hz. His tape-measure test: measure the distance from the sub straight to the microphone, then from the sub to the far side of the car and back to the mic. If the difference is close to 7 ft (about 2.1 m), the dip is most likely the cross-car mode. Then move the mic to the other front seat and see whether the dip follows. In one member's dual-sub box, the dip disappeared once the subs were crossed over at about 60 Hz.

4.5 #4 · Under-seat woofers in the BMW layout are "optimum".

Center-of-car woofers don't blast rear passengers (who usually get bass 6–9 dB too hot), the seat hides their distortion, and left/center/right steering works over their range. In his MS-8 BMW build they played ~80–200 Hz (200 Hz because the 4" doors couldn't go lower loudly), with the sub below ~60 Hz; keep an octave of usable bandwidth around each crossover. [RARE] Sources: DIYMA 2013-03-10 · DIYMA 2021-08-04

In plain English: Many late-model BMWs have woofers under the front seats, and he calls this layout "optimum". Because the woofers sit in the middle of the car, they don't blast the rear passengers (who usually get bass 6–9 dB too loud), the seat hides their distortion, and a processor can steer sound left, center or right across their range. In a BMW he set up with the MS-8 (a JBL processor from his Harman years), the under-seat woofers played about 80–200 Hz. The top was 200 Hz because the 4" door speakers couldn't play any lower loudly; the sub covered about 60 Hz and below. His rule: keep about an octave of a speaker's usable range around each crossover.

4.5 #5 · Multiple subs: one amp, parallel wiring, and EQ, not mode-smoothing arrays.

Wire separate woofers in parallel (one DVC woofer's coils in series is fine); at very low frequencies two subs EQ'd behave like one. Mixed impedances split power unevenly: in parallel the lower impedance takes more (~67% for a 2 Ω DVC beside a 4 Ω coil), in series the higher. Mode-smoothing arrays suit subs only; above that they break the image. [RARE] Sources: DIYMA 2012-03-10 · DIYMA 2022-02-11 · DIYMA 2015-06-07

In plain English: For more than one sub, he uses one amplifier, parallel wiring and EQ. Wire separate woofers in parallel; inside one dual-voice-coil (DVC) woofer, wiring its two coils in series is fine. At very low frequencies, two subs that have been EQ'd behave like one. If you mix different impedances, power splits unevenly: in parallel the lower-impedance load takes more (about 67% for a 2-ohm DVC woofer next to a 4-ohm coil), and in series the higher-impedance one takes more. Mode-smoothing arrays are groups of woofers placed to even out bass across seats. He says they suit subs only; above the sub range they break the image (the sense that the music sits on a stage in front of you).

4.5 #6 · Not enough bass? His four options, in order: vented instead of sealed, lower impedance, more speakers, a more efficient driver.

Lower impedance adds little because amps don't double into half the load; "bottoming" may be amp clipping, and either way you need "more power or more woofer". For a sub amp, take the most power on offer (~800 W in one case) and bridge a 2-channel amp for single-coil GB woofers. Sources: DIYMA 2010-03-16 · DIYMA 2021-07-10

In plain English: If a system lacks bass, he lists four fixes, in order: change a sealed box to vented, lower the impedance, add more speakers, or use a more efficient driver. Lowering impedance adds little, because amplifiers don't double their power into half the load. What sounds like the sub "bottoming" (hitting the end of its travel) may really be the amplifier clipping (running out of voltage and flattening the tops of the waveform). Either way, he says, you need "more power or more woofer". For a sub amp, take the most power on offer (about 800 W in one case he advised on). Audiofrog GB woofers have a single voice coil, so a 2-channel amp should be bridged (its two channels combined into one) to drive one.

4.5 #7 · Blown subs: suspect reversed DVC coil wiring and clipping, not DC.

Out-of-phase coils make little bass, so owners crank gain and boost into clipping; it takes ~28 V DC at 1 Ω to make 750 W, and no loud pop at amp shutdown means no meaningful DC. Sources: DIYMA 2011-09-16

In plain English: When subs keep blowing, many people blame DC (direct current) leaking from the amplifier. He says to suspect reversed coil wiring and clipping instead. Picture a dual-voice-coil sub with one coil wired backward: the coils fight each other, so it makes little bass, and owners crank the gain and bass boost until the amp clips. A clipped signal carries more average power, which overheats the coil. DC is unlikely: making 750 W into 1 ohm would take about 28 volts of DC. If you don't hear a loud pop when the amp shuts off, there is no meaningful DC.

4.6 Midbass capability and the midbass/sub split

What this part is about: Midbass is the upper bass just above the subwoofer's range, usually played by woofers in the doors or kick panels (the panels in the footwells). This part covers how low midbass speakers should play, where to split the bass between them and the sub, and where to mount them.

4.6 #1 · Crossing door midbass at ~40 Hz to pull the bass forward is "the worst mistake".

