• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Rja4000

Major Contributor
Forum Donor
Joined
May 31, 2019
Messages
2,755
Likes
4,673
Location
Liège, Belgium
Naïve question:
How far can one go with Roon ?
I mean, if you compute those complex filters, is that possible/easy to implement them with Roon software DSP?
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
@Trdat both Acourate and Audiolense produce excellent results and you can't go wrong with either. I do indeed use both interchangeably mostly depending on what is the easier path for the workflow required :)

The Visaton 890 MK3 looks very nice. I am not sure of the off-axis response as each device has it's own different polar response, but I could not say how they blend. Also, we are talking loudspeaker design where the whole is more than just the sum of the parts as it needs to be designed/engineered as a system. I.e. box volume, ported or sealed, vent size, baffle step response, cabinet edge diffraction and on it goes... Not the least of which is the Thiel/Small parameters for the drivers, including each drivers bandpass operation. Unless you are well versed in loudspeaker design, I would stick with the recommended design.

The main limitation of a 15" woofer is the XO point to the next driver (and it's fs). So if the 15" is only good to 700 or 800 Hz and if crossing over to a horn or waveguide, you need a larger horn and a larger exit compression driver to be able to go down low enough to crossover. Again there are so many variables it is difficult if not impossible to say whether a pair of 10" drivers would be better (at least has a higher crossover point) than a single 15". Are they short stroke, stiff cones or long stroke...? I wish I could give you a definitive answer, but there are so many variables involved that it requires a system design/engineering and listening tests to determine what's best for what you are looking for. If you don't already, diyAudio is a great place to hang out and discuss loudspeaker designs. Good luck with your system!
 

Ron Texas

Master Contributor
Forum Donor
Joined
Jun 10, 2018
Messages
6,226
Likes
9,353
@Trdat I personally would be afraid to modify the kit. Everything has been worked out for specified drivers. There is a lot more to it than just cutting holes in a different place. The supplied crossover probably will not work right and would have to be redesigned as well as cabinet size and the port. If you yearn for a 15" woofer, then find a design which starts out that way. In the long run, I believe it will work better.
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
Naïve question:
How far can one go with Roon ?
I mean, if you compute those complex filters, is that possible/easy to implement them with Roon software DSP?

Not naïve at all. Good question! The answer is yes, all the way :)

Roon has a convolution engine https://kb.roonlabs.com/DSP_Engine:_Convolution and supports many FIR filter designer software programs.
I use it, sounds great and I haven't had any buffering issues or dropouts.
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
968
Likes
397
Location
Yerevan "Sydney Born"
@Trdat I personally would be afraid to modify the kit. Everything has been worked out for specified drivers. There is a lot more to it than just cutting holes in a different place. The supplied crossover probably will not work right and would have to be redesigned as well as cabinet size and the port. If you yearn for a 15" woofer, then find a design which starts out that way. In the long run, I believe it will work better.

After some more deliberating I will have to agree with you and Mitch. Although, I will be bypassing the internal crossover and using DSP, I think keeping the original kit would take in consideration all the variables with speaker making. Using DSP to do the finishing touches and filters. I can't expect DSP to fully design the speaker crossover(digital) without the speaker having some design considerations, I mean at least not yet till I learn more and how much DSP can do in terms of offsetting all the variables. I know enough to design a speaker(at least on amateur level) and but the question is not, knowing enough, it's designing the perfect speaker, I should practise on a cheap version and leave the high end's such as the Visaton 890MK3 with the original kit.

Also, the Visaton is two 12 inches not 10inches and its a 3 way so for now I am heading in that direction. I will post some more threads around to see if someone can provide more insight between what two 12inches can do compared to a 15 inch. Also, curious about the off axis vertical response its about 30 degrees and feel that might be a really small sweet spot?


Otherwise, Mitch thanks for your reply as well. I think I have to analyse your articles and there should be enough to design crossover filters with Audiolense.
 

Ron Texas

Master Contributor
Forum Donor
Joined
Jun 10, 2018
Messages
6,226
Likes
9,353
@Trdat you can always add a huge sub or two later.
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
Hi @Trdat you are on the right track! I took Bob Crite's 3-way Cornscala kit as a finished design and bypassed the passive XO's and used digital XO's with excellent success. I ended up replacing the HF compression driver and horn to a much wider dispersion waveguide. The results, plus on and off axis responses are in my book.

You should be able to do this with any loudspeaker design that you like. While you can use the existing XO points from the passive XO as a starting point, you will be in a better position when measuring with Audiolense as you will get to see the passband of each driver and zero in on the ideal XO point. In the case of the Crites speaker above, the passive XO was 500 Hz from the 15" woofer (which sounded great in that ported box design down to 34 Hz - absolutely kicked butt) to the 2" compression driver and JBL copy waveguide. However, on close inspection, while the woofer could make it up to 1 kHz on and off axis, the combo CD and waveguide combo had a glitch at 800 Hz. I have attached the plot to show the before and after driver linearization to show that the glitch is still there, meaning it is an issue that DSP can't fix. So I changed the XO point from 500 Hz to 850 Hz and moved the glitch out of the passband with the steep linear phase XO. Not only measured better, but sounded better.

