[jtalden]

Ok John,
That's good news!!! Now adding the delay of 0.58ms plus the 4.083 that is already in would give me a total delay of4.663 ms correct?

Jay,
Probably, but...
Are sure the current setting is 4.083? It sounds wrong based on my experience using the DCX2496. The delay is normally only adjustable to 2 decimal places. Even then the second decimal place usually skips some the numbers due to limitations of a 96000 Hz sample rate. There should be no real need to show 3 decimal places, but Your 4080 box may indeed do that.

So:
> If it really reads 4.083 then yes we want 4.083 + 0.58 = 4.663 ms [and yes then 4.666 ms is good in that case.]
> If it actually reads 4.83 we want 4.83 + 0.58 = 5.41 ms and thus the data in Post 40 is not usable.

Please reconfirm the actual settings and adjustment are correct before I do the analysis of the data in Post-40.

 

[Jay Compton]

John,
I just sent you an email confirming the numbers.

 

[jtalden]

Thanks, I will analyze the new data.

 

[Jay Compton]

Ok great I'll be waiting for your results...

 

[jtalden]

Good!
This time we hit the exact target of post 26.

The only data adjustment I needed to do was to invert all the drivers in the right channel to match the left channel. Make this change in the 4080.

Below is the resulting comparison of the right and left channels. It appears the level of the Right TW is slightly different from the left so one level could be adjusted so that they match.

We are done with the confirmation of Step-2.

Your next steps are EQ related so measure the 2 channels individually from the LP in order to EQ the bass range. It is a good idea to average several measurements around the main LP for this purpose particularly if you intend to make any EQ changes above 300 Hz. A window 2' high by 3' wide one reasonable option. You can EQ the 2 channels independently <300 Hz. To the extent the bass is mixed as mono, it may be a better idea to EQ based on the both channels operating together. It is probably situation dependent.
> A good option is to EQ the <300 relatively flat initially. A well setup system may not need any significant bass boost.
> A good option is to leave the EQ unchanged >300 Hz even though at the LP there may be some peaks and dips due to room effects. We have already set a good EQ on the direct sound.
> There is nothing wrong with trying to adjust >300 if you like for comparison, but changes there should be broad and modest in level.
> You can also experiment with different house curves if the overall balance is not to your preference.

Let us know how it works out.

 

[jtalden]

I intended to intended the mdat of the 2 channels:

View attachment ja Left and Right.mdat

 

[Jay Compton]

Ok John,
Thanks brother I will adjust the left TW to match the Right TW looks like -1db to me, and then I will setup to measure the speakers separately from the LP...

 

[Jay Compton]

John,
You said earlier to match the polarity of the right channel to the left correct? They were already the matched in the Xilica. The attached file is what happens when I flip the right side to opposite of what they were when you looked at the files.

 

[jtalden]

Is the mic at the LP now? This looks good except...

The woofers are at much higher SPL level relative to the horns when measured here at the LP. They way too high to EQ.

Remember, I had you reduce the MR and HF horns 8 dB in the 4080 in an early post when the mic was close to the speakers. I appears for the LP position we now need to increase the horns back up the about the same 8 dB. Just start there and see how it looks. If necessary adjust again as needed. The objective is to have about the same level from 300-2000 Hz. Once that level is found then you can start work to EQ the <300 range as needed.

 

[Jay Compton]

Ok John,
Agreed they are much louder now then earlier, I will the adjustments...

 

[Jay Compton]

John,
Here are the adjustments made to only the gains.

 

[jtalden]

What happened?
The 2 channels are opposite polarity of each other now. The polarities were correct in Post-48. Did you change one of the channels?

That's a very big sag/dip from 50-80 Hz. EQ will not be able to help with that. You may want to consider moving the LP position forward or backward to see if it can be improved.

 

[Jay Compton]

John,
Here are the first measurements from today, my notes are attached in the note section. The green trace is the left channel with all drivers in the inverted position. The red trace is the right channel all drivers in the inverted position, the blue trace is the right channel all drivers in the positive position.

As for the sound, the phantom center sound is present when all drivers from both channels are in the inverted position, when I put all the drivers from the right channel in the positive position I loose the phantom center channel sound.

 

[jtalden]

Jay the good news is that the charts confirm your impression of the phantom center. The channels are in phase when the left channel is negative and the right channel is positive. Below are impulse and phase charts:

The bad news is that there is a settings or wiring issue that is causing all the confusion.

Does the 4080 allow you invert the input channels? If so, one of the 2 inputs may be inverted by mistake. Maybe there is a channel mixing page that has wrong settings. I would look at the 4080 first since that is the new element in your system and I assume you did not previously have this problem.

 

[Jay Compton]

John,
Yes the 4080 allows for changes in polarity on the input side. I thought of that too and checked this what I found, all of the polarity's on the input side are set to positive +...

 

[Jay Compton]

John,
I can't tell you how much I appreciate your help, specially getting this wiring problem fixed!!! I was able to grab a new set of cables today for in the future... Looking forward to our next session maybe Friday or Saturday before I can get back to it.

 

[Jay Compton]

Hey John,
So I have the new cables and adapter and I am ready to continue, what is the first thing you want me to check?

 

[jtalden]

I suggest...

With:
> Mic at LP
> Loopback timing active
> All 6 of the 2040 outputs set to Negative.

Measure:
> Left channel alone
> Right channel alone
> SWs alone

That will confirm the polarity of the 2 channels now measures correctly with the new cabling and also allow us to determine the best polarity and delay timing for the SWs.

 

[Jay Compton]

Ok John,
Here are the measurements you requested, all the drivers are in the negative position. Left and Right and then the Left and Right subs.

 

[jtalden]

Okay. The phase chart of left and right channels now show the correct phase polarity (per the common convention) and the same phase rotation as it should. If the phantom center is present when you listen this way then finally we have the measurements agreeing with the stage imaging.

With these settings the best timing for the SWs is obtained with the current SW polarity and an increase of the SW delay of 5 ms. [An alternate very good setting is obtained by inverting the current SW polarity and using the current SW delay setting.]

Both of this options still result is a pretty big sag in the response 50-120 Hz. EQ can help smooth this out, but I suggest you also shift the SW-mains XO point from 50 Hz to 100 Hz so see if that significantly improves the result. To do that, just make the same series of 3 measurements with the XO freq changed to 100 Hz.

Below are the results of the from the current 50 Hz XO setting showing the overall responses with the current SW polarity and +5 ms SW delay.
SPL with SWs:

Phase with SWs:

Phase Tracking through the 50 Hz XO range: