Jump to content

Xilica XP active crossover integration into my system


Chris A

Recommended Posts

If you had a tweeter, it looks like it is lagging the midrange (and down about 20-30 dB), so you need to put ~5.9 ms delay on the midrange channel, and 5.3 ms on the tweeter channel.

Link to comment
Share on other sites

15 minutes ago, Chris A said:

If you had a tweeter, it looks like it is lagging the midrange (and down about 20-30 dB), so you need to put ~5.9 ms delay on the midrange channel, and 5.3 ms on the tweeter channel.

 

With the above corrections. Better, though not sure how to interpret the bottom zig zag.

689974124_left1.jpg.0cef11a17b7ee4e0f11842a48cc06443.jpg

Link to comment
Share on other sites

Recommend toggling the vertical frequency axis to show logarithmic frequency, then you'll be able to see the response below 1 kHz (and also up to 20 kHz if you're "Limits" are set correctly).

Link to comment
Share on other sites

By the way, it now looks like 5.3 ms delay is a bit too much, maybe by 0.5 ms on the midrange and tweeter channels, judging from the bottom of your last spectrogram posted. The dotted line needs to smoothly bend away from the vertical axis from the midrange to bass--and if you play with it a while, you'll also see that the big spikes in the group delay plot will significantly decrease (which is good, because those group delay spikes are audible).

Link to comment
Share on other sites

31 minutes ago, tromprof said:

What is this telling me?

Left zoom.jpg

 

Can you tell me what drivers/horns are playing and where you've set the crossovers? It looks like you're crossing over to something at 100-200 Hz that's lagging the bass bin by 7 ms.

 

It would be really nice to see the vertical scale open up so you can see 10 Hz to 20 kHz.

 

Chris

Link to comment
Share on other sites

Here is a composite of the spectrogram that I posted earlier, the controls dialog and the limits dialog open so you can check the settings and correct them to show the same limits, etc.:

 

1357767045_Spectrogramlimitsandcontrolssettings.thumb.JPG.12cbc293d794c95aa1f049cff7503a09.JPG

 

The screen shot function is on the upper left hand corner that looks like a camera and is annotated by "capture".  You can set the size of the captured window in that dialog box that pops up.  For spectrograms, I recommend ~1200 pixels image width.

Link to comment
Share on other sites

I also wonder if the odd readings on the bass bin could be from the Integra pre/pro I use. It switches output on, with an audible mechanical click from the unit,  when it detects a signal. There is a slight delay, maybe affecting the beginning of the sweep?

Link to comment
Share on other sites

I spent several hours today playing test tones and measuring with REW. After getting the LR channels sounding superb with the Xilica I decided to tackle the other channels in my HT. I am using a pair of Ashly 3.24cl units which are not nearly as user friendly or capable but having learned A LOT over the last day the results are impressive. Speakers have location-ally vanished.   Everything is super clean, music and film sound awesome!

Many, many thanks to Chris A for sharing his knowledge. I owe you a beer or 3 for sure.

  • Like 4
Link to comment
Share on other sites

  • 3 months later...
On ‎6‎/‎7‎/‎2019 at 11:18 AM, Chris A said:

The same measurement, instead looking at the excess group delay plot:

 

1228710662_ExampleGroupDelayTimeMisalignment(BassBin4ms).thumb.jpg.3535e91246cda1dfdc74fd22b2fe3bd9.jpg

 

Chris

 

 

How is this read? On my marked up version below is it the red circles where the Excess Group Delay (white line) goes from horizontal at zero to where to goes vertical or is it the aqua circles where the white and blue lines don't track evenly or possibly something else? What does a trace look like when the proper delay is applied.

 

ChrisGDQuestions.JPG.c22488ee0958521bc830e325699a1bbe.JPG

 

Without having a Spectrogram to fist look at how is below read? Meaning what is the delay and is it applied to the HF or LF in a 2-way system?