The Audiolense digital XO's are linear phase so that they sum perfectly in amplitude and phase. Further, the driver's passband acoustic slopes are "convolved" into digital XO crossover slopes, so this is how they still sum perfectly from an acoustic perspective as well. You can't go wrong :)

Wrt to double 12" comparing to 15", will be interesting to get feedback. There are so many variables... Most of the 15" that I use are "short stroke" stiff cones in (very) large boxes. I have found this is what gives the "kick" or "chest punch" sound. The JBL M2 woofer is like this as well. This still may apply to the 12", but will be good to get some feedback from folks that have built the kit you are looking at. When it comes to smaller drivers, I have found "long stroke" woofers can get the kick sound to, like Purifi's new PTT6.5 (sub)woofer can do, albeit at a lower SPL than the large cones.

Let us know what you find out.

Happy Holidays!
 

Attachments

  • midrange compression driver and waveguide raw and linearized response.png
    midrange compression driver and waveguide raw and linearized response.png
    479.7 KB · Views: 209

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
968
Likes
397
Location
Yerevan "Sydney Born"
@Trdat you can always add a huge sub or two later.

I got Dual DIY subs which I will be replacing with the Dayton RSS315HO which will turn my current cabinet into a .45 QTC aligned sub. To be honest, my current Dayton Classic Series DCS305 which is the entry level driver in the dayton range plays fine up till 180hz with say techno, no muddiness. Of course with finer music I have to drop the crossover to around 130hz but all in all I am surprised what a affordable driver can do in a descent cabinet with a descent QTC. I can only imagine the new driver with its heavier cone(more impact), heavier magnet(provide enough power to keep up with the heavy cone) and .45 QTC is going to do. Looks promising!
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
968
Likes
397
Location
Yerevan "Sydney Born"
Hi @Trdat you are on the right track! I took Bob Crite's 3-way Cornscala kit as a finished design and bypassed the passive XO's and used digital XO's with excellent success. I ended up replacing the HF compression driver and horn to a much wider dispersion waveguide. The results, plus on and off axis responses are in my book.
Well, I don't know much about waveguides only started my research on them a little while back. I am considering seeing if the exponential waveguide part of the kit can be enlargened(I won't be purchasing the waveguide I will be DIY'ing it with CNC) somehow to give a wider response vertically. Obviously ill talk to Visaton get an idea of what they think is possible or other options available. To be honest first things first, understanding beam angle technical specification given with waveguide by Visaton.

You should be able to do this with any loudspeaker design that you like. While you can use the existing XO points from the passive XO as a starting point, you will be in a better position when measuring with Audiolense as you will get to see the passband of each driver and zero in on the ideal XO point.

I have some basic experience measuring my subs with Audiolense attempting to compare transients with impulse response and trying to figure out the highest frequency it will play according to the measurements. So yes, I will measure and then understand how the each driver with its particular cabinet interacts with the room acoustics and see if the XO's need tweaking.


However, on close inspection, while the woofer could make it up to 1 kHz on and off axis, the combo CD and waveguide combo had a glitch at 800 Hz. I have attached the plot to show the before and after driver linearization to show that the glitch is still there.

Great I will look into this great example thanks.

The Audiolense digital XO's are linear phase so that they sum perfectly in amplitude and phase. Further, the driver's passband acoustic slopes are "convolved" into digital XO crossover slopes, so this is how they still sum perfectly from an acoustic perspective as well. You can't go wrong :)

Actually, I was thinking of writing back and asking the "complexed" question of "will I be able to get good results not using the original crossover"? and the above seems to answer my question. I suppose if we are digital crossover guys(I will get there soon) we should just drop all preconceptions of digital crossovers and put the hard yards in.

Most of the 15" that I use are "short stroke" stiff cones in (very) large boxes. I have found this is what gives the "kick" or "chest punch" sound. TI have found "long stroke" woofers can get the kick sound to, like Purifi's new PTT6.5 (sub)woofer can do, albeit at a lower SPL than the large cones.

This at least points me in the right direction, obviously reading your previous post I knew that short stroke must of had something to with the kick if you mentioned it as a variable. Now, I can at least do some research without the preliminary starting point that us amateurs are usually stuck in for a while till asking the right questions. I will use DIYAudio and see if that helps...

I do have a second option with a 15 inch speaker called Calpamos from Humblehomemade HiFi its not PA but its sensitivity is high, who have given a great explanation in a PDF on their choice on the 15 inch Faital Pro 15PR400. Its combined with a typical 2 inch compression driver and waveguide, this seems like a great option have no qualm about it. But the Visaton for one, is a 3 way and the Faital Pro drivers are expensive with only parts express having it for cheap but don't know for how long. Point being, either I am going to go ultra high end with a few bob more, or I will go with the Calpamos when I can find the drivers on special.

Let us know what you find out.

Happy Holidays!
 

jorgen

Member
Joined
Sep 3, 2018
Messages
8
Likes
4
Hi Mitch

Can't find your email adresse on your website!

