Posts run forward in time from this point.


Throughout the life of this blog I will constantly discover new spelling mistakes or find better ways of sharing my thoughts. Try reloading the page occasionally. Or run CCleaner regularly as I do.

The default spelling checker I use here is UK English but is rarely working properly as I rattle away at the keyboard. Only later do all the red lines appear like bright poppies in the fields of text.

If the text seems too small for comfort try holding down the CTRL key and then pressing +. This enlarges the page including text and images and is very useful. CTRL and 0 (zero or nought) will return the text to normal size. CTRL and - will reduce it.

By the way: The dates of all posts are completely fictitious. I deliberately date each new post so that it follows a logical chronological order. When I started blogging it was impossible to run a blog "backwards". i.e. So that new posts appeared at the bottom of the chapter list rather than at the top. Dating posts backwards into the past to achieve my aims has continued.

If you find yourself running out of new chapters to read try clicking on "Home" at the bottom of the page. Clicking on"Older Posts" will usually provide the next chapter or chapters.

I chose the blog format because it is so flexible, provides enlargeable image hosting and can be continuously updated and corrected. It is also free. It's strange editing problems are a trial but beggars can't be choosers.

Blogging seems unlikely to suffer the sudden death common to ISP's free web pages as more avaricious companies buy them up. I lost 5 websites to changes of ISP ownership ending in a final toxic dose of agent Orange.

Although it doesn't seem very likely this blog actually gets more interesting as it goes backwards in time! If you have already dismissed March '07 as out of date then you may like to know that this was written in mid summer '09 just as I doubled the size of my IB to 8 x 15".

If you are regular visitor do remember to reload the pages regularly. I go back and change things all the time. Often adding new images as appropriate.


The IB Array

My first attempt at a True Infinite Baffle Subwoofer proved to be slightly flawed. It performed incredibly well compared with my large commercial SVS cylindrical subwoofer. It was just that the array seemed to lack the giant-killing power that I had read about in posts from other delighted IB subwoofer owners on The Cult of the Infinitely Baffled. (A specialist Internet forum dedicated to True Infinite Baffle Subwoofers)

The major problem was the flexibility of my baffle wall. This old planked-over-vertical-studs wall forms part of an even older timber-framed house with later additions. When the four big 15" driver cones moved ... the wall moved the opposite way! Newton's Law of action = reaction was biting back with a vengeance.

The original array had a remarkable performance but had shaken the wall and adjacent double doors far too much for comfort. In fact the whole house seemed to shake from the mechanical vibration of the drivers. The bass sometimes made me feel nauseous in the bathroom below the IB enclosure! Yet despite the available power and four large drivers the output on programme material was just a little disappointing. It failed to vibrate the floor on films no matter how much I played with BFD filters. Despite an easy 120+dB capability and 8Hz extension on test tones the IB just did not deliver quite as expected. I also had absolutely no headroom to play the subwoofer louder ("hotter") than the speakers on programme material. Something had to be done!


The IB Enclosure.

This is what the 4 x 15" vertical array looked like from the enclosure side.
I think you will agree that it looked very impressive.
The IB enclosure is a large prism, triangular in elevation and has a volume of about 600 cu.ft or 17 cu. meters. This small room was once an open balcony in the gable end of our old house. After the Danish Storm of the Century in late 1999 I had closed off the gable end with large, double-glazed windows to make a pleasant sitting room with a view out over our garden.

My wife suggested this space for the IB enclosure since it avoided clambering into the roof space and cutting holes in the planked ceiling of the AV room. Luckily the space offered a suitable volume and very easy access from the listening room via the original French windows. (glazed double doors)

The basic concept of an IB subwoofer is very simple indeed. The large drivers (loudspeaker units) are fixed into a wall (the baffle) between two adjacent spaces. The listener sits in one space and the IB subwoofer enclosure is the entire space beyond the shared wall. The loft can also be used as the enclosure with the ceiling acting as the necessary baffle. Underfloor IBs are also popular using the listening room floor as the baffle and the crawl space (or cellar) as the enclosure.

The black box on the steel table in the image above is a Behringer Europower EP2500, dual mono, power amplifier. Providing 450 watts RMS per channel into 8 Ohms. (650 Watts RMS into 4 Ohms) Designed for professional musician's instrument amplification it offers remarkable output for mere pocket money in domestic hifi terms. It closely matches the power requirements of my IB subwoofer. Its incredibly noisy original cooling fan has been replaced with a much quieter one! This is a popular and easy modification of these amps for domestic audio use.


Inside the box.

The term "infinite baffle" really means an infinitely large, rigid panel with a loudspeaker drive unit fixed somewhere in the middle. The great size of the baffle forces the sound waves from the driver to to travel a very long way before they reach the edge of the baffle. When the sound waves reach the edge of the baffle they would be of opposite phase which would cause cancellation at a frequency depending on the length of the wave which matched the distance from the driver. The positive pressure wave from the front of the driver cone cancels the negative partial vacuum of the rear wave. Since frequency is directly related to wavelength the baffle must be made very large indeed so that cancellation can only occur at very low frequencies. Hopefully at a frequency below that required to reproduce music and film effects with full power.

In real-world situations it is far easier to use a large, enclosed rear volume as an IB enclosure. The infinitely large baffle then becomes the wall between the listening room and the rear enclosure volume. The edges of the baffle have been effectively wrapped around to seal off the rear space from the AV room. Since the sound waves from the front and rear of the cones cannot interact with each other there is no out of phase cancellation at any frequency. Because the rear volume is so large there is no back pressure on the drivers as there would be in a small sealed box. This lack of back pressure allows the IB to reach very low frequencies indeed at high output levels while maintaining low distortion and very reasonable efficiency. When the drivers are housed in a box manifold sealed into the dividing wall the IB can be thought of as a local folding of the (infinite) baffle.

The listening room itself must be sealed from the IB enclosure to avoid out-of-phase sound wave cancellation. Open doors, serving hatches or windows between the two spaces would cancel the bass at certain frequencies depending on the wavelength at which the sound waves met each other at the opening in the baffle wall. Airtight sealing is not that important though. The large volumes of air movement required at very low frequencies would struggle to squeeze around the narrow gaps of a closed door for example. A perfectionist might like to draught seal the door so they can sleep soundly at night. If nothing else it will help to stop the door from rattling against its frame. You'd be amazed how a door can flex when driven by cyclic low frequency pressure waves.

A view inside the open manifold from the AV room: Click for a larger image. Backclick to return to the text.

The drivers in my own IB are now fixed in pairs on either side of a manifold (box). The box is completely open on only one side to the listening room. The open area (or vent) of the manifold is calculated to match the total cone area of the 4 x 15" drivers (or 4 x Sd) to avoid any compression effects in the manifold. (In my own case the opening is 40" x 13.5")

The driver cones in an IB must all move in the same direction (towards the open side of the manifold ) when a signal is applied. If any cones moved in opposite directions they would cancel each other's output. A torch battery connected briefly to the speaker cables can be used to check that all the cones jump in the correct direction. The cables can then be inserted into the power amplifier speaker terminals observing the correct polarity.

The unusual two-facing-in, two-facing-out arrangement of the drivers in my manifold in this image is supposed to cancel out some distortion components. The drivers which face into the box will not produce exactly the same audio response as the drivers which face out of the box. This imbalance helps to cancel out any variations between them. It is considered a better arrangement than having all the drivers facing inwards or all facing outwards. (Where unwanted driver effects would be additive)

Opposing the drivers on either side of the box cancels mechanical vibration due to cone movements.

Mechanical vibration was the weakness of my original vertical, line array IB. The downside is that a rather large box has to be housed next to the baffle wall. Whereas an array takes up no space in either room. The box can be hidden in the enclosure as I have done. Or it can be placed in the listening room. In which case it can be made to look like a large box subwoofer. This type is usually referred to as an "outie". My manifold could have been made smaller but I wanted to avoid any chance of compression effects in the box by making the opening into the AV room as large as possible. The opening into the room can be covered in breathable cloth without affecting the IB's performance. Some IB builders use decorative screens to cover the manifold opening. Grills are more practical with underfloor IB manifolds where the manifold opening might be walked on accidentally.


The manifold box.

The lack of "killer" bass from my IB required a serious rethink. The array was sitting in a weak wall which would never improve no matter how much I tried to reinforce it by fixing on more 4" x 2" timbers with 6" long screws.

So after much procrastination I finally decided to rebuild my IB as an opposed driver manifold. (box) Just to see if mechanical vibration really was my IB's Achilles heel. The box would be made of 3/4" plywood and I would use a single layer. I would judge how it performed first before beefing up the box with a second layer. I was still full of doubt abut the advantages of a manifold at this point.

The desired manifold size could be cut from a single sheet of 18mm plywood with little or no waste. I carefully selected a nice, flat sheet from the local timber merchant then brought it home in the car trailer. It didn't take long to cut everything to size and rout the mounting holes for the drivers.

Having built my heavy manifold in my garden workshop I suddenly realised that I couldn't get the large box up the narrow stairs to my attic AV room! I should have glued, clamped and screwed the box together right where it was going to be sealed into the baffle wall.

