A 3 way design study

A 700Hz ish crossover version :)
1712678462255.png

1712678471829.png


Reverse null
1712678498157.png


I am just restricting 5 biquads per channel due to the constraints on the DSP board for now.
The rest of the 8 biquads can only be applied as global EQ
 
Hey 👋🏼

I like your thread very much. Your project with the sica coaxial inspired me to build my own(Pics) , thanks

Maybe you write it, but can t find it.
What are the paramet. EQ in 2925Hz & 3733Hz are for?
Is it hearable If you turn it off?

Thanks, best greetings Daniel
 

Attachments

  • PXL_20240221_094109882.PORTRAIT.ORIGINAL.jpg
    PXL_20240221_094109882.PORTRAIT.ORIGINAL.jpg
    634 KB · Views: 70
  • PXL_20240221_094624094.PORTRAIT.jpg
    PXL_20240221_094624094.PORTRAIT.jpg
    539.6 KB · Views: 72
  • Like
  • Thank You
Reactions: 1 users
Thanks @Eggg3r
Your project looks very nice too.. :)

The Satori WO24P's have a very nasty breakup peak.. This driver is unlistenable without enough suppression.
With this lower frequency crossover, it is ok compared to a 1.2kHz crossover which I had earlier in another system (for which I made that EQ).

Crossover without those break up suppression EQs
1712682062180.png

1712682101788.png


With EQ
1712682187824.png
 
For this the DI is very helpful as a simple inidicator of the overall quality, certainly not an artifact - it describes pretty well how the source radiates sound, which is what we do listen to. It's all tied together.
Good point. And the word "artifact" was not the right word. I should have said that DI is a derived metric.

My point is that a visually abrupt change in the slope of the DI curve is not very important, as long as the DI curve continues to increase with frequency, which in this case it does.

j.
 
  • Like
Reactions: 1 user
Honestly, I had some concerns trying out this system given that the horns have slightly less beamwidth than I am used to. Also, given that the current trend is to have beamwidths hanging around 100+ degrees over the full frequency range..

But I am blown away by the performance of this system. Given the quickly cobbled up crossover, the Bluetooth connection between the source and the amp and everything that happens above 13kHz. It blows away the ST260 KVAR and everything other system configuration I have ever had or heard at home. I don't know if this has something to do with how the horn+system interacts with this particular room.. And given that I am a little biased towards dynamics compared to a speaker that has a very wide soundstage but lacks the dynamics .This system is just phenomenal.. :D

(If I were to nitpick, the male vocals are a tiny bit "thicker" compared to my reference headphones but that just Probably needs a minor tonality adjustment lower down in frequency..)
 
  • Like
Reactions: 2 users
Some people prefer a uniformly sloping PIR curve with a slope of 0.8 - 1.0 dB/octave. However, I prefer one which slopes down by about 5 dB from 100 to 1k, and then is nearly flat above 1k, with a slight downward slope. Your PIR curve is just about exactly how I like it.
I find this to be linked strongly to listening level. The flat above 1K with rise below seems to be a better proposition the louder the level is. The constantly falling response seems to me to be preferable at lower volumes, turn it up too much though and it can start to sound objectionable. Turn the flat from 1K down too much and a similar thing happens.
 
  • Like
  • Thank You
Reactions: 2 users
Yes I agree. That is a good observation. When I do serious listening, it is loud according to my wife. I listen with my eyes closed, and if she wants to talk to me during a loud passage, she has to come over to me and touch my shoulder... I would not hear her call my name. Once my neighbor across the street asked me if I was playing drums in the house.
 
  • Like
Reactions: 2 users
Just for fun, I made yet another video of the system playing 'The chain' by Fleetwood MAC and took some in room response measurements from around 1.2m away from the left speaker and mic at MLP
1712935972651.png



The video probably also shows that for serious recording of anything, I shouldn't be doing it holding the recording device (phone) in hand and also shouldn't move the hand while recording :D
 
  • Like
Reactions: 1 users
Listening to these videos is a good reminder of what an awesome job your ear and brain does of filtering out a lot of room interaction when you are there in the space with the speaker. The trick fails miserably through youtube and you hear the whole room and what it is doing.
 
  • Like
Reactions: 2 users
@fluid: Can you give explain a bit more about examples of a what in this video can be attributed to the room interaction. I want to learn that so that next time I can check for whether it is there or not..

And the biological signal processing system of our brain and ears is awesome regarding the filtering part.. But people still keep asking me to upload videos when I say I made a new speaker. So I still keep doing the same knowing that it is mostly pointless.. :D :D
 
  • Like
Reactions: 1 user
@fluid: Can you give explain a bit more about examples of a what in this video can be attributed to the room interaction. I want to learn that so that next time I can check for whether it is there or not..
It is hard to do without sounding negative. To me it sounds like the speaker is in a very reflective room, which it no doubt is. I am quite familar with the kick drum in that song and how it can sound with different processing and treatment. I am sure than in person it sounds much better than in the video.

But people still keep asking me to upload videos when I say I made a new speaker.
It is OK to say no :)
 
  • Like
  • Thank You
Reactions: 2 users
With such (great) waveguide some of early reflections from side walls could be eliminated by rotating speakers a bit more into the room, so their axis is crossed before the listening place.

In those videos from vineethkumar01 I appreciate clarity and non-colored middle and upper range. I hear similar sound (minus reflections and bass) if I play the same song on my system, that has also waveguide designed by MaBat.
 
  • Thank You
Reactions: 1 user
Anyone (with ears) who has recorded his own stereos knows how different the sound is in a video. Late reflections, echoes are terrible... even in a wooden frame house. Just guess how I know!

I don't actually understand the popularity of videos, but humans are primitive. Curiosity will get satisfied but evaluation is impossible in fact.
 
  • Thank You
Reactions: 1 user
Thanks a lot @fluid for the suggestions and feedback.. :)
This is exactly the kind of feedback that I was looking for.. Negative feedback as constructive criticism has always helped me (and probably most others too) :)
The room is probably pretty terrible here.. with full-wall-sized glass windows on one side and just a carpet and sofa doing the job of whatever room treatment is possible here.

I somehow knew that you know about the kickdrums on this track better than most others I know. So, I was expecting some feedback about it from you if you got a chance to see the video. :)
Since the start, I have had doubts about the tonality of the bass to lower mids on this system. The system sounds "darker"/ "thicker" (for the lack of better words to convey what I am hearing) in this region of the spectrum than when I play the same track on my PC and compare it with what I hear on my Truthear Crinnacle Zero headphones. The kickdrums sound slightly over-enthusiastic in the video and in real life compared to the headphones. I have been thinking about what to do regarding this (i.e., whether a simple EQ tweak would fix it or it is an unavoidable room-related artefact). Since the DSP board in this case is not as easily tweakable as DSP on my PC, which I have been using in my previous speakers, I have left the issue as such till now. Maybe I need to experiment a bit more and check if something can be done. Regarding reflections, there is probably nothing that I can do right away due to the living room situation.
Some tech details about the situation

Frequency response at MLP
1712994112406.png


The current RT 60 Topt and EDT plots look like below
1712994084286.png


Spectrogram
1712994182747.png


The peak energy dotted line in above curve keeps wiggling around till about 500Hz. I probably need to do something about it too...
 
  • Like
Reactions: 1 user