Small Size Open Baffle Question

[Gunawan W]

Hi all,
this is my first post, after 14 days searched and compared the best speaker enclosure in the Web, finally I choose open baffle system for my first project, you all know why i choose this type!
My goal is: make open baffle system simple and small enough to produce sound from Fs=90Hz above and subwoofer to handle below 90Hz.

My plan is using Fostex allrange driver FE 164 with baffle size: 40x40cm.
I run the simulation using xbaffle.xls written by Thorsten Loesch and the result is: -6dB at 200Hz.

To flatten FR, is it right using passive filter with 6 dB step reduction for freq. above 200Hz? as John L. Murphy said? http://www.trueaudio.com/st_diff1.htm

How about folded baffle, like this: \_/ seeing from top, with 15cm for front baffle and 15cm for both folded wing with 135degrees angle? Is it helping to shift down -6dB from 200Hz?
is there any formula to calculate folded baffle?

I open for your suggestions to make the best of it.
Thank's guys.

Gunawan.

 

[norman bates]

there's a couple more things........

Firstly I assume the baffle is 2' off of ground and 2' from side wall and 5' from front wall.

Doing this will lessen the boosting above 100-200 range, making for slightly deeper response (using program). (100-150hz closer to average output)

Using the baffle step filter changes the qes of the driver (and thusly the qts)........ because at low frequencies the inductor is in series with the driver.........
To have a baffle correct that low, the inductor may be 3 or 4mH, these have > .7 dcr (adding a .7 ohm in series or more)................ Which raises the q......... You can find the formulas online (I'm not at work and won't have them for a while)


I've not found much on baffle step on an open baffle, but I've found it is less than a normal baffle...........

Try and post results !!!!!!!!!!!!!!!

Good luck !!!!!!!!

 

[ThorstenL]

Konnichiwa,

Hi all,
this is my first post, after 14 days searched and compared the best speaker enclosure in the Web, finally I choose open baffle system for my first project, you all know why i choose this type!
My goal is: make open baffle system simple and small enough to produce sound from Fs=90Hz above and subwoofer to handle below 90Hz.

Look for a driver with an Fs a little below 90Hz (say 75-80Hz) and a Qt of 0.7-1.0. Then make the baffle at least 35cm wide and extend the baffle to the floor, should work fine.

My plan is using Fostex allrange driver FE 164 with baffle size: 40x40cm.

NEVER, EVER under ANY circumstances make the baffle square or round.

Seconly, the FE164 is not suited to open baffle operation nor is the current replacement FE 166 (the 164 is discontinued). The FE164 has a Qt of around 0.35, which means that teh driver is suited to neither open or infinite baffles, if flat response is desired.

To flatten FR, is it right using passive filter with 6 dB step reduction for freq. above 200Hz?

This may work up to a certain point but the drivers has little excursion capability and thus would likely overload if corrected electronically.

How about folded baffle, like this: \_/ seeing from top, with 15cm for front baffle and 15cm for both folded wing with 135degrees angle?

This will likely appear to have a response similar to a 20cm wide plan baffle. please try it yourself.

is there any formula to calculate folded baffle?

Not really as the system is fairly complex. But the LF output is similar to that from the front baffle plus the addition of the actual time the sound takes inside the reac cavity to reach the termnus of this short "transmission line".

I open for your suggestions to make the best of it.

Simple, use a driver suited to open baffle operation, for example the Diatone 6.5" or the Pioneer PM-16 seem suited.

Sayonara

 

[Gunawan W]

Hi,
thanks for your quick response.
Norman, you said 5' from front wall or back wall? actually 5' from back wall is to far for my room setup, I only get 2' max. What search words should I try in Google to find the formulas you mention?

KYW, to raise Qt, somewhere in this forum you said:
"Drive it from a high source impedance. The easiest is to just use a resistor, but that looses a lot of power. The smarter way is to build an amplifier with a controllable positive output impedance , something that is laughably simple."
Can you show some schematic diagram how to do "a controllable positive output impedance"?
Is it possible to use 4 Ohms dirver with 8 Ohms amplifier to raise Qt?