The heavy 48" x 24" x 16" box had to be run across the garden on a sack truck. Then slid up a gently sloping ladder and in through an open window. Straight into the IB enclosure. After a lot of heavy panting (and rest for a cup of coffee) I could push the manifold between the vertical timbers of the stud wall. A piece of advice: Wall studs are rarely straight without some twist or bend. Make your manifold width slightly undersized or be prepared to hack some studding away with a large chisel! This noisy exercise will not endear you to your partner. Who will assume that the confidence you had in your careful calculations was completely unfounded. They will think that the IB will be a total waste of time and money! It will not be, but don't give them any rope to hang you by at this critical stage. Build a slightly undersized manifold and then use wood or plywood packing strips. Or strips of foam plastic pressed well in. These deliberate ruses will impress your better half no end. You will appear to have everything completely under control no matter how poor your woodworking skills.

The maintenance of the balance of power is vital in an HT/AV/Hifi household. Diplomacy before a war of words will avoid expensive concessions which might be much better spent on equipment upgrades. Being allowed to make large holes in the house to install subwoofers goes completely against the nesting instincts so tread very carefully to avoid breaking too many eggs.

After the struggle to get the manifold between the twisted old timbers I had drilled holes for the speaker cables and fitted the drivers. Depending on your driver orientation it can help to fix a small batten to support the driver rim safely while you get the first fixing screw in. Once achieved you can take your time as you go around fitting the rest of the screws in place. Drivers are strangely unbalanced objects and can often benefit from a bottom fixing screw first rather than a seemingly more logical top one. It all depends on whether you fix your drivers from the front or rear of the basket rim and whether you have somebody handy to help you at this stage. I found myself having to climb right into my manifold while my wife gingerly handed me the drivers. A situation of the utmost gravity!

Once safely in place the drivers could be wired up to the amplifier again and double checked for phase. The drivers must all move towards the centre of the manifold or they will cancel each other out. Having managed this error a couple of times when moving the drivers around or rewiring after REW tests I can vouch for the lack of bass if one gets the polarity wrong.

The effect of changing to a manifold from the array was quite simply staggering! I nearly fell over with shock the first time I fired it up on a favourite bit of bassy rock music! I had mistakenly used the same amplifier settings as the baffle array. I suddenly had the bass that an IB subwoofer is famed for. The floor shook like a diesel truck on test tones. (or testosterone if you prefer)

Now I had to reduce the bass considerably to match my floor standing speakers. With the array I was always fighting for more SPLs from the IB to match the speaker output. Now I had far more bass than I could tolerate without quickly losing my nerve. It really was a night and day difference.

More importantly the wall and old French windows no longer vibrated an inch back and forth! The structure around the manifold remained almost completely free of vibration regardless of output level or frequency. The single layer of 3/4" plywood seemed quite stiff enough until I tried a 100dB(C) sinewave @ 10Hz. Even then there wasn't enough flexure to worry me unduly. The bass out in the IB enclosure also seemed to have dropped considerably and was very much less audible in the rooms below.

Just in case there is still any doubt, the manifold in an IB is not like a sealed speaker box. The manifold is certainly not the IB enclosure. The manifold merely holds the speakers rigidly in place. One side of the box is completely open to the AV room. All the drivers cones are exposed on the other side to the large IB enclosure. (Which is about 17 cu. meters or 600 cu feet in my case) The open side of the box behaves like a single, large, compression driver. The air vibrating in the mouth of the manifold produces all the bass in the AV room. Doing so with the combined power of all 4 x 15" loudspeakers simultaneously. A safe minimum number of drivers for an IB which can be exceeded at will.

The large volume of the IB enclosure means that the drivers' natural resonant frequency is not raised by an increased air pressure cushion pressing back on the cones. This allows an IB to perform to much lower frequencies at much higher SPLs than any other conventional subwoofer design. It also uses much less power into the bargain. The IB has the unique and often enormous advantage of not taking up any space (at all) in the AV room. Those who take their music and films seriously are often torn between building a huge subwoofer and their partner's natural reluctance to having such things standing in "their" living rooms. Under such circumstances the IB may be your giant subwoofer get-out clause.


Equalisation and frequency response.

Equalisation of my IB subwoofer response was carried out using the popular DSP1124 Behringer Feedback Destroyer. Popularly known as the BFD. The REW test software was downloaded from The Home Theatre Shack Forum along with a calibration file for my Galaxy 140 SPL meter. The Galaxy was used as the test microphone at the listening position. It has proved in tests to be highly repeatable and accurate down to 5Hz using the matching calibration file. Far better than the commonly used Radioshack SPL meter.

I used a +16dB BFD filter at 20Hz to lift the extreme bass using the maximum filter bandwidth of 120/60. (two octaves) The BFD has a lowest frequency filter option of 20Hz. This heavy boost was necessary because my original drivers had an Fs of 32Hz. Fs is the Thiele-Small parameter for the free air resonance of the driver.

The graph below shows the frequency response of my IB and my Mission 753 Freedom speakers after subwoofer equalisation. The curve has since been slightly improved after repeated test sweeps with the REW software and changing the filters slightly on the BFD parametric equaliser. Note that my IB subwoofer reaches 12Hz at full power before falling away quite rapidly. This extension into the infrasonic is rarely matched by any commercial subwoofer. The bass is deliberately louder (hotter) than the output from the speakers. An 80Hz 24dB/octave active crossover is used between the IB sub and the speakers. (Behringer CX2310)

My Mission 753F speakers are driven by a Naim NAC72-Hicap- NAP 180 stereo system. Surround sound (DTS & Dolby 5.1) for films is provided by an old Yamaha DSP E800 processor/amplifier. The resulting sound quality leaves affordable AV Receivers for dead.

Click on the graph for a larger image.

My 27 feet long attic AV room is very heavily damped acoustically. Because of a central chimney the listening triangle is arranged across only one half the room. The whole area behind the speakers (concealed by a curtain) is used for clothes storage with bookshelves behind which greatly adds to the room's acoustic damping. There is an open stairwell in the floor between the listener and the main speakers! All room surfaces are boarded. All this combines to absorb a very wide range of frequencies totally devoid of echo.

No attempt to fill the obvious troughs in the response by altering phase or crossover frequency has had any real effect so far. The system still sounds wonderful to my ears. The quality, massive weight, lack of colouration and startling reality of the bass is beyond all expectation. Whether thundering on a great organ pipe or the "War of The Worlds" film DVD I could not be happier with the sound quality. There is simply no comparison with any commercial subwoofer I have ever heard. The IB totally outclasses them in every respect.


New AE IB15 drivers

Acoustic Elegance kindly sent me some new drivers to replace my originals which had the wrong specs. John, the very knowledgeable owner of AE had suffered some personal tragedies over the last couple of years. This had affected his control over the company's products. He has now restarted his business on a more secure financial footing with far better quality control and redesigned products

These new AE IB15 drivers are currently available in sets of four for $400US plus freight. Four is probably a safe minimum with these 15" drivers in an IB. I found that muting any two of my four drivers makes a serious difference in output. That means more excursion is required to handle peaks. More excursion may be beyond the limits of only two driver's in an IB.

Warning! Click this image at your peril! The enlarged image is 400kB so don't attempt it on a slow modem! I posted this large image to show the sheer quality of these AE drivers. Smaller images just don't seem to do them justice.

The new drivers have a lower Fs of 16Hz which means they should produce more deep bass straight out of the box. I had been using maximum boost of +16dB on the BFD at 20Hz on my old drivers which had a measured Fs of 32Hz.

The graph below clearly confirms the massive improvement. The blue trace represents just two new drivers in the manifold. The other curves show the old drivers without bass boost in pairs and all four. Levels have been matched to make comparison easier. The 11-12Hz peak is a room boundary effect common to both old a new drivers. As it is free bass and not requiring extra power from the amplifier, or extra cone excursion, I think I can safely ignore it. Or even consider it a bonus. There aren't many subwoofers which can claim 11Hz at well over 100dB(C-weighting).

I had no problem getting 100dB(C) at 10Hz [uncorrected] on my Galaxy 140 SPL meter except for the physical discomfort. Though completely inaudible the feeling of pressure in the room is intense. The Galaxy cal. file adds nearly 11dBs of correction to the REW response at 10Hz. This suggests that my IB is producing a true 110dB(C) (and over) at 10Hz without reaching the driver excursion limits. The room gain factor at 11Hz is probably helping here though I have no idea where this gain is coming from. There seems to be no linear dimension between boundaries which would have this free and beneficial effect unless the sound waves are folding at the stairwell. It is also possible that the room below is acting as a Helmholtz resonator with the open stairwell as its port. The IB manifold opening is quite close to the stairwell and may be driving the room below like a giant reflex loudspeaker enclosure.

The new drivers show obvious signs of much greater care in manufacturing and have had a complete facelift. The cones are now made of reinforced paper instead of the original poly material. The cones are now heavily flared compared with the straight-sided cones of the earlier models. The basket has been given a black finish with bright edges to the legs of the basket. The spider is also black instead of yellow. The earlier screw terminals have now become a sprung push-release type. Much easier to use, perhaps, but I would have liked much more room in the cross-hole for thicker cables. I'm using original 79 strand to connect the EP2500 amps to the drivers. The push terminals will not quite accept doubled wires no matter how tightly they are twisted together. The magnet backplates now have rounded shoulders and a smoothly flared pole vent. The quality of the machining is very much higher than on my old drivers.

