Question on mids for all you Dipole guys

[Sheldon]

There is a fair amount of useful discussion in the archives about appropriate T/S parameters for dipole woofers, but I couldn't find so much about mids or mid woofers. Does it not matter so much, because you simply cross the woofer higher with a higher Qt mid woofer?

Related question: it doesn't seem from first glance the most of the available speaker modeling software addresses dipole response. Any recommendations?

Sheldon

[sreten]

It does matter but not a great deal.

A second order high pass can be easily used to compensate
for the frequency response due to to the units Q.

At frenquencies above the baffle width wavelength the mids
response on the forward axis is independant of loading, i.e.
its the same in a box and on a baffle.

At lower frequencies response is heavily influenced by
the room and the position of the dipole in the room.

Anechoic response is modeled by a 6dB/octave rolloff.

/sreten.

[Sheldon]

Thanks,

It would be interesting to have a program to play with response, depending on baffle dimensions and speaker placement on the baffle. I would think that it would not be that hard to model. Or is it easy enough to experiment, so not worth the effort to develop the model? That seems one nice thing about using conventional drivers on an open baffle. Real easy to make a small baffle, then just extend it in various dimensions to evaluate response.

Sheldon

[Nuuk]

That seems one nice thing about using conventional drivers on an open baffle. Real easy to make a small baffle, then just extend it in various dimensions to evaluate response.

In a word, YES! Just have loads of different sized pieces of wood/MDF/plywood handy and you can change the baffle size/shape in a jiffy. Then sit back and listen for a couple of days before repeating the process.

[ThorstenL]

Konnichiwa,

Originally posted by Sheldon
There is a fair amount of useful discussion in the archives about appropriate T/S parameters for dipole woofers, but I couldn't find so much about mids or mid woofers. Does it not matter so much, because you simply cross the woofer higher with a higher Qt mid woofer?

Hmmm. Yes and no. The Q (Qts/Qms/Qes) determine the behaviour of the driver around it's resonance point.

If you are using (for arguments sake) a 5.25" Driver with a Qt of 0.3 and an Fs of 50Hz with a 500Hz X-Over on an open baffle the result will be fine, as by the time you reach 500Hz the drivers efficiency has reached the maximum.

If for arguments sake you where to use the same driver with a 100Hz X-Over you would have to equalise the driver heavily as there would be a substantial loss even at the X-Over frequency.

Originally posted by Sheldon
Related question: it doesn't seem from first glance the most of the available speaker modeling software addresses dipole response. Any recommendations?

Originally posted by Sheldon
It would be interesting to have a program to play with response, depending on baffle dimensions and speaker placement on the baffle. I would think that it would not be that hard to model. Or is it easy enough to experiment, so not worth the effort to develop the model? That seems one nice thing about using conventional drivers on an open baffle. Real easy to make a small baffle, then just extend it in various dimensions to evaluate response.

Well, I have written something LIKE that, though it is as of yet somewhat inaccurate (good enough in practice for basic feasibility studies though).

http://www.exdreamnet.de/download/Xlbaffle.xls

It is an Excel Spreadsheet that runs under Microsoft Excel 2000 (probably also 97/XP but not tested) but NOT under any other Office Software like Star Office.

The baffle assumes ideal assymetric placement of the driver on the baffle (meaning making the distance to each and every edge non-divisible by any other) and takes account of room re-inforcement (inaccurate but close enough until I have time for a re-write) but not of room modes.

Sayonara

[Sheldon]

Thanks, makes sense. Goes with the impression one gets from looking at the freq response curves drivers that differ in Qt (that is, if such things are measured under similar conditions, which they don't often seem to be).

I did try to download the spread sheet yesterday, but the link on the thunderstone site didn't work. This one does, Thanks.

Sheldon

[sfdoddsy]

There's another spreadsheet here:

http://www.diysubwoofers.org/dipole/

[Jocko Homo]

It is an Excel Spreadsheet that runs under Microsoft Excel 2000 (probably also 97/XP but not tested) but NOT under any other Office Software like Star Office.

Why not Star Office? Some of usreally hate using MicroShaft anything. WinDoze is bad enough, without making things worse by installing Office.

Working on changing that feature.

Jocko

[Sheldon]

Originally posted by sfdoddsy
There's another spreadsheet here:

http://www.diysubwoofers.org/dipole/

Thanks, let me take the opportunity to ask a related question:

On the dipole woofer side, some installations use a w type frame with the woofers facing and the box openings facing the listener. This obviously shouldn't affect Lf sound, but what is the upper cuttoff where internal reflections begin to become audible?

Sheldon

[ThorstenL]

Konnichiwa,

Originally posted by Jocko Homo
Why not Star Office?

Where I work we use Mickeysoft so I use iut too. I'm not at all fussed. StarOffice is as bad bloat and bugware as MS Office and I get both for free. MS tends to work more reliable and has more features I actually need.

Just being pragmatic, no axe to grind....

Sayonara