DIY monster sub for HT use

Hi-Fi 4 Sale wrote:Friend,
This is amazing work. 6 months on and we see the realisation from concept to tangible.
Multi-faceted skills in electronics, woodworking & testing.

tq bro. learning new things everyday. posting some form of documentation so that it would help others contemplating on starting.

sorry for the long silence. my laptop acted up. photoshop to be exact. refused to start, complaining about lack of memory.

here what i've learned so far. from my reading, testing, experimentation, building, listening. if anyone is lazy to read the entire thread, at least read this.

• you can't judge how a driver would sound like by looking at the spec.

  ---

  a driver could have a perfect spec but could still not sound like how you expected it to be.
  
  
• fs (free-space resonance) is not the only number that matters.

  ---

  low fs is always preferable, but cone mass / moving mass is also important. lightweight cone would probably translate to thin sounding, fast beat with more emphasis on midbass. heavier cone would probably translate to deeper bass, albeit slower response, and less sensitivity.
  
  
• a simple cross bracing is never enough. every panel should be braced every 6 inch.

  ---

  large panel will resonate especially in sealed alignment, and with a heavy driver/driver with heavy cone. ported enclosure, not so. but still it'll vibrate if not properly braced. use window bracing wherever possible.
  
  
• 1" mdf is better in making an inert cabinet.

  ---

  but it'll be freaking heavy to lift all by yourself. 3/4" mdf is perfectly fine, as long as it's properly braced.
  
  
• power rating tell a lot about the driver.

  ---

  if it's rated with a low power handling, high chance it's not meant to be used in sealed enclosure. 'low' for me is below 500w for 15" driver and below 300w for 12" driver. you need a powerful motor to fight the negative pressure inside a sealed cabinet.
  
  
• opt for system Q (qtc) of less than 0.7 for sealed enclosure.

  ---

  0.5-0.6 would be nice. if you opt for 0.7 and try to reduce the q by stuffing, high chance it wouldn't work.
  
  
• it's perfectly fine to use lower internal volume for ported sub.

  ---

  i used to think that i'll need a huge cabinet, the largest volume to get a nice and flat freq response without eq. turned out some boost to help correct the freq response is fine, with acceptable result. though +3db is about as much as you want to go. +6db translate to quadruple the required power and for most driver, it's way past the rated power handling.
  
  
• 35m/s port air velocity is acceptable (for me).

  ---

  i tried playing some regular movie at maximum level that i'm willing to listen and there's no audible port noise. also tried playing some low frequency test tones at maximum level bordering insane and still no audible port noise.
  

though i have to say, it's all based on my own personal conclusion and might not be in line with other's experience and opinion. feel free to disagree.

got photoshop sorted out. so pictures could continue.

using the calculated 4.39 cubic feet of nett volume, simulated spl is as below:

green is 20hz tuning, yellow at 16hz, red at 11hz and orange in sealed config.
the sub need some boost at the tuning frequency to help achieve maximum performance (in term of excursion). for the ported alignment i added +3db boost at the tuning to get the spl curve you see above. sealed alignment are left with no eq.


cone excursion are kept to relatively 'safe' distance. a couple of mm past the rated xmax is still safe for most driver as xmax is just rated linear excursion and not the mechanical limit. above the xmax the driver would just be operating past the linear performance curve and should produce distortion exponentially.

the dvc15 driver that i'm using have a rated xmax of 15mm, xmech is guesstimated at 25-27mm. IIRC a guy measured his dvc15 driver in free air and the tinsel lead slap the cone at ~1 inch excursion (~25mm) (still safe) and the cone hit the bottom plate at around 30mm (not safe). pretty large excursion from such a cheap driver.

the sub should be safe for ~400w of power without the risk of breaking anything. yeah i did tried the smaller, previous 3cuft mfw15 at 1kw and i didn't break it. the magnet did get hot though. and i do mean HOT. got me surprised when i remove the driver after playback. probably not safe for long operation.


port size is 3 inch. 3.5" is kindda too large for me.
air velocity are kept to below 35m/s for all 3 tuning.

amplifier loading is at a maximum ~470W. will be powering them with a ~600W class d amp so it should be enough to handle the power demand.



calculated enclosure resonance (for an unbraced cabinet) are well above 120Hz. not gonna be much of a problem as it'll only be playing <120Hz.

once everything are finalized and i'm happy with the simulation & design, the panels are extended by 4mm inside sketchup.

the panels will be routed flush after gluing. i choose 4mm as 5mm is usually a bit too much wood to route and 2mm usually don't provide enough clearance.

the design are then exploded and arranged to fit inside a 4'x8' mdf sheet.

by arranging and rotating the panels you could minimize wastage and use the board to the max. you could see in this picture, there's only a couple of centimeters of unused wood.


detailed cutting guide are also extracted to help transfer the cut location on the cut panel.

dry assembly looks like there's not gonna be major problem. all the panel fit each other to within 1mm.

in this build i decided to use just one large main brace and the 4 side braces are constructed of smaller panels glued to the main brace. this will reduce the number of large panels that i have to cut and reduce waste. not to mention the possibility to use some leftover panel that i have from past build. hopefully it'll still be as strong.

first gluing process is to construct the bracing. you can see in this pic the front brace being glued to the main brace.


front baffle is next. pvc pipe is there to test fit the clearance. planning when to install the ports during the gluing process.

initial plan is to make the ports removable so that i could change the tuning by changing the port length. turns out there's not enough space to maneuver the pvc pipe. there's no way i'm gonna be able to remove the pipe once all the panel are glued. there's too much brace and too small of a cut hole.

note to self: next time plan ahead on how to remove the pipes.

been a while without any update. reason being, i'm drowning in my doctorate.
i'm about neck's deep in this debacle (which feels more like knee's deep since i'm sure my head is somewhere at the bottom haha :p)

the dual large rear sub is finished. but with a whole lot of port noise even at average excursion. baffling really, since the previous slot ported sub is simulated to have about 5m/s higher port airspeed. and i can't hear any port noise from it.

need to try and add a flare/donut to the inside port. the noise might be from the port edges or from the 90 degrees turn. if it's the edges a simple flare should be able to shut it up. praying it's not because of the bend else it'll be harder to rectify.

Very nice bro. Keep up the good work.