Interesting comment. Can you elaborate on the advantages of arrays for the midrange? What do you mean by average out?
Does it need to be floor to ceiling for it to have that effect?
Regards
One driver will have a floor reflection that creates a dip at the listening position.
Two drivers above one another will each create a dip at slightly different frequencies. But due to one driver filling in part of the dip for the other, they will be less deep. An array has each driver at a slightly different position in regard to floor (and ceiling). The floor and ceiling are a virtual extension of that array, it will seem to act like a larger array.
It works best with a floor to ceiling array, as you won't be able to find the dips usually created by the floor and/or ceiling. The thing to worry about is parallel planes as reflections of different drivers in the array can still be at similar points there.
It made it possible for me to get this result (measurement made with APL_TDA, DSP done with DRC-FIR and REW):
Measured at the listening position in a relatively normal living room this is the sum of a stereo pair.
This took 3 damping panels on parallel planes (at first reflection points). There's still reflections to be seen but they are way down in level.
The room:
Behind each curtain is a damping panel made with a combination of fiberglass insulation and (real) wool felt, behind the listening position is another panel using the same materials, disguised as a poster (a print on Kona cotton).
Two drivers above one another will each create a dip at slightly different frequencies. But due to one driver filling in part of the dip for the other, they will be less deep. An array has each driver at a slightly different position in regard to floor (and ceiling). The floor and ceiling are a virtual extension of that array, it will seem to act like a larger array.
It works best with a floor to ceiling array, as you won't be able to find the dips usually created by the floor and/or ceiling. The thing to worry about is parallel planes as reflections of different drivers in the array can still be at similar points there.
It made it possible for me to get this result (measurement made with APL_TDA, DSP done with DRC-FIR and REW):
Measured at the listening position in a relatively normal living room this is the sum of a stereo pair.
This took 3 damping panels on parallel planes (at first reflection points). There's still reflections to be seen but they are way down in level.
The room:
Behind each curtain is a damping panel made with a combination of fiberglass insulation and (real) wool felt, behind the listening position is another panel using the same materials, disguised as a poster (a print on Kona cotton).
Excellent post. Thank you!
I remember your build thread for those arrays. They look outstanding.
I'm curious why you don't have the first reflection point of the front wall treated. You tested and it didn't make a big difference? Aesthetic reasons? Other?
I hadn't thought about the reflections from an array, so this is very educational. I guess below 300Hz or so it's more about distributed bass sources along the lines of Earl Geddes, and above is about evening out reflections as you said.
Would love to get your thoughts in the context of the speakers I'm building: active system with DSP, time-aligned, etc, from 2kHz up with TPL-150H, maybe an array for midrange, direct radiators for midbass and subs. The TPL has the 80x30 degree horn, so minimizing floor and ceiling reflections. I could build a midrange array to the side of the TPL or above/below (sort of MTM). My guess is the floor/ceiling reflections up to 2khz, and minimized reflections above 2kHz would sound odd. What do you think?
Erik: I apologize for derailing the thread. Your question was thought provoking I guess, yet maybe in a different direction from initial intent. Maybe I should start a different one.
I tried a whole lot, so much that I can track down every bump in that graph above to where it comes from. The wall behind the speakers didn't contribute to mayor or minor reflections. I still know of a few places that could benefit from damping but I had to make a promise to my girl not to add any more.
Combining array and point source could sound odd due to SPL differences at different distances to the speaker. If you always listen at the same distance this would not be as big of a problem. It probably deserves it's own thread to take it a little further.
My post about the averaging effect that eliminates the floor and ceiling reflections (or at least reduces them to a minimum) was targeted to support the positive reviews on this thread about the IRS. I wish I had heard them myself.
Last edited:
This looks like duelund driver (the late duelund was involved in the first gryphon speakers) the drivers are no doubt based on scan speak parts.
Duelund recently started making these available
Mono and Stereo High-End Audio Magazine: Duelund speaker driver NEW
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.