Push Pull Subwoofer Build?

[almikel]

9 hours ago, Primare Knob said:

500x500x500mm external as a maximum

that's 102 litres assuming 32mm sides

using a Linkwitz Transform spreadsheet

for 1 x Dayton RSS460HO you need around 10.2dB of gain to stay flat to 20Hz

for 2 x Dayton RSS460HO you need around 15dB of gain to stay flat to 20Hz (but with 2 drivers you'll need less power for the flat part above rolloff)...

...I really should learn to drive WinISD properly

 

Staying with 1 driver per box, rough calcs indicate 500W would hit your LF target

2 identical subs would be ample

 

Mike

[Primare Knob]

that's 102 litres assuming 32mm sides
using a Linkwitz Transform spreadsheet
for 1 x Dayton RSS460HO you need around 10.2dB of gain to stay flat to 20Hz
for 2 x Dayton RSS460HO you need around 15dB of gain to stay flat to 20Hz (but with 2 drivers you'll need less power for the flat part above rolloff)...
...I really should learn to drive WinISD properly
 
Staying with 1 driver per box, rough calcs indicate 500W would hit your LF target
2 identical subs would be ample
 
Mike

I am not sure if there is a flaw in your spreadsheet, when calculating with 32mm sides I come to an 82 liter of volume. 102 liters is with 16mm sides.

I think there might be a flaw in that it needs to take of 64mm (32mm each end).

With my box size I think I will roughly end up in between 60-70liter of netto volume.

But as@davewantsmore suggested, 2x12" could do the job as well, but I haven't been able to model that, as I don't know how much room gain he is expecting. At the moment I am not finding it with my current 12" subs.

I don't mind to go for a bit of overkill so that I could redo the project if my room allows for some bigger boxes in the future.

[Primare Knob]

Off topic but look at Martycube's over on AVS. I have built 4 and am extremely impressed

Thanks but that is already bigger than what I want. The smallest 3.75 is already 106, liter and they probably mean a netto volume.

[almikel]

29 minutes ago, Primare Knob said:

am not sure if there is a flaw in your spreadsheet, when calculating with 32mm sides I come to an 82 liter of volume. 102 liters is with 16mm sides.

oops - yes you're correct

[Wozza_Lee]

The microcube is a 52cm cube. Just outside of your stated size requirements

[Primare Knob]

The microcube is a 52cm cube. Just outside of your stated size requirements

The 3.75cu ft are external sizes then.

Edit: did you build it with 18mm MDF?

[Wozza_Lee]

Yeah but I did use 2 internal window braces instead of one that the design calls for

*Note I have used the martycube (61cm cube) not the micro. But reports are that the micro is probably not far enough behind to warrant the size difference

[davewantsmoore]

18mm walls would be OK, but no thinner.    The amount and design of bracing is more important.    Avoid areas of unsupported panels of larger than 15-20cm

 

3 hours ago, Primare Knob said:

82 liter of volume

I get roughly this too.   125L minus the driver, minus substantial bracing, comes out at around 80L.

 

So... 80L for 105dB@20Hz

 

80L is a good volume for the RSS460HO 18" woofer.    I set the woofer to have a high pass response of 24dB/octave (instead of the typical 12dB/octave) ..... and then I kept the SPL = 105dB, and moved the high-pass response progressively higher until the driver was under ~xmax...  and the power required for that came out at 500w.

 

The -6dB point of the SPL is 15Hz  (which is a -3dB point of ~20Hz).    I would expect you'll get all this back from room gain.... making the response measure flat in room to ~15Hz.

 

2 or 3 of these boxes, would provide a good buffer.

 

3 hours ago, Primare Knob said:

At the moment I am not finding it with my current 12" subs.

?

 

 

[almikel]

40 minutes ago, davewantsmoore said:

2 or 3 of these boxes, would provide a good buffer.

[Primare Knob]

Ok, this roughly the size of the sub including a compartment for the amplifier module, 125 litres internal without bracing, driver and 18mm walls. Sitting in the left (or right) seat makes it all a bit uncomfortable and awkward.

I think I need to scale things down a bit, or come up with a clever design. Or a new position. One at 1/4 way in at the front wall of the room, and one at 1/4 way in at the back wall of the room. This will place one sub substantially closer to the LP than the other, but was another possibility besides placing them halfway the room.

[Wozza_Lee]

How about half way front back?

I have a friend that did a TV unit that was actually a sub enclosure with 2, 15" drivers and sounds/looks great

[Nada]

 

This room has a built in gift for your subwoofer needs you might have under estimated.  

 

The spaces between the rafters in the ceiling are just begging to have tapped horns installed.  Advantages include the efficiency of the tapped horns mean good output with smaller drivers and amps with potential cost saving,  excellent sub-bass capacity and freeing up your floor space.  

