A 3 way design study

[vineethkumar01]

Vineeth has 2x9" woofers, perhaps it has smoother vertical response than a single 10"+horn?

About 10 years ago k made several speakers with big horns and cardioid bass to get good sound at home (with minimal reflections), but later he noticed that investing on acoustics and wider smooth directivity speakers give more pleasant sound. This is my interpretation, he sometimes gives a bit mystical comments...

Thank you @Juhazi
I had been fascinated with Kimmosto's techniques experimentation etc since last 3 years or so..
I have his old website details/speaker details taken from 'wayback machine' and stored somewhere on laptop..
My main problem is investing in room treatment. I have been living in a rented accommodation since last 3 years and may need to move from here soon. Probably need to move once every 2 years (at least till I have/keep my current job)... I have had relatively bad experiences with wide radiating speakers of 6.5inch + 1inch format 2 ways in the past (none mentioned in this thread) when kept in untreated rooms and when turned up a bit loud.. Hence my newfound fascination with horn speakers and relatively bigger drivers..
Maybe I will move to somewhere else soon.. But I am not sure.. so just hanging on to whatever I have for now for learning more about all this and enjoy this hobby/addiction..
I quit academia after a PhD in wireless since I was fed up of all paper publishing (which these days has become more important than enjoying research, learning for learning's sake and all). I thought the industry might suit me better/hands-on experience and all. Of late I have been finding that is not probably the case.. So I have to do something else next...
By the way, this is me..
https://scholar.google.co.in/citations?user=sh3qFZoAAAAJ&hl=en

 

[Juhazi]

Thanks for telling about your life circumstances, Vineeth!

Yes, modern hard walled smallish room are very difficult for hifi. That's basically why waveguides have gained so much popularity - and already 30 years ago here in Finland. When I lived in such apartments, I liked good headphones more than speakers and simply just couldn't listen loud. That was before wg/horn home speakers existed...

I think that it is ok to have different tastes and preferences of hifi sound. Dedicated acoustically treated rooms are random luxury - and often too highly dampded IMO! Recording/mastering studios have also several different speaker sets, just because of this.

https://www.linkedin.com/in/juha-sirkka-848a377b/

 

[fluid]

Also, Kimmosto's comments on the ASR about how a 10inch woofer + 1 inch (titanium diagphragm) CD on a horn is sort of like the worst case scenario is making me wonder what I am missing... I want to know more about it.. Requesting everyone for opinions/comments about this aspect..

Be careful not to miss the part about choosing something else for a highly reverberant room which you look to have. You have also smoothed the directivity step between the woofer and waveguide and apart from a bit of a wiggle it is generally smoothly climbing. This is mitigating some of the complaint. The 10" driver size is likely due to where the directivity kink would occur and the titanium diaphragms have a history of sounding bright which could be made worse by using a high directivity waveguide to throw it straight at you.

I really don't know why I am attracted so much to speakers, their drivers, the science behind them all, and the overall music-listening experiences on different systems.. It has sort of become like an addiction of late..

Welcome to the club

For in room measurements measure from the listening position with both a single point and moving mic as they will both give different information. Have a look at the single sweep through a frequency dependent window of varying cycles from 4 to 15. Any big dips leave alone or look to mitigate through position or treatment. Otherwise EQ to be flat or a slight rise depending on taste.

 

[vineethkumar01]

Thanks for telling about your life circumstances, Vineeth!

Yes, modern hard walled smallish room are very difficult for hifi. That's basically why waveguides have gained so much popularity - and already 30 years ago here in Finland. When I lived in such apartments, I liked good headphones more than speakers and simply just couldn't listen loud. That was before wg/horn home speakers existed...

I think that it is ok to have different tastes and preferences of hifi sound. Dedicated acoustically treated rooms are random luxury - and often too highly dampded IMO! Recording/mastering studios have also several different speaker sets, just because of this.

https://www.linkedin.com/in/juha-sirkka-848a377b/

Thank you @Juhazi..
I too used headphones mostly till my grad school days due to lack of space to accomodate speakers in hostel rooms and all.. But I kept missing that physical impact of bass and the overall feeling of sort of sound washing over the body or something like that.. I tried a lot of headphones from dynamic driver ones to little higher end planar magnetic ones (not electrostatic ones). But I kept missing the sound of speakers.. Also most of them sounded different.. like speakers..
So pretty soon came back to speaker land..

I keep hearing from friends that Finland is a great country.. in terms of work-life balance etc.. Someday I want to travel to all these places.. Recently, both my wife and I sort of got close to getting post doc offers (in wireless) from a Professor at University of Oulu. But we didn't pursue it further since I was and still am not sure about whether i want to stick on to the same domain.. (I find speakers more interesting ) Anyway, I will stop rambling and get back to topic..

