Pair of small subs, TH?

Hey guys,

I'm thinking of making a pair of small subwoofers, for a bedroom system. My mains are a pair of Fostex FE126eNs, driven by either a 6EM7 SET amplifier, or an amp6-basic from 41Hz. The Fostexes cut off around 80Hz, so that'd be an ideal starting point.

I have available 4 ofthese, so I'd preferably use 2 per side. I know they may not be ideal, but if I could get something decent out of them, that'd be good.

I have drivers such as the TB W6-1139SG available to me, too (though at £70/pr, they'd need to be right for the job, not just something that might work).
Anyway, I'd like them to be ~700mm tall, less than 300mm wide, 300mm deep at a maximum. I'd like extension to cover 30Hz comfortably, SPL isn't a huge issue, as it'd need to keep up with a practice guitar amp (played at sane levels), not much more, in a small room (2.3mx2.8mx3.4m hwd). The speakers are ~half way along the longer wall, ~1.5m apart.

I've never built a tapped horn before, but I think it'd be something interesting to try.

I've tried Hornresp a few times, but never been able to get anything remotely usable out of it. If someone was to post sims, a brief explanation of the reasoning behind it would be much appreciated. I've read through a few tutorials, but none of them have explained it as you would to a teenager such as myself.

Nearly forgot - the amplifier would be initially a 25w/ch stereo amp (rated, before I upgraded the PSUs), bridged with a large PSU (2x200VA transformers, 2 bridge rectifiers and 18000uF of capacitors for each rail ~+/-34v - basically 2 near identical PSUs in parallel). The plan is to get a more powerful (likely switching, likely stereo) amplifier later on.

Musical interests vary, but I've tracks that go to 30Hz, so that's the initial target. The end result should beat a pair of these in a 40L cabinet with Linkwitz to get an f3 of ~28Hz (severely SPL limited in the lower regions). In reality, they go into harmonics before reproducing such frequencies at a decent level, so an upgrade is in order.

Anyway, if you've read this far, thank you very much. I'm looking forward to suggestions.

Chris

PS - the wood will likely be CNC cut, ply or MDF. The latter is preferred due to budget reasons...

Edit - the cabinet could be physically deeper, to maybe 500mm at a push.
 
Last edited:
..But I'd like to know if I can get decent results with the Visaton drivers I already have..

Hi, Your room is small and if comes with one closed door and widow: The passive 'room gain' could be wide-band and large and low end contributions from the Visaton's are IMO more than enough: See the first picture.

FYI, a couple of simulations:

b:)
 

Attachments

  • Criss661-room.JPG
    Criss661-room.JPG
    95.2 KB · Views: 696
  • Visaton-W130S8.JPG
    Visaton-W130S8.JPG
    760.7 KB · Views: 696
Bjorno, thanks for that.

A couple of things...

1) - is there any chance they could go lower? I know there's a few conflicting rules about how low THs can go. Some say half Fs, others say 0.7*Fs... Anyway, ideally I'd be able to use them for movies, too, so I figure the extra extention could be useful.
2) - the graph for the Tang Band (that I posted above) looks less-than-optimal... I've looked around a bit, but can't seem to find anything suitable. Maybe I'm just looking in the wrong places... What would be your opinion of making the one I listed, then playing with eq'ing out the 25Hz peak?

In think in the end, I'd like to mess around with some bonkers drivers, so the TBs are likely to be the one's I'll get...

Cheers again,
Chris
 
Disregard the above...

I'm going to build the recommended Th for the Visatons.

If it works, then I've spent a minimal amount of money for something decent. If it doesn't work so well, I haven't wasted much money (just the wood).

So, a couple of questions on that design.

1) The schematic shows only one driver. Would I be right in thinking that I could simply double the area for the second driver?

2) How should I go about folding this design?

I think I've worked out what all the numbers mean. It's not L23 (as in twenty-three), it's L Two-Three, ie, the distance between the indicated areas of S2 and S3.

3) I suspect I ought to use a double folded design (look at me, using the proper language...), as a single fold would result in a design that's too tall (Mum already complains occasionally)... Am I over-complicating things, or is this feasible?

4) (and it's only a minor one, I'll probably speriment anyway), would stuffing/lining be needed anywhere? I've seen some with lining on the folds, others without.


Thanks all,
Chris
 

tb46

Member
2006-01-09 7:04 pm
Texas
Hi chris661,

As to the Visaton W130S: the old way of specifying Xmax=(Winding Height minus Front Pole Plate Height)/2, in other words 2.25mm, and even the Excursion limit (Xmech?=8mm) is not very big. I feel that this may be a little limiting for a sub. It is real easy to bottom out such a sub, even when the overall levels are quite benign. Bye the way, just in case you are not aware of it: bjorno's Rec. enclosure (on the left) is not a tapped horn.

The best way to learn about tapped horns I can think of is to take known working designs and plug them into Hornresp, then make changes and note the results. Take bjorno's tappedhorn (the one on the right), and change it into a four section model, and play around with it until you have a flat response from 34 to 100 Hz, or, for example look at (admittedly, one of my favorite threads) geitmans' thread on a dual 8" tapped horn:

http://www.diyaudio.com/forums/subw...ped-horn-th-spud.html?perpage=25&pagenumber=1

,here you have a Tom Danley design, and a good example of how to build, and measure a dual tapped horn; then plug that one into Hornresp, and see what you can do with it to make it smaller, how different drivers affect the response, etc..

Anyway, its the journey ....

Regards,
 
I had failed to take into account that the sims posted above were the maximum SPL the drivers would do, while the 25Hz TH is at 1 watt, so the driver would have plenty of excursion reserves. This puts a new light on things, so I'm unsure now of what to do.

Despite my protests, Hornresp isn't giving out the same graph as bjorno's recommended design, though I've checked all the input parameters twice. My design gives a peak at ~32Hz before dropping off. Changing the design to look like the schematic helps a little, though the cutoff has gone up to 40Hz.

confused...
Chris
 

MaVo

Member
2006-01-07 12:40 pm
Hi, Your room is small and if comes with one closed door and widow: The passive 'room gain' could be wide-band and large...

b:)

Or you could say it this way: Your room is small and has little leaks, the resulting room modes will lead to deep notches in the response, lowering the usable output by a large amount (what use have those pesky high q spikes in the response, that look like room gain with a low enough frequency response resolution, if the nulls inbetween are deep?). Now, whats the truth? If it lies somewhere in between, then that means you could simply ignore room gain and get reasonable results. The principle at work here is: Never be too optimistic, concerning the results of your work, or you may have to redo it completely. Thats why one should think of room gain as a nice addition, but not as granted.

While you are in the planning stage, i highly recommend googling "harman white paper" and look at the subwoofer placement ones. Great stuff to know, easy to implement, makes the resulting speaker setup sound much better.
 
I've heard great tales of room placement making or breaking a subwoofer.

As this is a bedroom system (read, has furnature in, including desk, drawers, wardrobe, bed, etc), it's very difficult to play with placement. These subs are going to act as stands, so they're unlikely to able to be moved far at all.

I know my room's not going to be great, and as I'm still (after all these 16 years) living with my parents, room treatment will be off the agenda.

It's a shame, I know. Ideally, I'd borrow a pair of ideal, 20Hz subwoofers, run some sweeps and figure out where the room causes problems, and find a way of eq'ing it to death. This is unlikely to happen, so I'll do the next best thing - make my own subwoofers, and hope that, by being a distance apart, they'll iron out most of the dips themselves. If there's anything major left, I'll add some eq on the input of the subwoofer amps.

Chris, now fairly set on the TB subwoofers.
 
Well then, today I received the pair of TB W6-1139SG drivers, and the stuff to make an AMP1B.

The drivers have been bedding in a little ("lets see how far these things really move" - the answer was very far), including a brief stint in the cardboard postage box they arrived in. The rubber surround seems pretty stiff, so I'm giving that chance to loosen off. For the most part, though, they were sat on the table, rattling stuff. Awesome.

Currently awaiting a forum login confirmation at 41Hz so I can access the building instructions. Still, there's some pretty tiny SMDs included, which may make things interesting, with such a large soldering bit.
Anyway, when things get going, I'll post some pictures up. Woodworking will likely happen later this year, or in the new year. I remain optimistic.

Chris

PS - if there's spare room on the sheet(s) of wood, then I'll try the drivers in a small ported box, just because.
 
Last edited:
room treatment will be off the agenda.

not necessarily. there are several homebrew treatments that could be incorporated into even a small bedroom that you might try. I don't have links to them here at work but I'm sure someone will chime in and save you a search. Since the labor is free and materials are dirt cheap(and re-usable), you can hardly put off this learning opportunity.
 
Back again with some pictures.

I've not started gluing everything together yet, but a dry-fit and some hole cutting has taken place...

[IMGDEAD]http://sphotos.ak.fbcdn.net/hphotos-ak-ash1/hs794.ash1/168493_1700882116694_1077163384_2342433_6699536_n.jpg[/IMGDEAD]

[IMGDEAD]http://sphotos.ak.fbcdn.net/hphotos-ak-ash1/hs760.ash1/165169_1701856821061_1077163384_2343564_6724400_n.jpg[/IMGDEAD]

Apologies for the picture quality, still using the iPod camera.