At first I was running the new drivers in series pairs per channel because that was how the old drivers were wired. The new drivers have an 8 Ohms impedance which meant the amplifier was seeing 16 Ohms per channel. Despite hitting over 112dB on WOTW "machine rising" I felt the new drivers sounded a little quieter than the old ones. The old drivers were 4 Ohms each in series pairs for 8 Ohms and 450 Watts per channel. I have now rewired the new drivers in parallel pairs for 4 Ohms and 650Watts per channel or 325 watts per driver. This has increased output nicely over the former series pairs arrangement. Output now matches the older driver with plenty of weight on my favourite classical organ CDs and the bass-heavy pop and rock numbers I use to confirm bass quality and quantity.

This is an REW frequency response trace measured at the listening position using the Galaxy 140 SPL meter as test microphone with its matching calibration correction file loaded.

Here my BFD has been reprogrammed to match the new driver's output in the very deep bass. The trough between 12Hz and 25Hz has been filled with a small but wide boost filter at 20Hz. Further small filters at higher frequencies have smoothed the response nicely. The constantly rising slope to 12Hz could hardly be called a house curve. More of a ramp. This final curve seems to suit my ageing ears and my room. The image above is a subwoofer only response graph from REW. [Room Eq Wizard] This test software is available as a free download to members of Home Theatre Shack forums. John Mulcahy, the brilliant and generous software author, is usually able to answer specific questions about the software itself. General questions about its use are usually fielded by the very knowledgeable moderators of the specialist REW forum at the HT Shack.

Below are the manufacturer's T/S Parameters for the new AE IB15. Based on my own experience with these new IB drivers I heartily recommend them in sets of four for IB use. The sheer quality and quantity of bass they provide in a manifold IB is staggering using an EP2500 amplifier and 80Hz crossover. Your main speakers are much more likely to be the limiting factor in any sound reproduction system involving an IB. Given the amount of crossover overlap at high listening levels, even with a 24dB/octave crossover, requires a good quality floor stander with 110dB capability and at least 40Hz extension.

Thiele-Small parameters of the all new Acoustic Elegance IB15.

Fs: 16Hz
Qms: 6.8
Vas: 439L
Cms: .45mm/N
Mms: 220g
Rms: 3.239
Xmax: 18.5mm
Xmech: 25mm
Sd: 830sqcm
Vd: 3.07L (p-p)
Qes: .78
Re: 5.5ohm
Le: .33mH
Bl: 12.49
Pe: 500W
Qts: .7
1WSPL: 86dB
2.83V: 87.3dB

Here is one of the best online lists of Thiele-Small parameters and their meanings.

Car Audio: Thiele-Small Parameters

Click on any image for a larger version. Back click to return to the blog.


IB performance on organ music.

I have been listening to my favourite classical organ CDs all morning to assess the new drivers.

These include: Amongst many others of:

"Bombarde!" French organ classics: Liverpool Cathedral: Ian Tracey. Chandos CHAN 9716

"For Weddings" Chichester Cathedral : Kevin Bowyer. Nimbus NI712

"Tsar of Instruments" Organ music from Russia: Winchester Cathedral: Iain Quinn. Chandos: CHAN 10043

Cesar Franck's Complete Organ Works , Jean Guillou, organ at St.Eastache, Paris. Brilliant Classics 92282. On 2 Cds this 1989 recording is simply stunning for its clarity and capture of the building's reverberation at all frequencies. The bass quality and weight is phenomenal with the timbre of pedal stops and pipes laid completely bare by a good subwoofer. The inner detail of the big pipes and their complex interactions is entrancing.

The deep tones from the great pipes sound suitably heavy and effortless with fine rendering of timbre. The way the pipes stop and start is vital to enjoyment in my experience. Some pipes have a slow way of breathing themselves gently into audibility. Others chuff or start and stop abruptly. Some change pitch as they play their tone. All have different tonal qualities and this should be heard without effort. It is never enough to just reproduce low frequencies. The reproduction must be done with low distortion and great subtlety of tone, timbre and dynamics. It is no use just filling the room with loud sinewaves. (or square waves for that matter) That is not what organ music is all about.

The tones one hears must be nimble to match the footwork of the organist on the pedals. Most cathedral organs have various ranks of great pipes which can be played alone or in unison. The 16 foot pipes sometimes dance on a foundation of 32 foot stops. These huge pipes can produce inaudible tones which modulate the sounds of the shorter pipes. Sometimes the great pipes beat together producing a "tremolo" or "warble" effect with varying frequency. Sometimes the beating has clear gaps between the beats. Which themselves are constantly changing in speed and strength. I have heard beating as slow as a half second cycle. Putting one's head in the manifold brings out even more detail suggesting that there is much more to be had if only it could be brought out into the room. The danger is in overemphasis of the bass by playing the IB too hot. Very impressive but a little unrealistic.

All this detail must be reproduced for the listener's ears without artefact or distortion and without making him or her work hard to hear it all. The only way to do this well seems to involve many large drivers working in unison in an IB.

A single 12" driver can reproduce low frequencies in large quantities particularly if placed in a large, ported enclosure. Though it does so with too many false harmonics. Which often mask the fundamental with unwanted distortion. All detail is lost to a muddy rendition of loud sound waves which have little to do with the original musical performance. It could be described as a black and white cartoon rendition of the music. Bold, but completely lacking in the fine detail which we take completely for granted in the cathedral, hall or church when listening to live organ music.

The IB subwoofer fills in the details like a colour cine film. Everything is constantly moving, changing in tone, colour and contrast. One can hear the great pipes start, sustain and decay. The IB adds huge weight to the sound no matter how briefly a pipe may speak. The incredibly deep tones dance nimbly on the jumbled infrasonic ice field hidden just beneath the more audible surface.

Here's an image of the new drivers installed in the same IB manifold. The drivers have been brought magnet to magnet inside the manifold to avoid the metalwork "sweating" in cold Danish winters. My IB enclosure is unheated, though frost free and subject to rapid temperature changes in afternoon sunshine. A perfect solar oven were it not for the white curtains drawn across the glass gable end. Endlessly delayed plans to fit exterior louvres to block summer sunshine have never matured into reality.

Warning! : Large 300KB image if you click on the above: Not for slow modems.

It has been pointed out that my T-nuts are not properly bedded into the plywood to lock them firmly into place. This is because I originally planned to line the box with OSB or double the thickness of the plywood. As I may have to replace the original manifold I haven't bothered to secure the T-nuts properly yet.

The next plan is to build another manifold to stack on top of the existing one. This new box will support the four original drivers. Bringing the tally up to eight fifteen inch drivers. The Total Vas ratio to enclosure volume will fall to around 4.5:1. It will be interesting to hear how this arrangement performs when driven by the same amplifier. I may need to examine other installation ideas if it fails to please. These plans have been put on hold while I examine further alternatives for the four old drivers. I might build an open backed baffle over on the right hand side of the main speakers. The idea is to boost upper bass/midrange to allow higher levels and dynamic range. With the present performance level of the IB I don't need to do anything in reality but would like to use these older drivers for something useful.

I am examining ways to avoid the magnets coming so close together. I may stagger the drivers relative to each other in a new box manifold. To make the magnets alternate from one side to the other. This will make cross bracing between the drivers difficult or impossible though. Not that I've ever noticed much flexure in the manifold sides so far despite the use of only single layer of 3/4" (18mm) plywood.

It would not be too difficult to arrange something to go between the magnets to stiffen the structure further. Any inter-magnet bracing would need to allow the pole vents to function normally. It will have to wait for warmer weather before I am really tempted to start making more sawdust. In the meantime I shall be considering different ways of doing things. I am an inveterate modifier and a martyr to perfectionism.

There is little doubt that, should they ever suffer the misfortune of my arrival in any sort of heaven, I shall be offering ideas for improvement from day one. True perfection would be intolerable for me to enjoy. Others sometimes think I am being negative or criticising when the complete opposite is true. I automatically seek alternative and possibly better ways of doing things. This is the habit of a lifetime. Ideas are simply thrown up in the air. To be shot down or caught and given a bit more spin before being lofted again.

If I should ever earn a proper gravestone I shall have a very simple message carved there:


And carved just below: "No tools to be left near this grave!"


More thinking aloud

If the infrasonics really are the most important part of the IB's performance then adding a high pass filter at (say) 25Hz should cripple the IB's advantage over box subs. In fact when I engage the 25Hz filters on my CX2310 crossover there is only a very slight loss of impact on low frequency, electronic programme material.

I tried Bass Outlaws: Illegal Bass: "Extreme Woofer Test" at a steady 95 and then at 100dB(C) read straight off the Galaxy 140 SPL meter at the listening position. The non-filtered output was only a little more full with a few more rattles from the room's surfaces. So much for my essential IB infrasonics theory! There was zero visible cone movement with the 100dB "Woofer Test" yet the whole room was shaking and most of the house along with it.

My guess is that the IB's SQ advantage lies with very low distortion and remarkable transient performance. The large cone area is able to respond instantly to programme input due to the very small linear cone movements usually required. The individual cones each have adequate motor power to closely follow whatever is called for with minimum air disturbance.

A single heavy cone in a sealed box is constantly fighting against the resistance of the air in the box and/or the inertia of that in the port. The single cone must move many times as far as the multiple IB cones to achieve the same output. The single cone is under constant heavy acceleration which must make life difficult if it wants to add smaller nuances on top of the major acoustic output frequency. So detail is inevitably lost unless serious cone breakup comes into play. At low frequencies it is difficult to imagine the amount of breakup required to achieve simultaneous low frequency sine waves at different frequencies. The air movement required for each frequency is considerable at high levels. It is no wonder the multi-driver IB is so capable since it can move so much air with no visible come movement whatsoever.

