"DS: If you look at it in pure engineering terms then it can be distilled down to: Let’s design a speaker. It needs to play X dB loud at X meters (defining power handling, distortion). The audience width and depth will be X (defining dispersion). Etc. Viewed in those terms the product isn’t pro or domestic, it just needs to meet a particular performance spec.
First principles. Commonly ignored in DIY designs. As is what naturally follows . . . design for the room. That's where the (potential) benefit of DIY resides . . .
I just thought.. what kind of room? Like little chalkboarded flat or spacious loft with hard walls? Treated or non-treated? Where the actual benefit of designing for particular room would come from? How would it differ from engineering approach?
I don't know how it will differ from "engineering approach" - but speakers I've heard that were designed wit the room in mind (including crossovers) have always sounded pretty good too me, and have been praised by others.
If you know what room your speakers will be in, design for that room. There is plenty of talk and questions about that on this forum. I disagree that it is "Commonly ignored in DIY designs."
If you know what room your speakers will be in, design for that room. There is plenty of talk and questions about that on this forum. I disagree that it is "Commonly ignored in DIY designs."
Elias - sorry I missed your questions back int he summer. I don't know the answers to your questions, so will look into this week.
Happy New Year.
Wouldn't designing the room and the speakers together to optimize them both make the most sense? That's what I do.
Understood, but still it is the only way to achieve the ultimate in what is possible. You can tweak speakers and amps all you want but if the room remains a given nothing is going to overcome the problems with a bad room/speaker setup.
Hello Earl
Most of us don't have a choice about the room. With that as a given how do you work around the room you have and some how integrate that into a DIY speaker design? Where do you start?
One obvious idea would be too use multiple subs and only design your speakers with enough low end to meet the subs. Beyond that where else would you go??
Rob🙂
Exactly.
It's no surprise that SL's designs work well for me . . . my listening room is the same size as his and has similar acoustics. Not to mention our similar tastes in music. If my design target was movies in a 600 square foot room or rock and roll the way I used to listen to it I'd build horns.
Draw the room.
Draw the desired listening area within that room.
Draw the loudspeaker placement and polar that will uniformly illuminate that listening area with both direct and reflected sound.
Size drivers to give the desired maximum SPL in the listening area.
Build enclosures that give the desired polar with those drivers.
The rest is trivial.
I've never been in favour of the sub/satellite approach - eg main speakers with inadequate bass extension crossing over to subs exclusively producing the bottom octave or two...and still believe that the main speakers should be big enough to be run full range, even if they're not producing ALL the bass by themselves.
However I have had to temper my dislike of subs in general by the simple fact that after years of trying I've come to the realisation that no placement of two full range main speakers will ever give a totally satisfactory in room bass response regardless of how good the bass response of those speakers is.
Thanks to room modes and boundary cancellation effects, with only two sources at the opposite side of the room from the listener you will almost always have notches that you can't eliminate and peaks that will be highly location dependant thus resistant to effective EQ. (I actually find the notches more annoying than the peaks, at least you can notch out the peaks at the ideal listening position - you can't do anything about the notches which can lead to a "hollow" sounding bass)
That's even if you're willing to sacrifice imaging by positioning the speakers relatively close to boundaries to optimise the bass response, if you position the speakers a bit further out for optimal imaging the bass situation deteriorates even further. Bass traps can reduce the severity of the peaks and dips but don't fundamentally solve the problem.
Multiple distributed subs is the only easy solution in a typical small room, and in my opinion having the main speakers big enough to run full range down into the low bass (at least down to 30-35Hz or so anyway) is beneficial because if you have 3 distributed subs in the room but your main speakers can also go down to the low 30's and are fully overlapped with the subs then you effectively have 5 subs distributed around the room not 3 - up to a point the more distribution around the room the better.
Although I haven't measured it I don't think modal smoothing with 3 subs operating from 25-80Hz exclusively and mains only from 80Hz up would work nearly as well as mains also going down to around 30Hz. Below 30Hz you don't need as many sources because the wavelength is so long and notches etc don't form, but in the mid bass region the more sources the better up to a point.
I think my ideal target would be two main speakers that go down about 30-35Hz on their own, two subs at the main speaker end of the room near the corners running from 20Hz to 160Hz and two at the back (listener) end of the room going from 20Hz to 80Hz.
Running the front pair of subs as high as 160Hz would (with careful positioning) help smooth the notches in the 100-160Hz range which can often be quite severe (The subs at the back end of the room can't be run that high as they're too far away from the mains to avoid localisation issues) while all 4 subs contribute to modal smoothing below 80Hz.
Last edited:
We can't expect monophonic bass from recordings so any optimization based on monophonic bass will be more or less off. The solution would be to cut bass from L and R, sum it and send it back to L, R and the sub(s).
Simon
Running my room modes software will validate everything that you are saying. I would just differ on how low the mains should go (I don't think that it is necessary to go below the first few modes). I completely agree on overlapping the mains with the subs and NOT using a crossover. The software validates that as well.
Markus
The issue you raise has always concerned me as well. Unknown how serious a problem is really is however.
Running my room modes software will validate everything that you are saying. I would just differ on how low the mains should go (I don't think that it is necessary to go below the first few modes). I completely agree on overlapping the mains with the subs and NOT using a crossover. The software validates that as well.
Markus
The issue you raise has always concerned me as well. Unknown how serious a problem is really is however.
Agreed that not all bass is monophonic, I have quite a bit of music that isn't - some jazz has the bass off to one side and a few pop/electronic songs I have even have some out of phase bass for some bizarre reason. (Not equal amplitude though so there isn't complete cancellation, causing it to sound like its coming from the side)
I don't agree with your solution though as by summing the bass to mono you're removing information from the recording. It's why my preferred solution would be a 4 sub layout with a sub near each corner and symmetry down the centre line - as well as allowing for a higher cutoff frequency for the front subs than the rear, you can easily drive the two left (front and rear) subs from the left channel and the two right subs from the right channel.
If your recording has mono bass all speakers would work together as in your mono summing scenario but if there is any non mono bass the left and right sidedness and symmetry of two stand alone speakers is maintained but still with a large degree of modal smoothing. (The main and its two subs still provide for three overlapping distributed sources of bass even for the individual channel)
4 subs may not be needed (over 3) for modal smoothing but if you want the symmetry needed for stereo bass reproduction then its the logical way to go if the corners are all available.
Last edited:
^
My solution would be a headphone based binaural loudspeaker renderer. I sound like a broken record.
My solution would be a headphone based binaural loudspeaker renderer. I sound like a broken record.
While in ear earphones or headphones can produce very smooth and extended bass free of room effects they're never going to have the same dynamic impact or realism as speakers in the bass region because only the ear drum is stimulated while speakers vibrate your entire body in a more natural way.
I find I just don't get the same "buzz" from listening to headphone bass that i do from listening to loud (but clean and smooth) bass in a room - regardless of how loud the headphones are. Dynamic range compression and intermod between bass and higher frequencies is a clearly audible problem with all but the best headphones too.
For these and other reasons headphone listening will never replace actual speakers for me. (At the moment headphone listening is all I have, and I really miss the full speaker set up)
Last edited:
I agree with Simon. Headphones, to me, are to be used only when speakers are not possible, like traveling. They will never be a serious solution to me. (Although I do use them quite often.)
- Status
- Not open for further replies.