Car Audio Explained, Part 4: Door and Midbass Installation, Sound Deadening and Speaker Locations (Andy Wehmeyer's Lessons)

3 days ago

41 min read

This is Part 4 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.

2. Installation craft and hardware choices

Andy Wehmeyer's installation advice is applied physics: put speakers where the car helps rather than hurts, mount them solid and sealed, cross every driver where it is safe and still disperses widely, and stop spending effort (door boxes, foam, aiming, show fabrication) that doesn't change what reaches your ears. Points come from his DIYMobileAudio (DIYMA) posts (anonymized accounts "14642", 2008–2016, and "156546", 2017–2022, both his), Audiofrog's blog, manuals and product pages ("COMPANY" marks unbylined company text), and his podcasts and videos. Jargon used throughout: IB (infinite baffle) = a speaker mounted in a large closed space such as a door, with no box; Fs = a driver's free-air resonance (Fc once it is in a box); Q/Qts = how peaky a driver is around resonance; Vas = the air volume equal to its suspension stiffness; Xmax = linear excursion; LR4 = a 24 dB/octave Linkwitz-Riley crossover.

What this section is about: This section covers the hands-on side of a car system: where the speakers go, how they are mounted, what sound-deadening material really does, and which hardware suits which spot. His approach is practical physics: put speakers where the car helps rather than hurts, mount them solid and sealed, cross each one over where it is safe, and skip work that doesn't change what reaches your ears. Where a point is marked COMPANY, the words come from unsigned text by Audiofrog (his company), not from him by name. A few terms used throughout: Hz (hertz) measures pitch, with low numbers meaning deep bass; dB (decibels) measures loudness, and +6 dB means double the sound pressure, an easily heard increase; a high-pass filter is a crossover setting that stops low notes from reaching a speaker.

2.1 Doors and midbass: baffle, not box

What this part is about: How to mount and choose the mid-size bass speakers ("midbass", usually 6 to 8 inches) that sit in car doors. His message is that a door needs a good seal rather than a box, that the door and the car's corners already add bass for free, and that the deepest bass belongs to a subwoofer.

2.1 #1 · A door midbass needs a sealed baffle, not a box.

[RARE] With any high-pass, a door box is "a waste of time and money": the electronic filter limits excursion as a box's air spring would, so you only need front and back "pretty well isolated". Seal the speaker's front to the trim panel, or put a couple of layers of Dynamat over the sheet-metal holes (both is fine). A door box's only benefit is stopping rattles; if you build one, size it like a sub box. Sources: DIYMA 2009-03-29 · DIYMA 2008-11-22 · DIYMA 2015-08-19

In plain English: A baffle is the surface a speaker is mounted on. Its job is to keep the sound from the back of the cone apart from the sound from the front, because the two are opposite and cancel when they mix. A box does that too, and its trapped air also acts like a spring that limits how far the cone moves. His point is that once you use any high-pass filter, the filter already limits the cone's movement, so the door needs no box. Front and back just need to be "pretty well isolated". You can seal the speaker's front to the door's trim panel, or cover the holes in the door's sheet metal with a couple of layers of Dynamat (a sound-deadening sheet). Doing both is fine. Many people believe door midbass needs a box. He calls one "a waste of time and money" whose only benefit is stopping rattles, and says that if you build one, size it as carefully as a subwoofer box.

2.1 #2 · Tiny sealed boxes behind midbass drivers are "bass prevention".

[RARE] A small box raises the resonance and Q, cuts low end, needs more power for the same bass and adds a peak at resonance. A high-pass limits excursion just as well, and EQ supplies any boost the box would have given. Sources: DIYMA 2014-01-30 · DIYMA 2010-07-04

In plain English: Every speaker has a resonance, a frequency where its cone moves most easily. Below that, its bass output falls away. "Q" describes how peaky the sound is around that resonance. A small sealed box adds the stiffness of its trapped air, which pushes the resonance and the Q up. He says this cuts the deep end, makes the speaker need more amplifier power for the same bass, and adds a bump (a peak) at the new resonance. That is why he calls tiny boxes "bass prevention". Many people box midbass drivers to protect them. His point is that a high-pass filter limits the cone's movement just as well, and EQ (equalization, the electronic tone controls in a processor) can add back any boost the box would have given.

2.1 #3 · In a door, air volume and sealing matter; Fs and Q barely do once the driver is high-passed.

[RARE] A door's volume is similar to or larger than a car driver's Vas, so it hardly changes the driver (IB-type drivers have Qts about 0.5–0.8). With a high-pass above Fs plus EQ, "does it suit IB?" debates are pointless. A mid and a midbass can even share one door cavity: it is huge and "really lossy". Sources: DIYMA 2015-10-10 · DIYMA 2015-07-24 · DIYMA 2014-04-01

In plain English: Vas is the amount of air that is exactly as springy as the speaker's own suspension. A space much smaller than Vas stiffens the speaker and changes how it behaves; a space as big or bigger barely does. He points out that a car door holds about as much air as a typical car speaker's Vas, or more, so the door hardly changes the driver. Drivers meant for infinite-baffle (no-box) use have a Qts of about 0.5–0.8. So once the speaker has a high-pass set above its resonance (Fs) and some EQ, arguing over whether a driver "suits IB" is pointless. What matters is the door's air volume and how well it is sealed. A midrange and a midbass can even share one door cavity. He says no good engineer would do that in a home speaker, but in a car it is no big deal: a door is huge and "really lossy" (its air does not act as a tight spring).

2.1 #4 · The door doesn't change the speaker; the cabin changes what reaches you.

A driver flat from 100 Hz to 10 kHz stays flat in a door, and no driver can be designed to cancel the car: the gap from its anechoic curve is reflections, which EQ handles. So pick drivers for the job their data shows (he says a Purifi 4" is plainly built for a small vented home 2-way). Sources: DIYMA 2013-10-30 · DIYMA 2021-12-24

In plain English: An "anechoic" measurement is taken in a room with no echoes, so it shows the speaker's own sound. His point is that mounting a speaker in a door does not change the speaker. If it measures flat (even) from 100 Hz to 10 kHz, it is still flat in the door. When your in-car measurement looks different, the difference is the car's reflections, and EQ is the tool for those. No speaker can be designed to cancel out what a car does. So pick a driver for the job its published data shows it was designed for. His example is the Purifi 4", which he says is plainly built for a small vented two-way home speaker.

2.1 #5 · Ported midbass: generally no, and never vent a door box into the door.

A port big enough to avoid noise is "huge", and a grille in front of it shifts the tuning; a long-coil sealed/IB driver with power and EQ is easier. The one exception: a high-Fs driver you are forced to use, vented to lower its −3 dB point. A port firing into the door throws its output away, leaving less bass than sealed. Sources: DIYMA 2011-10-18 · DIYMA 2008-11-22 · DIYMA 2022-01-13

In plain English: A ported (vented) box has an opening, the port, that adds bass around a chosen "tuning" frequency; a sealed box has no opening. He is generally against porting midbass. A port big enough not to make air noise would be "huge" for a midbass, and a grille in front of the port shifts its tuning. He finds it easier to use a sealed or IB driver with a long voice coil (so it can move further) plus enough power and EQ. His one exception is a driver with a high resonance (high Fs) that you are forced to use, for example because nothing else fits. A vent can lower its −3 dB point, the frequency where its bass has dropped to half power. Never vent a door box into the door. What you hear from a ported box is the speaker plus the port, so a port firing into the door throws its output away, leaving less bass than a sealed box would give.