Even a low frequency sinewave at higher levels is demanding in itself. The cone must move to the maximum excursion required for the acoustic output demanded. It must do so in both directions without undershoot or overshoot. How closely the cone can follow the signal is vital to the waveform reproduced. Would you trust multiple, large driver cones (each with their own motor) moving a tiny fraction of an inch? Or a single cone moving well over an inch in each direction and trying to follow and reproduce a number of perfect waveforms all at the same time at various sound pressure levels? (SPLs)

It is interesting that some authorities deny that subwoofers have variable speed depending on the design. They suggest that the rise time for the signal is far too slow for a subwoofer to know the difference. They obviously haven't felt the sharp edge of an IB exciting the floor through acoustic pressure alone on an LFE pulse.

Doppler distortion would seem very likely with a single long-throw cone in comparison with multiple cones moving over tiny distances in unison.

Mechanical noise and turbulence must be higher in a long-throw "air beater". Imagine a small fan whizzing round to move as much air as a large fan moving much more slowly. You can't even hear the large fan as it effortlessly moves hundreds of cubic meters of air per minute. The small fan quickly drives you out of the room with its racket. Or you try to turn down the speed to get some peace. Only to lose the required air movement to keep you cool. So what do you do when you get fed up with fan noise? You put up a huge, slow, ceiling fan for a bit of quietness. The large ceiling fan doesn't even need carefully shaped blades to achieve its low noise cooling effect. It can make do with flat plates arranged at a suitable angle of attack. No wonder a multi-driver IB wins over a single cone.

Ignoring displacement, the question is whether matching cone area by purchasing a stack of large box subs adds up to an IB on sound quality. My immediate thoughts would be that any colouration would be additive with the boxes. The cost of going this route would of course be many times that of the IB with similar cone area. The power requirements of multiple box subs would need a special cable to be run. All of these potential gremlins may be subjectively inaudible of course. So it is just as well this is just a thought exercise and none of my own money changes hands.

If reality follows my musings then the Thigpen fan subwoofer should have a much larger but much slower moving fan. To kill the unwanted noise. The fan blades of this clever subwoofer must deliberately change their angle of attack with frequency. But each angle of attack must require a completely different aerofoil section to avoid stall and/or turbulence. Carefully shaped cowl rings on both sides of the baffle in which the fan rotates would no doubt help reduce noise. Even if I had a fan sub to play with my home is just too structurally noisy to tolerate such serious output at deep infrasonics frequencies. Just playing "The Ultimate Woofer Test" at serious levels already finds every rattle and point of flexure. This strange track has vastly more serious LF content than any other track I have found so far. It sounds completely different and almost innocuous through headphones. No doubt it sounds very different through box subs and car subs too. Played at 110dB through my IB it has the impact of a wrecking ball! Rather like the server room scene in the film "Pulse" but considerably more musical.


Subjective performance

I have been watching a lot of films and listening to lots of music since I fitted the new drivers.

The new drivers have greatly improved the performance at the bottom end. The IB is also incredibly quick on LFE impacts. I watched Elektra for the first time and the shocks through the floor were almost painful to my feet. Brutal hammer blows with zero overhang! Since my IB is built into the wall this is purely an acoustic effect of pressure waves acting on the floor. The manifold itself does not move physically to driver reaction forces so the floor is not receiving any vibration by mechanical conduction from that source. Nor is the floor well connected to the IB enclosure.

The much smaller BFD filters I need now are not sapping the power as much either. I have gained about 8dB over my old drivers which had used massive bass boost at 20Hz. (+16dB) The new drivers are also seeing more power because they are in parallel pairs for 4 Ohms impedance per channel instead of series pairs per channel for 8 Ohms. This has offered a very worthwhile increase in output. On a couple of films with really "hot" bass levels I have even backed off the bass by a further 6dB to increase comfort levels and headroom. Relative bass levels are very much a matter of taste. Which depends a great deal on programme material and one's mood.

I love the way the IB locks onto the bass drum and bass guitar on rock music. There is absolutely no sense of the subwoofer being off-centre. Listening to Satriani CDs at a steady 100dB(C) is certainly entertaining these days. The clarity and impact on drums is excellent. The IB cones aren't visibly moving at these levels. Perhaps a fraction of a millimetre. It can be felt with a fingertip on the cone surround but it is not usually visible even watching from close to.

Only film LFE provides the levels and frequencies which get the cones moving and then only rarely to a serious degree.

An update:

Well, I seem to have found my new drivers' absolute limits. The four AEIB15s seem to be capable of nearly 120dB(C) (Galaxy 140 SPL meter uncorrected) measured at 8 feet from the manifold. Watching the Fantastic 4 DVD I heard a couple of deep knocking sounds when the heat seeking missile exploded on the river barge. When I checked the levels again (after reducing the subwoofer gain by 6dB for safety) I hit 112.8dB(B) on Max hold/Fast on the SPL meter during this scene. Adding the 6dB back on, there is no doubt that the IB subwoofer was giving out well over 120dB(C) if the REW SPL meter correction factors are also added in. Until then the whole room had really been shaking. There were times when the floor went completely soft beneath my feet and the air itself could be felt shaking vigorously.

I approached the manifold at one point in the film with a torch to check the drivers' excursion and found I could hardly catch my breath during LFE rumbles. The acoustic vibration at the manifold opening was incredibly fierce yet the drivers were hardly moving. This reminds me of a kind of forced air heater used to warm or dry out large halls and buildings after a flood. These industrial, forced air fan heaters produce vast quantities of VLF and infrasonics. The effect of standing near the manifold is rather similar. In fact an acoustic furnace is quite a good description of the manifold during action films.

Occasionally I notice some strange transient cone movements on films which produce well over 1/2" (12mm) of excursion but nothing audible is heard. There seems to be nothing on-screen to suggest any reason for these rather odd transients. The speaker cones shoot out simultaneously and then return slightly more slowly.

Needless to say I am absolutely delighted with the performance of my IB despite the slight possibility of having bottomed the drivers. Though it may well have been a structural rattle. The bass levels were truly shocking until that point and I see no reason (whatsoever) to criticize this level of performance.

Since I seem to have found the safe limits of my IB I ought to press ahead with the second manifold to gain some extra VLF headroom. The LFE effects I like. Risking these amazing AE IB15 drivers is not my purpose. John E Janowitz the designer and builder of the AE speakers claims that these drivers can't be physically bottomed so the knocking sounds I heard may be sympathetic structural movements. The double doors out to the IB enclosure are the most likely culprits here. It is amazing to see glazed wooden doors bend back and forth by an inch as they used to do with my array. Leaning my full weight on them and pressing hard with my shoulder or hands had no effect at all on this flexure. It may be that the doors still flex to sudden acoustic pressure waves on film LFE transients.


The Electronics.

Here are illustrations of some of the kit in my system.

The Behringer CX2310 active crossover sitting on top of the "infamous" DSP 1124 BFD equaliser. The BFD [Behringer Feedback Destroyer] is used by many home AV enthusiasts worldwide to improve the performance of their subwoofers by adjusting the in-room frequency response. Behringer was so uninterested in this vast customer base that they discontinued it! Then resurrected it again. One popular HT forum is trying to get Behringer to offer a specialist equaliser for the home subwoofer market. Don't hold your breath...

The image was cropped from my own rather poor photograph of the entire rack using flash. The Behringer CX2310 24/dB/octave active crossover is set at around 80Hz to split the analogue stereo signal to the Main/Stereo speaker channels. The High channel goes to the Naim power amp and the main/stereo speakers. The Low channel goes to the EP2500 and the IB subwoofer. It is difficult to tell where 80Hz is exactly on the marked frequency scale. Note the rack mounting ears and the total lack of rubber feet for domestic use. Removing the ears leaves open holes in the case which could expose high voltages to probing children and idiot adults.

A view inside my Naim NAC72 preamp. Though not obvious in this image the sub circuits (or cards) stand up on edge towards the viewer.

The view inside the Naim NAP180 stereo power amplifier which drives my Mission 753 Freedom floor standing loudspeakers. 90 wpc RMS into 4 Ohms is claimed. Neat workmanship with the large smoothing capacitors hidden away under the PCB. I find my Naim amps offer a rather hard sound on massed orchestral strings. Perhaps this is just transparency to the CD63SE source?

Inside the Naim Hicap power supply. It weighs and costs a ton! It replaces the internal power supply offered by some (but not all) of the Naim power amps. Naim claims improved sound quality from driving their pre-amps with plenty of clean power. Most owners seem to agree.

The view inside the rather untidy Behringer EP2500 Europower amplifier. Note that the cooling fan shroud has been reversed to show where the fan is situated inside the backplate. The shroud should form a continuation of the heat sink. [EDIT] As supplied, it pushes cooling air through the tubular heat sink and out through the front panel slots. Many owners prefer to replace the 24Volt fan with a much quieter model. Some like to reverse the airflow to pull it through the heat sink from the front panel. [/EDIT] I have never noticed more than a couple of degrees above ambient temperature whatever the film or musical punishment is thrown at this amp. 110dB(C) continuous for an hour on bassy rock music and car audio "tuning" tracks had little effect on temperature.

The manufacturer claims 650 WPC RMS into 4 Ohms. This inexpensive pro-power amplifier has proved very reliable for IB use provided it isn't given foolishly low impedances to work with. 4 Ohms per channel or 8 Ohms bridged is a sensible load for any amplifier. Not just the Behringers. Competitors against the EP2500 are considerably more expensive for the same power output. The slightly cheaper EP1500 offers lower power for little saving in money except for those on a very tight budget.

