Open Baffle Desktop Speakers

[borispm]

Hi guys,

I just came up with an idea of utilizing a Tangband W6-789E woofer and a Tangband W3-1318SA full range driver for making a pair of small open baffle speakers, but there's a problem I can't really find a way to solve, and that is the crossover problem.

The baffles will be 8" wide, so I expect the woofers, without filter, to start its 6dB/Oct roll off at 570Hz (-3dB point). To compensate this roll off, I can use an inductor (maybe around 12mH) to low pass the driver at 6dB/Oct after around 100Hz. With the inductor in place, I will get a 6dB/Oct roll off above 570Hz.

Assuming the full range driver is capable of running open baffle without major problems and roll off at 570Hz (-3dB point) too, does that mean I do not need to high pass the full range driver? Will I get a 3dB peak at 570Hz? It seems that the -3dB point of the filtered woofer roll off is above 570Hz, according to my very simple graph plotting... Should I roll off the full range with a cap?

I know a 3dB peak is unlikely to be that disturbing as the tolerance of most speakers out there can do a lot worse, but I want to be engineeringly correct... Do you think I need an LRC notch filter to iron that out?

Please point out any mistake. Thanks.

 

[borispm]

Opps, I'm sorry... It shouldn't be a 570Hz but somewhere higher, I suspect it's around 850Hz as it seems to be half an octave higher.

Any input, guys?

Thanks.

 

[Chaz H]

I'll bet gainphile can help you; he's all over the desktop, nearfield open baffle as I recall. See for example:

http://www.diyaudio.com/forums/mult...design-using-seas-8-tang-band-full-range.html

 

[60ndown]

if i were at a desk any amount of time a month id want great sound there.

 

[borispm]

Hi Chazdrumzalot,
Thanks for giving me the link. I was actually inspired by Gainphile's design but I don't want to mess up with bi-amp, equalizer and electronic XO, so I am going for a passive version. I am not so familiar with asymmetrical slope XO so I am not sure whether I should add a high pass to the full range driver or not...
Thanks anyway.

 

[picowallspeaker]

I suggest you to try a push-pull of two TB W6-789 with a regular box ( even the OB baffle is too tiny ) and the little FR on top . P-p will help in making the low-pass coil littler (mutual movement conditioning ,same mass ! ) and the little FR will follow its natural rising response with an easy HP .

 

[borispm]

Hi Picowallspeaker,
Will I end up with asymmetrical slopes between the high pass and the low pass?
And I can't really understand why I can make the low pass "littler". Will I trade bass extension for a "littler" low pass?

Thanks.

 

[picowallspeaker]

Because in a push-pull ,the woofers are tied together but the air mass in between can couple to the membrane movement only at low frequencies .
Plus , it requires a 'slimmer' inductor for a 4 Ohm load !

 

[borispm]

Picowallspeaker,
I am sorry I still don't quite get it... My fault for being dumb in physics, would you mind telling me more about the benefits of push-pull dipole and why it is better? Some guys at the OB Gallery in the Multiway side said it is a bit stupid to get push-pull dipole (due to the cost?)... I don't really know.

Thanks.

 

[picowallspeaker]

No ,no ,I was talking about two woofers in a regular box ,vented on the back , with the woofers positioned nose-to-nail (in axis ) or with magnets sharing the same plane ,but with one cone firing frontally and one firing in the vented volume .
For an OB you'd need bigger woofers and larger baffle .

 

[borispm]

That would be a normal isobaric setup, right? But It seems the woofer needs almost 40L to breathe, even if I use isobaric, I still need about 20L... That would be killing the whole idea of having a desktop speaker as my desk is not that deep... It's just about 60cm deep, and I am planning to get some open cell foam on the wall to kill the loading effect...

 

[picowallspeaker]

In a misaligned magnet-to-magnet p-p arrangement, box depth would be approximately driver depth X 2 .
Do you want my© design ?

 

[borispm]

Hi picowallspeaker,

Of course I would like to see your design, but the problem is still there...
Space is the biggest problem here. Putting a pair of 30L+ cabinet is almost suicidal... If I make them thin, they must be either very wide or very tall, that isn't really a good idea, especially for a small room...