2.1 #6 · Pull out foam cup "speaker baffles"; at most keep the top as a rain shield.

[RARE] For a "lifeless and hollow" system his first instruction was to remove the foam baffles (then check polarity, set gains and recalibrate). In 2021: cut off the bottom of the cup and keep only the upper rain-shield part. Sources: DIYMA 2013-05-19 · DIYMA 2021-09-21

In plain English: Foam "speaker baffles" are foam cups sold to sit behind a car speaker. A forum member described his system as "lifeless and hollow". The first thing Andy told him to do was take those foam baffles out. After that: check polarity (that every speaker's + and − are wired the same way round), set the amplifier gains (input levels), and recalibrate (rerun the processor's setup). In 2021 his advice was to cut off the bottom of the cup and keep only its top part, as a shield against rain water running down inside the door. The sources don't spell out his reason.

2.1 #7 · Foam rings go on the bracket, never on the speaker; they stop rattles, they don't "make midbass".

[RARE] Returned speakers often arrive caked in foam-ring adhesive, sometimes stuck to the surround. His adapters carry closed-cell foam about 3/4" wide on the bracket, never touching the speaker, to stop the door panel rattling. Sources: OFCAP Ep. 38 @34:02

In plain English: Foam rings are strips of foam placed around a car speaker. Many people believe they "make midbass". He says their only job is to stop rattles. Never stick them to the speaker itself. Speakers sent back to his company often arrive caked in the rings' glue, sometimes even stuck to the surround (the flexible edge around the cone). Audiofrog's mounting adapters instead carry closed-cell foam (a dense foam whose bubbles are sealed) about 3/4 inch (19 mm) wide on the bracket. It never touches the speaker, and it stops the door panel rattling against the bracket.

2.1 #8 · The door and the car's corner give midbass its bass: about +6 dB per halving of the radiating space.

[RARE] Below about 500 Hz a 6" radiates everywhere; the baffle reflects its rearward-going front sound forward (+6 dB) and a corner adds another +6 dB. Midbass always sits near a corner (door/floor or kick panel), so the location boosts bass on top of cabin gain. Sources: Audiofrog blog Jun 2021 · DIYMA 2017-08-17

In plain English: Low notes spread out from a speaker in every direction, like ripples spreading in a bathtub. If a wall blocks half of that space, the sound that would have gone backward is bounced forward, so it gets louder in front. His rule of thumb is about +6 dB each time the space the speaker radiates into is cut in half. Below about 500 Hz, a 6" speaker radiates everywhere. The door (its baffle) reflects the part heading backward toward the front (+6 dB), and a corner adds another +6 dB. A midbass always sits near a corner (where the door meets the floor, or in a kick panel), so its location boosts its bass. That comes on top of "cabin gain", the extra deep bass a small closed space like a car naturally adds.

2.1 #9 · So choose a flat, low-Q midbass; that is why the GB60 has no bass hump.

[RARE] A high-Q "European sound" hump helps in a home box but becomes a peak in a car corner, and high Q raises distortion at Fs; flat midbass also blends more easily with the sub (a reviewer measured Qts 0.452, Fs 54.8 Hz). Don't chase ultra-low-Fs "door subwoofer" midbass if you have a sub: one was nearly 10 dB less sensitive in the midbass, "a lot of power". Sources: Audiofrog blog Jun 2021 · DIYMA 2017-08-17 · DIYMA 2021-11-02

In plain English: Some speakers have a high Q, which gives them a bump of extra bass near their resonance. He calls it the "European sound", and it helps in a home speaker box. His point is that in a car corner, which already adds bass (see the point above), that bump turns into an unwanted peak, and a high Q also raises distortion near the resonance (Fs). A flat, low-Q midbass also blends more easily with the subwoofer. That is why Audiofrog's GB60 is designed with no bass hump; a reviewer measured its Qts at 0.452 and Fs at 54.8 Hz. If you have a sub, don't chase "door subwoofer" midbass drivers with an ultra-low Fs. He judged one nearly 10 dB less sensitive in the midbass, which means about ten times the amplifier power for the same loudness: "a lot of power".

2.1 #10 · Cross door midbass at about 80 Hz and the sub at about 60 Hz, both LR4; 40 Hz is the worst mistake he sees.

[RARE] Door drivers' Fs is around 60–70 Hz; below resonance only the spider controls the cone, distortion soars, the spider wears "like an old shoe" and the car becomes "virtually untunable". The half-octave gap keeps sub and midbass in phase after the 9–12 dB sub boost cars need; at 80 Hz a GB60 moves under 3 mm at 115 W. Sources: OFCAP Ep. 19 @1:45:06 · DIYMA 2020-12-30 · DIYMA 2021-07-24

In plain English: A crossover frequency is where one speaker hands the music over to another. LR4 is a common, steep type of crossover filter. The spider is the fabric ring behind the cone that holds it centered and acts as part of its spring. He says door speakers resonate around 60–70 Hz. Below that resonance only the spider controls the cone, so distortion soars, the spider wears out "like an old shoe", and the car becomes "virtually untunable". That is why crossing door midbass at 40 Hz is the worst mistake he sees. Instead, cross the midbass at about 80 Hz and the sub at about 60 Hz, both LR4. Cars need the sub boosted 9–12 dB, and the half-octave gap (an octave is a doubling of frequency) keeps sub and midbass in phase (pushing air together, so they add rather than cancel) after that boost. At 80 Hz a GB60 moves less than 3 mm at 115 W.

2.1 #11 · Door woofers can't replace a sub in a loud system.

[RARE] Cars need about 6–12 dB more bass than midrange; getting it from doors means big boosts near 50 Hz, "exactly what your door drivers hate", and the coil leaves the gap (odd-order distortion). Add a 10" sub: if space is short, swap the spare tire for Fix-a-Flat and build in the well, or vent a bandpass box into the trunk. One driver from sub-bass to 1 kHz can't be built and isn't needed. Sources: DIYMA 2013-05-22 · OFCAP Ep. 19 @1:48:08 · DIYMA 2017-07-05

In plain English: He says cars need about 6–12 dB more bass than midrange. Getting that from the door speakers means big boosts near 50 Hz, "exactly what your door drivers hate". The voice coil is pushed out of the magnet's gap, which causes "odd-order" distortion (a kind that adds tones at odd multiples of the note, such as three times it). His fix is to add a 10" subwoofer. If you are short of space, he suggests replacing the spare tire with a can of Fix-a-Flat (tire sealant) and building the sub box in the spare-tire well. Another option is a bandpass box (one where the sound leaves only through a port) that vents into the trunk. He adds that one driver playing from deep bass up to 1 kHz can't be built, and isn't needed.