I will continue the experimention on speaker configurations and keep learning from all of you and post my updates/observations here

 

[vineethkumar01]

Be careful not to miss the part about choosing something else for a highly reverberant room which you look to have. You have also smoothed the directivity step between the woofer and waveguide and apart from a bit of a wiggle it is generally smoothly climbing. This is mitigating some of the complaint. The 10" driver size is likely due to where the directivity kink would occur and the titanium diaphragms have a history of sounding bright which could be made worse by using a high directivity waveguide to throw it straight at you.

Thank you @fluid for this explanation..
Is there a waveguide like ST260 that has such smooth response/directivity down to even lower in frequency like 800Hz.. I am not sure of the ST280.
Something that works with either a 1.4inch CD or a 1 inch CD like the BMS 4550? This is to experiment with my 15inch 15PR400 driver..

Welcome to the club

Thank you..

For in room measurements measure from the listening position with both a single point and moving mic as they will both give different information. Have a look at the single sweep through a frequency dependent window of varying cycles from 4 to 15. Any big dips leave alone or look to mitigate through position or treatment. Otherwise EQ to be flat or a slight rise depending on taste.

Sure. I will do this next.. Can you please suggest me any resources to learn more about the moving mic technique?

 

[fluid]

Is there a waveguide like ST260 that has such smooth response/directivity down to even lower in frequency like 800Hz..

CE360 might work or the 15" size designs Tom Kamphuys made in the Ath thread. A 1.4" driver is more capable of getting to 800Hz in general but of course with SPL limits in place many 1" drivers could still get there. The bigger waveguides are harder to print though.

Sure. I will do this next.. Can you please suggest me any resources to learn more about the moving mic technique?

https://www.audiosciencereview.com/forum/index.php?threads/rew-moving-microphone-method-help.12641/

https://www.ohl.to/audio/downloads/MMM-moving-mic-measurement.pdf

 

[tktran303]

I've lived in 6 houses in 8 years. From listening rooms of 3x3m to 6x9x3.6m. Make the best out of a bad situation, to make lemonade out of lemons. Look for the silver lining in the cloud.

Short of modifications to your room/acoustics, the best thing you can do is to try Dirac Live for Windows/MacOS.
The biggest game changer and I've been using it on/off (depending on the room) since 2015...

 

[vineethkumar01]

Thanks a lot @tktran303 for pointing out Dirac live.. This is very interesting
I will try out the free version first and the free alternatives mentioned here.. https://www.audiosciencereview.com/...corrections-in-rephase.31655/#js-post-1117185

It seems like the technique used is some sort of deconvolution algorithm along with appropriate filtering.. This is an area I am comfortable with If these sort of algorithms works out in my case, I can even request my wife to collaborate and help me study/develop algorithms like these even using machine learning techniques if needed as she is an expert in that area..

 

[tktran303]

I'm not sure if I've shared this with you previously but the first speaker I measured with Dirac Live (version 1) is this one:

https://www.htguide.com/forum/forum...ts-ii-a-musical-evolution?p=780336#post780336

Here's in-room amplitude response measurements:
https://www.htguide.com/forum/forum...ts-ii-a-musical-evolution?p=784927#post784927

 

[hifijim]

To be honest I am really liking this system. My job pressure is restricting the time I get to work on this project but maybe by christmas time I would have determined which one I like more.. The wavecor system or this horn system

Really great performance. In your high-reflective environment, you may very well prefer the high directivity horn.

 

[vineethkumar01]

I'm not sure if I've shared this with you previously but the first speaker I measured with Dirac Live (version 1) is this one:

https://www.htguide.com/forum/forum...ts-ii-a-musical-evolution?p=780336#post780336

Here's in-room amplitude response measurements:
https://www.htguide.com/forum/forum...ts-ii-a-musical-evolution?p=784927#post784927

Fascinating stuff.. I haven't seen these posts earlier..
Although from what I understand so far, we are implementing a 1-dimensional solution (preprocessing at signal level) to a 3 dimensional problem (waves radiating and hitting boundaries in 3D space and the resultant reflected waves screwing up response at listening position).
However, if the directivity of the speaker itself is good enough and as long a we get good enough data about the room (sort of map out at least major reflection causing boundaries in the room w.r.t listening position), we can use even simpler deconvolution algorithms to get good results. The scope of these results (area in which the good/predicted frequency responses after correction) can be obtained again depends on the acoustics of the whole problem I guess (including speaker directivity, room size and listening position).. But having predictable results even in a relatively small are is good I think.. I'm going to work on this..

