Gavin's DIY Home Theatre (Hardcore)


After a long time, I finally got my DIY setup in the new place to a state where the screen, projector and 7.1 setup can go up for a test run.

Took a while DIY , it’s about 13 months since I moved into this new place. :sweat_smile:

Room is ~5m x 7m, with irregular ceiling that is also quite low 2.2m on the right and back. Also there’s a pillar… :sweat_smile:

Front stage consists of a 16:9 (edit :150in) AT from XY Screens, their 4K material.

The mains are using existing DIY speaker from previous setup. Haven’t built the new mains to match the new center. Edit : 1 in compression driver from Peerless/Tymphamy, Seos 12 WG, Beyma 10 inch midbass, and SB Audience Nero 18 pro sub, main subs from my previous HT.

Behind the screen is a baffle wall, with center speaker, and LFE subs. Consisting of push pull opposed Fi IB318 for low bass, and SB Audience Nero 21 for midbass.
Center speaker is 2.5 way design with a large JBL horn and CD, 12inch midbass, and a top bottom 12 in woofer in vertical WTMW configuration.

Edit : One more pic with yours truly for scale.

Surrounds are Diysoundgroup Vortex 12 and Vortex 15 kits. These are coaxials, one of the best configuration for surround field.

I’m currently using my JVC X5900 projector that I kept from my previous setup.


Love takes time…
Great works Bro!!



Massive big room!
Congratuations time to enjoy the fruits! :slight_smile:

Curious bro

  1. It’s hdb 2 rooms hacked? or landed :slight_smile:
  2. How big is the AT screen and viewing distance away?

Thank you.

Hey Ron,

It’s a 150in screen, about 3.5-3.6m distance. ~50deg viewing angle.

Attic room, hence the low ceilings and pillar.



This project has been a long long time in the making. i try to incorporate as much theory and learnings accumulated over the years into the build.

Starting with the LFE channel. i think everyone has heard of the Hoffman iron law when it comes to bass, or subwoofers. There are 3 factors - Size, Low bass extension, and Efficiency. One can only pick two out of the three.

If it is small and extends low, it will very low efficiency. This is the most popular trade off for commercial subs, for practical reasons. Power is cheap these days, relatively.

I chose to go with low bass extension, as well as efficiency. That means giving up size. The LFE subs fill up the entire baffle wall across the front of the room, some 6000 Liters in volume from my rough estimate.

The LFE subs have the following characteristics.

  1. Infinite baffle, IB for short- this is a bass alignment, like sealed, ported, transmission line. IB alignments are basically sealed, but with very large volume. The official criteria is for volume > 10x VAS, but that as usual it is not a hard threshold where one side is IB, one side is sealed. There will always be efficiency gains with larger cabinet volumes.

Subjectively, the sound of an IB is also different. Typically a subwoofer is built in an enclosure, and the optimal volume is determined by the VAS parameter of the sub. One can think of VAS as the volume of air that has the same ‘springiness’ as the subwoofer suspension. A very stiff suspension sub has low VAS, and vice versa.

For a sealed sub with cabinet volume ~ VAS, say 100L, the air spring in the cabinet is ~ equal to the the suspension. One can imagine that that the air spring is non linear, especially at large excursions. The air spring effect is also different on the outward vs inward stroke of the subwoofer. All these non linearity has an effect on the subjective sound. And suffice to say, there is a avid following for IB subs. They hang out at an old forum - Cult of the Infinitely Baffled.

Recently, IB subs seem to become more popular. The main barrier to entry is having sufficient space for the IB. Attics, basements, store rooms, equipment rooms, baffle walls are some of the ways to achieve IB setups. Subjectively, the sound quality of IB bass is 2nd to none.

  1. Single bass array, SBA. This is related to the Double Bass array , but with passive back wall absorption. Theoratically, it generates a plane bass wave and sends it down the length of the room. In my application, my room is not symmetrically rectangular, and i also had to offset the upper right source inwards. The SBA will theoratically cancel the width and height modes of the room, leaving only the length modes to contend with. More on this later.
    Double bass array - Wikipedia

  2. Push pull force cancellation Subwoofer drivers ideally have suspension that are fully symmetrical in the outward vs inward directions. Real subs do not have perfect symmetry, and the difference in output from the cone-out direction vs cone-in direction generates 2nd order harmonic distortion. Having one driver flipped basically cancels out this 2HD. This feature is commonly seen on push pull MK, KK and most recently Perlisten subwoofers.
    Edit : the difference in the KK, MK and Perlisten subs - they have 2nd harmonic force cancellation due to the push pull, but with the drivers 90deg rotated there is actually no force cancellation between the drivers.

