@vineethkumar01 what made you change the reference axis from 10 degrees to 20 degrees of axis?
In light of all these comments about toe-in,, balance etc, it may be wise to keep that reference or design axis in mind.
yeah and be open to change the reference axis after playing around how the system works with the room. When coverage is rather smooth and you have a DSP, the toe-in doesn't need to relate to frequency balance in anyway, just find toe-in that works for you for spatial effects you like, and then balance out the response for the direct axis you now happen to have.
@vineethkumar01
Did you get a chance to listen to music or TV with your linear phase crossovers? I tried a linear phase LR48 last night with Disney's Haunted Mansion 2023. The bad guy (Alistair Crump) has a really great low frequency voice with LR48 and LR24 but loses the magic with linear phase (example of Crump's voice 1:36:33 to 1:37:03). However, there was less overall coloration with the linear phase filters. The regular characters had more natural sounding voices.
Then I watched The Wonderful Story of Harry Sugar (Netflix 2023) which is constant male speaking speaking voice. That one was rough because my center channel is located on the floor beneath my TV. Each male actor seemed to present best with a different crossover; LR24, LR24LP, LR48, and LR48LP. Consequently, I tend to think the center channel would work better if it was cardioid.
Did you get a chance to listen to music or TV with your linear phase crossovers? I tried a linear phase LR48 last night with Disney's Haunted Mansion 2023. The bad guy (Alistair Crump) has a really great low frequency voice with LR48 and LR24 but loses the magic with linear phase (example of Crump's voice 1:36:33 to 1:37:03). However, there was less overall coloration with the linear phase filters. The regular characters had more natural sounding voices.
Then I watched The Wonderful Story of Harry Sugar (Netflix 2023) which is constant male speaking speaking voice. That one was rough because my center channel is located on the floor beneath my TV. Each male actor seemed to present best with a different crossover; LR24, LR24LP, LR48, and LR48LP. Consequently, I tend to think the center channel would work better if it was cardioid.
That is a lot of tips and suggestions.. Thanks a lot everyone 🙂
Finally my day job caught up with me and kept me busy yesterday, even today. So I didn't get a chance to listen to the speakers or do anything more with it yet. Also, recently my 6 channel audio interface, which I use for DSP crossover has been showing some issues in 2 of its output channels. I am seeing a constant 3rd harmonic distortion of 2 percent across the full bandwidth from 20Hz to 20kHz+. So it seems like either I might have to go with another DSP solution or use only 4 channels to drive the bass and mid+tweeter with a passive crossover between mid and tweeter. I have to think about this and find a solution (though I am gravitating more towards a full DSP solution).
@wesayso: I had made a mistake with the reference axis reporting in my intial post. What I thought was 10 degrees off axis was actually 30 degrees off axis ( it turned out that I scrolled my mouse too much in that reference axis determining field in Vituixcad initially and didn't notice it until later).
If you look at the Vituixcad files I have attached with a previous post, the sound power DI and early reflection DI curves are the smoothest straight flat ish lines if reference axis is taken as 30 degrees off axis. It looks ok ish at 20 degrees, slightly worse at 10 degrees and the worst at 0 degrees. Also at 0 degrees, and 10degrees, the notch around 10kHz due to the coaxial arrangement of drivers is much deeper even though the listening window and on axis curves follow each other across full bandwidth. At 20 degrees off axis, that notch is the mildest and the on axis and listening window curves track each other very closely. But at 30 degrees off axis, the listening window and reference axis response curves separate out by a few dBs as seen in first crossover pics. Some idea about all this can be got from below plot of tweeter raw response also, in which the diffraction related issues seem to be mildest around 20 and 30 degrees off axis.
The pic above shows the kind of listening environment that this speaker will have. The speakers will be pulled in closer to the PC on the left and right sides (and the CPU moved down) and the nearest wall is the back wall. The sidewalls will be about 2.5m+ away on each side. I will be sitting about 1m to max 2m away from the PC screen.
Hence I was thinking that since I will more be in the near to mid field of this speaker while listening, the reference axis and listening window responses should be smoothest compared to other curves. I have also been reading posts & papers by Dr Earl Geddes where he used to say that with his constant directivity speakers, he would optimize the speaker for best 22.5 degrees off axis response, along with his placement related toe in ideas, which @tktran303 also pointed out above. Even though I think the listening environment in my current case is much different from his (where the power response and other curves might also come into play much more),I was caught up with that 20ish degrees off axis idea. In fact my current horn system has been heavily toed in exactly as he pointed out and I have been liking it much so far. (Dont know if the toe in is visible in below pics well. One gripe I have with the room where the horn speakers are compared to the echo chamber room I had in previous house is the missing out on a bit of spaciousness feeling which I had plenty of in earlier house. This room feels a little too dead with the curtains, carpets, sofa, and absorption panels behind the sofa etc.. or maybe I am not used to it yet. The imaging is very nice but the feeling of spaciousness not so much or not as good as earlier)
These are the convoluted reasons why I chose 20 degrees off axis as the reference axis in my later posts 🙂
Please let me know if there are holes in the above logic. Offcourse till now it is all hypothesis and listening to the new speakers is yet to happen along with further adjustments based on it, if needed.
@CinnamonRolls: I am eager to listen to the speakers too.. So far my ears (and my wife's) are plenty tired from all the sine sweeps that I have been doing for measurements. Hopefully I will get a chance to listen to some real music on these speakers soon 🙂 Will report back once I do.
Finally my day job caught up with me and kept me busy yesterday, even today. So I didn't get a chance to listen to the speakers or do anything more with it yet. Also, recently my 6 channel audio interface, which I use for DSP crossover has been showing some issues in 2 of its output channels. I am seeing a constant 3rd harmonic distortion of 2 percent across the full bandwidth from 20Hz to 20kHz+. So it seems like either I might have to go with another DSP solution or use only 4 channels to drive the bass and mid+tweeter with a passive crossover between mid and tweeter. I have to think about this and find a solution (though I am gravitating more towards a full DSP solution).
@wesayso: I had made a mistake with the reference axis reporting in my intial post. What I thought was 10 degrees off axis was actually 30 degrees off axis ( it turned out that I scrolled my mouse too much in that reference axis determining field in Vituixcad initially and didn't notice it until later).
If you look at the Vituixcad files I have attached with a previous post, the sound power DI and early reflection DI curves are the smoothest straight flat ish lines if reference axis is taken as 30 degrees off axis. It looks ok ish at 20 degrees, slightly worse at 10 degrees and the worst at 0 degrees. Also at 0 degrees, and 10degrees, the notch around 10kHz due to the coaxial arrangement of drivers is much deeper even though the listening window and on axis curves follow each other across full bandwidth. At 20 degrees off axis, that notch is the mildest and the on axis and listening window curves track each other very closely. But at 30 degrees off axis, the listening window and reference axis response curves separate out by a few dBs as seen in first crossover pics. Some idea about all this can be got from below plot of tweeter raw response also, in which the diffraction related issues seem to be mildest around 20 and 30 degrees off axis.
The pic above shows the kind of listening environment that this speaker will have. The speakers will be pulled in closer to the PC on the left and right sides (and the CPU moved down) and the nearest wall is the back wall. The sidewalls will be about 2.5m+ away on each side. I will be sitting about 1m to max 2m away from the PC screen.
Hence I was thinking that since I will more be in the near to mid field of this speaker while listening, the reference axis and listening window responses should be smoothest compared to other curves. I have also been reading posts & papers by Dr Earl Geddes where he used to say that with his constant directivity speakers, he would optimize the speaker for best 22.5 degrees off axis response, along with his placement related toe in ideas, which @tktran303 also pointed out above. Even though I think the listening environment in my current case is much different from his (where the power response and other curves might also come into play much more),I was caught up with that 20ish degrees off axis idea. In fact my current horn system has been heavily toed in exactly as he pointed out and I have been liking it much so far. (Dont know if the toe in is visible in below pics well. One gripe I have with the room where the horn speakers are compared to the echo chamber room I had in previous house is the missing out on a bit of spaciousness feeling which I had plenty of in earlier house. This room feels a little too dead with the curtains, carpets, sofa, and absorption panels behind the sofa etc.. or maybe I am not used to it yet. The imaging is very nice but the feeling of spaciousness not so much or not as good as earlier)
These are the convoluted reasons why I chose 20 degrees off axis as the reference axis in my later posts 🙂
Please let me know if there are holes in the above logic. Offcourse till now it is all hypothesis and listening to the new speakers is yet to happen along with further adjustments based on it, if needed.
@CinnamonRolls: I am eager to listen to the speakers too.. So far my ears (and my wife's) are plenty tired from all the sine sweeps that I have been doing for measurements. Hopefully I will get a chance to listen to some real music on these speakers soon 🙂 Will report back once I do.
Attachments
I believe musical experience driven by 50% gear/speakers and other 50% your room. Are you planning to treat your room? Or get by with just eq.
Even this limited treatment in my very narrow space made night and day difference to my ht/musical experience.
Even this limited treatment in my very narrow space made night and day difference to my ht/musical experience.
Attachments
Last edited:
@krevilplays: I dont agree with the 50-50 rule. But I do agree that there is significant role played by the room along with speakers/gear. The percentage importance of onw w.r.t other varying from one use case to another 🙂
I do not have the luxury of applying treatment to room currently under discussion since it is a rental accomodation and I may have to move out of this house sooner or later. The only treatment I will be able to use will be some absorption panels on the back wall, if at all needed in this case. The EQ which is being discussed at length above is the speaker voicing EQ for crossover. Any room/environment/placement related EQ will be a layer applied on top of that crossover layer EQ later down the line. But to answer your question, I would happily go with EQ and less of room treatment, if at all that works out well for the application under discussion.