 

[fluid]

Thanks a lot @tktran303 for pointing out Dirac live.. This is very interesting
I will try out the free version first and the free alternatives mentioned here.. https://www.audiosciencereview.com/...corrections-in-rephase.31655/#js-post-1117185

You can do better than Dirac for free, my advice is to start here
https://www.diyaudio.com/community/...rical-loudspeaker-correction-networks.275730/
No self respecting tweaker can be happy with a closed algorithm

https://drc-fir.sourceforge.net/doc/drc.html

"Here it is a brief summary of what DRC does:

1. Initial windowing and normalization of the input impulse response.
2. Optional microphone compensation.
3. Initial dip limiting to prevent numerical instabilities during homomorphic deconvolution.
4. Decomposition into minimum phase and excess phase components using homomorphic deconvolution.
5. Prefiltering of the minimum phase component with frequency dependent windowing.
6. Frequency response dip limiting of the minimum phase component to prevent numerical instabilities during the inversion step.
7. Prefiltering of the excess phase component with frequency dependent windowing.
8. Normalization and convolution of the preprocessed minimum phase and excess phase components (optional starting from version 2.0.0).
9. Impulse response inversion through least square techniques or fast deconvolution.
10. Optional computation of a psychoacoustic target response based on the magnitude response envelope of the corrected impulse response.
11. Frequency response peak limiting to prevent speaker and amplification overload.
12. Ringing truncation with frequency dependent windowing to remove any unwanted excessive ringing caused by the inversion stage and the peak limiting stage.
13. Postfiltering to remove uncorrectable (subsonic and ultrasonic) bands and to provide the final target frequency response.
14. Optional generation of a minimum phase version of the correction filter.
15. Final optional test convolution of the correction filter with the input impulse response."

 

[fluid]

Although from what I understand so far, we are implementing a 1-dimensional solution (preprocessing at signal level) to a 3 dimensional problem (waves radiating and hitting boundaries in 3D space and the resultant reflected waves screwing up response at listening position).

This is true but if you do it properly it is surprising how well it works. Don't let the fact that an impulse is only valid for one point in space stop you from improving the sound you hear with it (often over a reasonable area)

 

[tktran303]

Very interesting that program. I haven’t used it. @fluid Have you seen or done a head to head comparison of those two?

Re: Dirac they keep upgrading the software making it more and more user friendly but less tweakable unless you look under the hood.
So by DL 3 it’s has a slick UI/UX maybe a bit TOO user friendly.

@vineeth: Dirac asks if you are listening from a chair, a couch or a theatre, and it takes 8-9 measurements over a narrow, medium or wide area respectively. I didn’t do a deep dive into how it works; but it’s the first equalisation program that makes things generally better, rather than generally mixed, or sometimes worse.


I’m aware of Acourate and other proprietary apps; but nothing in a single box- PC free; which is exactly why I like Dirac.

Digital Data in -> Dirac Live Black box -> Digital Data out

Done in 30 minutes. Engage or bypass.

Yes, my Windows brethren won’t use something so closed and proprietary, these buddies of mine like muck around with Android; I just use iPhone (since version 4)

For me; after equalising for ‘listening whilst on the treadmill’ position, well, I was sold!

 

[vineethkumar01]

You can do better than Dirac for free, my advice is to start here
https://www.diyaudio.com/community/...rical-loudspeaker-correction-networks.275730/
No self respecting tweaker can be happy with a closed algorithm

https://drc-fir.sourceforge.net/doc/drc.html

Awesome..
Thanks @fluid. I am definitely going to try this one out..
The impulse response/estimation/filtering/..... is a language that I understand a bit better compared to acoustics..

 

[tmuikku]

I dont actually know what EQ to dial in or which modes to cut in the above room responses... Please guide me about those aspects.

This was the comment from Kimmosto that I was talking about (In fact there is a lot of interesting discussion in that thread)
https://audiosciencereview.com/forum/index.php?threads/neumann-kh150.33454/post-1364894

Thanks, its nice kimmosto is posting again his amusings I see new term coined "acoustic resolution" which seems to relate to directivity and room positioning and acoustics, as well as the speaker itself. As suspected the notion that "1" + 10" is worst" is application dependent. I understood his message its about directivity vs. the room, "deep horn" makes narrow dispersion on top so very high DI, where as 10" will have very low DI to somewhere 1kHz, perhaps magnified with poor directivity match around crossover. I see a cure for something like this is to have lower or higher DI all the way through crossover region so that the whole bandwidth is uniform in a way. As narrow coverage angle is not so easy below 1kHz it would be perhaps better to have wide coverage angle all the way to top without a horn/waveguide. He mentions resolution mismatch and other issues besides coverage, like difference in dynamic capability to distortion splitting the full bandwidth sound in two in many respects.