  3. Plenum / Slot loading. This is approaching the more subjective aspect. The idea is that plenum loading the subs is akin to a short horn, and subjectively this provides desirable characteristics to the bass sound. Both 3 and 4 are something that i’ve been wanting to try for the longest time. In my plenum, i have about 3:1 compression, meaning the total opening of the plenum is 1/3 of the combined displacement area of the 2 driver cones.

There is a large impedance mismatch between air and the surface of the driver. If you imagine trying to push air with a flat (subwoofer cone) surface, the air just moves away to the side. Having the air constrained in the plenum improves the impedance match between the air and the subwoofer cones, and supposedly this is a desirable character when one is trying to move (pump) lots of air to generate bass.

3 and 4 combined gives Push pull slot loaded. PPSL. There are various threads, including this : A Thread for those interested in PPSL enclosures | diyAudio


can’t wait to visit and experience your DIY subs

IB subs require a uniquely optimised set of parameters. There are a few options for dedicated IB drivers, including the following.

For comparison, i will use the ever popular Dayton Ultimax 18.

The IB subs are simmed in 1000L, while the UM18 is in a common 120L (4cuft) sealed enclosure. All are at at 100W input power. For comparison, i have also simmed the UM18 in 1000L IB. it is the red line that has the slow rolloff and matches the IB subs at 10Hz

At 20Hz, All the IB subs generally have 5-6dB more low bass output compared to the sealed alignment. the UM18 in IB alignment has 3 dB more at 20hz vs sealed, and ~6dB more at 10Hz.

6dB may not seem a lot, but it requires 4x power to achieve 6dB of increased output. This is not a big problem at 100W vs 400W, but at 500W vs 2000W, or even 1000W vs 4000W, this is not so easy.

The big subs like seatons, JTR, deep sea sound commonly use SP4000 and SP6000 amps for this reason. Even if the amp can supply the power, there are mechanical linearity and thermal issues for the driver. All the tech requirements go up immensely.

My LFE weapon of choice was then the Fi IB 318V2, chosen for its bang for buck. Fun fact - Fi used to make the drivers for JTR, during the Captivator models. Not sure who supplies the drivers for the new RS1 /RS2 models.

Nice carbon fiber cone that is unfortunately totally hidden in my alignment.

Very long stroke motor, but otherwise not very impressive. IB drivers have higher Qtc, which means relatively weak motors. This was actually an area of concern for me, as strong motors are generally very important for subs.

Very long voice coil, 34mm Xmax!


@Wechnivag bro

Super nice subs u had there: ) Look solid!
This cone lines here really look alike my JTR RS2 Driver!
Power 34mm Xmax! yeh the latest JTR RS2 is 36mm Xmax :slight_smile:

The poweress and weight of the driver cone can really give the tactile feel and weight for HT experience.

Will pm u 1 day we can visit one another for the experience cheers :slight_smile:

1 Like

I actually do not need anywhere near the output of 8x18 subs… 4 would have been plenty. But, i needed 4 sources for the SBA, and i knew i had to do opposed force cancellation for ULF subs, otherwise the vibration will be very difficult to manage. And for that simple reason, i ended up with 8…

For this build, i knew i had to get a sub amp that is fitting for the drivers, and it can only be the Speakerpower SP16000.

According to the sims, ~220W gets me to reference 115dB at 20Hz. The full 6000W gets a silly 129.5dB at 20Hz. This is at 1m i believe, before any room gain adjustments etc.

Just below 6mm excursion for reference output. At full 6kW, it stays just below the 34mm Xmax.

In DIY circles, overkill is a must. Suffice to say, i killed it dead. :laughing:

The subs are mounted in holes cut in the baffle wall, which is 1 inch plywood on frames, built by my carpenter as part of reno.

The build started with building the manifolds. As usual, i like to size everything super tight, also to achieve that 3:1 compression ratio.

This was a test fit of the first manifold.

Another shot of manifolds. Hardly any baffle remains after the circles cutout.

4 ready to go.

1 Like

Getting the subs assembled in the room and loading into manifolds.

A 12in and a 5in driver for scale

Doing some running in with a small amp

Weight wise, these are just barely manageable alone to shift. The drivers are about 14kg each, and the manifold is just over 10kg. Total just about 40kg per assembly.