2.1 #12 · No need for 8" or 10" door woofers; with a sub, put a 6" high in the door's top corner.

[RARE] With subs low-passed at 100 Hz a user heard no difference at full volume with door panels removed ("Nope, no need"); in 25 years Andy heard one good car with very large front woofers. A 6" in the top corner (the factory grille area) makes left and right far more alike; with correct crossovers most of the image then sits at tweeter height. Sources: DIYMA 2011-08-28 · DIYMA 2021-12-03

In plain English: Many people think bigger door speakers (8" or 10") give better front bass. In the forum thread, a user had his subs low-passed at 100 Hz, meaning the subs played everything below 100 Hz. He took his door panels off and heard no difference at full volume. Andy agreed: "Nope, no need" for 10" door woofers. Andy said that in 25 years he had heard only one good-sounding car with very large front woofers. With a sub, he suggests putting a 6" high, in the door's top corner, around where the factory grille usually sits. That makes the sound from the left and right speakers far more alike. With correct crossovers, most of the stereo image (where you hear the music coming from) then sits at the height of the tweeters.

2.1 #13 · Bigger isn't "slower", and punch doesn't live in the midbass.

[RARE] The motor, not cone size, sets "speed". Punchy midbass depends on flat response from about 1 to 4 kHz (impact near 3 kHz: a bass player's thumb slap, the snap of toms). Test it on headphones through a 100 Hz, then a 200 Hz low-pass: there is no punch down there. Sources: DIYMA 2011-08-27 · DIYMA 2015-05-09 · DIYMA 2021-10-13

In plain English: Many people believe big woofers are "slow" and small ones are "fast" or punchy. He says cone size doesn't set "speed"; the motor (the magnet and voice coil that drive the cone) does. And the "punch" people want from midbass actually depends on even response from about 1 to 4 kHz. The sense of impact sits near 3 kHz: a bass player's thumb slap, the snap of tom drums. His test is to listen on headphones through a 100 Hz low-pass filter, then a 200 Hz one, so you hear only the notes below that point. There is no punch down there.

2.1 #14 · A midbass and a sub are optimized differently; a "sub that plays midrange" mostly doesn't exist.

[RARE] A good midbass has low moving mass, about 1/4 a sub's inductance and a very compliant suspension for low Fs without added mass, so it suits a door or rear deck, not a small box. Low Fs plus high efficiency means a huge Vas and a fragile driver (Hoffman's Iron Law); the closest thing, for horn systems, is a large pro woofer in a vented box. Sources: DIYMA 2022-01-13 · DIYMA 2020-08-24

In plain English: Moving mass is the weight of the cone and coil. Inductance is an electrical property of the coil that holds back higher notes. Compliance is how soft the suspension is. He says a good midbass has low moving mass, about a quarter of a sub's inductance, and a very soft suspension. That gives it a low resonance (Fs) without adding weight, but it suits a door or rear deck (big air spaces), not a small box. A driver that has both a low Fs and high efficiency needs a huge Vas and ends up fragile. He ties this to Hoffman's Iron Law, a speaker-design rule that you cannot have deep bass, a small box and high efficiency all at once. The nearest thing to a "sub that plays midrange", used in horn systems, is a large professional (pro audio) woofer in a vented box.

2.1 #15 · An 8" in a door is a woofer, not a midrange: the GS8ND2 case.

[RARE] About 3" deep for 6×9/8" door openings and tiny boxes (about 7 L sealed, 14 L vented, or IB: under seats, Lexus decks); a shorting ring lets it play to about 300 Hz, so cross it there and add a GS25/GB25. Dual 2 Ω coils: one woofer wired to 4 Ω suits a bridged 150–200 W channel pair; a pair at 2 Ω suits a 500–750 W sub amp. [Changed view: in 2016 Audiofrog would make no 8" and offered the GS690 6×9 instead; the GS8ND2 arrived in 2020, framed as a woofer.] Sources: DIYMA 2016-06-21 · DIYMA 2020-06-24 · CMA Networks 2021 @39:01

In plain English: The GS8ND2 is an Audiofrog 8" speaker; the GS25 and GB25 are its 2.5" midranges. A shorting ring is a metal ring in the motor that lowers inductance, helping the driver play higher notes. He describes the GS8ND2 as a woofer, not a midrange. It is about 3" (76 mm) deep, made for 6×9 or 8" door openings and very small boxes: about 7 L (liters) sealed or 14 L vented, or no box at all (IB), as under seats or in Lexus rear decks. The shorting ring lets it play up to about 300 Hz, so cross it there and add a GS25 or GB25 above it. It has two 2 Ω (ohm) voice coils. Wired to 4 Ω, one woofer suits a bridged pair of 150–200 W amplifier channels ("bridged" means two channels combined into one stronger one). A pair wired to 2 Ω suits a 500–750 W sub amp. The tag records a change: in 2016 Audiofrog said it would make no 8" and offered the GS690 6×9 instead; the GS8ND2 came in 2020, presented as a woofer.

2.1 #16 · Without a sub a 6×9 makes more bass; with a sub, the GB60 is the better driver.

The GB60 has a 2" coil, neodymium motor, shorting ring and more than twice the GS690's excursion; in a no-sub car the GS690 6×9 gives more bass. Sources: DIYMA 2021-08-08 · DIYMA 2020-08-25

In plain English: A 6×9 is an oval speaker, 6 by 9 inches, with a lot of cone area. Without a subwoofer, the GS690 6×9 makes more bass. With a sub handling the deep notes, the GB60 is the better driver. It has a 2" voice coil, a neodymium (strong, lightweight) magnet motor, a shorting ring, and more than twice the GS690's excursion (how far its cone can move).

2.1 #17 · Under-seat midbass: keep an octave of usable bandwidth around each crossover.

[RARE] In a BMW build he crossed the factory under-seat woofers at 200 Hz (as low as the 4" door speakers could play loud), so they ran about 80–200 Hz with the sub below 60 Hz. BMW under-seat boxes behave like aperiodic enclosures, borrowing frame airspace through small openings; high-Q drivers crammed in never sounded good to him. Sources: DIYMA 2021-08-04 · DIYMA 2020-06-24

In plain English: "Usable bandwidth" is the range of notes a speaker plays well. His rule is to leave about an octave (a doubling of frequency) of good performance on each side of every crossover. In a BMW build he crossed the factory under-seat woofers at 200 Hz, because that was as low as the 4" door speakers could play loudly. The under-seat woofers then covered about 80–200 Hz, with the sub below 60 Hz. He says BMW's under-seat boxes behave like "aperiodic" enclosures. An aperiodic box is deliberately not airtight: it breathes through damped openings, so its air is not a tight spring. These boxes borrow air from the seat frame through small openings. High-Q drivers squeezed into them never sounded good to him.

2.1 #18 · Factory-door fit is set by mounting depth and the factory grille.