My tall rack is always dimly lit as far as natural light is concerned. So has to be photographed with flash. Unless it is photographed at an angle the fascias reflect the flash back to the camera. Making it difficult to capture without distortion. There is a remarkable amount of barrel distortion in the heavily cropped image below.

Click on any image for a larger view. Back click to return to the blog.


An organ recital


Nearly ten years have passed since I last enjoyed live organ music so I thought I'd better remind myself of the real thing. An online search provided an opportunity to enjoy a half hour recital. It was mid afternoon only a couple of days later in a large church in the city. Having a free day I took along my Galaxy 140 SPL meter (and my wife) and placed myself half way along the body of the church. I read through the programme and details of the organ on the leaflets kindly provided as I waited for others to arrive. It seemed very odd to turn my back on the organ. So I reversed and sat on one of the chairs provided at the end of each pew.

Though certainly not of cathedral proportions the beautiful organ sported several 16' stops. Right on time the organist launched straight in and averaged 80dB(C) throughout. Only on the quietest moments did the sound level drop down to 70dB(C). Background noise level was 45dB(C) with occasional 50dB(C) peaks when a member of the tiny audience coughed. I believe we were only five in total which really did not do justice to the excellent playing! This Danish organ was quite unlike anything I'd ever heard in the UK. Very much brighter and considerably more nasal at times than I remembered. The Walther variations provided an amazing range of stops with many different voices on display.

My IB is certainly capable of playing much louder and deeper than this particular organ on the programme material we enjoyed today. The IB's detail and weight in the bass was easily on a par with this real organ. Very realistic indeed!

What I had not expected were the high average levels of the performance covering 4 composers and as many superbly contrasting works. Though I only saw a few maximum peaks at around 86dB(C) I now realise that I listen at home with the bass rather hotter than a real organ but with my average listening levels a few dB lower than reality. This makes for a much more comfortable experience. Particularly when spread over several hours rather than a short recital. Now I'd really like to find a cathedral organ with 32' stops to compare with my IB.


Another day, another organ.

Haderslev Cathedral main organ. My apologies for the monochrome image. I was still finding my way around a new compact digital camera.

My search for the authentic organ sound to compare with my IB brought me to Haderslev Cathedral in southern Jutland for a performance of Messiaen's "L'Ascension". Vibeke Astner was playing the organ for a moving rendition of a work which I had never heard before. I tend to shy away from 20th century composers just as much as I avoid the works of 20th century artists. The audience must have numbered perhaps 40 by the time the last, lost, elderly soul wandered in towards the end of the half hour recital.

The music was certainly different. I read the supplied leaflet to discover that Messiaen was something of a genius. He invented his own scales and sought inspiration in the music of other continents. He was particularly interested in birdsong and would use his own notation to describe the sounds they made.

I thoroughly enjoyed the piece from start to finish. The playing was flawless and beautifully expressive. The full power of the organ had my solid oak pew vibrating against my back. Even my chest vibrated to the pedal organ at one point. Like the last, this organ has only 16' stops. I had rather fixed the idea in my mind that it took a giant organ with a 32 foot stop to produce the really heavy effects. However after reading a fascinating organ website I discovered the effects of playing pipes a fifth apart produced a new tone an octave below the lowest pipe. This may well explain the results of running organ tracks through SpectraPlus software where one can often see lots of 8Hz content. There are very few real 64 foot pipes or even folded 32 footers but there are always subharmonics. One odd effect I noted was a slow, half second beating effect on the deep tones occasionally. Whether the organist was producing this deliberately or it was a natural acoustic effect of combining tones I wasn't sure. It is certainly something I have heard on some of my English cathedral organ CDs.

From this same website I even found that clever organ designers had really invented the Tricolumn, a folded horn in a simple pipe form full range speaker. A design I had enlarged considerably to make a subwoofer to listen to organ LPs in the late 1960s. I was using a Mullard 510 valve amp back then which produced about as many useful watts as a birthday cake candle to drive my 10" Whitely, cambric-coned, loudspeaker unit. Cambric was a fine, light cotton material which pre-dated Kevlar and carbon fibre reinforcement material for paper loudspeaker cones by half a century or more.

But I digress. The more organs I hear the more I am delighted with the bass from my IB. Both have much the same sound quality, the power, the breath and the exquisite detail as each other. There is still a brightness and airiness missing from my speaker's attempts to reproduce the smaller pipes. The reverberation and "airiness" within the vast spaces of the buildings which house the organs is missing to some degree. I was never fond of bright speakers so must learn to live with this for the moment.

I was able to buy a Vibeke Astner CD of Messiaen's "Les Corps Glorieux" "Sept Visions brèves de laVie des Ressuscités" on the Helikon label after the recital . [HCD1056]

The sound quality on this CD is simply stunning. Absolutely state of the art. The 7th track: "Le Mystère de la saint Trinité" is the last of these mostly rather quiet, meditative pieces. Here the upper voices are underpinned by the most amazing, deep, quiet bass tones one is ever likely to enjoy. The clarity is extraordinary and makes the CD worth obtaining simply for this gorgeous, reference-quality track. The other tracks are all beautiful too in Messiaen's own rather individual style. Helikon is a small Danish record label.

Another organ CD I must mention is Felix Hell's "Organ Sensation". The sheer weight of bass on this stunning CD is tremendous! Certain tracks found the resonant frequencies of my walls and floor and even shook the kitchen ceiling downstairs. There is considerable infrasonic content to be found on this superb album with (probably) very few subwoofers other than an IB able to make really good use of it. Felix has an excellent website which is a must for the pipe organ enthusiast. His videos are a unique opportunity to watch an organist actually playing a large, fully-exposed console. Normally the organist is hidden away. Often high up on a screened balcony. Where one might only see the top of someone's head swaying along to the music.

Click on any image for a larger version. Back click to return to the blog.


Revelation, at long last.

Another trip in the car and another organ. This time it was to Svendborg in southeast Fyn where a young organist showed off his talents at St. Nikolai's Church. It was immediately apparent that this organ had plenty of seriously deep bass compared with the rather more subtle organs I had recently enjoyed. Three 16' stops were listed this time. The audience was far larger than previous recitals and provided suitable applause to encourage the clever young man at the console. The 4 + 1 pieces played within the half hour recital produced a feast of exhilaratingly deep bass tones. Listening to the descending notes as they dug ever deeper was a revelation to my ears. I had waited for so many decades to hear such stunning clarity, weight, depth and sense of limitless power. Finally, I had a reference with which to compare my IB subwoofer back at home.

I knew immediately that had I allowed myself to enjoy live organ music on a regular basis that my speaker system would have left me deeply disappointed for every moment of those 40 long years. Until, that is, the last moment before I had fired up my IB manifold. My pair of DIY 6th order series subwoofer could do wonders but were just too muffled and coloured in retrospect. The big SVS cylinder had plenty of output down low but was just too muddy on organ music. The IB had been the breakthrough in bass quality which I had been seeking since the late 1960s. Until now I had nothing with which to compare it.

My IB speaks the language of the organ as far as my ears are concerned. It can chuff or breathe itself slowly into life. It can bellow with raw power and still whisper the subtle truth in the next moment of inky black silence. It provides the same immense weight and grandeur as suddenly and effortlessly as a real organ. Easily conveying the power of the great pipes which congregations and audiences have held in awe for hundreds of years. I am beginning to realise that I need new, more revealing main speakers to capture the clear, sharp tones and airiness of the smaller pipes in these vast spaces.

Click on any image for a larger version. Back click to return to the text.


Brace yourselves...

Seeing the massively thick IB manifolds illustrated on the IB Cult forum has finally nudged me into action. I bored two holes and fitted a piece of studding as a cross brace between the two sides of my 3/4" plywood manifold. The studding (or screwed rod) is 1/2" in diameter. (12mm) With very large, square, thick, galvanised washers (inside and out) to spread the tension and compression loads from the driver reaction forces in the side panels. I had carefully chosen a brace position which lay midway between the drivers and the open side of the box.

I carefully tightened the nuts hard without distorting the box and put on a favorite organ CD to judge the effect. The improvement seemed well worthwhile on the subjective level. With plenty of detail in the deep bass beautifully exposed. However, a quick check with REW produced zero difference between the nearfield response plots whether the nuts were tightened hard or completely loosened. Absolutely no change could be seen between output levels or response curves. Well, at least I made the effort and it took only a few minutes to gather the tools and materials to make the modification.

A standard of 2 x 18mm (2 x 3/4") layers for a laminated box manifold is still probably a very good idea for most IB installations. Ensuring a solid but very heavy box well able to resist reaction forces between the opposed drivers. Giving them the best chance of moving air rather than "flopping about" uselessly. (like my array)

My own manifold used an entire (metric) 8' x 4' sheet of 18mm (3/4") high quality plywood with very little left over. This gives a pretty good idea what the bare manifold weighs. Add the weight of another sheet to bring the manifold up to proper 1.5" thickness. Then add the weight of the four drivers and one should plan an IB installation very carefully. Fortunately my manifold sits on the floor and has easy access both inside and out. In a ceiling installation one should make certain that the joists can support the load without sagging.

When I think about it; the back, top and bottom of my manifold add plenty of stiffness on three sides of the driver panels. The open sides of the manifold are also being reinforced by the wall studs between which the manifold fits. The surface covering on the adjoining walls adds further resistance to bowing at the manifold mouth provided the manifold is well screwed to the studs.