You said that if I use folded baffle with total width 45cm, this will have the same FR as 20cm wide straight baffle? So additional 25cm wide has no effect?
I will do this project ASAP (time constraint), but in the mean time would you mind if I ask some more question?

How about using 15cm front baffle with 30cm each folded wing at 90degrees angle, so the total width is 75cm, is this will get some FR change? I ask this question is to keep the baffle as small as possible from front view.

If I make like this (top view) _|_|_ it's like 90degrees folded baffle with extended front baffle at both side, is this will extend the lower part of FR?

Diatone 6.5" or the Pioneer PM-16, do you use this driver yourself?

If any of you use this driver with great result, please let me know.

Thank you.

 

[ThorstenL]

Konnichiwa,

Norman, you said 5' from front wall or back wall? actually 5' from back wall is to far for my room setup, I only get 2' max. What search words should I try in Google to find the formulas you mention?

I believe Norman comments on the use of my spreadsheet (Xlbaffle.xls). This is still not as accurate as I wish, it was allways a first approximation based on matching observations. Very close location to walls overstates the LF increase.

I have sicne fully understood how Dipoles (open baffles) interact with the room and will update this section if and when I find time.

KYW, to raise Qt, somewhere in this forum you said:
"Drive it from a high source impedance. The easiest is to just use a resistor, but that looses a lot of power. The smarter way is to build an amplifier with a controllable positive output impedance , something that is laughably simple."
Can you show some schematic diagram how to do "a controllable positive output impedance"?

http://sound.westhost.com/project56.htm

Is it possible to use 4 Ohms dirver with 8 Ohms amplifier to raise Qt?

With most amplifiers the use of a 4 Ohm driver with an amplifier rated for 4 ohm load will not raise the Qt. Adding a series resistor in effect reduces the output at frequencies where the driver has no resonances while it does not reduce the output as nmuch around the drivers resonance. It means you will never get more SPL at resonance, you only get proportionally less SPL above resonance.

You said that if I use folded baffle with total width 45cm,

I said if you use a 15cm baffle with folded back wings.... The point is that folded back wings CANNOT be counted as simply making the baffle wider.

So additional 25cm wide has no effect?

Not quite. What you get is a 15cm wide baffle and a "transmission line" a little under 15cm in length. This transmission line will delay the rear output a little between driver and terminus (rear opening), leading to slight reduction in the F3 point. The key is to understand that a plan baffle has effects well past those caused by the delay of the soundpressure folding around it, when it comes to radiating sound.

How about using 15cm front baffle with 30cm each folded wing at 90degrees angle, so the total width is 75cm, is this will get some FR change?

Somewhat, but still not much.

I ask this question is to keep the baffle as small as possible from front view.

Then the 15cm is more or less what you get. There are two mechanisms at work with a baffle with folded back wings. One of them is that which is accounted for in my "Xlbaffle.xls" spreadsheet, the other is the transmission line effect.

For this effect to become relevant you need a fairly long Transmission line or rather a fairly deep box (still open back), which brings it's own problems with it. An interesting (and working) application of a folded back baffle is the Rondo from Auditorium 23 in Germany and the Shindo and Auris speakers which use the same kind of design principle.

If I make like this (top view) _|_|_ it's like 90degrees folded baffle with extended front baffle at both side, is this will extend the lower part of FR?

Why not leave the folded back parts out and make the baffle wide enough to work?

Diatone 6.5" or the Pioneer PM-16, do you use this driver yourself?

No, I use the Supravox 215 Signature Bicone, a driver expressly designed for use on open baffles. However these are over $ 300 each and the Pioneer (currently produced) and the Diatone (only old units via E-Bay and other scalpers).

