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

[FargateOne]

Hello,
our 5.1 system migrated from a small room in the basement to a vast room (cathedral ceiling, 16 feet x 26 feet floor surface, photo to come soon). When it was in the basement I had a lot of help from many people here and @jtalden among them.

Also I now have a minidsp 2x4 HD in the chain between the receiver and the sub.

The plan is the following:

1. with only one sub, fine tuned the delays to integrate the sub with the fronts at the best xo.
2. with the alignment tool decide the best place (among those possible waf intended!) for the second sub incoming.
3. when the second toy arrives, integrate both subs together and fine tuned the virtual one sub with the fronts.

I did the following homeworks for part one of the plan:

1. Find the best placement in the circumstances (the room is also the livingroom and it is open on the dining room...) for my 5 satellites (see mlp file)
2. Find the best xo between 100Hz and 80Hz;
3. Find the best of one of 2 distances setting (delay in the receiver) using polarise inverse for the sub if needed. I had to choose between 135cm which is the physical distance from the MLP to the sub or 285cm which was the old distance inthe ancient room.

The result was finally xo 80Hz and distance is 135cm which is the delay in the receiver.

Then I measured
FL alone,
Fr alone
C alone
sub alone at xo80
see second file

Now the thricky part and the existencial question: did I use correctly the alignment tool to find a possible improvement by adding 1 foot delay (-1ms) to the distance setting of the sub and inversing the polarity of the center?

If I am correct, I will try to use alignment tool to find where to put a second sub.

 

[jtalden]

Everything looks good!

I might have inverted the center channel speaker and adjusted its timing slightly to better track the phase of the front left and right speakers through the midrange frequencies. I would be hard to claim that this is necessarily better than your current setting however.

Note:
It is better to remove the FDW prior to calculating the 'Aligned sum'. That way we can view the charts of those calculated 'measurements' with, or without, a FDW applied to them.

 

[FargateOne]

Ok then, second sub placement choices coming soon.

 

[FargateOne]

Hello,
ready for part 2 of the plan. I moved my (for the moment) only one sub everywhere possible around the room to find the best place to put a second sub. The delay in the receiver was different of the one finally choosen (495cm instead of 135) and the timing refence was the center but it doesn't matter for the moment I think.
My choice is what seems to me the most interesting combination. I would like to know how to use alignment tool to make the final 3 best places?

file A must be openned first and followed by file B to get all places together

 

[jtalden]

I finally got a chance to review the files.

The SPL of the 2+9 combination in your 3rd file looks very good. I tried many other combinations from the other positions without finding a combination that was nearly this good.

This result when the 2nd sub is in place would be a very favorable 2 sub setup.

 

[FargateOne]

OK While I am waiting the second sub, let us imagine that I live in galaxy far far away where the spouses have a veto on certain sub placement. If I were living there it would be wise to find the second best spot. I notice that you choose the 2+9 only on the SPL.
Phase are not relevent for the moment isn't it?

I would choose the 2+6 combination, no?

Or is it too soon and it must be decide after real measurements?

 

[jtalden]

Yes, I thought the 3 middle options in your original file all looked like very favorable choices, i.e., the ones you marked 2+6, 2+9, and 2+11. The first and last are not quite as favorable, but probably workable. The other combinations I tried were significantly worse.

Real measurements with the new sub will match these predicted results unless the sub is a different model or some room changes are made.

 

[FargateOne]

Still waiting the 2nd sub...

 

[FargateOne]

still waiting the second sub maybe next week say SVS for Canada...

 

[FargateOne]

SVS sub back order untill July 22nd...is in a container somewhere it seems...

 

[FargateOne]

my bad! Still waiting...

 

[jtalden]

The pics and mdat don't open in this post for some reason. Others posts files do, so maybe just post them again.

 

[FargateOne]

The pics and mdat don't open in this post for some reason. Others posts files do, so maybe just post them again.

deleted

 

[FargateOne]

@jtalden and @John Mulcahy

OK after 2 hours to check every thing now I am quite sure of the following. It seems that I have weird result with my Center channel. What do I do wrong?

 

[jtalden]

I'm confused. I do not understand what is going on here. Measurement 3 is not the the sum of the first 2 measurements. I assume it was supposed to be?

If the objective is to find the delay needed for the center channel given the final SWs delay setting with the L and R then:

1. Measure the CC, the SWs and the CC+SWs. [Do not change the settings on the SWs. They must remain aligned with the left and right mains.]
2. Use the alignment tool to adjust the delay on the CC (not the SW's). Save that aligned sum. *
3. Enter the delay change needed into the system settings and then remeasure CC+SWs.

The aligned sum and the remeasured CC+SWs charts will agree.

We can also compare either if these to the original CC+SWs measurement to see the impact of the delay change made.