If one doesn't like the look of a naked threaded cross brace then a pipe or even a length of opaque hose could be cut to fit between the inner nuts. Then slid over the screwed rod as it is assembled in the box. Perhaps a plywood "window" brace might be better. Though if the brace is glued in the drivers become more inaccessible. It is usually so dark in my own manifold that appearance hardly matters and the performance is still superb. After a few days there still seems to be greater weight and detail in the bass I'm still not sure whether it is a purely psychological improvement, or not.

Here's the simple studding brace in place showing the big, load-spreading washers to resist crushing forces on the plywood.

Having had time to think about it I now realise that REW is a rising sinewave sweep at a fixed level. The reaction forces of the drivers when subjected to sudden acceleration from severe transients is hardly the same thing. So it is hardly suprising that I saw no difference in the REW response plots.

I really ought to know better and should build a more solid manifold because I so enjoy the subtleties and microdynamics in the bass. Anything which helps to resist driver reaction forces should aid sound quality at all levels of output. However, I am still uncertain how to proceed beyond this point with regards to adding more drivers for greater dynamic headroom. So the manifold remains at 3/4" (18mm) for the moment. If I beef up the box with another layer of plywood or MDF I might want to scrap it soon afterwards to build something else.

All images may be clicked for a larger version. Back click to return to the text.


More measuring.

Last time I found some rather odd variations between my pair of RS SPL meters and my Galaxy 140. I wanted some new nearfield comparisons so took this opportunity to do a completely new set of REW measurements. I also included my new ECM8000 microphone and Xenyx 802 mic preamp in the tests. It soon became clear that the REW Cal file for my two RS meters was seriously out of step with the Galaxy and ECM8000. So I deleted the 33-2050 cal. file and downloaded the 33-2055-4050 instead. The latter file gave curves much closer to the Galaxy and Behringer microphone. Now I could progress with my testing.

All devices were tested twice without being touched to exclude possible environmental variations. (wind, distant traffic, people moving around in the house, doors opening and closing, etc)

The red trace below is the ECM8000 and those above this line are the Galaxy and then the two RS meters respectively. All traces have the BFD out of circuit. All devices used the matching REW calibration file freshly downloaded from the HT Shack forum. The CX2310 active crossover remained in circuit with a 24dB/octave cut at around 70-80Hz. It is difficult to read 80Hz accurately on the frequency setting dial. The speakers were muted in all cases at the crossover. One of the advantages of having all those push buttons.

These nearfield curves are all quite smooth and each pair of tests per device match each other closely enough to give me confidence that there is nothing untoward in my test methods this time.

A cynic might argue that the ECM8000 is reading low in the deep bass since it does not match the three other devices which all match each other closely. However, the high cost of professional microphone calibration will not have me following that route in a hurry. For my subwoofer testing purposes they all do well enough. The Behringer mic and Galaxy SPL meter are better for fullrange speaker testing at higher frequencies. These two are also better at reading deep infrasonic output accurately since the correction factors are very much smaller than the cheaper SPL meters and the cal file extends down to 5Hz in the case of the Galaxy 140.

It should be noted that some of my other graphs show much closer correlation between all these measuring devices. I have convinced myself that any of these four devices is suitable for measuring most domestic subwoofers. Only when speakers or IBs are to be measured should the Galaxy 140 or the ECM8000/Xenyx combination be seriously considered.

All the devices were laid horizontally and individually on a cushion on the floor at the mouth of my IB for these tests. The small variations seen could well be down to slight changes in position or minor resonance in the bodies of the SPL meters. The very solid, metal-bodied ECM8000 is probably immune to this but compressed the cushion much more than the other devices because of its comparatively small cross section and considerable weight.

From Left to Right: RS 33-2050, 33-4050, Galaxy 140 SPL meter and ECM8000 calibration microphone. The ECM8000 needs a mic preamp with phantom power adding considerably to the cost of this inexpensive microphone. I believe some computer sound cards offer phantom power so that may be worth exploring.

I chose to buy a Behringer Xenyx 802 to go with the microphone because this was the unit suggested by the HT Shack experts. This offered a proven device with settings already worked out for the new user. Here I have plugged the microphone into the input XLR socket. Normally an XLR cable will fit in here and the microphone would be mounted on a tripod for testing. Note the positions of the control knobs on the Xenyx for REW testing. The HT Shack has a better image for confirming these settings.

The background noise level reading on the Galaxy in this shot is due to the proximity of my computer and the C-weighting option being selected. A-weighting background noise in my room is around 31.5dB(A) at the listening position or 43dB(C-weighting) . Both readings were taken with the Galaxy resting on my listening chair with the rather noisy computer running about twelve feet away.

The nearfield curves in the REW graph show little or no room gain. At the listening position the lower frequencies are boosted considerably by room gain. The nearfield curves are reasonably flat from 20Hz upwards and speak volumes about the quality of these AEIB15 drivers.

Later reading suggests that there is some spread to the frequency response of the inexpensive Behringer ECM8000 microphone. Those who wish to enlighten themselves or have their microphones properly calibrated could do worse than check the HT Shack threads on this subject.

Remember that you can click on all images for a larger version. Back click to return to the blog. .


Calculating IB enclosure volume.

I have received a comment asking for IB enclosure volume calculations. Thanks for the timely suggestion.

So: Total Vas x 10 = minimum enclosure volume for an IB subwoofer.

That's all there is to it really. But I suppose you want more? :-)

Until now I have largely avoided putting real numbers on my IB. Though, in fact, the design calculations are very simple indeed and can be (almost) ignored if one stays well within the usual IB design parameters. Which is just another fancy way of saying a 4 x 15" IB can be built to standard dimensions and it will work very well indeed. It will also stay safely within any cone excursion limits which might endanger the drivers. It will also have a reasonably flat frequency response down to well below 20Hz. Or even 10Hz if your room allows it.

First things first: An IB subwoofer is critically dependant on the chosen drivers: So choose from the small range of 15" or 18" IB-specific drivers and most of your design problems cease to exist right away. The specialised IB driver will offer a realistic maximum cone excursion. Which usually means a minimum of 18mm Xmax. Avoid car audio drivers with flashy paintwork and chrome finishes claiming ultra long stroke. Unless Xmax is clearly stated then any claims are meaningless sales hype. So, which drivers do the real experts use for competition car audio dB drag racing? Those manufactured by exactly the same companies which offer serious IB drivers at lower cost. Not the over-hyped, over-Xpensive paint and chrome items intended to appeal to the young, the inexpert and the gullible.

Driver displacement is vitally important in IBs. Cone excursion multiplied by the cone area multiplied by the number of drivers should come to around 10-12 litres minimum total displacement. For 15" IB drivers: 4 x 3 =12 litres. This displacement can also be achieved with two good 18" IB drivers. So we've already learned that 4 x 15" IB drivers = 2 x 18" IB drivers. Since Xmax is measured as a one way come movement we use 2 x Xmax x Sd to give the displacement for one driver. The larger cone area of the 18" means a smaller Xmax will provide a similar displacement to a 15" driver. Better still is a large Xmax giving the 18" IB driver around double the displacement of the typical 15" IB driver.The Fi Car audio IB18 offers 30mm Xmax!

What if you want to use 12" drivers? Then you must calculate their total displacement to achieve a minimum of 10-12 litres just to be safe. Displacement obviously favours large cones so your 12" drivers had better have a seriously long stroke and you'll need at least 8 (and preferably more) to compete safely at high sound levels at very low frequencies with a typical 4 x 15" IB. Twelve or even sixteen x 12" drivers is a better starting point but likely to be rather expensive using high quality drivers. Most 12" drivers wont have such a low Fs as proper 15" IB drivers.

BTW: For those who have forgotten: The area of a circle = radius x radius x Pi. (Pi = 3.142)
The squaring of the radius is the obvious clue to the rapid increase in driver effective area with increasing diameter.

Reproducing low frequencies need lots of air movement. Fanning the air slowly with a large board is obviously far more efficient than waving a postcard a yard backwards and forwards each way. So it is obvious that bigger drivers move far more air per stroke than smaller drivers. Big driver often play loud and low without any visible cone movement at all! So more cone area = higher efficiency at low frequencies. The IB is naturally efficient with falling frequency thanks to a lack of back pressure and large cone area. It also has a motor for each of those big cones to keep everything nicely under control.

A 15" driver is a reasonable size for true pistonic movement at low frequencies. Bigger drivers may suffer from cone breakup. Very large cones almost certainly will. Large plastic/vinyl cones are generally regarded as too soft to be ideal for a sub-bass driver. Particularly in straight-sided conical configurations. Steeply flared, reinforced paper cones are greatly admired by those in the know for their lightness and stiffness. Aluminium cones are highly dependent on the skills of the designer and manufacturer and disliked by some purists who claim the cones can "cry" audibly at higher frequencies. Only testing with software like REW would prove this to be the case. All the Cult recommended IB drivers have reinforced paper cones.

IBs are so efficient that with quite reasonable power they will go very loud and very deep on a typical musician's power amp. The Behringer EP2500 amplifier is a popular choice for IB use with 4 x 15" drivers. Two 8 Ohm AE IB15s in parallel is a comfortable 4 Ohms impedance per channel. The manufacturer claims 650 watts per channel RMS into 4 Ohms. So the drivers are seeing 325 watts each. The EP2500 has also been used satisfactorily with 8 x 15" drivers thanks to the greater efficiency of the larger combined cone area.