GB40: 41 mm (1-5/8"), fits late BMW doors; GB60: under 2-3/4"; GS8ND2: just over 3". About 85% of G62 coaxes go under factory grilles, so it ships without one; coax tweeters sit low (GS62 2 mm, GS42 3.75 mm above the surround), and kits add multi-fit adapters and gaskets (GS682 removable gasket, GS693 six strips, GS25 foam gaskets). (Specs COMPANY.) Sources: Crutchfield, Audiofrog packages · CMA Networks 2023 · GS series store collection

In plain English: Mounting depth is how far a speaker sticks out behind the surface it is screwed to. Many factory doors have little room. The depths given: GB40 41 mm (1-5/8"), which fits late-model BMW doors; GB60 under 2-3/4" (about 70 mm); GS8ND2 just over 3" (about 76 mm). A coaxial speaker ("coax") has its tweeter mounted in front of the woofer. About 85% of G62 coaxes fit under the car's own grille, so it ships without one. Coax tweeters sit low to clear grilles: 2 mm above the surround on the GS62, 3.75 mm on the GS42. The kits add adapters that fit many cars and gaskets (seals): a removable gasket on the GS682, six strips with the GS693, foam gaskets with the GS25. "(Specs COMPANY.)" means these specs come from Audiofrog's unbylined company text, not from a post signed by him.

2.1 #19 · Leave clearance beyond Xmax: a surround hitting the panel or grille is an install error.

[RARE] Xmax is only the linear part; the cone travels further through the top-plate thickness and on momentum (GB10 10": 17.6 mm rated plus about 8 mm), so keep grilles about 1.25 in from a woofer's front. His GB60 in a gutted factory Dynaudio frame hit the Volvo's door panel. A 4" that distorts at 150–400 Hz usually has a loose mount or a surround hitting the grille. Sources: DIYMA 2021-12-22 · DIYMA 2021-09-22 · OFCAP Ep. 38 @33:02 · DIYMA 2021-12-02

In plain English: Xmax is how far a speaker's cone can move while staying clean (linear). He warns that the cone can travel further than that. The coil can move beyond the thickness of the magnet's top plate, and the cone can carry on moving on its own momentum. For Audiofrog's GB10 10" sub, that is 17.6 mm rated plus about 8 mm more. So keep a grille about 1.25 in (about 32 mm) from a woofer's front. He calls a surround hitting the panel or grille an installation error. It happened to him: his GB60, mounted in a gutted factory Dynaudio frame, hit the Volvo's door panel. If a 4" distorts at 150–400 Hz, the usual cause is a loose mount or a surround hitting the grille.

2.1 #20 · Grilles are fine; just don't bury a mid in a tunnel.

[RARE] Home speakers are designed with their grilles on; for mids a grille over about 50% open area makes no difference, and Audiofrog tweeters are designed for theirs. Listen for cloth slapping the mesh under bass, never put a midrange at the end of a 6" tunnel, expose as much open area as possible and leave real space between cone and grille. Sources: DIYMA 2012-08-12 · DIYMA 2015-02-27 · OFCAP Ep. 38 @33:02

In plain English: Many people remove grilles, believing they spoil the sound. He says home speakers are designed with their grilles on. For midranges, a grille that is more than about 50% open makes no difference, and Audiofrog tweeters are designed to play through their own grilles. His cautions: listen for grille cloth slapping against the metal mesh when bass plays, never put a midrange at the end of a 6" (150 mm) tunnel, expose as much open area as you can, and leave real space between the cone and the grille.

2.1 #21 · Aiming a door 6" is pointless between 80 and 200 Hz, and door mounting doesn't cost depth.

It radiates everywhere there; the real problems are path-length differences and the sub-to-midbass blend, and a small mid above it removes any need to aim. Speakers placed further forward move where the stage sits, not its depth (he says IASCA judges confuse the two). Sources: DIYMA 2010-02-14 · DIYMA 2013-04-27

In plain English: Between 80 and 200 Hz, a 6" speaker sends sound in every direction, so pointing it at you changes nothing. He says the real problems at those notes are path-length differences (the left and right speakers are different distances from you) and getting the sub and midbass to blend. A small midrange above the 6" removes any need to aim it at all. Many people believe door speakers can't give a deep soundstage. He says placing speakers further forward changes where the stage sits, not its depth (the sense of instruments layered from front to back). He says judges in IASCA, a car-audio competition, confuse the two.

2.1 #22 · Measure before rebuilding a door location.

Measure near-field at the grille, then at the seat: lows won't differ, highs will, which separates a location problem from a cabin problem. A flat +6 dB above 250 Hz with the trim off is the car's door-open volume cut (check the door switch), not the panel. Near-field readings can't show panel vibration or cone breakup; those appear only beyond about 3× the driver's largest dimension. Sources: DIYMA 2010-02-18 · DIYMA 2022-01-30 · DIYMA 2022-03-16

In plain English: "Near-field" means measuring with the microphone very close to the speaker, so it picks up mostly the speaker itself. Before rebuilding a door location, he says, measure near-field at the grille, then measure at your seat. Low notes won't differ between the two; high notes will. The comparison shows whether a problem comes from the speaker's spot or from the cabin; the notes don't spell out the comparison further. If the response is a flat 6 dB higher above 250 Hz with the door trim removed, the trim is not the cause. Some cars turn the volume down about 6 dB when they think a door is open, so check the door switch. Near-field readings also can't show panel vibration or "cone breakup" (the cone flexing instead of moving as one piece). Those appear only beyond about three times the driver's largest dimension, roughly 20 in (50 cm) for a 6.5" speaker.

2.2 Sound deadening, rattles and road noise

What this part is about: What sound-deadening material is really for, and what it can't do. He treats deadening as a way to stop rattles and panel vibration, not as a way to soak up sound or change the car's acoustics. Road noise gets its own short recipe.

2.2 #1 · Deaden to stop rattles and panel movement, in a set order.

[RARE] Door priority: (1) the trim panel, the worst rattle source (a large passenger leaning on it shows it); (2) the inner skin; (3) the outer skin only with material left over. Foam anything that might rattle; wrap or tie down harnesses. In his C30 he damped doors and floor, not the headliner, with Ground Zero material, no buckshot. Sources: OFCAP Ep. 38 @1:13:01

In plain English: Sound deadener is sheet material stuck to panels to stop them vibrating. A door has a plastic trim panel on the inside, then two metal "skins": the inner skin (facing the cabin) and the outer skin (the outside of the car). He says the goal is to stop rattles and panel movement, and gives this order for doors. First the trim panel, the worst rattle source: a large passenger leaning on it makes a loose panel obvious. Then the inner skin. Then the outer skin, only if you have material left over. Put foam on anything that might rattle, and wrap wiring harnesses in foam or tie them down. In his Volvo C30 he damped the doors and floor but not the headliner (the roof lining). He used Ground Zero material and no buckshot (lead pellets some builders pour in to add weight).

2.2 #2 · Know what each material does: foam stops rattles, mass lowers panel resonance, and absorbing low frequencies takes feet of material.