Can’t imagine a sealed opposed 18, which needs to be at least 200L…

I’m on the 3rd floor with no lift…size and weight was a major design consideration :sweat_smile:


Completely amazed. 130dB at 20Hz! Will be curious what the final measurement is after the room gain. My HT is also in an attic and everything rattles. Is this a new construction or did you renovate the attic with the HT in mind?

What are those black structures on your ceiling?

I was one of the lucky guys who got to experience Gavin’s set up last week!

Enjoyed the demo! Thanks Gavin

The Nero 21” subs were fast and Hard hitting, doesn’t feel slow at all…

1 Like

Thanks for all the interest and kind words. The room is far from done and all visits so far have provided valuable feedback and food for thought.

The room was built during renovation for the HT. Quite a lot of work was done on the walls and structure to manage the sound energy I was planning. The ‘structure’ on the ceiling are hangar style bass trap. More on those, and the details of the room construction later. Many areas and details to cover… :joy:

As I shared in the Anthem thread, there is a length mode dip at ~30hz that is expected In SBA setup. After building a projector shelf to mount the projector properly, my next task is either to build the matching LR mains, or the back wall low bass treatment to manage that dip. Probably both… :grimacing:

In the meantime, here’s the current projector stand. :rofl::rofl::rofl:

1 Like

More on the SBA passive absorption.

This room has a massive 40hz dip. After the back wave absorption, it all but disappears. But it took a 2ft thick entire wall of rockwool to do that…

In the spirit of DIY , I am mulling different type of treatment that may be only 5-10cm thick across the back wall. Hopefully it will be effective (enough) , time will tell.

Here’s the link for more info on the OG SBA build that inspired me.


Here’s another build with SBA IB subs. A man after my own heart - form follows function.

in this case, builder chose not to mount the drivers Push Pull. It is in Push Push. Having some 2HD is not actually a bad thing, it is actually part of the sound character, the ‘growl’ if you compare the sound of JTR subs vs Rythmik servo, that some may find too lean for HT. I’ve pored over the AVSforum threads looking for subjective sound impressions, comparing HD plots on Databass…

i seriously considered not doing Push Pull for this reason, however the physical installation constraints actually decided it for me. There would be no way i could install the subs with the drivers in Push Push. More on this and photos later.

The other benefit of the Push Pull, is that the motor and frame structure of the driver in the manifold actually helps to mitigate the quarter wave resonance in the manifold. It acts as a quarter wave resonator, with the first modal peak determined by the depth.

For the dimensions that i need, 460cm diameter driver, 50cm deep manifold, i calculated the first quarter wave mode peak at ~171Hz. The comfortable guideline is for the first peak to be 1 octave higher than the passband. For LFE cutoff at 120Hz, that would be 240Hz. You can see i barely made it. Having the driver frame and motor in the manifold helps to further damp this mode.

In the Event horizon manifold, 15in subs are used. The manifold depth was not minimized, it looks to be perhaps 65cm deep. That gives a first modal peak at 132Hz. Not the end of the world, can be taken care off in EQ. But…

The devil is in the details…

Here’s another interesting reference, from the great Nelson Pass. His slot loaded open baffle project.

There’s no fancy math behind the 3:1 compression ratio for manifold. Good old monkey-see monkey do. :laughing:

Mr Pass also talks about the manifold / slot resonance issue. The peak occurs at quarter wave, the cancellation null occurs at halve wavelength. There’s double the margin to the first cancellation null, so i guess there’s plenty of buffer for subwoofers. Not so much, for woofers that are trying to cross higher.

Wow, v good space @Wechnivag. Woohoo! Finally u embarked on your HT project in your new home! This space is bigger than your previous den ya?

Yes, it is a nice space. There are some things that gave me a ton of pain, like that pillar in the middle of the room… and the low/uneven ceiling height…

Still dunno how to mount atmos… :sweat_smile: will cross that bridge when i come to it…

The manifolds get loaded into the baffle wall. I built supporting structure at the back, and the subs basically just sit on it on compliant sorbothane mounts
I specifically do not mount or fix it to the structure, to avoid structure borne noise. Otherwise, This would transmit vibration to the whole house and neighbors as well.

Shelf prepped

Loading up a manifold. Not much wriggle room

Another look at how close the fit is. It would never fit with a Push Push mount, both backside out.

Slowly rotating them into position

Shelf and baffle wall support structure. The baffle wall is somewhat intentionally flexy, as opposed to strongly overbuilt.

Manifold in position, before baffle is added

Another look from behind the baffle

And yes, I had to crawl into the space.

1 Like