New room, sub integration, xo, delay and second sub to come, help with alignment tool

[FargateOne]

At last!
@jtalden I finally placed the new sub . In post #5 we saw that combination 2+9 seemed the best. It is the placement choosen for the measurements here. Sub 1 (the new Pro) is at left side of the MLP and sub 2 (the old pc2000) is at the rightr side of the mlp.
Sub xo 100Hz
FL and Fr xo at 70 Hz
Center xo at 80 Hz
Sub 1 distance in AVR is set to 315cm and 5.6ms delay in miniDSP
Sub 2 distance in AVR is set to 315cm and 5.6ms delay in miniDSP

I have gain matched the 2 subs by putting the new one at the exact same place as the old one ( left side mlp) and adjusted the volume to get the exact same spl FR response. After i placed the second sub at the right of mlp and didn't touch the volume since then.

Mic at mlp is 316cm from front wall
FL distance is 255cm in AVR, the same for FR and CC.

Thanks again if you have time to comment.
Please note that I will be out of town next week.

 

[jtalden]

Good news! It arrived so quickly. Oh wait, maybe I am a little confused about that delivery performance.

I just spent the day with this new data. I recognized the same issues in determining a favorable timing that I had with the previous data. This time I finally resolved what I think would be the most favorable timing alignment. It differs from my previous suggestions.

The changes needed from the current settings are:

Oops, We cannot reduce the delay on the subs enough to achieve these new settings so some additional info is needed to see if we can make it work.

• Approximate actual distances to the LP for the 2 subs
• Current AVR distance settings and approximate actual distances to the LP for all surround speakers
• How were all the current AVR distances established? Automated setup routine? Manual Input by you?

We may be ably to achieve this setup, but it may require rerunning the automated setup routine in the AVR if that is what was used.
If there is no easy way to correct this I can select another favorable timing that does fit within the current range of settings.

 

[FargateOne]

Hi,
yes! It would have been be faster with a 8 horses WellsFargo stagecoach!

All distances settings in the AVR are manually input.

1.
Subs are cylendrics and downfiring. Physical distances from MLP at MLP level are:
Sub 1 left MLP 60 cm
Sub 2 right mlp 115 cm

Subs are 87 cm height x 42 cm diam. and the driver is at 6 cm from the floor.
If we need the distance from mlp to the drivers Pythagore must help us.Taking 81 cm for side a) and distance from mlp level for side b), hypothenuses c are:
Sub 1 101 cm
Sub 2 141 cm

In AVR the one sub output distance is set to 315cm and 5.6ms is added to the 2 subs via miniDSP

2.
Distances measured by laser tape from MLP and AVR settings:
FL physical dist 262cm setting in AVR 255cm
FR physical dist 262cm setting in AVR 255cm
Center physical dist 262cm setting in AVR 255cm

3.
Distances setting in AVR are by step of 15cm from 0 to 30 meters

 

[jtalden]

Sorry, I just notice that the 2 subs measured in Post-26 used the LFE channel (HDMI channel 4) so the sub response is not being impacted by the XO filter. This impacts the timing needed to get the most favorable XO handoff between the subs and mains. We should measure the subs using a main channel (HDMI channel 1 or 2) with that main channel speaker disconnected. The bass is then passed through the XO and the resulting redirected bass to the sub is measured.

Let me know if you want to remeasure the 2 subs for this analysis. If we use the current LFE channel data, we don't know how much it will impact the actual XO range SPL. I suspect in this case the timing difference may result in a relatively minor SPL difference, but cannot be sure. If the mic position is close to the same LP position and the REW and AVR settings are all the same, it is not necessary to remeasure the mains again, but that is an option also.

 

[FargateOne]

Sorry, I just notice that the 2 subs measured in Post-26 used the LFE channel (HDMI channel 4) so the sub response is not being impacted by the XO filter. This impacts the timing needed to get the most favorable XO handoff between the subs and mains. We should measure the subs using a main channel (HDMI channel 1 or 2) with that main channel speaker disconnected. The bass is then passed through the XO and the resulting redirected bass to the sub is measured.

Let me know if you want to remeasure the 2 subs for this analysis. If we use the current LFE channel data, we don't know how much it will impact the actual XO range SPL. I suspect in this case the timing difference may result in a relatively minor SPL difference, but cannot be sure. If the mic position is close to the same LP position and the REW and AVR settings are all the same, it is not necessary to remeasure the mains again, but that is an option also.

Hi,
maybe the following info is important before I take new measures (it's not a problem if needed) :

I disconnected the left speaker to measure the subs using HDMI channel 4.

So does it help ?

 

[jtalden]

That doesn't help because HDMI channel 4 is the Low Frequency Effects channel. It only contains bass content intended for the SWs alone. The XO settings have no effect on it.

The mains only play sounds from their dedicated channel. The XOs will split out and redirect any content on those channels that fall below the XO frequency and sent it to the to the SWs.

So, the SWs play their dedicated channel and also any bass from the main channels that was redirected there by the mains XOs.

 

[FargateOne]

OK I will remeasure all channel sorry for my mistake.

 

[FargateOne]

@jtalden I've been able to measure sooner than I thought. Here they are.

(P.S.: ref #7 we have position 2+6 also workable if the present one is too difficult)

 

[jtalden]

Thanks.
The data looks good and the chart below shows the SPL of 3 favorable XO timing alignments. Alignment ja5 is the conventional timing alignment. It looks good, but there is a large sag centered at 25 Hz. It is the result cancelation due to a room mode. There is not much content down there other than special effects, but most expect deeper bass capability from their 2 SWs. I found the other 2 options (ja1 and ja3) which involve inverting the polarity of one of the 2 SWs. Both of these provide deeper bass and still reasonably support the XO range SPL.