Acoustic Elegance made my IB15 drivers and they are deservedly popular in sets of four at $100US each plus freight. Even in Europe they make good sense despite the heavy freight charges and import and sales taxes. One day, global tariff agreement will be reached or tariffs will be scrapped altogether. Then we'll only have to pay for freight to Europe for US drivers without all the unwanted extras. Check with AE for latest pricing before placing an order though as this blog may eventually grow as many grey hairs as its author.

Fi Caraudio makes excellent 15" and 18" IB specific drivers. Four Fi IB18s would probably be bulletproof no matter how wild action DVDs may become and will probably leave nothing to upgrade for several decades. They also make more extreme models at higher, but still attractive, US prices. They are a very small company and each driver is handmade to order.

Other popular drivers are listed on the IB Cult's recommended driver page:

Driver manufacturers come and go. Their model range may also change. So check for the latest drivers and manufacturers before making any final driver decisions. Also look for Mach5's IXL18.4 and and the DIY Cable Tempest X15 for further popular models for IB use. The Tempest X15 is available from a European distributor. (Streaming Audio of Sweden) Prices are probably competitive in sets of four with the AE IB15s once freight and taxes are added to the AE drivers US prices. The Tempext X15 claims 27mm of Xmax for slightly more displacement at 4 liters and more headroom than the AE IB15's 3 liters. At a glance these two drivers look remarkably similar to each other in appearance except for the outer roll suspension. The AE claims lower distortion for their IB15 due to the careful motor design.

There is a link below to the list of drivers currently on the IB Cult's forum website. Use the US dollar price per litre of displacement as the surest guide to driver value for IB use in the USA. Elsewhere in the world you'll have to do the maths depending on local availability and any additional import and freight charges. These can easily double the US online purchase price.

Loudspeaker drive units (or 'drivers' as they are now commonly known from US usage of the term) are built to very exact specifications. Thiele-Small paramaters are used to describe drivers in precise detail and in units which don't mean much to most people. Here's the Wikipedia entry: You can safely ignore all the terrifying equations and scroll straight down to the list of Thiele-Small parameters list much further down the page under Qualitative Descriptions:

Some useful terms for IB design: (with simplified descriptions) Feel free to wallow in the fine details from Wikipedia and other online resources. You really don't need to know more than appears below to design an excellent IB. Remember that buying recommended IB specific drivers will save a lot of extra work and worrying over details. Choosing these IB drivers is a short cut to success where these recommended drivers are available and affordable.

Fs is the free air resonance of a driver and measured in Hz.

replaced cycles per second some years ago in honour of Hertz but the two terms mean exactly the same thing. Having an Fs somewhere between 15 and 20Hz is ideal for an IB driver. Go much higher in frequency for Fs and you might need to start boosting the bottom end to get serious infrasonic output.

Infrasonics = sounds below the normal frequency threshold of the human ear.

Though usually completely silent, at the levels we listen to at home, infrasonics are good for making the room shake on action films and deep organ music. The lower audible frequency threshold for adults is usually around 20-22Hz) A low, free air resonant frequency driver will avoid the need for low bass boost. Room gain will help. Leaving a nice flat, extended response to below 10Hz in most cases. (provided the 15-25Hz range for Fs is adhered to) The AEIB15's Fs = 16Hz = Ideal for IB use. The IB specific drivers mentioned above all fall safely within this range. This is another good reason to seek out IB specific drivers. Everything is taken care of by the designer saving you having to worry about specific details.

Provided we stick to standard IB driver designs and use a recommended number of IB specific drivers we only need to know the numerical values of the following T-S parameters: There is absolutely no need to understand how these numbers are obtained by the manufacturer unless it really interests you. It won't make a better IB but you may be able to impress your lucky visitors with more than your world class IB bass.

Vas= Equivalent air volume of driver suspension stiffness. Vas is used to calculate the size of the rear volume or backspace of the IB. In normal loudspeaker and subwoofer terms we'd call this back space the enclosure. The same holds true for IBs. IBs just use huge volumes for their enclosures.

(N.B. Enclosure size does not relate to the manifold volume. The volume of the manifold is largely irrelevant as long as it is big enough to house the drivers. A driver manifold is really just a local folding of the baffle with serious mechanical advantages which you should not ignore.)

Vas is given in litres by the manufacturer in their driver specification. The AE IB15 has a Vas of 439 litres. (15.5 cu.ft. in old money) If you are metrically challenged just divide litres by 28.4 to convert them to cu.ft.. BTW: Liters and litres are interchangeable names depending entirely on your own geography. Divide litres by 1000 if you want volumes measured in cubic meters.

Qts = Is a unitless measure of quality of a driver. It usually lies between .2 and .8 for loudspeaker units or drivers. The AE IB15 has a Qts of .7. Which is fine for IB purposes. You can go lower than .7 but avoid going much higher than .7 if you can. Experts consider .5 to be the optimum where electronic equalisation may push up the overall Q of the system. Q can be pushed up by accident or design. It cannot be lowered.

Vd = Cone displacement measured in litres = Cone area (Sd) x (2 x Xmax)

is calculated from the following two parameters:

Sd = Useful cone area in square centimetres (usually including part of the outer suspension roll surround which contributes to the driver's output)

= One way safe cone excursion measured in millimetres. (there are other factors involved but we don't need to delve deeply because we just use the manufacturer's claimed Xmax in millimetres)

All these figures should and will be provided by the high quality IB driver manufacturer. So all you need to do is pick off the numbers you need and you can go ahead and design your theoretically perfect IB.

Ideal IB enclosure minimum volume = Vas x Number of drivers x 20.

Enclosure volume for 4 x AEIB15s = 439 Liters x 4 x 20 = 35120 liters = 1236 cubic feet. (Wow! That's a 10.75 x 10.75 x 10.75 foot cube!) Say you only have a typical 7 feet domestic headroom you're going to need a space 13ft x 13ft to house your IB optimally.

A multiplier of 20 x total VAS is the calculated figure which will not raise the Fs (natural resonant frequency) of the IB drivers any higher than if they were measured in free air. As you reduce the rear volume (enclosure size) the natural F of the driver(s) will slowly rise. This might cause you to lose very deep bass and decrease system efficiency. This will theoretically require more amplifier power to reach the same output levels at very low frequencies and might also need some low bass boost. You may lose subtle sounds and dynamic effects in the deep bass which make an IB sound so real.

How low dare you go with an IB enclosure volume? Ten is quite a safe Vas multiplier but the risk is yours, not mine. My own IB enclosure is now down to a Vas multiplier of about 7. (at a very rough guess) Unfortunately the enclosure space is just too handy for storing packaging. Insecurity makes one keep the packaging in case the item needs to be returned to the dealer. In an ideal world all those boxes would be nested and left wide open. (To help reduce compression effects within the IB enclosure and to help scatter any standing waves) Do I notice anything from adding yet more empty boxes? Nope.

Note: My enclosure space is quite poorly sealed to the outside world which probably helps the cause. I also have 12"-16" of almost bare rock wool insulation above the sloping surfaces of the enclosure with only a DPM covering at the moment. Polythene DPM is probably quite transparent to very low frequencies if they are allowed to bounce around in the enclosure. Such a thickness of rock wool is quite good at absorbing very low frequencies. If you have a badly undersized concrete box for an IB enclosure you'd better not use it for storage of anything but fibreglass or rock wool wadding!

So, in summary: In a perfect world, you aim for at least 20 x Total Vas (Total Vas = Vas x No. of drivers) as your ideal enclosure volume. Then hope you can find enough rear volume in your crawl space, loft or cellar. As discussed earlier, the IB enclosure can be a spare room which shares a wall with your AV room. It can also be the underfloor (crawl) space or cellar. Or the loft above the AV room ceiling. Thankfully crawl spaces are usually very large areas covering the entire house floor plan. So the lack of depth isn't usually too much of a problem where volume is concerned. Lofts are often the largest room in the house when they go unused. They are often asymmetrical, triangular in cross section and well supplied with thick absorbent insulation into the bargain.

The real problems start when you really want to have an IB but you only have a small closet or cupboard available for your enclosure. Assuming you want to use 4 x AE IB15s and your closet height is only 7 feet then you need to think very seriously about alternatives. You could spread the necessary volume around between several smaller IBs in different places around the room if that helps you get more enclosure volume. The extra enclosures could be more cupboards, underfloor spaces or above the ceiling.

Some IB builders have removed the ceiling of an undersized walk-in closet to exhaust into the room above. Not ideal if it has to exhaust into the loft space though! Not unless the closet enclosure can be very well insulated in very cold climates. You must insulate or the heat in the house will wick rapidly away through the closet walls and probably cause massive condensation in winter. The opened closet ceiling can't be blocked with insulation or the IB drivers will still think they're trapped in the small closet. You wouldn't want to leave the open hole unguarded or people and wildlife will continuously fall into your IB enclosure. The exit hole can't be small either or the very low frequencies won't notice the hole is even there. Steel weld mesh over the hole down into the closet is a good start to avoid nasty accidents and wont affect the necessary cyclic airflow through the open closet ceiling.

Air movements are so large at VLFs that they need lots of room for free flow. Sound pressure waves move backwards and forwards. Not up and down like a skipping rope making waves. Imagine low frequencies as a string of huge balloons being cyclically stretched and then compressed back together. The scale is awesome. The usual crossover point of an audio system with a subwoofer is 80 Hz which has wavelength of around 15ft. 20Hz has wavelength of nearly 60 feet!

