Post Directiva r1 Active Crossovers here

[RickS]

Here is v20 as it is the minidsp HD one in Amir's review....

The Woofer peak filter was needed for his older woofer. It is not used for my newer one. Note that the woofer high pass filter is shorted out, but I use it and may be useful for those wanting to use r1 with subwoofers. The VituixCAD project is attached for simulating the review speaker too.

 

[TimVG]

Okay, been fooling around for a good 30 min.

Sidewall average/Horizontal ERDI + 0-90° horizontal

Vertical +/- 30°

 

[TimVG]

So I went a slightly more radical route using 8th order slopes, just because the miniDSP can. I also sacrificed some of the horizontal smoothness in favor of a flatter response and a better (or at least different) vertical integration.

 

[abdo123]

I also sacrificed some of the horizontal smoothness in favor of a flatter response

How is that ever a good compromise in an active design?

 

[TimVG]

How is that ever a good compromise in an active design?

"Some". It's still smooth. Also I respect the choices CTRL made. But I figured the goal here was to experiment. So I optimized some different aspects in the hope that Rick could share how he experiences the differences.

Edit: this would also be a good 'nearfield' x-over

 

[thorvat]

So I went a slightly more radical route using 8th order slopes, just because the miniDSP can. I also sacrificed some of the horizontal smoothness in favor of a flatter response and a better (or at least different) vertical integration.

It looks good. I would also consider bumping 1-2kHz area (centered at 1,5kHz) app 1,5 dB up.

 

[McFly]

Please post phase responses people, need to see how the drivers track each other. I had a crack at a passive last night, but didnt get anything conclusive. My skills are lacking, though.

 

[TimVG]

Please post phase responses people, need to see how the drivers track each other. I had a crack at a passive last night, but didnt get anything conclusive. My skills are lacking, though.

You could always open the file in Vcad and check.. But here's mine.

 

[TimVG]

Here's another one, with priority on the horizontal reflections.

Spinorama:

Horizontal behaviour (photo #2 shows on-axis, LW and sidewall reflection average + horizontal ER DI)

 

[headshake]

It would be neat to know the preference rating (the first one) when you change the XO.

Congrats again to the Directiva team for making a great XO!

I was able to get a 7.226 Pref. score by changing around the 5khz to 11khz a bit. This is such a tiny and petty change though.



edit: dotted lines are the original XO.

I am still tinkering with lower XO's but it is really hard to top the work that has been done.

edit2:
I don't know if I traced the right one and I did not save that iteration. Here is one I worked on a bit more. Pref. score: 7.303:

 

[ctrl]

Hi folks,
it would be very helpful if you could show the version measured by Amir (v20, see Rick's first post) as an overlay in the crossover suggestions.

This would look like this for example:

With one glance one can see that the low bass around 60Hz was raised a bit and the sound power dip was filled up a bit.

This makes it much easier to transfer the changes in their effects to Amir's CEA2034 measurement.
Unfortunately, due to various circumstances, the VCAD simulation deviates somewhat from the real measurement and one is forced to take an additional step in the evaluation of the crossover proposals and compare the changes in VCAD with the CEA2034 measurement.



So @TimVG and @headshake if it is still possible for you to adjust your posts accordingly, that would be very helpful.

 

[TimVG]

@ctrl

Done. Here they are seperately.

Actually despite me starting from scatch, my 2nd attempt where I prioritized a smooth horizontal DI, apart from some overall levels, isn't too far off from what you ended up on.

 

[peanuts]

why dont you simply upload the minidsp config files, where are they?

 

[abdo123]

why dont you simply upload the minidsp config files, where are they?

the config files would be both device and sampling rate dependent, it's much better that they share the PEQ filters with us and we enter it manually.

 

[kipman725]

could improve vertical polars using FIR brickwall filters?

 

[abdo123]

could improve vertical polars using FIR brickwall filters?

FIR crossovers take a serious amount of processing power so you would probably be limited to using a PC for the crossover, you also could run into ringing issues.

But i would love if someone could explore that.

 

[kipman725]

FIR crossovers take a serious amount of processing power so you would probably be limited to using a PC for the crossover, you also could run into ringing issues.

But i would love if someone could explore that.

I can do up to 1024 tap filters (without using too much processing %) in my Symetrix 8x8 DSP I just haven't had the time to take comprehensive enough measurements of my own speakers to be confident in using them. There is this brand new unit for £515 that does FIR:

the t.racks FIR DSP 408

Digital Speaker Management System With FIR filter, 4 Inputs 8 outputs, 96 kHz sampling rate, 32-Bit DSP, 24-Bit AD / DA converter, XLR inputs and outputs, USB and RS232 / 485 interface, Ethernet interface, Easy to read display, Format: 19"/ 1U,...

www.thomann.de

this free software can be used to define the filters:

rePhase

rePhase, a loudspeaker phase linearization, EQ and filtering tool. rePhase is a free FIR generation tool for building fully linear-phase active crossovers with arbitrary slopes. It also let you manually compensate for the phase shifts of your loudspeakers and existing crossovers, be it active or...

rephase.org

 

[mdsimon2]

I can do up to 1024 tap filters (without using too much processing %) in my Symetrix 8x8 DSP I just haven't had the time to take comprehensive enough measurements of my own speakers to be confident in using them. There is this brand new unit for £515 that does FIR:

the t.racks FIR DSP 408

Digital Speaker Management System With FIR filter, 4 Inputs 8 outputs, 96 kHz sampling rate, 32-Bit DSP, 24-Bit AD / DA converter, XLR inputs and outputs, USB and RS232 / 485 interface, Ethernet interface, Easy to read display, Format: 19"/ 1U,...

www.thomann.de

this free software can be used to define the filters:

rePhase

rePhase, a loudspeaker phase linearization, EQ and filtering tool. rePhase is a free FIR generation tool for building fully linear-phase active crossovers with arbitrary slopes. It also let you manually compensate for the phase shifts of your loudspeakers and existing crossovers, be it active or...

rephase.org

The 2X4HD has a total of 4096 taps at 96 kHz, it could probably implement a linear phase 96 dB/oct LR x-over at 2-3 kHz relatively easily. I haven't found much need for steep linear phase x-overs but would certainly be something to try just because.

Michael

 

[abdo123]

The 2X4HD has a total of 4096 taps at 96 kHz, it could probably implement a linear phase 96 dB/oct LR x-over at 2-3 kHz relatively easily.

Really? with just 1024 taps per filter?

 

[thorvat]

Really? with just 1024 taps per filter?

You don't need many taps for that. And no, such filter would not cause any pre-ringing. Not to mention that phase deviation of XO filter can also easilly be corrected in rePhase (in case you would use minimum instead of linear phase filter).