Best regards,
Jorgen
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
Hi Jorgen,

mitch (at) accuratesound.ca

Kind regards,
Mitch
 
Last edited:

murl

Active Member
Joined
Sep 21, 2019
Messages
118
Likes
20
Hi Mitch,

I sent an email about possibly getting my 8cs calibrated. Are you still offering this as a service?

Thanks
Ryan
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
Hi Ryan,

Yes, I am. If you are Ryan P I responded Jan 12th. Perhaps caught up in junk mail? If not, send me a PM and we can sort it out.

Kind regards,
Mitch
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
A couple of announcements:

We are offering a "turnkey" miniPC music server with built-in DSP room correction.

The server comes with our remote room correction service as part of the package. This offering is designed for folks who want to take advantage of room correction, but don't have either the equipment and/or know-how to set up and configure a complete room correction system in a timely manner. The number one support issue we see is the number of hours/days and sometimes weeks spent trying to take an accurate room measurement in both the frequency and time domain. Our solution comes complete with everything installed, configured, tested and ready to use:

miniPC music server with built-in DSP room correction.jpg


2nd announcement is a walkthrough of Dirac Live 2: https://audiophilestyle.com/ca/revi...al-room-correction-software-walkthrough-r884/

Hope you are enjoying the music!
 

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,194
Likes
16,916
Location
Central Fl
A couple of announcements:
If I neglected to tell you before Mitch, I want to wish you the best of luck and success with your new enterprise.
You deserve it for all the work you've put into the audiophile community over the years.
Cent' Anni
 

Omid

Member
Joined
Nov 8, 2019
Messages
22
Likes
14
Hi Mitch,

I was looking at your measured room response and was really impressed at your results (flatter freq response, with less dB correction than my setup). I figured I'd check out your speakers for interest, and noted your were selling them on canuck audiomart. I hope I'm not being too nosy but what are you upgrading to?

Omid
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
Hi Omid,

Not nosy at all. I had a momentary lapse of judgement, then I quickly came to my senses :) They are not for sale anymore, but can't take the ad down. If I was to upgrade, it would likely be to change the narrower JBL 2384 waveguide pattern from 90 x 50 degrees to say a JBL M2 waveguide with it's 120 x 110 pattern. Or perhaps an Earl Geddes design 90 degree OSWG design I have been eyeing over at diyAudio.

One of the reasons I am able to achieve a flatter response is due to being an active 3 way system. I can individually tailor the output level to each driver to better match the end target, so less correction is applied.

Kind regards,
Mitch
 

Omid

Member
Joined
Nov 8, 2019
Messages
22
Likes
14
I think my living room is my chain's main weakness, but I'm always curious about how to improve my system. I have a pair of Tannoy GR stirling speakers. I generally like how they sound; there is something about coaxial drivers, a single point source that works well for me.
However, a couple of things have prompted me to look at other speakers. First, I measure a steep roll off at 12,000Hz and up, in the far field (even though the speaker points directly at the mic). To be fair, in the near-field the speaker measure much better. I've used audiolense to get a treble boost to compensate for the roll off, but I wonder if I might be pushing the tweeter beyond its comfort zone when I play loud. I 'll do an REW measurement to see if there much distortion in the higher frequencies.
The second reason is that when I initially considered bypassing the crossover, I put an inspection camera in the speaker to see what it would take to disable the XO. I was horrified at the amateurish layout and solder on the crossover board. Not what I expected for the price I paid. I guess that's trivial as long as it does the job right. Looking at my response curve though, the frequency response curve is far from ruler flat (let's call it an idiosyncratic Scottish design!).
That's why I like Amir's reviews, it keeps the manufacturers accountable for their sloppy work. Anyway, I could disable the crossover, and reuse one of the caps on the board to protect the tweeter, but I'd have to unscrew the woofer out of the cabinet, and rewire things. I'm a bit reluctant given the price of the speakers and the fact that the results may not even be audibly better (I'm guessing Audiolense corrects the sins of the crossover in large part). Also not sure of there is a seal/glue around the woofer: I don't want to disrupt an air tight seal.
I'll look at the ML2. Cheers.
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
643
Likes
2,408
Re: Tannoy GR Stirling loudspeakers - very nice!! I have a soft spot for Tannoy's and coax's in general (I have a pair of LS50's). I used a variety of Tannoy's in the studio as a much (much!) better alternative to the Urei 813 time aligns.

I had a pair of Bob Crite's 3 way Cornscala's with passive XO in 2013. I then carefully disconnected the passive XO and went active with digital XO and never went back. To my ears and measures, the improvement was significant. When I sold the system, I carefully put back the passive XO and tested to ensure everything was up to factory spec. They were a good sounding speaker and the measurement results are documented in my book as that is what I used as an example.

Audiolense (and Acourate) do a good job with passive XO's for sure. But going all active with digital XO is another evolutionary step up in the loudspeaker optimization game. Aside from bypassing the passive XO, there are additional items to consider like a multichannel DAC and more amplifiers. However, you now have total control over each individual driver and to my ears and measurements, it is worth the effort.

Good luck!
 
Top Bottom