Speaker simulator with high-pass filter

[hitsware2]

I would like to find a simple speaker simulator program that
would allow graphing the response with a first order high-pass
on the input .....

[johnny5jz]

WinISD has that option. Just add a filter in the filter section. It will allow you to add a first order high pass.

[billfitzmaurice]

It allows a number of filters, orders and Q. The superiority of higher order filters is pretty obvious when you overlay them.

[JRT]

I would like to find a simple speaker simulator program that
would allow graphing the response with a first order high-pass
on the input .....

In Unibox, you can add the transfer function of an active filter and/or in much more limited manner can add some passive components.

[Chris Roemer]

To get an accurate response (using .frd files) you need an XO simulator.
WinISD can show various "transfer functions" - which are just generalities, AND do not take into account a driver's varying impedance (using passive components).

[fpitas]

Yes. Just adding a series capacitor and hoping the driver is resistive 8 ohms (for example) often gives surprising results.

[hitsware2]

Is there a way to ' whittle down ' WinISD to work with simply Vas, Fs, and Qts ?
( Where's BoxPlot when you need it ? )

[billfitzmaurice]

No program can work with just those specs.

[Chris Roemer]

Actually, WinISD "beta" (from 2014 - it's vers. 0.44) WILL plot a woofer's bottom end rolloff (and suggest a box size/tuning) using just those 3 main parms, but that's only 1/10th of the data that more advanced (i.e. "Pro") versions will show you.

What's the aversion to a more robust ISD vers?
You need to (really) enter about 9 parameters: Qes, Qms (Qts gets calculated), Fs, Vas, Re, Le, Sd, Pe, and Xmax.
This has little to do w/your orig. ? though ...

[billfitzmaurice]

You can't get a useful result from Vas, Fs and Qts alone. At the very least you also need Sd and Re.

[hitsware2]

Actually all I need to plot is infinite baffle response with a first order filter ( for my purpose )
Could probably do without Vas ....

> "beta" (from 2014 - it's vers. 0.44)
I'll look at that ...

[fpitas]

As we discussed above, actually obtaining a first order response may require a Zobel etc. Anywhere near the driver resonance there's an enormous reactive impedance peak.

[Chris Roemer]

[Chris Roemer]

Actually, if all you care about is a system's possible F3, just take the diff. between 0.40 and a driver's Qts (must be in a "reasonable" range, like 0.3Os through 0.6Os) and add/subtract that from its Fs ...

Given a set of drivers w/an Fs of 50Hz, you CAN get a ported box to reach (F3):
50Hz if its Qts = 0.40,
only 60Hz if its Qts is 0.10 lower (= 0.30, a stronger motor),
40Hz if its Qts = 0.50 (0.10 higher, so subtract 10Hz),
and down to (about) 30Hz if its Qts = 0.60.

This is VERY general, and won't give you a box size (or tuning) w/out Vas, but is still a pretty reliable rule-of-thumb.
If you choose to go sealed, your F3 will end up (nearly) an octave higher (and the box size will roughly halve).

[hitsware2]

What I want to do is model an open baffle ,
using the old way of looking at it .
I.E.....
The response is equal to that of the driver
in an infinite baffle ( big sealed box )....
With a first order high pass determined
by the size of the baffle .

[billfitzmaurice]

An open baffle isn't the same as an infinite baffle. Totally the opposite, actually. An open baffle allows the front and rear waves to meet, an infinite baffle doesn't. The roll off frequency of an open baffle is where it is one wavelength in dimension.

[Billet]

I think I see what you are trying to do, planning to use the first order filter to simulate the open baffle response?

I have modeled open baffles in a similar way, but instead of adding a high pass filter, I just add a large amount of baffle step to my response (low pass). I use a very large closed box, like 999 liters, and about -12 to -15 db cut above 100 Hz has worked for me.

[hitsware2]

• An open baffle isn't the same as an infinite baffle.
• Totally the opposite, actually.
• An open baffle allows the front and rear waves to meet,
• an infinite baffle doesn't.
• The roll off frequency of an open baffle is where it is one wavelength in dimension.

That is why it is called " infinite " ....
It is longer than all wavelengths .....
Like an infinitely wide baffle ...........
With the same frequency response
IF ( As I do ( for this exercise ) )
One chooses to disregard the comb filtering .

[billfitzmaurice]

Not quite. So long as the front and rear waves don't meet it's technically an infinite baffle. As of late the term has been mainly applied to large rear chamber installs, like in walls between rooms, in closet doors or ceilings, but going back to when the term was originally used it applied to sealed boxes of any size as well. Comb filtering doesn't enter the equation at all, as that applies to what can occur with multiple drivers/speakers of any alignment.

[hitsware2]

I think I see what you are trying to do, planning to use the first order filter to simulate the open baffle response?

I have modeled open baffles in a similar way, but instead of adding a high pass filter, I just add a large amount of baffle step to my response (low pass). I use a very large closed box, like 999 liters, and about -12 to -15 db cut above 100 Hz has worked for me.

How did You arrive at Your numbers ?

[Billet]

Open baffle losses ~= 13500 in/sec (speed of sound) / 12" (baffle width) / 3 (???) = -3db at 375 Hz. Falling at 6 db per octave: -9 db at 375/2 = 188 Hz (1st octave), -15 db at 188/2 = 94 Hz (2nd octave). That is how I did it, more or less. I spent some time adjusting it to my room afterwards, but at least it gave me some kind of reference point.

[JRT]

What I want to do is model an open baffle...

At the link is John P. Kreskovsky's ABC Dipole:
http://musicanddesign.speakerdesign...._C_Dipole.html

[jhollander]

You can also use the Edge program to see the open "baffle" response, but ABC Dipole has both pseudo driver and baffle response.