Linkwitz Greiner with 12 dB highpass filer

[Hylle]

I have read about the Linkwiz Greiner circuit on this page, I need a little help to understand it : TA Speaker Topics - Designing a Linkwitz Transform

They write about a 12dB high pass filter in the circuit, but I dont find it. Where is this circuit, or is it performad by some of the components in this circuit already, then please point them out to me. Because as I see the circuit it only performs the fs and Q box change, or am I wrong?

[Davey]

It looks like the 12db filter curve was added to the spreadsheet just for informational (viewing) purposes.

It would be up to the user to add the necessary (extra) circuitry to actually implement it.

Dave.

[PLB]

Hi Hylle,

If you were to examine the frequency response of a closed box loudspeaker, you will find that at frequencies well below its low frequency cut-off, the response should fall at a rate of 12dB/Octave. At frequencies well above the low frequency cut-off, the response should be flat. These two characteristics are also true of a 2nd order high-pass filter. In the paper you linked to, John Murphy (the author) is actually equating the low frequency acoustic response of a closed box loudspeaker with the response of a 2nd order high-pass filter.
So there is no external/separate 2nd order HP filter per say, he is just referring to the equivalent response of the speaker. Hope this helps.

Regards

Peter

[Davey]

Yes, the ultimate roll-off of a closed-box system would be similar to a 12db/octave high-pass filter. However, he talks about a "2nd order high-pass filter to reduce infrasonic boost." You can't reduce boost unless you implement some kind of circuitry to accomplish it.

I agree with Hylle. It's puzzling.

Cheers,

Dave.

[PLB]

Hi Hylle/Dave,

I have to confess, I skimmed through John Murphy’s paper twice after reading your most recent post and I still couldn’t find any part where he mentions the words “reduce infrasonic boost “. So I copied/pasted his whole text into MS Word and ran a search for “infrasonic” and at last I found it in the “download box” with the smallest type in the whole document – it went right by me --- I must be getting old.
My apologies for not seeing this first time round.
I agree with you that to implement a rumble filter, you will definitely need an extra HP filter stage in addition to the Linkwitz transformer.

Regards

Peter

[tinitus]

ESP have had something on the subject

[Hylle]

Thank all of you! Yes the text is a bit tricky. I could not see any HP filter anywhere, so i guess we all agree that additional circuit is needed. I will check the ESP site.

[event horizon]

Actually you don't need a high pass filter when using a Linkwitz transform on a sealed box. The whole idea of the Linkwitz Riley transform is to extend the bass response & correct for a driver that may be too high in the way of Qcb in a sealed enclosure.

When you select the low frequency cutoff of the Linkwitz transform the whole idea is that you make it sensible so the driver won't move more than the linear excursion. So assuming you boost the response to say 20Hz you give the transform a Q of say 0.7 at 20Hz. At that point there will be no more boost below that frequency (check it out on WinISD Pro by modelling something), & as we are dealing with a sealed box it doesn't matter about excursion limitation as it can't produce enough power to move the driver further as the response will roll off just like a standard sealed box would.

However, the same Linkwitz transform can be used to correct a reflex enclosure... If you do this then you'll certainly need to implement a high pass filter below cutoff.