Door midbass Fs is typically 60–70 Hz; below resonance the spider does the work, distorts most and wears out, and every car's cross-car mode near 70 Hz needs another fix. When told mids are crossed at 40 Hz he "stops listening"; making doors play 50 Hz is "the worst possible advice". [RARE] Sources: OFCAP 19 @1:45:06 · DIYMA 2018-02-09 · DIYMA 2022-01-14

In plain English: Many people make door midbass play down to 40–50 Hz so that the bass seems to come from the front. He calls crossing door midbass at about 40 Hz "the worst mistake". Door midbass drivers usually have their resonance (Fs) at 60–70 Hz. Below resonance, the spider (the springy ring that centers the cone) does the work; that is where the driver distorts most, and the spider wears out. Every car also has a cross-car mode (a cancellation across the car's width) near 70 Hz, which then needs a separate fix. When someone says their mids are crossed at 40 Hz, he "stops listening", and he calls making doors play 50 Hz "the worst possible advice".

4.6 #2 · Cross every driver except the sub at or above its resonance: about 70–80 Hz for a 6" midbass.

"Resonance IS distortion"; 80 Hz raises output, lowers distortion and lengthens suspension life. Each octave lower costs ~2× power and 4× excursion, so don't run a small speaker (e.g. a GB40) down to 100 Hz either. [RARE] Sources: DIYMA 2017-08-17 · DIYMA 2021-09-11 · DIYMA 2021-12-03

In plain English: His rule: cross over every speaker except the sub at or above its resonance frequency. For a typical 6" midbass that means about 70–80 Hz. He says "resonance IS distortion", so crossing at 80 Hz raises output, lowers distortion and makes the suspension last longer. Each octave lower (each halving of frequency) costs about twice the power and four times the cone travel. The same thinking applies to small speakers: don't run one such as the GB40 (a small Audiofrog speaker) down to 100 Hz either.

4.6 #3 · A door or kick is a corner worth ~6 dB, so car midbass should be low-Q.

Each halving of radiating space adds ~6 dB below ~500 Hz; home 6" drivers fake bass with high Q ("That higher Q IS distortion"), and the GB60 sounds thin on a display board but full in a car. [RARE] [Changed view (partial): 2021, crossed above resonance a midbass's Q "is not really important".] Sources: Audiofrog blog Jun 2021 · DIYMA 2015-12-13 · DIYMA 2021-09-11

In plain English: Below a few hundred hertz, a speaker sends sound in all directions, so nearby surfaces reflect its sound and add to it. He says each halving of the space a speaker radiates into adds about 6 dB below about 500 Hz, and a door bottom or kick panel works like a corner, worth about 6 dB. So car midbass speakers should be low-Q (well controlled at resonance). Home 6" speakers often fake bass with a high Q, and he says "That higher Q IS distortion". That is why the Audiofrog GB60 sounds thin on a store display board but full in a car. The tag marks a partial change of view: in a 2021 forum post he said that once a midbass is crossed over above its resonance, its Q "is not really important". So the two positions are that car midbass should be low-Q, and that Q matters little when crossed above resonance.

4.6 #4 · Ultra-low-Fs (~35 Hz) midbass drivers chase a crossover you don't need once there's a sub.

One was nearly 10 dB less sensitive in midbass than midband ("10dB is a lot of power"), and "50 Hz isn't midbass". A sub isn't a good midbass either: midbass drivers use compliant suspensions and ~¼ the inductance for efficiency (Hoffman's Iron Law), so a "sub that plays midrange" mostly doesn't exist. [RARE] Sources: DIYMA 2021-11-02 · DIYMA 2022-01-13 · DIYMA 2020-08-24

In plain English: Some midbass drivers are built with a very low resonance (Fs around 35 Hz) so they can be crossed very low. He says that chases a crossover you don't need once you have a sub. One such driver was nearly 10 dB less sensitive in the midbass than in its midband, and he notes "10dB is a lot of power" (10 dB takes ten times the power). He also says "50 Hz isn't midbass". The reverse doesn't work either: a sub is not a good midbass. Midbass drivers use more compliant (softer) suspensions and about a quarter of a sub's coil inductance to be efficient. Hoffman's Iron Law is the standard rule that a speaker can't be small, efficient and deep-playing all at once. So, he says, a "sub that plays midrange" mostly doesn't exist.

4.6 #5 · With a sub and an active high-pass, door enclosures are unnecessary, and never port a door box into the door.

The high-pass does what a box would; damp the door and mount free-air. Vent into the door and you hear only the woofer, less bass than sealed. GB60 band ~70 Hz–3.5 kHz; in a GB four-way, midbass high-pass 60–80 Hz and low-pass ~300 Hz. [RARE] Sources: DIYMA 2016-06-17 · DIYMA 2022-01-13 · CMA 2022 @57:02

In plain English: Many people build boxes inside doors for their midbass. He says that with a sub and an active high-pass filter (a processor or amp setting that keeps deep bass away from the midbass), door enclosures are unnecessary, because the high-pass does what a box would do. Damp the door (deaden it) and mount the speaker "free-air", with no box. And never port a door box into the door: you then hear only the woofer, losing what makes a ported box worthwhile, and get less bass than a sealed box would give. Audiofrog's band for the GB60 is about 70 Hz–3.5 kHz; in a GB four-way system, the midbass high-pass is 60–80 Hz and its low-pass about 300 Hz.