[RARE] Absorbers start working at about a quarter wavelength, so a 200 Hz reflection needs about 1.5 ft: padding a console is wasted. Real absorbers are membranes (MLV's job in walls) and fiber, where friction makes heat. A "deflex" pad or Dynamat just adds mass, lowering panel resonance; foam absorbs almost nothing. Sources: DIYMA 2011-11-24 · DIYMA 2015-11-15 · DIYMA 2022-01-14

In plain English: He says each material does a different job. Foam stops rattles. Mass, such as a "deflex" pad (a heavy pad stuck behind a door speaker) or Dynamat, just makes a panel heavier, which lowers the note it vibrates at (its resonance). Foam absorbs almost no sound. Real absorbers are membranes (that is MLV's job in walls; MLV is mass-loaded vinyl, a heavy flexible sheet) and fiber, where friction between the fibers turns sound into heat. Absorbers only start working once they are about a quarter of a wavelength thick, and low notes have long waves. So absorbing a 200 Hz reflection takes about 1.5 ft (45 cm) of material. Many people pad a center console to tame reflections; he says that is wasted effort.

2.2 #3 · A 250 Hz peak is usually a reflection to EQ, not a panel to deaden: prove it with a tone.

[RARE] A true resonance is mass and springiness trading energy (a door skin); a cabin peak at 250 Hz is usually a dimension reinforcing by reflection. Play a 250 Hz tone: if nothing buzzes, it's constructive interference, so cut it with EQ. Sources: DIYMA 2021-09-07

In plain English: A "peak" is a note that measures louder than its neighbors. He says people often blame a peak near 250 Hz on a vibrating panel and reach for more deadener. A true resonance is a mass and a springiness passing energy back and forth at one note, as a door skin does. A cabin peak at 250 Hz is usually something else: one of the car's dimensions makes a reflection arrive in step with the direct sound, so the two add up. His test is to play a 250 Hz tone. If nothing buzzes, it is constructive interference (two sound waves adding together), not a panel, so cut it with EQ.

2.2 #4 · Road noise: deaden, then buy quiet tires, then make per-road presets; never EQ for noise above 300 Hz.

[RARE] "The best midbass upgrade I ever made" was swapping aggressive Goodyears for Continentals. Record noise on your usual roads and build a couple of DSP presets (in 2008 he kept one for a coarse-pavement highway). Sources: DIYMA 2013-04-25

In plain English: His order for dealing with road noise: first deaden, then buy quieter tires, then make processor settings for particular roads. He calls swapping aggressive Goodyear tires for Continentals "The best midbass upgrade I ever made". For the last step, record the noise on roads you usually drive. Then build a couple of DSP presets: saved settings in your digital signal processor that you switch between. Back in 2008 he kept one for a highway with coarse pavement. His limit: never use EQ to fight noise above 300 Hz. The source doesn't spell out why he chose that limit.

2.2 #5 · Don't treat the cabin to kill reflections: mats, films and carpet barely work, and it's whack-a-mole.

[RARE] Fibrous dash mats absorb only the top octaves; a film neither absorbs nor diffuses (the only two ways to reduce a reflection). Chasing each reflection with shapes and foam is "like pulling weeds in the jungle". To identify one, read the impulse response: delay after the first arrival × 1,132 ft/s gives the extra path (in a Mini, 4.9 ms was the windshield). Sources: DIYMA 2013-09-14 · DIYMA 2010-05-10

In plain English: A reflection is sound that bounces off a surface, such as glass or the dash, and reaches you a little after the direct sound. He says there are only two ways to reduce one: absorb it or diffuse (scatter) it. Fibrous dash mats absorb only the top octaves (the highest notes), and a film on glass does neither. Many people try to tame each reflection with shapes, foam and carpet; he calls it "like pulling weeds in the jungle". To find which surface causes a reflection, read the impulse response. That is a measurement over time that shows the direct sound arriving first, then each echo. Take an echo's delay after the first arrival and multiply by 1,132 ft/s, the speed of sound (about 345 m/s). That gives how much further the echo traveled; for example, 1 ms (a thousandth of a second) is about 1.1 ft. In a Mini, the 4.9 ms peak was the windshield.

2.2 #6 · Heavy absorption sounds accurate but small.

[RARE] The most absorptive car he heard (thick panels over almost all glass, speakers low or in under-dash horns) was loud, dynamic and rattle-free, like great headphones in front of you, but lacked space; he saw it only as a base for added multichannel ambience. Sources: DIYMA 2013-09-15

In plain English: Many people believe the fewer reflections a car has, the better it sounds. The most absorptive car he ever heard had thick panels over almost all the glass and speakers mounted low or in horns under the dash. He found it loud and dynamic with no rattles, like great headphones but with the sound in front of you. But it lacked a sense of space. He saw it only as a starting point for adding multichannel ambience, meaning extra channels and processing that recreate the feeling of a larger space.

2.2 #7 · Rattles pull the bass image backward.

[RARE] For bass to seem frontal, the sub must not make distortion or rattles (and must be in phase with the fronts at the crossover), so rattle control is an imaging fix. His own IB sub hangs from the package shelf on rubber isolation mounts to avoid rattles. Sources: Tuning PDF v1 (~2018) · DIYMA 2017-12-09

In plain English: "Imaging" is where you hear each sound coming from. A subwoofer usually sits behind you, in the trunk or rear deck, yet bass should seem to come from the front. He says that only happens if the sub makes no distortion or rattles, and is in phase with the front speakers at the crossover (both pushing air together there). Rattles therefore pull the bass toward the back, so stopping them is part of fixing the image. His own IB sub hangs from the package shelf (the shelf behind the rear seat) on rubber isolation mounts, to avoid rattles.

2.2 #8 · Build sub boxes rigid and airtight; beyond that the box matters only near resonance.

A couple of octaves above resonance a box has little effect unless it "is leaky, flexes a lot or buzzes". Line it with about 1" of fiberglass or dacron and model the stuffing (he used LEAP); polyfill mainly absorbs high frequencies far above a sub's range. Sources: DIYMA 2008-05-26 · DIYMA 2013-10-30

In plain English: His basic rule for subwoofer boxes is to make them stiff and airtight. Beyond that, he says, the box only shapes the sound near the speaker's resonance. A couple of octaves above it (for example, from about 160 Hz up if the resonance is 40 Hz), a box has little effect unless it "is leaky, flexes a lot or buzzes". He says to line the box with about 1" (25 mm) of fiberglass or dacron (polyester fiber) and to include the stuffing when you model the box in design software; he used LEAP. Polyfill (pillow-type stuffing) mainly absorbs high notes, far above the range a sub plays.

2.2 #9 · Prove what treatment does by recording it.

[RARE] Audiofrog suggests binaural recordings of treated and untreated cars to show treatment's effect, and recording a car's ambient noise to set "masking equalization". (COMPANY.) Sources: Audiofrog Binaural page

In plain English: A binaural recording is made with two microphones in the ears of a dummy head, so that playback on headphones sounds as if you were there. Audiofrog's company page suggests two uses. First, record treated and untreated cars this way to show what sound treatment actually changes. Second, record a car's background noise to set "masking equalization". Masking means a steady noise makes quieter sounds of similar pitch harder to hear; the page doesn't explain the method further. "(COMPANY.)" marks this as unsigned Audiofrog company text, not a post by him.