I do not have the luxury of applying treatment to room currently under discussion since it is a rental accomodation and I may have to move out of this house sooner or later. The only treatment I will be able to use will be some absorption panels on the back wall, if at all needed in this case. The EQ which is being discussed at length above is the speaker voicing EQ for crossover. Any room/environment/placement related EQ will be a layer applied on top of that crossover layer EQ later down the line. But to answer your question, I would happily go with EQ and less of room treatment, if at all that works out well for the application under discussion.
I was actually giving less weightage to room. Unless you have a very large space, I would say even bigger impact than 50%. Well you have to hear it to believe it and I have. Although eq will alleviate your headache from boominess something does not sound right with heavily eqd system. Been there done that. Not my cup of tea for the experience I am looking for. You will never be able to fully fix your time domain response with eq. Look at your waterfall and spectral density graphs.
Just my few paise 🙂. Cheers!
Just my few paise 🙂. Cheers!
Last edited:
Also worthy of note with eq you are fixing or optimising only around your personal LP . Its going to potentially mess up sound in other parts to even worse degree to the point of brain spasms.
Last edited:
@krevilplays: At this point I don't know what to reply regarding your above posts other than you may find at least some of the 60 odd pages of this thread interesting if you read it from the beginning 🙂
But thanks for sharing your experience.
But thanks for sharing your experience.
Vineeth brutha,
Whatever happens sonically, I love the black and red motif!
Whatever happens sonically, I love the black and red motif!
Surely when I find some time of my busy schedule. Just was making sure you won't get disappointed by your room after all the work. Otherwise nice work overall. Cheers.
Well this is considered very small room from acoustic treatment requirement perspective.
Even this one "This room is 3m in width, 7+ m in length and about 3m in height" which is very similar to mine Although I do have open staircase leading 2 floors above and 2 floors below that adds to "openness".
Would be very curious to see some waterfall and spectral density graphs from final LP and speaker placements, if not posted already, for my room case studies. Thanks!
Last edited:
About HT - it is VERY important tho have all three front speakers L/M/R with exactly same topology, no of drivers and xo type and F!
If (most likely) the mid is different, imaging will never be right!
If (most likely) the mid is different, imaging will never be right!
I have moved from that house and moved away from those speakers. So I don't have any data about that room or those speakers other than what is posted in this thread, somewhere.
Somewhere scattered in this thread, I had also posted the measurements of my then used listening room, my minimal room treatment experiences, and Room EQ results and all associated graphs I had made at the time.
Everyone’s entitles to an opinion, but there’s some point of fact to a 3D space that aren’t open to debate……take it from a career veteran of the music industry from live sound engineer to recording/mix/mastering engineer……the room is EVERYTHING.…..there’s acoustic drum rooms with 6 month waiting lists for sessions for the simple fact the room does matter just as much as the drums and mics used.
That’s a sound approach when you build/design a speaker that may have to reside in different environments, but EQ only addresses a small portion of the overall performance factor. Yes, Geddes understands this better than most, but then made some assertions along the way that most people with ears on the sides of their heads knew were objectively without support.….scattering shrill HF content through a maze of foam doesn‘t sit well with folks who know that when power compression IS NOT a performance factor, a silk dome tweeter is much more pleasing to listen to than any compression driver you can find.
I haven‘t followed this thread all the way nor read anything close to enough context but please correct me if I’ve drawn an incorrect conclusion?……your building a 3way speaker for near field critical listening in a live 3D space where placement is limited to within 2ft of the wall behind the speaker?
^^ having listenend to untreated and moderately treated room, I agree I am never going back to untreated small space.
Past couple of years is what I would consider grinding and eye opening experience building 11.4 channel fully DIY speakers and analog room eq. The amount of energy when everything running is insane in a acoustically small room.
Having said that what I achieved so far is just a baseline. There is still lots of tweaking from speakers to treatment enhancements to be done. But then we need to also draw a line somewhere provided limited space, time and to the fact with zillions of other hobbies and work-life. Cheers!
Past couple of years is what I would consider grinding and eye opening experience building 11.4 channel fully DIY speakers and analog room eq. The amount of energy when everything running is insane in a acoustically small room.
Having said that what I achieved so far is just a baseline. There is still lots of tweaking from speakers to treatment enhancements to be done. But then we need to also draw a line somewhere provided limited space, time and to the fact with zillions of other hobbies and work-life. Cheers!
@mayhem13: yes you are correct about the intention & placement scenario regarding this particular 3-way speaker build discussed in the last two pages of this thread. And this speaker's crossover design is what we are trying to refine now.
I do agree with everything regarding the importance of the room as well. Ideally, I would have used a well-enough treated room for my speakers but due to various restrictions in the living space, I am very limited in what I can do with room treatment for this speaker
(The title of this thread was for another 3-way speaker which underwent several different iterations over time and is currently in the form of a 2.5-way system with a relatively small waveguide/horn at the top and sits in my living room)
I do agree with everything regarding the importance of the room as well. Ideally, I would have used a well-enough treated room for my speakers but due to various restrictions in the living space, I am very limited in what I can do with room treatment for this speaker
(The title of this thread was for another 3-way speaker which underwent several different iterations over time and is currently in the form of a 2.5-way system with a relatively small waveguide/horn at the top and sits in my living room)