4.6 #6 · A 60–80 Hz dip in only the near-side midbass is likely the cross-car mode: cross above it and let the sub play there.

Near-field measurements can't show door-panel vibration; near field turns far field at ~3× the driver's largest dimension. [RARE] Sources: DIYMA 2022-03-16

In plain English: If only the midbass on the side nearer you shows a dip between 60 and 80 Hz, he says it is likely the cross-car mode (the cancellation across the car's width). His fix: cross the midbass over above that dip and let the sub play there. He adds two measuring notes. Measuring very close to the speaker (the near field) can't show the door panel vibrating. And the near field turns into the far field at about three times the driver's largest dimension.

4.6 #7 · The passenger kick panel measured as the strongest midbass spot; the door's top corner evens left/right.

His old Jetta tests matched a member's; with a sub, no 8" is needed. Sources: DIYMA 2021-09-16 · DIYMA 2021-12-03

In plain English: His old measurements in a Jetta, like a forum member's, found the passenger kick panel (the panel in the footwell) to be the strongest midbass location. Mounting the midbass in the top corner of the door makes the left and right responses much more alike. And if you have a sub, he says you don't need an 8" midbass.

4.6 #8 · At moderate levels a 4–5" can blend with a sub; dedicated stereo midbass matters at real volume.

It needs separate L/R parametric EQ; modern mixes carry stereo bass (mono bass was a vinyl limit). Sources: DIYMA 2008-11-20 · DIYMA 2021-08-20

In plain English: At moderate volume, even a small 4–5" speaker can blend with a sub, given separate left and right parametric EQ (an EQ where you choose each band's frequency, width and level). At real volume, dedicated midbass speakers in stereo (separate left and right) matter. Many people believe bass is mono, so stereo midbass wouldn't matter. He says modern recordings carry stereo bass; bass was mono on vinyl records only because of the medium's limits.

4.6 #9 · Most cars have too much 60–200 Hz; thick midbass is his pet peeve.

MS-8 usually turns that band down "because all cars have way too much of that". Sources: DIYMA 2012-03 · OFCAP 19 @3:31:01

In plain English: He says most cars have too much sound between 60 and 200 Hz, and thick, heavy midbass is his pet peeve. The MS-8 (a JBL processor from his Harman years) usually turned that band down, "because all cars have way too much of that".

4.6 #10 · Midbass power needs are modest (product-specific).

At 115 W with an 80 Hz crossover a GB60 never reaches 3 mm of excursion. Without a sub, a GS690 6x9 makes more bass; with a sub, the GB60 (2" coil, more than twice the Xmax). Sources: DIYMA 2021-07-21 · DIYMA 2021-08-08

In plain English: This point is about specific Audiofrog products. Midbass speakers need modest power: at 115 W with an 80 Hz crossover, the GB60 never moves as much as 3 mm. If you have no sub, a GS690 (a 6x9" oval speaker) makes more bass; with a sub, he picks the GB60, which has a 2" voice coil and more than twice the Xmax.

---

The full series

1. Start here

2. Part 1: How Stereo Imaging Works in a Car, Reflections and Dispersion

3. Part 2: Aiming and Where Each Speaker Goes

4. Part 3: One Seat vs Two, Center Channels, Rear Fill and Upmixing

5. Part 4: Door and Midbass Installation, Sound Deadening and Speaker Locations

6. Part 5: Pods, Tweeter Mounting and Choosing Drivers

7. Part 6: Speaker Design, Driver Failures, Tweeter Protection and Wiring

8. Part 7: His Step-by-Step Tuning Process and Measurement Tools

9. Part 8: Mic Placement, Target Curves and EQ

10. Part 9: Crossovers, Polarity and Phase

11. Part 10: Time Alignment, Levels, Listening Tests, Processors and Common Tuning Mistakes

12. Part 11: Bass in a Car: Cabin Gain, Front Bass, Enclosures and Subwoofer Placement

13. Part 12: Power, Clipping, Gain Setting, Amplifier Specs and Noise

14. Part 13: Wiring and Integrating with Factory Systems

15. Part 14: Car Audio Myths He Debunks (1 of 2)

16. Part 15: Car Audio Myths He Debunks (2 of 2)

17. Part 16: Reading Specs and Reviews, and Planning a System

18. Part 17: Choosing a Shop, the Car Audio Business and SQ Competitions

19. Part 18: Recordings, Listening and Who Andy Wehmeyer Is

20. Part 19: Glossary and Where His Material Lives

Download the complete guide as one PDF (351 pages, every point with its explanation).