2.2 #10 · Not found in anything he said:

CLD coverage percentages, MLV as a car noise barrier, closed-cell decoupler layers, door water management, a step-by-step rattle hunt, or brand-versus-brand deadener comparisons. The OFCAP 38 order above is his only stated deadening procedure.

In plain English: This is a researchers' note, not one of his claims. They searched his material and found nothing on these topics. CLD coverage percentages: how much of a panel to cover with constrained-layer damping, the foil-backed sheets such as Dynamat. MLV as a noise barrier in cars. Closed-cell "decoupler" foam layers. Managing water inside doors. A step-by-step hunt for rattles. Brand-versus-brand deadener comparisons. The order he gave on the OFCAP podcast, Episode 38 (point 2.2 #1 above), is the only deadening procedure he stated. So advice on those other topics should not be credited to him.

2.3 Where midranges and tweeters go

What this part is about: Where to put midranges (mid-size speakers for voices and most instruments) and tweeters (small speakers for the highest notes). His main idea is that the windshield, dash and side windows bounce sound back. Those bounces cancel certain notes, so a speaker should sit either right in a corner or surface, or well away from it, never hanging a short distance off it. The part ends with the products and the cars he built.

2.3 #1 · A-pillars are "absolutely the worst places to mount speakers", especially mids, and delay can't rescue them.

[RARE] A mid a few inches from the windshield/dash/side-glass apex makes each surface a second source with its own comb filter; you can't delay a reflection apart from its source. A pod about a foot off that corner puts a 2-ft reflection path and a massive hole near 250 Hz. The only gain is height, which a correct crossover gives anyway. Sources: DIYMA 2012-09-03 · DIYMA 2013-11-20 · DIYMA 2021-08-26

In plain English: An A-pillar is the slanted post on each side of the windshield. Many people put mids in A-pillar pods to lift the soundstage. A comb filter is what happens when a sound mixes with a slightly late copy of itself, such as a reflection: some notes cancel and others add, in a regular pattern like the teeth of a comb. He says a mid a few inches from the corner where windshield, dash and side glass meet turns each of those surfaces into a second source, each with its own comb filter. Delay (time alignment) can't fix this: a reflection is made by the speaker's own sound, so you can't delay one apart from the other. A pod about a foot (30 cm) off that corner gives a 2-ft (60 cm) reflection path and a massive hole near 250 Hz. The pillar's only gain is a higher stage, which he says a correct crossover gives anyway.

2.3 #2 · If you must use the pillar, push the pod into the base of the corner, flush, and never fire mids up at the glass.

[RARE] Flush in the pillar firing across the dash is fine with a low enough crossover, and closer to the corner is better. Mids fired at the windshield or ceiling become omnidirectional horizontally, smearing toward the near side glass: "the most dense comb filter possible". Sources: DIYMA 2021-09-07

In plain English: "Flush" means mounted so the speaker's face is level with the surface instead of sticking out. If you must use the pillar, he says, push the pod into the base of the corner and mount it flush. Flush in the pillar, firing across the dash, is fine if the crossover is low enough (the source gives no number), and the closer to the corner the better. Never fire mids up at the windshield or ceiling. Aimed that way, at midrange notes they spread sound in every horizontal direction. The image then smears toward the side window nearest you, and he calls the result "the most dense comb filter possible".

2.3 #3 · Use a tape measure before building: distance to the nearest surface sets where the reflection dip lands.

[RARE] A null falls where the reflected path is half a wavelength longer than the direct one: a mid 6" from a surface (about 1 ft extra path) dips near 500–550 Hz, repeating near 1.5, 2.5, 3.5 kHz; 1 ft away, near 280 Hz. Nulls can't be EQ'd and smear imaging. To find a culprit, match the null to a surface's extra path and block it to confirm. Sources: Audiofrog blog Jun 2021 · DIYMA 2020-10-07 · DIYMA 2021-11-06

In plain English: A null is a deep dip where sound cancels. A reflection travels further than the direct sound. At the note where that extra distance is exactly half a wavelength, the two arrive opposite, one pushing as the other pulls, and cancel. So the distance from a speaker to the nearest surface decides where its dip lands, and he says to check it with a tape measure before building. His numbers: a mid 6" (15 cm) from a surface has about 1 ft (30 cm) of extra path and dips near 500–550 Hz, repeating near 1.5, 2.5 and 3.5 kHz. A mid 1 ft away dips near 280 Hz. He says nulls can't be fixed with EQ, and they smear the imaging. (Boosting a note boosts its reflection too, so the two still cancel.) To find which surface is to blame, match the null's frequency to a surface whose extra path fits, then block that path to confirm.

2.3 #4 · Tuck mids into a corner or flush in a surface; keep tweeters away from surfaces.

[RARE] Close to a surface pushes the null above the mid's low-pass; mounting in the surface removes it (why dash corners work great). Pillars fail because they are near several surfaces but flush in none: spreading chaos isn't reducing it. Position relative to glass and dash matters more than mid-to-tweeter spacing, which forums obsess over. Sources: DIYMA 2013-08-10 · DIYMA 2021-12-24

In plain English: This follows from the point above: the closer a speaker is to a surface, the higher its dip. He says to tuck a mid very close to a surface so the dip moves above its low-pass, the top of the range the mid plays, where it no longer matters. Mounting the mid in the surface (flush) removes the dip, because that surface becomes part of the baffle instead of a reflector. That is why dash corners work so well for mids. Tweeters go the other way: keep them away from surfaces, so the dip falls below the notes they play. Pillars fail because they are near several surfaces but flush in none; spreading the problem around is not the same as reducing it. Many people on the forums obsess over the distance between mid and tweeter. He says the mid's position relative to the glass and dash matters more.

2.3 #5 · Best front layout: tweeter in the sail (mirror) panel, mid at the top of the door below it.

[RARE] It moves both away from the windshield and makes door and window the baffle (bigger baffles mean fewer nearby reflections). A sail tweeter about 1.5 in from the glass dips near 2.2 kHz, low enough not to matter, and gives a wider stage than dash or pillar. Sources: DIYMA 2011-11-05 · DIYMA 2017-10-31

In plain English: The sail panel is the small triangular panel at the front corner of the front door window, where the mirror usually mounts. His preferred layout is the tweeter in the sail panel and the mid at the top of the door, just below it. That moves both away from the windshield, and it makes the door and window act as the baffle. A bigger baffle means fewer nearby surfaces to reflect from. A sail tweeter about 1.5 in (38 mm) from the glass has its reflection dip near 2.2 kHz, which he says is low enough not to matter. He also finds sail tweeters give a wider stage than dash or pillar ones.

2.3 #6 · Stack the tweeter directly above the mid, as home speakers do.