The following table shows the new settings for each option in the last 3 columns. The new settings to AVR SW distances are shown in bold. Just set the closest AVR setting to these values. Also note that 'yes' indicates that a SW has its polarity reversed. The DSP delay values for the 2 SW is not changed. They remain at 5.6 ms.

The new CC AVR distance setting is 315-120 = 195 cm. I did not list it in the table above as it is the same for all options. It provides the most favorable XO SPL support, but is no longer is timed well in mid and high frequencies. To match the timing of the tweeter frequencies the original 315 cm setting was good. The broad midrange frequencies are best matched with a setting between these 2 values.

 

[FargateOne]

Thank you so much.
I did my first tries this pm and I found that to raise the gain of sub1 by 2dB could help but you didn't mention that possibility. Any reason?
I will continue with the alignment tool to find how you have reached your options.

 

[jtalden]

The gain of SWs should be adjusted to best fit the target house curve in preparation for EQ. I usually don't know what what the house curve target is so I cannot make any gain recommendations. I usually just suggest the SW level be adjusted to individual preference. To answer your particular question though, I did not investigate changing the gain balance of the 2 SWs this time. I have rarely found it to be helpful, but it can be in some cases. I am not hung up on matching gain or on matching SPL at the LP. If a change to relative gains helps SPL then go for it. Improving the SPL smoothness seems more important to me.

 

[FargateOne]

I see I was not clear in my post #28 about the subs output in my AVR. The AVR has 2 sub output but only ONE delay setting for both outputs .
It is why I wanted to find a delay in ms to put in miniDSP and a distance in the AVR the get a margin of manouver.

I try to translate your recommandation in the following board in column 6 and 8. Is it correct?

But, should it be simplier to set distance at 0cm or 15cm in the AVR and, first time align the 2 subs in miniDSP and after that adjust delays into miniDSP to integrate the dual sub response with satellites?

 

[jtalden]

It was my oversight. I forgot that constrain.
Increasing the distance of the SWs in the AVR will reduce the delays applied just as your table is setup to do. I agree with the result of ja5 and the difference in ja3 is trivial. I just choose the AVR distance of ja 3 a little higher to avoid a small negative delay in the DSP. I don't agree with ja1. It looks like you possibly just switched the 'new DSP' values in ja1 for the 2 subs?

I revised my table to provide for a single SW distance:

It would be prudent to confirm one of these setups to assure that there are no other mistakes.

 

[FargateOne]

Just for the sake of our followers, about ja1. Current distance 315cm, new distance 831cm (831 -315=516cm /34 cm/s= -15ms to go ) Increasing distance from 315 to 630 = 9 isshh ms in AVR + 5ms in DSP = -15 approx as needed. You were right!

In your last post, the values in the fifth column (Change required AVR(cm)) are different of those in your post #34 that I religiously copied in my table; my explanation is that you took into count the new distance (AVRcm) setting for the AVR.

I will confirm soon all this by real measures. Stay tuned

 

[FargateOne]

HI,
here are the results in real life

 

[jtalden]

There is something wrong with the mains levels between these 3 traces. A mono measurement (L+R) should provide the same mid and high frequency traces. Something went wrong.

I compared the this data by calculation:

The comparison is a little confusion in that your 6 db mains level reduction has been implemented in these new measurements. I can still see that the proper groupings however. I am sure there is close agreement with the expected results given this change.

 

[FargateOne]

There is something wrong with the mains levels between these 3 traces. A mono measurement (L+R) should provide the same mid and high frequency traces. Something went wrong.

I noticed it. If I average the response of L + R and LSub + RSub and sum the average, it seems to reconcile with measurement of L+R+2sub.
I don't know what happens. I will try again.

 

[jtalden]

The measurements should be with REW signals sent to both channel 1 and 2. There is no need for the timing channel. [No REW measurement signal should be sent to channel 4. The XOs will handle redirecting the bass portion of channels 1 and 2 to the 2 subs.]

So:
Select HDMI channels 1 and 2 in the REW measurement dialog and start the measurement.

 

[FargateOne]

Please find new measurements, OK I think. I deleted the files in post #40

 

[jtalden]

Looks good. The SPL the XO timings are as expected.

 

[FargateOne]

I want to thank very much you again. I will use the tool to follow your path and to get the same result that you suggested. You took a lot of time I am sure but it will be longer for me

(...)

 

[FargateOne]

Hi,
just for the fun of it, I checked the Center timing at xo and found the following delay could be better (from mdat #33 and ja3 setttings):

to add 1,93ms delay i.e. substract 60cm distance in the AVR

am I correct?

 

[Bernard]

Hello,

Juste pour s'amuser, if T° = 20° C, C (speed of sound) = 331.3 × sqrt (1 + (T°/ 273.15)) = 343.21 meters/second ou 0.34321 m/s.
So, 1.93 ms is equal to 1.93 x 0.34321 = 0.6623953 m = 66.3 cm.
Am I right ?

 

[jtalden]

Hi,
just for the fun of it, I checked the Center timing at xo and found the following delay could be better (from mdat #33 and ja3 setttings):

to add 1,93ms delay i.e. substract 60cm distance in the AVR

am I correct?

Yes. That setting looks good for SW XO handoff.

 

[FargateOne]

Hello,

Juste pour s'amuser, if T° = 20° C, C (speed of sound) = 331.3 × sqrt (1 + (T°/ 273.15)) = 343.21 meters/second ou 0.34321 m/s.
So, 1.93 ms is equal to 1.93 x 0.34321 = 0.6623953 m = 66.3 cm.
Am I right ?

Le problème was not 1,93 ms = 60cm but how to set it in the AVR since my Center isn't connected to the miniDSPThanks quand même!