If I have missed understood the objective and questions you intended here, if so, please explain what was done and your questions in more detail.

* I did use the first 2 measurement to calculate the favorable timing for the XO range. The best timing was a 2 ms reduction in the CC delay setting with the CC polarity reversed. This may not be the best setting however as it didn't seem to agree as expected with the G main. It appears the speaker design and thus the phase rotation of the CC are different than the left and right mains; 2-way vs 3-way designs? If so, a compromise setting for the CC is needed. We can either match the phase through the XO or range or match the phase through the midrange. The HF will deviate either way. This is not likely to be a significant sound quality problem either way as many home setups have this situation.

 

[FargateOne]

Dear Sir,
I am so sorry for the confusion I caused. I wasn't trying to find a better alignment for the center. The plan is still the same and when the second sub will arrive I will try one of the 3 places identified in post #7 and with your help will integrate them. (BTW there will be 2 svs PC 2000).
But in the mean time I was "playing" with REW and it happens that the same measurements gave me 2 different results with the alignment tool. In the mdat file you will see 2 copies of the same measurement of the Center channel and 2 copies of the sub xo at 80 . The sum of measurement 1 (C) + 2 (sub) is the contrary of the sum of measurement 3 (same C) + 4 sub (same sub) ...
Then I was confused. And was searching for an explanation. I found in the measurement info of measurement 2 and 4 (the same sub measurement) the following difference:
IR Starting time 0,0010ms vs IR starting time 7,6656ms.
No clue where or why it is different nore how it happens.
3+4 agree with the current delay I set for the sub (165cm) and the center (255 cm positive polarity).
Hopefully this clarifies that ?

 

[jtalden]

Better.
Since 3 and 4

Dear Sir,
I am so sorry for the confusion I caused. I wasn't trying to find a better alignment for the center. The plan is still the same and when the second sub will arrive I will try one of the 3 places identified in post #7 and with your help will integrate them. (BTW there will be 2 svs PC 2000).
But in the mean time I was "playing" with REW and it happens that the same measurements gave me 2 different results with the alignment tool. In the mdat file you will see 2 copies of the same measurement of the Center channel and 2 copies of the sub xo at 80 . The sum of measurement 1 (C) + 2 (sub) is the contrary of the sum of measurement 3 (same C) + 4 sub (same sub) ...
Then I was confused. And was searching for an explanation. I found in the measurement info of measurement 2 and 4 (the ame sub measurement) the following difference:
IR Starting time 0,0010ms vs IR starting time 7,6656ms.
No clue where or why it is different nore how it happens.
3+4 agree with the current delay I set for the sub (165cm) and the center (255 cm positive polarity).
Hopefully this clarifies that ?

Better.
Just FYI - Since 3+4 is the current settings. I checked what change I would recommend if this was the final data.

• Center - Delay change = +6 ms
• Center - Reverse its current polarity setting.

This change would align the XO range well, but would leave the midrange phase of the center different from the left and right mains.
I hope you found this same change. We will see what is really needed when the 2nd sub is in place.

 

[FargateOne]

Better.
Just FYI - Since 3+4 is the current settings. I checked what change I would recommend if this was the final data.

• Center - Delay change = +6 ms
• Center - Reverse its current polarity setting.

This change would align the XO range well, but would leave the midrange phase of the center different from the left and right mains.
I hope you found this same change. We will see what is really needed when the 2nd sub is in place.

Thank again!
1-Please find the result mdat file. We must not forget that adding 6ms is substracting distance setting in the AVR Rotel. Center delay was 255cm - 224cm= 30ish. I didn't had the time to listen if the difference is audible. For the moment I will keep th delay as iti is.

2- In order to be prepared when the sub arrives, and knowing that I have put a MiniDSP 2X4 hd in the chain, I thought that to integrate 2 subs together and after that the dual sub with the mains it will be easier to work with the delays setting in the MiniDSP only. So I wanted to see how to replace the 165 cm sub delay in the Rotel by the right delay in MiniDSP.

I found that I must to set sub dealy to 0 in the AVR and 4.6ms delay in MiniDSP concurs with sub delay of 165cm and Cente delay of 255cm. It is the second mdat file. Please note that I didn't had the time to do the usual measurements to confirm with alignment tool nor to try to add 6ms to MiniDSP. I will send soon the sub only, C with no sub etc.

 

[jtalden]

1-Please find the result mdat file. We must not forget that adding 6ms is substracting distance setting in the AVR Rotel. Center delay was 255cm - 224cm= 30ish. I didn't had the time to listen if the difference is audible. For the moment I will keep th delay as iti is.

In practice it is the relative delay differences that are important. We can set distances or delays in and AVR or DSP or a combination of the two.

2- In order to be prepared when the sub arrives, and knowing that I have put a MiniDSP 2X4 hd in the chain, I thought that to integrate 2 subs together and after that the dual sub with the mains it will be easier to work with the delays setting in the MiniDSP only. So I wanted to see how to replace the 165 cm sub delay in the Rotel by the right delay in MiniDSP.

I found that I must to set sub dealy to 0 in the AVR and 4.6ms delay in MiniDSP concurs with sub delay of 165cm and Cente delay of 255cm. It is the second mdat file. Please note that I didn't had the time to do the usual measurements to confirm with alignment tool nor to try to add 6ms to MiniDSP. I will send soon the sub only, C with no sub etc.

I'm not sure what you observed and concluded from this data or if you have with any questions.

General Observations
In the first file:
Measurement 1; Current settings
Measurement 2; My calculated recommendation for the center channel settings.
Measurement 3; Agrees with my recommendation for the center channel settings.

In the second file:
All 4 measurements do not have the polarity of the center channel reversed; no issue - just observation.
Measurement 1; Appears to be a faulty measurement or ?? as there is no resemblance to the response of the center channel.
Measurement 2; Current settings.
Measurement 3; Appears to have poor XO delay timing and SPL reflects it. It adds unnecessary group delay.
Measurement 4; Appears to have less than ideal XO delay timing, but the SPL is still resonable. It adds significant unnecessary group delay.

 

[FargateOne]

The first measurement in the second file is the spl of the surround right which is timing reference channel.

I'm not sure what you observed and concluded from this data or if you have with any questions.
(...)

The second file wasn't the best for what I wanted to try.

I wanted to learn how to use the delay in the 2X4 HD for one sub before to get down the rabbit hole with 2 subs. I understand that it is the relative delay that counts. The second file was not the best for that. I found that to put 5ms (equal to 165cm more or less) in MiniDSP, I must first set the correct distance for the sub in the AVR to get 0 delay with the timing reference. In my case to put 315cm distance setting in the receiver. After that, I put 5ms in 2x4HD.

 

[FargateOne]

SVS PC sub will be back order for an othe 3 weeks ! unbelievable! I would have cancel my order if I wasn't oblige to get 2 subs identical. What a shame! 3 months...later and nothing yet!

So I restarted to finf what delays to set in MiniDSP for the sub instead of 165cm in the receiver.

As I said the second file in my post #18 was not correct for what I wanted to try and you were right about it:

In the second file:
All 4 measurements do not have the polarity of the center channel reversed; no issue - just observation.
(...)
Measurement 3; Appears to have poor XO delay timing and SPL reflects it. It adds unnecessary group delay.
Measurement 4; Appears to have less than ideal XO delay timing, but the SPL is still resonable. It adds significant unnecessary group delay.

I followed this procedure:

1-Set 2 front speakers and sub distances to the same value (255cm);
2- with FL as timing reference, I run measurement for the FR and the sub;
3- I added the difference in cm to the sub channel in the receiver to get as close as posssible to 0 delay between the timing reference and the sub.

It gives me 315cm to set in the AVR for distance setting of the sub. Then I set 5ms (approx 165cm) in MiniDSP 2x4 HD. The mdat file compares the front left channel before and after.

Now since I have time... it's been a while that I want to compare a single xo vs different xo per channels. It seems important to know which is best in my context. I can set a main xo for the sub channel and a different one for the fronts, the center and the surrounds. @jtalden I completly understand that, when it will the time (!) you will not want to examine more than one xo settings. I made my homework to the best of my knowledge and give you my conclusion. It is the second file.

The following settings were in the receiver:

Sub xo 100Hz and distance setting in the receiver 315cm
6 ms delay in the sub output in MiniDSP

Front Left and Front Right xo set to 70 Hz (1 octave above there drop off low fr)
Center xo set to 80Hz (same logic)

after many attemps I found nothing to change when I used mixed xo...do you agree?

And judging by the comparison, the mixed xo seems better for the fronts?

 

[jtalden]

The delay timing looks favorable for D and G channels in both files. The timing on the C is not as favorable at the XO as it could be. At least it is not the delay timing I would chose for it. That doesn't mean the SPL would be significantly different however so from a practical perspective if is probably fine.

There is no issue using different XO frequencies in the AVR for the different channels. The XO delay may, or may not, need to be adjusted with an XO frequency change. It depends on several factors.

 

[FargateOne]

Thanks. Now I can try what you were suggesting in your post #17 for the center Center channel because now I have a margin to work with.

 

[FargateOne]

Nothing new on the SVS PC2000-Pro back order front...still WAITING

 

[FargateOne]

@jtalden update: SVS PC-2000 pro SHOULD arrive next Monday!!! Still hope for the dual sub integration project! Stay tuned!