Well, only way to understand and figure out stuff like this is to listen the options and take notes. We can read his message like bible or listen various systems by ourselves and figure out whats it all about in our environments Its very hard to connect dots without having auditory experience to go with words and visuals. When there is experience of various systems, then it would be easier to interpret and relate to what he is saying. Or anyone else for that matter. Phase matched xo is mentioned again like in his c-c stuff few years back, so what we learn here at least is that low order phase matched xo is good for sound, and the physical system should be made so that it works in a room with such xo (or FIR to have good step response).

For the room EQ, I'm not pro at it, there is plenty of resources on the forums and for example REW doc. This I know though, getting low / high balance right is important and it depends on the room and what you like to listen to. When anechoic axial response is flat the sound at listening position is probably not, unless its outdoors. You get something like the power response and possibly great peaks for the lows depending on toe-in and positioning to boundaries. You'd probably also spot any issues in measurements used in simulation, or errors in DSP implementation and all sorts of things. Opportunity to match left and right to improve stereo image and stuff like that. Balance response to toe-in. All kinds of stuff that affect perceived sound quality. What ever you do is good, it helps to connect sound to the graphs and gain knowledge what sounds better.

I found out that the speaker matching was off as said earlier, need to measure both and simulate both sides if want very smooth response or leave out micromanaging. Another thing was, after balancing out the bass section all of a sudden there is lots of clarity in the mid as less masking I think, also its now evident my bass boxes aren't that good, I think I'm hearing how the boxes sound (both are different size prototypes). Well, all in all a good learning exercise for measurement techniques, listening, connecting graphs to sound, translating speaker properties like directivity to sound and so on. I used just simple MMM technique with REW "Psychoacoustic" smoothing enabled to tune in the balance, adjusting DSP settings some to bring the sides closer together and fine tune the "slope". I have not much experience on this so better listen to fluid or others what one should do My advice is just get messing with it so understanding the advice gets easier.

 

[Juhazi]

Jean-Luc Ohl is the pro about multimic room analysis. You can use sine sweep or noise for measurements, overlays or averaging. It soon comes apparent that high Q corrections are not for good, except perhaps for the lowest mode (longest distance between walls in the room typically)

https://www.ohl.to/

I don't like the idea of Dirac either, and haven't installed it. Basically, regarding timing/GD, it tries to correct the highest amplitude of response to same arrival. But it can't do anything about "tails" ie. modes/reverb unlike room treatment or positioning of speakers and listener. But I'm not an expert...

Wavelet of a room response in REW https://www.roomeqwizard.com/help/help_en-GB/html/graph_spectrogram.html

 

[fluid]

Very interesting that program. I haven’t used it. @fluid Have you seen or done a head to head comparison of those two?

Dirac asks if you are listening from a chair, a couch or a theatre, and it takes 8-9 measurements over a narrow, medium or wide area respectively. I didn’t do a deep dive into how it works; but it’s the first equalisation program that makes things generally better, rather than generally mixed, or sometimes worse.

Dirac is meant for people who pretty much want to push a button and have it done for you. I do not believe that this is the best way to go. This whole process is still more alchemy than anything else and as far as I am concerned it needs to be evaluated and modified to get it right.

I’m aware of Acourate and other proprietary apps; but nothing in a single box- PC free; which is exactly why I like Dirac.

Which is exactly why it doesn't have the horsepower to do much with low frequencies. The PC doesn't have to be the last link in the chain though, the player can still be a wireless end point.

 

[tktran303]

Thanks for that link fluid. Mitch and I corresponded privately a lot about the state of the art in RoomEQ when he released his PDF book on Amazon back in 2016.

I had already dropped in/out of roomEQ since 2014-2015 with Dirac Live and didn't come back, so I've been ignorant.

I see the improvements with more flexibility in using a general purpose CPU or SOC and having more taps allow more FIR filters for EQ under 100Hz where's it may be most needed, not being locked into proprietary solution or having to buy multiple license for each and every DSP hardware product, which is how Dirac is funded eg. PC software & miniDSP since 2014, then automotive, most recently AVRs.

Having said that I recently traded my nanoDigi for a Flex 8 and prefer it because it accepts Toslink, coaxial and USB and Bluetooth 5, (AAC, aptX, etc).

Sometimes there's DIY and sometimes the spouse just wants to listen to music via her phone or that old CD player, without having to press 10 buttons to get things going.

 

[Juhazi]

Your new academic career here? Combining interest in signal processing and audio... Climate is much better than in Oulu, but rents/houses are more expensive. Sea, lakes, wildwoods and international airport are near!

https://www.aalto.fi/en/department-of-signal-processing-and-acoustics
https://en.wikipedia.org/wiki/Espoo