Now try to squeeze the vast balloons through too small a hole and the balloons will object. Greatly increasing in friction until no movement is possible. You might even think of the squeals of protest from the giant balloons as harmonic distortion components. This is why you don't want to port your IB enclosure. The IB can do bass like no other enclosure. "Gilding the lily" with a port is not a good idea. Using the same number and size of drivers (as in your intended IB) the enclosure size for a reflex design would have to be completely different. (usually very much smaller than an IB enclosure) The port would have to pass so much air that it would be difficult to avoid turbulence, noise and distortion. Moreover the ported enclosure would have to be airtight and rigid enough not to flex unduly to work as an efficient Helmholtz resonator. The air in the huge port required must have some deleterious effect on the sound quality due to turbulence and inertia.

The IB doesn't mind if you make its enclosure much larger than 20 x Vas. Nor how leaky it is. You are not really tuning an enclosed volume like a ported box. With an IB you are merely blocking the rear pressure waves from meeting the front pressure waves being produced by the driver cones.

In theory you could have your IB open to the outside world. Just fix your drivers into an outside wall and turn up the volume! Fine, if you live in the woods and are half a mile from your nearest neighbours. Not a good idea (at all) if you live in town. It's best to cut down some of the energy in the rear waves before they reach your neighbours right next door. So enclosing the rear waves is a vital exercise for most IB builders. Containing the IB in an enclosure also limits the bass leakage to other rooms in the house. Obviously this will vary and the rear waves aren't the only culprit. Any powerful subwoofer will share its output with the rest of the house unless you have thick concrete walls and floors, brick or block walls and heavy soundproof doors.

IB design in a nutshell:

A regular box of any shape has a volume of Breadth x Length x Height .

A roof prism shape (attic or loft) has a volume of 1/2Base x Height x Length.

A trapezium (attic or loft) with a flat ceiling width A and floor width B has a volume of 1/2 (A+B) x Height x Length.

Let's work out an example to check my own IB enclosure volume. It is a simple roof prism which goes right up to the peak of the roof without any form of flat ceiling.

Height is 10 feet. Base = Total width across the floor = 18 feet. Length between the two triangular walls is 6'6".

Volume = 1/2 x B x H x L = 10 x 18/2 x 6.5 = 10 x 9 x 6.5 = 585 cubic feet.

It comes to exactly the same answer if you multiply all the dimensions together then divide by two. (B X H X L)/ 2 = (18 x 10 x 6.5)/2 = 585. You can halve any one of the dimensions of a roof prism and still get the same answer. 1/2 x Base is the same as half x Height or Half x length provided you halve only one of the dimensions in the formula for volume.

Don't forget to multiply the Vas of your chosen drivers by the number of drivers to obtain Total Vas. Or your rear volume calculations won't be of much use. Using Vas for only one driver in the volume calculations seems to be quite a common mistake and leads to unfounded optimism about the minimum allowable volume of an IB enclosure.

If you can only find enough rear volume using Total Vas x 5 then I'd probably still go ahead. You can put some thick absorbent material in the rear space which will fool the drivers into thinking they have a little more space available. It's not ideal but you'll still have a subwoofer which leaves most others for dead on sound quality, frequency extension and sheer power. The downside with such a small enclosure is that the enclosure walls will need to be solid or well braced in comparison with a much bigger volume. The walls of a bigger space are not subject to such fierce cyclic compression and rarefaction effects as the smaller rear volume. Though the surfaces will still vibrate if they are flexible enough even in a very large space. Wall vibration may be out of phase and cancel your bass at particular frequencies. Unwanted structural noise produced from vibration is bad because it draws attention to itself. Spoiling the pleasure of listening to music or a film soundtrack.

Simple Basic Rules for IB construction:

A: Use an opposed driver manifold unless you can build a very stiff and/or massive wall for an array. If you are determined to have an array then keep drivers low down in a horizontal row just above the floor. So the stiffness of the floor plate directly contributes resistance to driver movement. Or put them in a horizontal row up high. Using the lateral resistance of the ceiling joists in the same way as a floor. Or arrange your drivers vertically up a corner between two walls to gain the resistance of the side wall. Worst case scenario is when an array is arranged vertically in the middle of a lightweight partition wall. Even worse is having a door beside the array! The drivers will sh-sh-sh-shake the wall and door like you would not believe! Been there. Done that. Didn't like it! Have you seen a door literally bending rapidly back and forth in its frame by a full 1" from flat? I have. That's why I now have a manifold! Do not underestimate the reaction forces from 15" or 18" drivers!

B: Use Vas x Number of drivers x 20 for your enclosure volume. Cheat on volume if you must. We won't tell if you don't. But big is still beautiful. 10 xVas is fine in practice.

C: Use (at least) 4 x 15" or 2 x 18" IB specific drivers for a minimum of 10-12 Litres of total driver displacement. Make no excuses and you won't bottom your drivers on the first action film you watch! As ThomasW of the IB Cult says: Drivers make poor (but expensive) paperweights. Say you cheat and try to use only two x 15" drivers and break the first two. Then you have to buy four more to be safe next time! That's six drivers you have to buy in all! Even I can understand the maths of such an economic disaster. Ouch!

D: An impedance load of 4 Ohms per channel is optimum. Eight ohms per channel is acceptable but will limit output a bit. Two Ohms per channel is far too hard on many amplifiers. Bridging an amplifier is the same thing as halving the impedance per channel seen by the amplifier. So eight ohms bridged is okay. While 4 ohms bridged is very hard work and overheating or instability may occur in the amplifier. Some amplifier manufacturers will specifically tell you not to wire your drivers for a 2 Ohms load per channel or 4 Ohms bridged. It really is far safer simply to avoid 2 Ohms per channel or 4 Ohms bridged. If your amplifier goes into heavy clipping it will burn your voice coils with nasty distortion. More paperweights! Even if you can afford all these paperweights try and think of the planet's finite resources.

E: Two, four and eight drivers are usually easy to wire up for a sensible amplifier load. Three or six or nine drivers each of 4 or 8 Ohms impedance are difficult to wire up sensibly. You can't wire them in series, parallel or series-parallel to obtain a suitable impedance around 4 Ohms for your amplifier to drive.

F: Feed your IB drivers with a popular/reliable musician's power amp for clean headroom. If you want more output then double the number of IB drivers rather than doubling the RMS wattage. Avoid plate amps if you can. Most lack the power you need and are not such good value for money as pro power amps anyway. Your AVReceiver will usually act as the crossover between subwoofer and speakers. So the plate amp's crossover will be bypassed and becomes superfluous.

G: Test your proposed IB location with a conventional subwoofer and REW before you start making large holes in your home. If a smaller subwoofer suffers from massive nulls or peaks at the listening position then it is certain an IB will suffer from them too! Moving a finished IB is not the easiest thing in the world and loses you major house points with your partner. So check your proposed manifold position first even if you have to borrow a box subwoofer and learn to use REW. Or build one of your IB drivers into a box and use that for testing. If you are really confident you want an IB could build your manifold and use that with the opening blanked off or resting on a carpet or rug to obtain a seal. It really will be worth the hassle to test the proposed IB location first. If the conventional subwoofer sounds okay to your ears at the chosen spot then you may be okay. Though using REW is very much safer because you can see exactly where particular frequencies are missing or peaking in the response curve and this isn't always easily audible. Not unless you listen very carefully to a wide range of different material during your tests and know what to listen for.

H: Last but not least: Join "The Cult of the Infinitely Baffled" forum. All of the above information has been gleaned at the feet of ThomasW, the IB Cult's high Wizard. He has forgotten more about audio than I will ever know. (or can still remember) So if you are really serious about building an IB join the IB Cult and get stuck into the IB Cult's priceless FAQs. Each of which is individually hand beaten from 21st Century alchemist's gold by the fabled high wizard himself. Seriously though, Thomas has popularised the IB subwoofer almost single handedly with his IB Cult forum. His decades of audio experience remove much of the hearsay and bullshit you will read elsewhere. Reading some of the nonsense posts on others forums where the members try to talk about IBs from a standpint of total ignorance can be more excruciating than amusing! There really is a lot of ignorance surrounding IBs even by those who try to build them without knowing any of the simple rules.

I advise you to read the Cult's FAQs right through first. Then look at the galleries for lots of working examples. Only then ask questions on the forums. It's only polite, given the enormous effort ThomasW has put into them.

And finally: Don't ask about car audio infinite baffles on the IB Cult. There are plenty of car audio forums for discussing all kinds of subwoofers. The IB Cult is for discussion of properly designed, HT/AV/Home Audio, True Infinite Baffle subwoofers. Whether for film, or music, or both, the IB will deliver bass sound quality and clean power like you have never heard before. Provided you follow the simple rules.

In case you have any lingering doubts about an IB's sound quality: The Cult is littered with examples of owners who have built IBs to replace stacked or multiples of expensive, commercial subwoofers. The IB has always sounded better except in one or two rare cases. Most IB builders are absolutely delighted with their results provided they followed ThomaW's advice and adhered to the basic rules.

FOOTNOTE: This post was published quite some time ago now. The links, prices, drivers, charts and availability have all changed or disappeared. Join the IB Cult for up-to-date information and/or browse for manufacturers mentioned here for their latest offerings and prices.

For Fi Caraudio IB3 series IB drivers in Europe contact Blade ICE. (UK) They have a solid reputation with AVForum members. They can also supply many other drivers, of course.