Open Baffle Midrange Peak?

[croat47]

So, my situation is this:

81mm midrange driver
2m separation between speakers
2 meter distance from speakers is the listening position

I am modeling a 178mm x 914mm baffle with the drivers 68mm off of the edge to the outside (away from the other speaker) and 711mm off the floor.

If I am using the Edge correctly, I have placed the mic 2m from the baffle, and at a horizontal location 1m to the "inside".

This gives me a situation where I could cross at ~500 and ~2300Hz...just right for the drivers I've got to support the bass and treble.

Does that look reasonable to you, Rudolf?

Again, thanks for all the input!

[croat47]

I played around some more and got a broader and flatter range by lowering the driver to 0.68m off the floor.

So, 450/2500 crossover points look possible. Please advise if those are okay for target points.

Do you have better results with 2nd order? 4th order electrical? The real question is: what filter types work well for OB?

[LineSource]

Tapering one side like the Apogee Ribbon Monoliths, or both sides for a trapezoid are also popular dipole baffle shapes. Rear wings on just the bottom for bass extension also seems popular, plus the wings add strength and stability.

[Rudolf]

Originally posted by croat47
I played around some more and got a broader and flatter range by lowering the driver to 0.68m off the floor.

Aaron,
did you look at the scale of the y-axis in EDGE? You are equalizing a 1 dB "hump"!

So, 450/2500 crossover points look possible. Please advise if those are okay for target points.

Anything between 400-500 Hz is perfect for the highpass. Xover to the tweeter can be anything from 2000-3000, depending on the tweeter capabilities.

Do you have better results with 2nd order? 4th order electrical? The real question is: what filter types work well for OB?

I don´t believe in crossover sound. And definitely there is nothing special for OBs. It´s hard to design acoustically symmetrical crossovers without measurements. So I don´t think I can be of much help here.

For the integration of all drivers on a single baffle LineSource already gave some good advice.

Rudolf

[croat47]

Originally posted by Rudolf

Aaron,
did you look at the scale of the y-axis in EDGE? You are equalizing a 1 dB "hump"!


Anything between 400-500 Hz is perfect for the highpass. Xover to the tweeter can be anything from 2000-3000, depending on the tweeter capabilities.


I don´t believe in crossover sound. And definitely there is nothing special for OBs. It´s hard to design acoustically symmetrical crossovers without measurements. So I don´t think I can be of much help here.

For the integration of all drivers on a single baffle LineSource already gave some good advice.

Rudolf

I wasn't really "worried" about the 1dB bump. I was just moving things around to see what effects driver placement made. The 450-2300 range looked flatter when I put things as stated in my last post. So...wasn't "trying to eq"...just happened.

I am looking to do something like Tony Gee's Modulus

http://www.humblehomemadehifi.com/Modulus.html

So, the tapered baffle falls out. But, that gives me an idea on how to use some leftovers of drivers and lumber...

The mid and tweeter that I am using have been crossed over around 3200 in another design. Too high? I posed the question about crossover topology in the event that something about open baffle behavior required a certain type. On that subject: will using a 2nd Order L-R to the bass give a 18db/octave slope when combined with the 6dB slope from the baffle loss?

Again...thank you!

[Rudolf]

Originally posted by croat47

The mid and tweeter that I am using have been crossed over around 3200 in another design. Too high?

Following that Modulus design a 3 kHz crossover should be almost identical for a box and an OB. Taking the crossover from the original design would be a good strategy.

I posed the question about crossover topology in the event that something about open baffle behavior required a certain type.

None that I would know about. Just the usual precautions: Since dipoles generally are excursion challenged, first order highpass is critical in many cases.

On that subject: will using a 2nd Order L-R to the bass give a 18db/octave slope when combined with the 6dB slope from the baffle loss?

Yes.

[croat47]

Rudolf:

The wave guide that you mention...is it a pre-fab unit or did you chamfer the front side of the cut out for the tweeter?

Is this setup shown somewhere on your site? Headed there to look...

Also: in another thread about MJK's article on the Jordan/H-Baffle, the cross section is 16"x16" and the length 16" (8" either side of the centered driver mount). Which drives the 1/4 lamda response? That is: if one made a 10" driver version of the same, does an 11"x11" section with the same length have the same resonance.

Read about TL theory...just not polished yet.

Thanks!

[dcathro]

I am doing a partial open baffle design with a refex bass and am also concerned about midrange peak and optimum diffraction.

I thought it would be easier to ressurect this thread rather than start a new one.

Does anyone know what the effect of curved OB wings would be?

I am thinking of having a flat wedge shaped baffle not much larger than the drivers, and then large radius ( 6" - 8" ) wings that sweep back with a depth of perhaps 5" at both the sides and the top - kind of like a cross between a theil a Stadavari, but open baffle. The open baffle would also have a base ( similar to Tony Gee's Modulus ).


Background:

I will be using a Supravox 215 EXC full range on a smallish open baffle sitting on top of the reflex box ( similar to Tony Gee's Modulus). There will be a support tweeter and a 15" bass.

I want to keep the crossover as simple and unobtrusive as possible, and want to optimise the acoustic behaviour of the open baffle.
The Supravox will have no LP or HP filters, and will use the 6dbl roll off of the OB at around 250Hz to 300Hz to match into the Bass which will have a 1st order LP filter. The support tweeter will come in around 6KHz to just add to the Supravox's output.

From modelling flat OB's, I had anticipated that I would need a broad series notch filter to take out a 3db midrange peak from about 300Hz to 3K, but would like to avoid this if possible.

Thanks

David