[davewantsmoore]

45 minutes ago, Nada said:

rafters in the ceiling

 ???

 

46 minutes ago, Nada said:

excellent sub-bass capacity and freeing up your floor space.  

Tapped pipes do not go low unless they are very (!) large.

 

Unlike sealed boxes, its impractical to EQ them to get additional low end...... tapped horns of the size which would fit up there are going to have a strong high pass response at 40, or maybe 30 Hz.

 

 

All that being said, you could put a small one up there, and blend it with other subwoofers .... but it's not going to take you down to very-low frequencies.

[almikel]

1 hour ago, Primare Knob said:

think I need to scale things down a bit, or come up with a clever design

there are no "clever" designs with sealed subs, but the driver only cares about volume, so change the dimensions any which way that suit, and reduce the volume as much as you want knowing that you'll need more power...

...Hoffman's Iron Law - choose any 2 of 3 - small box, efficiency, low bass...

 

4 minutes ago, Nada said:

The spaces between the rafters in the ceiling are just begging to have tapped horns installed.

I'm a big fan of tapped horn subs - but I've never designed one - my T20 was designed by @Paul Spencer.

This thread has been an eye opener for me on how much power and excursion is required for sealed subs.

I've often considered taking the 2 x 12" Rythmik drivers out of my T20 sub and running them sealed - but this thread has convinced me not to...it reminded me why I went for the T20 tapped horn in the first place - deep bass at reasonable power at the expense of a large box

http://redspade-audio.blogspot.com/2012/02/how-does-sensitivity-of-tapped-horns.html

 

Careful though...tapped horns are effective only in a narrow frequency range - my T20 operates below 60Hz and works down to 15-20Hz "in room" - there's no free lunch.

 

As much as I love my tapped horn sub, there's no way it could it get above 60Hz or so to cross to your ESLs - tapped horns have a hard "upper frequency" limit where the enclosure design causes resonances that need to be avoided - my T20 has a massive resonance around 90Hz (refer the link above).

I cross my T20 sub to stereo Acoustic Elegance TD18"s for mid bass - the stereo Acoustic Elegance TD18"s have plenty of flexibility to run them down as low as required (with power and EQ) to cross to the T20 sub around 60Hz...

 

How low do your ESL's go?

 

IMHO stick with sealed subs and accept you need lots of cone radiating area and power to hit your SPL and LF targets...

...those rafters do lend themselves to a tapped horn enclosure...or limp mass treatment...or?

...perhaps an ULF tapped horn underneath the sealed subs (another Xover required)

 

Mike

 

 

 

 

[almikel]

33 minutes ago, davewantsmoore said:

Unlike sealed boxes, its impractical to EQ them to get additional low end

agreed - and could easily destroy the driver - below the tuning frequency the excursion of a tapped horn goes nuts.

37 minutes ago, davewantsmoore said:

tapped horns of the size which would fit up there are going to have a strong high pass response at 40, or maybe 30 Hz.

if you made it project down enough (below the rafters) not to constrict the flare, surely it's only about the folded length that determines how low a tapped horn could go (assuming an appropriate driver)?

 

Tapped horns will always be narrowband, so if it goes low, it won't go high - 1.5 octaves is about their max.

 

They're tricky to design, and not all drivers suit a tapped horn design.

 

As a lazy DIYer, sealed subs are quantums easier to get right than any other design - ported is hard, tapped horn harder again.

Thanks to Paul my T20 tapped horn is awesome - no way I could drive HornResp well enough to design it myself.

 

Mike 

 

[davewantsmoore]

58 minutes ago, almikel said:

surely it's only about the folded length that determines how low a tapped horn could go (assuming an appropriate driver)?

Yes, the length and the volume.    A box to get to 20Hz is going to be 300L+.... It's going to have to protrude out below the rafters quite sub-stantially, I would think

 

 

Size goes down dramatically as you move the lowest frequency up....   but it's wise to seriously consider not only the potential for annoying vibration in other rooms  ... but literal damage to the structure of the house.

[Primare Knob]

@Wozza_Lee Center sub positions aren't going to work as this will not create a nice SPL response in my room, and definitely not across the outer seating positions. At the moment with the subs halfway the length of the room I am getting the flattest SPL graph which doesn't need much EQ. Also the reponse on the outer seats are rather good and could probably be solved when I apply some EQ on each sub itself instead of a general EQ.

 

@almikel The position of LP and main speakers is creating a rather annoying null from 70-90Hz, and a massive +18dB and wide Q at 35Hz, that seems to overshadow everything up to 50Hz.  Although the woofers can go low, they aren't anywhere near subwoofer performance and the speaker cabinet isn't very rigid. With clever design I meant visual design. If big is needed could I design something that is more attractive and visually less intrusive, make it part of the room of sorts.

