[hum4god]

hi

i want to use the natural rolloff of an OB and need to know what the formula is to calculate it ?
i have some contradicting information and need clarification please .

what is the rolloff slope of the baffle ?

i want to use a fullrange driver on a smaller OB which determines its low crossover point and need to know how to calculate the width of the baffle to achieve a certain Frequency .

any help appreciated

[Rudolf]

AFAIK there is no simple formula to calculate that. JohnK has given some advice what you need to consider:
http://www.musicanddesign.com/Equivalent_Baffles.html

But that study does not account for the driver parameters. Maybe you will be best off by buying his "A B C Dipole Design Guide".

[gitarretyp]

It's not a simple calculation for conventional baffle shapes. You should download some simulation software, such as Xlbaffle and The Edge. They'll at least give you a somewhat reasonable approximation of your baffle response.

[scorpion]

In fact there is a very simple formula for this: rolloff frequency = 34000/(bafflewidth * 2) = 17000/bafflewidth (EDIT: measured in centimeters, I should have stated), in practise this will not hold sharp.
You will normally have bafflesupport further down in frequency than the formula would predict. But if you for instance simulate with the Edge program,
http://www.tolvan.com/ you will see that this rule is implemented in that program. After the corner frequency the roll off will be 6 db/octave down to
the elements resonance frequency below which the combined roll off will be 18 db/octave.

Here is a simulation with Martin J Kings Math-Cad  models: http://www.quarter-wave.com/ of Monacor Fullranger SP-130/X8 which I use in my 'Volks-OB': http://www.audiocircle.com/index.php?topic=46951.0 . Element on a baffle 100 cm high and 40 cm wide, element center situated 85 cm up the baffle and 15 cm in from the side:



As you can see there will be baffle support down to about 150 hz. I have however chosen to crossover at 300 hz to have on average a little better efficiency and above all much better
power handling. You will face trial and error or simulate with a good simulation programme.

/Erling

[hum4god]

wow thats a lot of stuff to look into .
thank you very much , i guess i will have to do some homework .

one more thing : is there a measuring device that is recommended to measure the freq. response in room .

any pro gear that could be rented with microphone .
this way i could  actually measure the results on the baffle .

thanks again

[gitarretyp]

In fact there is a very simple formula for this: rolloff frequency = 34000/(bafflewidth * 2) = 17000/bafflewidth, in practise this will not hold sharp.
You will normally have bafflesupport further down in frequency than the formula would predict. But if you for instance simulate with the Edge program,
http://www.tolvan.com/ you will see that this rule is implemented in that program. After the corner frequency the roll off will be 6 db/octave down to
the elements resonance frequency below which the combined roll off will be 18 db/octave.

It's not quite that simple. The problem is that the baffle width is not constant, except for a circular baffle with the driver centrally mounted. The equation you site will get you in the ball park. Low enough in frequency the roll-off does begin to behave more like a typical open baffle. However, once you start getting that low, boundary reinforcement starts taking effect and things behave differently again. Your safest bet (aside from building a test baffle and measuring, which i strongly encourage), is to use a combination of the edge and xlbaffle to get an idea of the response behavior.

[hum4god]

what would be a good and simple way to measure a test baffle ?

[gitarretyp]

what would be a good and simple way to measure a test baffle ?

I'm not sure of what you ask. Do you mean what hardware, software, location,... or all of the above?

[hum4god]

i mean measure the frequency in room with a microphone by some sort of device like behringer perhaps ( don't know really how they work )  which would show a graph of the actual performance .
hardware and software if necessairy .

[markC]

The simplest way, (albeit perhaps not the most accurate), would be test tones and an spl meter. At least this would allow you to see the response in the room with the effect of room reinforcement.

[gitarretyp]

Room EQ wizard and an spl meter is the simplest, semi-accurate method. More advanced would be a decent mic, eg Behringer ECM8000, a mic preamp or usb soundcard with mic preamp, and any of a variety of software. SynRTA is free and easy to work with. Speaker workshop is free but with a much steeper learning curve. Then you have the pricey choices like ARTA, LspCAD, soundeasy, ...

I'm not sure how accurate the RTA feature on the behringer EQ is, so i can't comment on it.