[RARE] Vertical stacking keeps mid/tweeter alignment at every horizontal angle (only vertical angles cause lobing); side-by-side works much worse. With good crossovers you hear each pair at the higher driver, so width doesn't come from the mids unless the tweeter is crossed absurdly high (about 7 kHz). His own doors are built this way. Sources: DIYMA 2021-09-07

In plain English: Home speakers stack their drivers in a vertical line, and he says car speakers should too: the tweeter directly above the mid. Stacked that way, the mid and tweeter stay in step at every left-to-right angle. Only up-and-down angles cause "lobing", where the two drivers partly cancel each other near the crossover because their distances to you differ. Side by side works much worse. He explains that with good crossovers you hear each pair of drivers as coming from the higher-pitched one, so a mid/tweeter pair sounds as if it comes from the tweeter. So stage width doesn't come from the mids, unless the tweeter is crossed absurdly high (about 7 kHz). His own doors are built this way.

2.3 #7 · Kick panels: better geometry, not a cure.

[RARE] They cut the left/right path difference from about 28" (center cancelling near 250 Hz) to about 6", "Better, but not fixed", and sit far from the windshield corner. But the footwell often makes a big low-midrange dip (EQ can't fill it), legs block them, midbass boxes are hard to fit, and they fail when you sit high above them (over about 20° down: minivans, trucks). The passenger kick measured as the strongest midbass spot. Sources: Audiofrog blog Dec 2020 · DIYMA 2016-04-27 · DIYMA 2021-09-16

In plain English: Kick panels are the side panels in the front footwells, near your feet. The driver sits much closer to the left speakers than the right. With door mids the difference is about 28" (71 cm), which makes left and right cancel each other for center sounds near 250 Hz. Kick panels cut that difference to about 6" (15 cm): "Better, but not fixed". They also sit far from the windshield corner. The downsides he lists: the footwell often causes a big dip in the low midrange that EQ can't fill, your legs block the sound, and midbass boxes are hard to fit there. They also fail when you sit high above them, more than about 20° downward, as in minivans and trucks. In his measurements, the passenger kick panel was the strongest spot for midbass.

2.3 #8 · Dash mounts get big free low-end gain: cut it, don't boost the "missing" top.

[RARE] The dash cuts the radiating space by 3/4 or more (+6 dB per halving) and the windshield adds more, so a 2.5" in the dash can play down to 200–250 Hz with care (OEMs give it only 15–20 W). The apparent high-frequency roll-off is really a low-frequency hump: cutting it, even 12 dB, just saves power and excursion. Sources: Audiofrog blog Jun 2021 · DIYMA 2020-08-25 · CMA Connected Oct 2021 @34:22

In plain English: This uses the rule from point 2.1 #8: about +6 dB each time the space a speaker radiates into is halved. He says a dash cuts that space by 3/4 or more (two or more halvings), and the windshield adds more on top. So a 2.5" (63 mm) dash speaker can play down to 200–250 Hz with care, even though carmakers give it only 15–20 W. On a measurement, a dash mid can look as if its high notes are fading. He says that is really a hump at the low end. Many people boost the "missing" highs. He says to cut the hump instead, even by 12 dB, which only saves amplifier power and cone movement.

2.3 #9 · A small dash mid aimed at a raked windshield is a "poor man's tweeter", which is why the GS25 exists.

[RARE] Its narrow treble beam bounces off the glass to everyone, so beaming stops mattering (not with upright glass, as in a Jeep). GM, Toyota, Chrysler, Bose and Harman use this "middler"; the GS25 drop-in does the job with far more power handling (high-pass 200 Hz or above, 12 dB/oct or steeper). A real tweeter is still better. [Conflict: this uses the windshield reflection his custom door 3-way (2.6 #6) minimizes; it is a factory-location strategy.] Sources: DIYMA 2022-01-03 · CMA Connected Oct 2021 @32:43 · GS25 page

In plain English: "Raked" means sloped back. Higher notes "beam", meaning they come out in a narrow beam rather than spreading, so you only hear them fully if the speaker points at you. He says a small mid in the dash, aimed at a sloped windshield, sends its narrow treble beam onto the glass, which bounces it to everyone in the car, so beaming stops mattering. It doesn't work with upright glass, as in a Jeep. GM, Toyota, Chrysler, Bose and Harman use this small dash speaker, the "middler", and he calls it a "poor man's tweeter". Audiofrog's GS25 drops into that spot and handles far more power. Use a high-pass at 200 Hz or above, at 12 dB per octave or steeper (the slope says how fast lower notes are cut). He adds that a real tweeter is still better. The [Conflict] tag notes the tension in his advice. This trick relies on the windshield reflection, which his custom door 3-way design (point 2.6 #6) tries to keep small. The tag files it as a strategy for factory speaker locations.

2.3 #10 · For a dash mid under a factory grille, put a small tweeter under the same grille rather than adding pods.

[RARE] Clear the plastic with a rotary tool; this works if the mid still responds to EQ up to about 4 kHz. Keep the tweeter base off a 5/8 in gap to the windshield (big dip near 10 kHz), and don't leave a mid 3/4 in behind a grille piece: mount it flush or cut the plastic out and use mesh plus cloth. Sources: DIYMA 2021-09-05

In plain English: Many people add separate tweeter pods on the dash. For a dash mid under the factory grille, he suggests putting a small tweeter under the same grille instead. Clear away the plastic behind the grille with a rotary tool (a small high-speed cutting tool) and mount the tweeter there. This works if the mid still responds to EQ up to about 4 kHz, so the tweeter only has to fill in the very top. Avoid placing the tweeter's base 5/8 in (16 mm) from the windshield: that gap causes a big dip near 10 kHz. And don't leave a mid sitting 3/4 in (19 mm) behind a piece of grille plastic. Either mount it flush, or cut the plastic out and use mesh covered with grille cloth.

2.3 #11 · A top-of-dash center: once a "frequency response nightmare", later the place to be.

Firing at the glass, a dash L/C/R gives "three left speakers, three right speakers and two center channels" and confuses auto time alignment (try reversing the center's polarity); make sure the baffle doesn't leak. By 2021 he advised mounting it as close to the windshield–dash junction as possible: its broad dip near 2 kHz mimics the dip a phantom center has, so it sounds familiar. [Refinement, not reversal: the 2010 "nightmare" remark was about response ("though it does raise the stage"), and the MS-8 FAQ (2011 or earlier) already named a big dash mid firing at the windshield his preferred center ("This works great"; 1.10 #5); 2021 adds placement at the junction and the 2 kHz rationale from Shively and House AES papers.] Sources: DIYMA 2010-08-15 · DIYMA 2021-09-22

In plain English: A center channel is a mid-dash speaker for centered sounds, such as a singer; "L/C/R" means left, center and right. In 2010 he called a top-of-dash center firing at the glass a "frequency response nightmare". With all three dash speakers bouncing off the glass, you get "three left speakers, three right speakers and two center channels", and that confuses automatic time alignment (processors that set delays by themselves). His tip was to try reversing the center's polarity, and to make sure its baffle doesn't leak. By 2021 he advised mounting it as close as possible to where the windshield meets the dash. The reflection there gives a broad dip near 2 kHz. A "phantom center", the center image you hear between two stereo speakers, also has a dip near 2 kHz, so the real center sounds familiar. The tag explains this is a refinement, not a change of mind. The 2010 remark was about frequency response, and even then he said it "does raise the stage". His MS-8 FAQ (2011 or earlier), written about the JBL processor he worked on at Harman, already named a big dash mid firing at the windshield as his preferred center ("This works great"). The 2021 advice adds the exact spot and the 2 kHz reasoning, taken from audio-engineering (AES) papers by Shively and House.