The Pioneers are available from here:

http://www.eifl.co.jp/index/export/pe16m.html

More details on this driver:

http://audionautes.com/Altriprodotti/pioneer/datape-16m.htm

From Eifl for 250 Bux per pair, that said, if it was my money I'd go Supravox and pay the balance.

Sayonara

 

[Gunawan W]

Hi KYW,
wow, thank's for the reply, I should chewed all this information not to fast to prevent chocking.

"With most amplifiers the use of a 4 Ohm driver with an amplifier rated for 4 ohm load will not raise the Qt."
How about: 4 Ohms driver with amplifier rated for 8 Ohms?

So with folded baffle, the result is more complicated and unpredictable?
Like my first OB speaker test below:
I use Kenwood 5" driver taken from my existing system with fs=74Hz and Qts=0,41.
I made 90degrees folded baffle with front baffle wide=15cm, folded wing=25cm each, baffle high=60cm. It placed at 100cm measured from driver centerline to the floor, 50cm from side wall and 60cm from back wall.
No eq. on preamp and amplifier.
Quick hearing result from 1m distance is like this: begin hear sin wave sound at 65Hz and raised to half loud at 78Hz then rolled down and barely hear at 140Hz then raised again and at 250Hz full loud - flat - up to 10KHz and rolled down again and no sound at 16KHz.
It's different from your xlbaffle.xls spreadsheet result.
There was a notch at 140Hz, is this only happen on folded OB or normal for all type of OB?

Any suggestion to improve this, please?

Thank you.

 

[ThorstenL]

Konnichiwa,

"With most amplifiers the use of a 4 Ohm driver with an amplifier rated for 4 ohm load will not raise the Qt."
How about: 4 Ohms driver with amplifier rated for 8 Ohms?

Same thing. Mosty amplifiers are designed for Zout >> 0Ohm (Output Impedance approaches 0 Ohm).

So with folded baffle, the result is more complicated

Yes.

and unpredictable?

No, it is quite easy to predict, but requires a different math than that in my spreadsheet. Simply "unfolding" the baffle and entering that number in my spreadsheet gives highly unrealistic predictions, suggesting much lower reaching bass than present in the implemented speaker.

I made 90degrees folded baffle with front baffle wide=15cm, folded wing=25cm each, baffle high=60cm. It placed at 100cm measured from driver centerline to the floor, 50cm from side wall and 60cm from back wall.
No eq. on preamp and amplifier.
Quick hearing result from 1m distance is like this: begin hear sin wave sound at 65Hz and raised to half loud at 78Hz then rolled down and barely hear at 140Hz then raised again and at 250Hz full loud - flat - up to 10KHz and rolled down again and no sound at 16KHz.
It's different from your xlbaffle.xls spreadsheet result.

You must understand that my Spreadsheet attempst to account for the radiation angle, but NOT for room resonaces.

I suspect the 65 & 78 Hz may very well be room modes (check dimensions). Using your parameters I agree with around 250Hz as the point where the speaker works properly, below this it rolls off and any output below around 100Hz would be much shelved down (the "wings" do provide some extension to the LF), possibly enough to partially energise room modes.

There was a notch at 140Hz, is this only happen on folded OB or normal for all type of OB?

This happens with most speakers (floor notch) and how deep it is depends upon listener position. You really need descent measurement gear using non-sinewave signals to get a handle on what the speaker does.

Any suggestion to improve this, please?

Use a baffle 60cm wide and 1m tall, standing on the floor. Narrow baffles do not work, not only in dipoles, but equally in boxes. Narrow baffle boxes invariably sound anaemic in the lower midrange even IF the on axis response is corrected.

Sayonara

 

[Sheldon]

How about using 15cm front baffle with 30cm each folded wing at 90degrees angle, so the total width is 75cm, is this will get some FR change? I ask this question is to keep the baffle as small as possible from front view.

So why not fold the side panels back when not playing and bring them out when playing? You don't really lose anything with this design, and you can play with the fold angles as much as you wish. You can also make the folding panels with clear plastic (or tempered glass) to reduce the visual space impact.