 

GroupDelayNothing.jpg.7634fe90bc573a263c1b91f72e42d049.jpg

Link to comment
Share on other sites

Aqua for the low frequency, red for the high frequency (the HF is usually around zero ms group delay).  You're looking at the excess group delay curve some distance away from the crossover region.  In your plot with the red trace, your woofer/bass bin is trailing by 12-15 milliseconds relative to the HF driver/horn.

Link to comment
Share on other sites

1 hour ago, Chris A said:

Aqua for the low frequency, red for the high frequency (the HF is usually around zero ms group delay).  You're looking at the excess group delay curve some distance away from the crossover region.  In your plot with the red trace, your woofer/bass bin is trailing by 12-15 milliseconds relative to the HF driver/horn.

 

So are these next two statements for two different crossover points correct....

 

In the Green/White graph there is not much group delay around the 450 crossover point? Should I apply any delay, maybe because of the 182 difference?

 

450XOGroupDelay.jpg.81c5c8c26de8cb960ad63b9f212a7e2c.jpg

 

 

In the Blue/White Graph there is aprox  3ms around 280Hz and/or 6ms around 184Hz with the 350 crossover point

 

350XOGroupDelay.jpg.bb8442fd3ece1e8a3fa2f1ffe99009a7.jpg

 

 

If you were to go by the Red, Green, Blue group delay plots alone and no listening would you pick the 450 crossover point since it appears to not need any correcting?

Link to comment
Share on other sites

You're showing about 8 ms delay on the low frequency channel (and it's unknown where your crossover frequency is located in this plot).  I'd try 8 ms on the HF channel and iterate by 0.1 ms steps around that value until you find a happy medium (i.e., in-phase so that there aren't any cancellations in the crossover region--as seen in the spectrogram). 

Link to comment
Share on other sites

9 minutes ago, Chris A said:

You're showing about 8 ms delay on the low frequency channel (and it's unknown where your crossover frequency is located in this plot).  I'd try 8 ms on the HF channel and iterate by 0.1 ms steps around that value until you find a happy medium (i.e., in-phase so that there aren't any cancellations in the crossover region--as seen in the spectrogram). 

 

Red = 200 crossover

Blue = 350 crossover

Green = 450 crossover

 

Which one are you saying 8ms and around what frequency?

 

As for the green one the XO point is 450 but the Group Delay looks pretty good (to me) there. It is back around 182Hz that I can put the cursor on and see/read the following differences 42.3 - 36.8 = 5.5ms. But again to me...and I know nothing, it looks pretty good around 450. It is a 6dB Bessel, so 1st order with lots of overlap, right? If I were to account for the 5ms delay around 182 wouldn't that, perhaps, produce more group delay in other areas like 200-600 which seems pretty good?

 

450XO182HzWhite.jpg.e27d0010c5271118ce07792de00a2c61.jpg

 

450XO182HzGreen.jpg.93604c0b615a6f18a18d359e5ea185ca.jpg

 

Link to comment
Share on other sites

  • 1 year later...

I listen to 100% digital sources. I have a Parasound P6 playing middle man before the Xilica DSP in my system.

 

Digital source (Roon)  - converted to analog (Parasound P6) - converted to digital by the Xilica internally for processing - finally back to analog to Jubilees

 

I noticed that the XD-4080 offers a AES/EBU I/O.

 

Could I get something like a PI2AES to go digital from a Raspberry Pi into the XD-4080? I'd dial in the overall gain on the XD4080, and make fine adjustment within Roon.

https://www.pi2design.com/store/p19/PI2AES_-_PRO_AUDIO_SHIELD.html

 

If it worked it would end up with:

Digital source (Roon) - Digital PI2AES straight into the Xilica XD4080 - final Analog conversion to Jubilees

 

Do you think using AES into the XD-4080 have any improvement on sound? Am I being OCD trying to reduce the extra digital to analog conversion in my current setup?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...