I have attached a REW mdat. Levels set around 80dB. All speaker and sub distances set to 0. No crossovers used when taking sweeps 1-7. I routed the sub channels through the main right speaker channel. Sweep No 8 'front IB.' I put this though JRiver's bass management just to show this can be done.
Please let me know if you require any more sweeps at this stage.
Wull, I don't think we can do much with those measurements except align the subs for timing. In order to align the combined subs to the mains the XO has to be active for all measurements.
We can:
1. Remeasure all 6 speakers with the XO active. That way we can determine the timing needed for the SW's and also time align the timed group of SW's with the 2 mains. The final result will match the results you will get if you remeasure with the recommended timings applied.
2. We can do this is two steps. I can use this data to confirm the timings needed of the 4 SWs. We can then remeasure the 4 SW together with those timings and also the measure the 2 mains. All 3 of these measurements need to be with the XO active. That will determine he timing needed between the SW group and the mains.
Either way we end up at the same place; the best timing between the 4 SW's and the best timing between the SW group and the mains.
I will now be away working on other matters for several hours today, but can work on 1 later today with this data. If you want to provide the needed data for 2 instead, I will just wait for that data to be available. Either approach is good.
Analyzing the 4 SW for timing was interesting. There was consistency in the SW measurements between the 3 XO sets of measurements.
I first spent some time reviewing the pros and cons of choosing the better XO frequency. My first preference would be the 120Hz XO because the SPL level and rolloff rates appear to be more compatible between the mains and subs. That is, I would expect the XO range SPL to be smoother. The other 2 XO points appeared they may result in a significant sag in SPL in the 100-120 Hz range, but It is not possible to be sure without complete analysis of all 3 options in the final configuration.
Some individual SW measurements differed somewhat from the previous (post 17) measurements. Possibly the locations of some were changed. The as-is timing between the subs was good. It did differ from my normal time aligned target, but the resulting 4 sub group was likely even a little better than time aligned. With multiple subs it's possible to have several different timing/polarity settings that provide good results. It requires automated analysis to find the most attractive ones. Possibly you used MSO or just happened upon this favorable alignment. In this as-is setting, 3 of the subs are time aligned and the 4th (back-IB) is leading the others by 6 ms and is also inverted polarity from the others. The analysis indicates the resulting SPL and phase is somewhat similar to time aligned, but the as-is setting providing more SPL <18 Hz (a questionable benefit in my opinion). The more interesting result is that the as-is timing results in shorter reverberation time.
Some Comparison Charts:
Do you want me continue on and provide 120 Hz XO as-is timing to determine the 4Sub group vs mains timing.
Would you prefer I do the mains timing analysis using other options?
Wow. Gulp. Amazing!.......That nearly went completely over my head jtalden. A lot to take in at first...Thank you
Yes I had moved the front left & right 15 sub's a little to improve there response. The mic was also placed 12" away from when I first measured (post 17) for best seating arrangement.
No I haven't used MSO. (just got lucky by the sounds of it) I have read about MSO but that's as far as I have taken it.
Yes, more than happy to continue with the 120 Hz XO as-is timing to determine the 4Sub group vs mains timing if that's OK with you
I just started looking at the L and R. A Puzzle!
> The 2 tweeters are in phase.
> The phase thus tracks closely from 20k Hz down to about 1.5 kHz.
> The phase is 180° different <1.5 kHz.
Below are charts; SPL and phase tracking:
> The SPL chart shows the L + R vs. L/with polarity Inverted + R. It's easy to see that when the polarity of L + R is the same the SPL is stronger above 1.5k and weaker below 1.5k when compared to reversing the polarity of L before summing the 2 mains.
> The phase chart of L vs R shows close phase tracking above 1.5k and a 180° difference below that. The L channel shows the abrupt 180° shift at 1.5k and the R shows a smooth continuation of phase through the entire range.
So:
Can you add some background? What speakers are these? Is there an XO between drivers at about 1.5 kHz? Is there any other reason to think there may be a problem with the L speaker?....
Having looked at both L & R speaker XO's I cant see anything wrong. Both XO's look the same. The components have all been soldered the same way round etc. So I have swapped the XO's from one speaker to another just to see if the problem moves. I have attached a REW mdat for you to peruse. I'm finding it very difficult to read the phase chart if I am completely honest.
I will look at what you provided, but want to point out first that for issues like this, the direct sound phase is much easier to read with mic closer to the speakers. With the mic about 1 m from the center horn, the phase is more easily read. Setting the FDW window to 6 cycles/octave will also clean up the trace a lot. [Don't use loopback or acoustic timing.]
I would first use a 1.5 V battery on each speaker and confirm that all woofers move the same direction on both speakers when using the same battery polarity. If they do, then there is no problem.
Looking at the data:
The Left Speaker (as original) is the only one that appears to be wrong.
When the XO was swapped to the right speaker the measurement was fine.
I suspect the wire to the wooferswas reversed at the XO board from what was intended.
With the swapped XO boards they now all measure correctly, so all is well.
FYI, the process to get these charts was:
> Set the large impulse peak to 0 ms for all traces using 'Set T=0 at cursor'.
> Set the FDW window for all measurements to 2 cycles/octave.
The new data matches well to the old for timing and I will be able to use the older sub group timing with the new swapped L and R to determine the timing needed between them. The mic was very close to the same position. I will work through that analysis now. No new measurements will be needed.
Few, that's good to hear. I carried out the 12V battery check and this proved OK. I also ran left & right sweeps with the mic 1 mtr from the horn as you suggested. 'Attached mdat'. Not that this matters now. All the same it's good for me to educated myself. Cheers
The timing change needed to achieve close phase tracking of the direct sound is:
> Reduce the 4subs delay by 9 ms, or increase all the mains delays by 9 ms.
Plus -
> The 4subs need to be reverse polarity, or else, all the mains polarities should be reversed.
Below is the resulting SPL:
Below is the resulting direct sound phase:
Below is the phase at the mic including the room effects:
As commonly the case, there is a different impact from the room on the 2 mains resulting is different SPL responses in the XO area.
Attached REW sweeps after making your adjustments:
"Reduce the 4subs delay by 9 ms, or increase all the mains delays by 9 ms. Plus - > The 4subs need to be reverse polarity, or else, all the mains polarities should be reversed"
I mentioned also that that adjustment would make little difference so either setting is okay.
To time align the 4 subs make that change. Below are 2 options for timing the mains. Option 1
Invert the Subs (or mains) reduce the 4sub delay by 9 ms. Option2
Reduce the 4sub delay by 15 ms. (don't invert the 4subs)
All of 3 of these settings will provide very similar SPL. The sound quality is likely to vary a little between them however. They all are good overall phase alignments.
Note that measuring L + R impact the upper frequencies significantly. It better to measure L and R separately. The lower frequencies are okay to look at however so see how a mono signal is impacts the XO range.
The 3 option compared:
Option 2 still provides very good phase tracking within the XO range. Below 60Hz it has less phase rotation and thus slightly less group delay. This is another common alignment.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