 

The rafters are reserved for acoustic treatment and a binary slat pattern.

[Nada]

12 minutes ago, Primare Knob said:

.....At the moment with the subs halfway the length of the room I am getting the flattest SPL graph which doesn't need much EQ. Also the reponse on the outer seats are rather good and could probably be solved when I apply some EQ on each sub itself instead of a general EQ.

 

 

What subs do you have now?

 

Are you saying all you really need is to put your current subs in the rights spots and equalise them?

[Wozza_Lee]

Coffee table sub(s) then?

Again over on AVS there is a guy running martycube (bigger than mircomarty) down firing as a coffee table on either side of his recliner couch

[Primare Knob]

 
What subs do you have now?
 
Are you saying all you really need is to put your current subs in the rights spots and equalise them?

Not directly, I have to apply an EQ cut at 100Hz as I haven't found a solution for that node yet, an EQ boost in the low region to flatten the response and compensate for the sub design and some minor EQ to smooth it all out. That is for 2 subs EQ-ed as one. In my outer left and right position there are some minor variations which probably have to be dealt with individually.

The current micro single 12" with passive radiators subs don't produce enough output to reach 105dB probably due to the placement halfway the length of the room to cure my 35Hz room node, and not adding any room gain. They also don't go down to 20Hz or below, probably due to design restrictions and filters by the sub.

[davewantsmoore]

13 hours ago, Primare Knob said:

The position of LP and main speakers is creating a rather annoying null from 70-90Hz, and a massive +18dB and wide Q at 35Hz

Post data.  They cannot be creating a +18dB (that's impossible) - something else is going on.

 

 

Here is a chart you sent me, that doesn't show what you're talking about  (so am confused).

 

11 hours ago, Primare Knob said:

Not directly, I have to apply an EQ cut at 100Hz as I haven't found a solution for that node yet

Have you confirmed that 100 Hz is not SBIR?    (It's probably not, but you must check).     Assuming it is a mode you can just -6dB EQ it .... but remind us where you plan the low pass filter?

11 hours ago, Primare Knob said:

and not adding any room gain

?

"Room gain" is the same where ever you place the woofer in the room.

 

 

[Primare Knob]

@davewantsmore That graph was from my subs and not my main speakers, see pic. The peak at 35Hz is from my room length node, did also also appear on the subs graph when I had them placed in the front of the room. The whole reason for the subs being placed half way the length of the room is to cure this node.

 

Up to now I have just been tweaking the cross over settings/ HP LP filters to get a smooth response, and probably aren't the best as I have a slight rise in group delay around the cross over frequency that I can't get rid off by adjusting delay time of the subs. Cross over is set at 100Hz for all speakers at the moment, but that might change if acoustic treatment is able to solve some issues.

 

It seems that it has fixed some of the troubled areas of the main speakers as there is no longer a null between 70-90Hz.

 

Yes the 100Hz is a room node, 2nd width, as it appears on all speaker graphs

[davewantsmoore]

2 hours ago, Primare Knob said:

@davewantsmore That graph was from my subs and not my main speakers, see pic

Right, I see.

Quote

 

 The peak at 35Hz is from my room length node

The peak from the room can really only be +6dB.

 

In this chart, I would suspect you have some sort of bass EQ, lifting the response below about 60 or 80Hz.

 

Quote

Yes the 100Hz is a room node, 2nd width, as it appears on all speaker graphs

It can only be ~+6dB ... so it is likely a combination of effects you're seeing in this data.    More charts of things in different places could start to separate them out.

[hollowboy]

10 minutes ago, davewantsmoore said:

It can only be ~+6dB ... so it is likely a combination of effects you're seeing in this data.

This is my go-to page for how much the room can change LF response.   Nice graphs, lots of detail about the test conditions. Graphs 3-6 show some 12dB peaks vs the same driver outdoors.

https://data-bass.com/data?page=content&id=80#!prettyPhoto

 

Maybe you mean it is the average gain that's capped at 6dB?

[davewantsmoore]

2 hours ago, hollowboy said:

This is my go-to page for how much the room can change LF response.   Nice graphs, lots of detail about the test conditions. Graphs 3-6 show some 12dB peaks vs the same driver outdoors.

That is a different proposition, that comparing.

 

1.  The driver inside a ficticious room which has "room gain"... but no room modes.

2.   The same as #1, but with modes.

 

The effect of the mode will give +6dB.

 

What you're looking at is outdoors (no modes, no room gain) ..... vs in a room (modes and room gain operating together.

 

2 hours ago, hollowboy said:

Maybe you mean it is the average gain that's capped at 6dB?

No - I mean the gain from a modal peak alone.