 

[borispm]

BTW, one update to this current project:
First, I think I am going to downgrade the W3-1318A to W3-871SC or W3-1319A due to the cost and the similar level of performance. These drivers are more affordable as an experimental build.

Second, I am playing around with the idea of offering these speaker systems as DIY kits if they happen to sound great (which I guess would be). I hope I can build them this summer... I will post some Sketchup models later.

 

[borispm]

Finally a simple sketchup model

Hi guys,

I have finally finished my simple sketchup model. All the boards shall be screwed in place instead of using PVA glue, just like the Tangband DIY Kits.
Still figuring out where the large 12mH laminated core inductor should be placed...

 

[DYNABLASTERTUNERS]

don't take the W3-871, for me it's one of the worst TBs, it has no TB magic

 

[borispm]

Hi Dynablastertuner,

The silver cone W3-871 was praised by a lot and it seems many still miss it (I think it's still available directly from Tangband). How would you describe the sound of it? Are there other Tangband alternatives that you would suggest?
Thanks.

 

[chris661]

Hi guys,

I just came up with an idea of utilizing a Tangband W6-789E woofer and a Tangband W3-1318SA full range driver for making a pair of small open baffle speakers, but there's a problem I can't really find a way to solve, and that is the crossover problem.

The baffles will be 8" wide, so I expect the woofers, without filter, to start its 6dB/Oct roll off at 570Hz (-3dB point). To compensate this roll off, I can use an inductor (maybe around 12mH) to low pass the driver at 6dB/Oct after around 100Hz. With the inductor in place, I will get a 6dB/Oct roll off above 570Hz.

Assuming the full range driver is capable of running open baffle without major problems and roll off at 570Hz (-3dB point) too, does that mean I do not need to high pass the full range driver? Will I get a 3dB peak at 570Hz? It seems that the -3dB point of the filtered woofer roll off is above 570Hz, according to my very simple graph plotting... Should I roll off the full range with a cap?

I know a 3dB peak is unlikely to be that disturbing as the tolerance of most speakers out there can do a lot worse, but I want to be engineeringly correct... Do you think I need an LRC notch filter to iron that out?

Please point out any mistake. Thanks.

Hi,

The reason you have a peak in output at crossover frequency is the speakers are still working at full volume at the crossover frequency, so you'll get more output (because both are contributing) that elsewhere.
The solution: move the crossover frequencies apart a little, so the responses of each driver are ~3dB down when they combine. That'll give you a much flatter response.
If you wanted to get the woofer to roll off faster higher up, you can add a capacitor in parallel (after the inductor). You'll need to play around with the value of the capacitor to find one you like, but I'd expect 33uF would be a decent start: it'll begin to roll the woofer off around 500Hz.

The sketchUp model looks good... Go for it, let us know what you think.

Chris

 

[borispm]

Hi Chris,
Long time no see!
Thanks for your advice. I think I should try to roll off the woofer faster too, but I'll be dealing with asymmetrical slopes, which worries me a bit as I have absolutely no experience in that (and that's why I seldom read stuff at the multi-way forum!;-))

If I have a first order XO at 100Hz, and by adding the cap in parallel to the woofer after it, will that be a first order slope at 100-500Hz and a second order slope above 500Hz, or will I end up with a second order slope in the middle instead of 2 slopes? That's troubling me.
Can someone explain it?
This foreign policy stuff is a little frustrating...
I think I am unprepared for war....
I know the woofer and full range driver can coexist peacefully...

I know there's another way of dealing it, that is to have a tapered baffle, but that will be even harder to work with... Duh...

Thanks, and looking forward to seeing more replies and aids.

 

[DYNABLASTERTUNERS]

Hi Dynablastertuner,

The silver cone W3-871 was praised by a lot and it seems many still miss it (I think it's still available directly from Tangband). How would you describe the sound of it? Are there other Tangband alternatives that you would suggest?
Thanks.

here is my description

http://www.diyaudio.com/forums/full...ver-works-best-bookshelf-sized-speaker-2.html