Sheldon

 

[Rudolf]

Question for Kuei Yang Wang

Originally posted by Gunawan W:
How about folded baffle, like this: \_/ seeing from top, with 15cm for front baffle and 15cm for both folded wing with 135 degrees angle?
Answer by Kuei Yang Wang:
This will likely appear to have a response similar to a 20cm wide plan baffle. please try it yourself.

Hello Kuei Yang Wang
I have done a quite similar folded baffle for myself: 10 cm frontbaffle and 22 cm for each folded wing (45° back). I believe the effective width is much more than the 40 cm projected width, as seen on axis.

Are you aware of
www.linkwitzlab.com/frontiers.htm#L
and the discussions with John Kreskovsky on the madisound forum about his U-frame dipoles?

I´m with you that we can´t simply add the length of the folded back baffles to the baffle width, but the delay caused by the "depth" of the baffle is something that must be accounted for.

Anyhow, I´m looking forward to your new evolution step of the dipole spreadsheet.

Rudolf

 

[ThorstenL]

Re: Question for Kuei Yang Wang

Konnichiwa,

I have done a quite similar folded baffle for myself: 10 cm frontbaffle and 22 cm for each folded wing (45° back). I believe the effective width is much more than the 40 cm projected width, as seen on axis.

Maybe so, but it remains much less than that of a 54cm wide plan baffle (done the experiments).

Are you aware of

Yes.

I´m with you that we can´t simply add the length of the folded back baffles to the baffle width, but the delay caused by the "depth" of the baffle is something that must be accounted for.

Absolutely. With your "wings" at 45 degrees the actual depth is 15cm, causing a delay of 0.43mS. This will be added to the projected 40cm (1.15mS), giving a cancellation at around 630Hz, which means it actually does not even feature in the LF behaviour of the dipole.

Anyhow, I´m looking forward to your new evolution step of the dipole spreadsheet.

This strictly applies to the room interaction of a baffle standing on the floor. I will not bother with any folded baffles, you get close enough assuming a slightly wider baffle than that presented by the unfolded part using my spreadsheet.

Sayonara

 

[Gunawan W]

KYW,
in my case, according your advice it's better using a driver with fs=75-80Hz and Qt>0,7, but the Vas affect SPL on your spreadsheet result. It's hard to find 6" driver with big Vas to get SPL around 95dB.
I read somewhere that Vas can be ignored in calculation because there is no box, is it true?

Sheldon,
good idea even better with motorized baffle wing!

See you.

 

[ThorstenL]

Konnichiwa,

in my case, according your advice it's better using a driver with fs=75-80Hz and Qt>0,7,

Yup.

but the Vas affect SPL on your spreadsheet result.

Of course it does. My spreadsheet calculates the absolute SPL for 2.83V applied and 1m distance. Most Box Calculation programes only provide a relative SPL calacuation. A larger Vas meas a more efficient Driver.

It's hard to find 6" driver with big Vas to get SPL around 95dB.

I am not surprised. You are running up against the little thing called Physics.

I read somewhere that Vas can be ignored in calculation because there is no box, is it true?

Of course not.

Sayonara

 

[Gunawan W]

Hi,
apparently I reach the limit of small OB, with 6" driver and 40x60cm baffle size (folded), passive filtering, the result is not efficient, limited flat FR (200Hz to 12000Hz), so I drop to build it, am I give up to soon?

Happy New Year 2004, to all you guys, may audio be with you!

 

[ThorstenL]

Konnichiwa,

Hi,
apparently I reach the limit of small OB, with 6" driver and 40x60cm baffle size (folded), passive filtering, the result is not efficient, limited flat FR (200Hz to 12000Hz)

That comes from using unsuitable drivers (Low Qt and low sensitivity) and way too small size baffles (remember, folding gains you only minimal LF extension over the front section of the baffle on it's own.

Do the job right (Suitable driver, sufficiently large baffle) and the result will be very good.

Sayonara