2.3 #12 · Every location has its dip; choose which one you can live with.

[RARE] Mids low in his doors accept a side-glass dip for width; all-in-the-dash loses width and adds a windshield dip; kicks add a midbass dip ("It's always something"). For height, move the mid to the door top; for width plus height, everything in the doors and a center in place of a dash gauge pod. A 300–500 Hz dip is usually the door window; don't hunt endlessly. Sources: DIYMA 2013-08-31 · DIYMA 2021-11-02

In plain English: He says there is no perfect speaker location; each one has its own dip, so choose the one you can live with. His own car has mids low in the doors, which accepts a dip caused by the side glass in exchange for stage width. Putting everything in the dash loses width and adds a windshield dip. Kick panels add a midbass dip. "It's always something." If you want more height, move the mid to the top of the door. For width plus height, put everything in the doors and add a center speaker where a dash gauge pod (a housing for extra dials) would sit. A dip at 300–500 Hz is usually the door window reflecting, and he says not to hunt for it endlessly.

2.3 #13 · Map candidate locations with a speaker at the driver's head and a mic at each spot (acoustic reciprocity).

[RARE] Mic and speaker swap roles where dispersion is wide, so it works for sub, midbass and mid spots but not tweeters. His rig: a GB25 in a 3D-printed pod at head height and a 3D-printed boundary mic at dash top, pillar, kick, door tops and trunk. From 2023 he taught "system design" this way rather than guessing locations. Sources: OFCAP Ep. 38 @36:00 · Absolute Electronix 2023 @1:00:14

In plain English: Acoustic reciprocity is a physics principle: the sound path from point A to point B behaves the same as from B to A, so a speaker and a microphone can swap places. He uses it to test spots before cutting anything. Put a small speaker where the driver's head is and a mic at each place you might mount a speaker, then compare the measurements. It only holds where the speaker spreads sound widely, as a mic hears from all directions. So it works for sub, midbass and mid locations, but not tweeters, which beam. His rig: a GB25 in a 3D-printed pod at head height, and a 3D-printed boundary mic (a mic made to sit flat on a surface) placed on the dash top, pillar, kick panel, door tops and trunk. From 2023 he taught "system design" this way instead of guessing locations.

2.3 #14 · Which Audiofrog set for which location.

GB25 2-way: A-pillars or center; GB15 + GB60: "the primo set" for kick panels (the GB15 is designed for the GB60); GB10 + GB25 + GB60: the ultimate active system for pillars with door or kick midbass; GB10/GB15 + GB40: doors (BMW) or center; GB25 in any 3-way when only a small mid fits; G62 coax at the bottom of the door or in the rear deck. (COMPANY and retailer.) Sources: Crutchfield, Audiofrog packages · GB25 page · GB15 page

In plain English: This is a product guide from Audiofrog and a retailer (Crutchfield), not his signed posts; that is what "(COMPANY and retailer)" means. In Audiofrog's naming, the GB10 and GB15 are tweeters (1" and 1.5"), the GB25 is a 2.5" mid, the GB40 is a 4" mid, the GB60 is a 6" (165 mm) midbass, and the G62 is a coaxial speaker. The pairings: a GB25 two-way for A-pillars or a center. GB15 plus GB60, "the primo set", for kick panels; the GB15 was designed to go with the GB60. GB10 plus GB25 plus GB60 as the ultimate "active" system, where each driver has its own amplifier channel and processor crossover, for pillars with midbass in the doors or kicks. GB10 or GB15 plus GB40 for doors (as in BMWs) or a center. A GB25 in any three-way where only a small mid fits. The G62 coax at the bottom of the door or in the rear deck.

2.3 #15 · His Harman-era Volvo S60 (c. 2008–10) shows the priorities.

[RARE] Dual-coil 6" midbass at the front of the factory lower grille, nothing in the factory 4" spot behind it ("a stupid place for a midrange"), a 3" mid ahead of the door handle, a 1" tweeter in a sail he built, a 6" plus tweeter in the dash center, two 10" IB subs in the deck. He ranked center channel, door-mid location, lots of EQ and center steering highest, and praised cars built for audio (big door openings, trunk battery, deck room). Sources: DIYMA 2010-02-19 · DIYMA 2009-05-27

In plain English: This describes his own car from about 2008–10, when he worked at Harman, the company behind JBL and Infinity. He put dual-coil 6" midbass speakers at the front of the factory lower grille, and nothing in the factory 4" spot behind it ("a stupid place for a midrange"). A 3" mid went ahead of the door handle, and a 1" tweeter in a sail panel he built. In the middle of the dash he put a 6" plus a tweeter as a center channel, and two 10" IB subs went in the rear deck. He ranked the center channel, the door-mid location, lots of EQ and center steering highest. Center steering is processing that sends center sounds to the center speaker. He also praised cars built with audio in mind: big door openings, the battery in the trunk, and room in the rear deck.

2.3 #16 · His Audiofrog demo cars use the same recipe.

Mercedes (CES 2016): identical active GB10 + GB25 + GB60 3-ways in all doors and sides, GB10 + GB40 deck pairs, a machined-aluminum center pod, a custom 24-channel amp. Volvo C30 (2023, 7.1, 18 speakers): door 3-ways, a GB60 + GB10 center so it can play loud (2" coil), two GB10D4 subs in a small sealed box, a head unit that kept the factory HVAC display so the dash stayed uncut, timed turn-on sequencing. Sources: Ricardo Rangel, CES 2016 · OFCAP Ep. 38 @9:02 · Absolute Electronix 2023 @51:00

In plain English: Audiofrog's demonstration cars follow the same plan as his own. The Mercedes, shown at CES 2016 (a large electronics trade show), had identical active three-ways in all the doors and sides: GB10 tweeter, GB25 mid and GB60 midbass, each on its own amplifier channel. It also had GB10 plus GB40 pairs in the rear deck, a center pod machined from aluminum, and a custom 24-channel amplifier. The Volvo C30 (2023) is a 7.1 system (seven main channels plus one bass channel) with 18 speakers. It has three-ways in the doors and a GB60 plus GB10 center, so the center can play loud (the GB60 has a 2" voice coil). Two GB10D4 10" subs sit in a small sealed box. Its head unit (the radio and source unit) kept the factory climate-control (HVAC) display, so the dash stayed uncut. The equipment also switches on in a timed sequence.

The full series

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