To make a removable back, I've doubled the panel, plan to rebate one piece and to fit a frame inside the subwoofer to create something to screw the panel to...
I haven't explained that very well, so I shall post pictures to illustrate when I can.

Chris
 
Reading through the thread I on the AVS forum, the concensus was that the lining made minimal if any difference, so I left it out.

One up and running now, certainly an interesting experience building the thing.
It's currently hooked up to an old PA amp, as the Tripath amp has gone nowhere.

So, sounds strange to me. Haven't heard anything like it before, so I'm still getting used to the sound. I'll post proper impressions later, but it goes low and loud (100w amp used)...
 
Reading through the thread I on the AVS forum, the concensus was that the lining made minimal if any difference, so I left it out.

One up and running now, certainly an interesting experience building the thing.
It's currently hooked up to an old PA amp, as the Tripath amp has gone nowhere.

So, sounds strange to me. Haven't heard anything like it before, so I'm still getting used to the sound. I'll post proper impressions later, but it goes low and loud (100w amp used)...

Sorry I missed your thread...

Think I know that design....:D

Anyhow - my measurements suggested that the lining makes little difference, so, in my opinion adding it is not worth the added effort.

Can you measure impedance? Air leaks in the wrong spot can kill the performance of these. SPL measurements will also indicate if something is amiss.

How did you seal the driver to the baffle?
 
Ah yes, had lots of fun with air leaks.

Used a foam gasket to attach the back, screws every ~4". The driver was sealed using draught excluder, with some overlap (went around the circumference, then a bit).
Before adding those, there wasn't much output. Now, if nothing else, it'll flap trousers at ~26Hz.

The sound seems better now. I haven't had a sub in use for months, so it's a case of re-acclimatising (if I spelt it right). Very impressive for such a small speaker, though I keep looking at the size of the box and counting the number of 12" drivers I could stick in the side...
Still, I'm a teenager (nearly 17 :D), so spending lots of money on drivers won't be a possibility any time soon.
I can measure impedance (well, AC current - easy to see where the impedance peaks are though), and shall do that at some point.
For now, I'm just enjoying it.

Quick playing around with a signal generator on my iPod Touch (because I have that sort of thing on my iPod), there's definitely output down to 30Hz. After that, my hearing falls off so quickly that it's barely sensation. I have some over-ear headphones that won't play 25Hz loud enough for me to hear it, they just bottom out :D.
Anyway, there's still enough air movement at the mouth to feel it down to 22Hz, but bad noises have started by then.

With regards
Chris

PS - in my attempts to find something it won't do annoyingly well, I hit it with 15 Step by Radiohead. Not a problem there.
The only thing left is Bass I Love You, but that's just cruel...
 
Last edited:
Ah yes, had lots of fun with air leaks.

Used a foam gasket to attach the back, screws every ~4". The driver was sealed using draught excluder, with some overlap (went around the circumference, then a bit).
Before adding those, there wasn't much output. Now, if nothing else, it'll flap trousers at ~26Hz.

The sound seems better now. I haven't had a sub in use for months, so it's a case of re-acclimatising (if I spelt it right). Very impressive for such a small speaker, though I keep looking at the size of the box and counting the number of 12" drivers I could stick in the side...
Still, I'm a teenager (nearly 17 :D), so spending lots of money on drivers won't be a possibility any time soon.
I can measure impedance (well, AC current - easy to see where the impedance peaks are though), and shall do that at some point.
For now, I'm just enjoying it.

Quick playing around with a signal generator on my iPod Touch (because I have that sort of thing on my iPod), there's definitely output down to 30Hz. After that, my hearing falls off so quickly that it's barely sensation. I have some over-ear headphones that won't play 25Hz loud enough for me to hear it, they just bottom out :D.
Anyway, there's still enough air movement at the mouth to feel it down to 22Hz, but bad noises have started by then.

With regards
Chris

PS - in my attempts to find something it won't do annoyingly well, I hit it with 15 Step by Radiohead. Not a problem there.
The only thing left is Bass I Love You, but that's just cruel...

The latest version of REW v5 Beta will do nice impedance sweeps if you make up a set of test leads. That is all explained in the program's online help file.

As far as Bass I Love You - that is beyond cruel. They're just little woofers....:D

Seriously - they don't do much below 25 Hz. Above 25 Hz, for what they are, they do just fine.
 
It'd be much quicker to use an AC ammeter (because it takes 2 seconds to set up), so I'll try that first. I'll record the input voltage (maybe use a scope for that, leave it connected), then watch the current as I change the frequency (Hz at a time...).

If that doesn't give good results (ie, no real pattern emerging, just a wiggly line), I'll try REW.
Cheers,
Chris
 
Last edited: