Understanding sub phase/group delay for placement

[FargateOne]

Hi,
I would like to know if I am confused or not. I have only one sub. I have two choices to place it. It is important that, at the crossover the satellite and sub are in phase (I think...). Looking at my graphs, I am correct to think that, based on the phase, sub 1 place is a better choice because it will be easier to match the phase crossover for satellites? (see FR of center speaker in mdat file). I checked also the group delay, but I am not sure what it means for my case.
Thanks for your help!

 

[FargateOne]

So, somebody...maybe my question doesn't make sense, don't hesitate to tell me.

 

[jtalden]

The sub phase is very similar in either location. There are only minor phase differences in either position due to room effects. The posted charts only look different because the 2 impulses are not closely aligned together at 0 ms. The attached chart shows the phase response after aligning them more closely.

The delay timing of the sub XO cannot be accurately determined until the chosen XO is active when the sub and main speakers are measured. The REW acoustic timing feature also needs to be active. The current measurements did not have the XO activated or acoustic timing activated.

The current measurements are suitable to select a favorable XO range, one that hopefully minimizes the impact of the room nulls. The attached SPL chart suggests to me that an XO target between 60 and 80 Hz may be favorable and that the second sub position may be more favorable to the initial position. The only way to be sure however is to actual analyze all the various options. This can be a lot of work with no strong expectation of finding a significantly improved setup.

I suggest selecting the sub position and activating the XO and then remeasure the sub and main. That data can be used to determine favorable delay timing.

[My approach is to use the FL, FR mains for this process rather than the CC main. There often are very significant differences between the 3 front mains and I tend to prioritize XO settings that work best for the music content rather than voice. The above process is similar for either approach.]

 

[FargateOne]

I select position 1 (sub origine for waf reason), and xo at 80 and I will come back soon with measurements.

 

[FargateOne]

Hello,
I, too, prefer to prioritize for music. But, just to be precise, in the mdat joined, CC polarity has been reversed. (I can't reverse polarty of the sub). Let me know if you need more infos. Thank you so much.

 

[jtalden]

Summary:
The 5 mains all look good as currently configured. The mains to sub timing look questionable, but this cannot be confirmed with this data.

This data shows:

• REW acoustic timing was properly applied to all measurements
• L, R, arL, arR speakers are time aligned well as their impulses all lay on top of each other.
• The C speaker is inverted polarity. Its impulse location is acceptably close to the others. The phase tracking with the L and R through the XO range is good. The inverted polarity and current timing are good to achieve close phase tracking through the XO range as shown in the phase chart attached. This current timing results in the phase diverging from the L and R mains at frequencies >800 Hz. If the C delay was increased by 0.20 ms the higher frequencies would be better aligned up to about 4 kHz. This additional delay can be added to C, but I would not expect any impact to the sound quality. Minor mic location changes can easily have this type of impact. I consider this to be trivial.

This data does not show the timing of the SW to the mains. There is some question as to the current timing because the delay of the sub appears to be more than is normally seen. Its impulse rise trails the mains impulse rise by 9 ms as shown in the attached chart.

New measurements would be required to confirm the sub XO timing is correct:

1. Sub with XO (same as #5 measurement in this data).
2. L with XO (but with sub turned off).
3. R with XO (but with sub turned off).

 

[FargateOne]

Hi,
thank you again. Here some infos before to open the mdat file.
1- The receiver allows to change delays by 15cm segments or (if I am correct 0,44ms) so, I could move backward the CC by 7cm, isn't it?
2- The receiver allows also to HP filter the satellites by pair (front mains together, center, surrounds together). In the mdat file joined I made the measurements you asked for and added measurements with the sub xo at 80 and the mains hp_filtered at 60.
3- The tape measures the distance between sub and mic at mlp is 255cm. The delay I choose for the sub is 165cm in the receiver.

 

[jtalden]

1- (move CC backward): Yes, but as mentioned, this is a trivial change with no practical benefit. I was just explaining why the CC impulse did not need to exactly overlay the other main speaker impulses.
2- We may not be understanding each other on the measurements needed. I will better explain my thinking. The receiver applies the XO frequency chosen. It divides the signal sent to a main speaker between the sub and that main speaker. The measurements needed In order the evaluate a favorable delay timing (distance setting) for the sub when the receiver XO chosen is 80 Hz are:

1. FL only with receiver XO set to 80 Hz (sub off)
2. FR with receiver XO set to 80 Hz only (sub off)
3. Sub only (EQ on is okay) with receiver XO set to 80 Hz [measure this by choosing to measure the FL speaker (output 1). Disconnect the wire to the FL speaker. That way the receiver XO will still sent the low frequencies of the FL to the sub, but the high frequencies of the LF will not be measured. We are thus measuring what is called; 'redirected bass'.

The FL and FR measurements are correct. You measured the LFE channel (output 4) when you measured the sub. That channel bypasses the receiver XO and sends the full range signal to the sub. I think you have set the low pass filter (LPF) on the sub to 80 Hz thinking that it was a part of the XO. That setting will filter out high frequencies coming to the sub, but it impacts only the channel 4 signal (the LFE channel) rather than the redirected bass from the main channels. It is fine to leave the LPF on the sub at 80Hz as it has no impact on what we are doing. We are trying to get a smooth handoff between the redirected bass frequencies and the higher frequencies of the main speaker - sorry for being unclear about the measurements needed. We can take these same 3 measurements at 2 different receiver XO frequencies if you are unsure whether you want to use a 60 or 80 Hz XO in the receiver.
3- The 165 cm setting is fine for these measurements. We will determine if an adjustment is needed to this setting to provide a more favorable handoff.

Note: The FL and FR measurements provided are just what is needed. If your mic setup is the same or close to the same it would not be necessary to measure them again. A measurement of the redirected bass from the FL channel to the sub is all that is needed for each of the 2 receiver XO frequencies (60 and 80 Hz). I can easily combine those 2 measurements with the FL and FR measurements already provided.

 

[FargateOne]

Thanks for your patience and great explanations. Hopefully, here what we need!

 

[jtalden]

I looked at this data in several ways and found several good optional timings. The new REW ‘Alignment tool’ allows the analysis of many different XO timing alignments very quickly. In general, the 80 Hz XO options look slightly better than the 60 Hz options. The difference is small, so a 60 Hz XO is likely to be okay in practice. The following comments address only the 80 Hz XO. There are parallel options at 60 Hz XO, but the timing changes would be slightly different.

I usually suggest finding the closest phase tracking when using an FDW window on the measurements. This type of timing is preferred by most experts for higher frequency XOs as from a midrange to a tweeter. For a sub XO I normally find the SPL support to be about the same as using the unwindowed measurements. In this case there is a small SPL advantage by instead using the unwindowed data so that is what was done here. The FL and FR timing were initially looked at separately, but it became clear that the best tradeoff between the FL and FR timings would be found by vector averaging the FL and FR in this case, so that is what is shown below.

Two 80 Hz timing options:

1. Invert sub polarity and reduce the current delay timing about 7.37 ms (Increase sub distance by 254 cm)​

This timing provides the closest phase tracking throughout the entire XO range with no window applied. The group delay is also reduced significantly.​

2. Retain positive sub polarity and reduce the current delay by about 2.24 ms (Increase the sub distance in the receiver by 77 cm)​

Phase tracking is very good through the significant portion of the XO range with no window applied. The group deal is marginally reduced from the original timing.​

The sound quality may differ between these 2 options more than the similar SPL support suggests. You may want to try them both out and select the one you prefer.

Current Green, Option 1 Red, Option 2 Blue

Option 1 Phase Tracking with no window applied

 

[FargateOne]

Thank you so much for your time and detailed response. My SVS PC-2000 doesn't have a switch to reverse polarty nore the receiver. Is shifting phase by 180 degrees does the same?

 

[jtalden]

I don't know for sure, but I expect not. Another sub with a variable phase control only shifted the delay time. That is not the same thing and does not help. You could measure the sub with phase at 0° and at 180° to learn exactly what is does if you like. Use redirected bass (FL) and leave the REW acoustic timing engaged just as before.

 

[FargateOne]

There are the new measurements. I don't know if they are what was expected or if I made a istake. Thank you so much again

 

[jtalden]

I was looking for just 2 measurements; Sub Phase 0 and sub phase at 180 to see the exact impact of the phase control. I was still able to learn a lot from this data however.
> The phase control at 180 is a good functional way to achieve a similar result to inverting the sub polarity. It is not exactly the same as inversing the polarity as at the very low frequencies phase is not completely inverted. That very minor difference is trivial to the practical impact throughout the frequency range of interest.
> Thus your measurements 9 and 10 (Option 1) are inverted from each other, but are functionally equivalent in all relevant respects.
> Measurements 9 and 10 show a deeper null at 120 Hz than was predicted in my analysis, but I believe that it is just the result of my analysis using FL + FR instead of the FL only.
> Measurements 4 and 6 also provide good SPL support in the XO range similar to 9 and 10.
> Measurement 10 is the only one that provides a marginal improvement in group delay and only below 40 Hz. I would think this is trivial.
> While these 4 setups are very similar in SPL and other measurement characteristics, you may have a preference for one of them regarding sound quality.

 

[FargateOne]

I was looking for just 2 measurements; Sub Phase 0 and sub phase at 180 to see the exact impact of the phase control. I was still able to learn a lot from this data however.
(...)

I would never have dreamed that you could learn something from an amateur like me but, it was my intent to send you the most infos possible! I was scared with the dip at 120Hz when I saw the graphs but not you. That's the difference between an amateur and someone who knows his stuff. Thank you very much for your time. I will experiment the 2 settings and report if I ear an audible difference and which one I prefer.

 

[FargateOne]

Oh boy, what to say. I gave a short listenning session to option 2 ( positive polarity for sub and delay at 240cm). With dolby degital decoded from cable box and from blue-ray 7.1 source DTS-master hd downsizes to 5.1.
The least I can say: there is a very easy audible difference.
Hopefully my brain doesn't trick me.
SS&I seems more open, wider, spaciousness more evident, less sound image shifting to the right, speakers disappeared more, bass more around all the room.
What it mystifies me is that the IR for 165cm delay or option 2 are quite the same.
And CC channel doesn't seems to suffer of those changes based on FL and FR.
I don't understand.
Stay tuned more to come later...

 

[FargateOne]

If I go with option 1 (invert polarity of the sub or in my case phase shift by 180°) wouldn't it be logical or better to come to positive polarity for the CC ?

 

[jtalden]

No, the CC was inverted in order to have its phase tracking match well with the other 4 mains (FL, FR, SL and SR). Thus the adjustments we made to make the sub to match well with the FL and FR will also apply to the CC, SL and SR.

 

[FargateOne]

No, the CC was inverted in order to have its phase tracking match well with the other 4 mains (FL, FR, SL and SR). Thus the adjustments we made to make the sub to match well with the FL and FR will also apply to the CC, SL and SR.

Hi,
first Happy new year everyone here!. I've been authorized (guess by who!) to move out some furnitures of the room which gave me the opportunity to try a new location for the sub: right beside the main listening position on rear wall. The physical distance from the mic at mlp and the sub is only 60 cm.
I've tried many delays (75cm to 165cm) and I think that 90cm is the best.

If you have the time, I would be gratefull if you could take a look and maybe confirm my finding.

To me , this new place is better than what we discussed before, isn't it?

I joined mdat :
FL only (sub off) xo80
FR only xo80
sub only xo80 (center disconnected)
FL & FR sub on xo 80
and the sme but for xo 100.

 

[jtalden]

General Notes:

• The FL measurements with 80 Hz XO has much lower SPL above 2 kHz than the FR or the FL with 100 Hz XO. I presume you made an adjustment to the FL TW output. This will have no impact to the sub timing recommendation below.
• There was no FL and FR individual measurements at the 100 Hz XO so delay timing analysis was not done for that XO.

80 Hz XO analysis:

• The SPL support was not very favorable when delay timing was adjusted based on the direct sound. For this reason, I decided to try to find better timing using the unwindowed LP response.
• There is a significant difference between the unwindowed phase response between the FL and FR speakers in the 80 - 120 Hz range. For this reason, the FL and FR were aligned and averaged to find the best delay timing for SPL support in the XO range. This led to strong SPL support across the XO range at the expense of a few milliseconds of additional sub delay. The resulting increase in group delay would not be expected to have any significant negative impact. This is thus likely to be the best tradeoff.

Charts:

 

[jtalden]

Oops, you probably want to know what the timing adjustment was.

The sub was delayed an additional 2.4 ms (2’-8” less distance in an AVR).

 

[jtalden]

I found an error in the SPL chart above. I accidentally summed the FR and FL instead of averaging them. That made the SPL support look a little better than it actually was. Below is the correct SPL chart. This has no impact on the timing adjustment, or the other 3 charts.

 

[FargateOne]

The FL measurements with 80 Hz XO has much lower SPL above 2 kHz than the FR or the FL with 100 Hz XO. I presume you made an adjustment to the FL TW output. This will have no impact to the sub timing recommendation below.

Thanks again. The tweeter was coverred with a towel (!) Ooops!.

Oops, you probably want to know what the timing adjustment was.

The sub was delayed an additional 2.4 ms (2’-8” less distance in an AVR).

The delay in the receiver was set at 90cm (36 inches). I want to be sure. You suggest to set delay at 36 inches - 32 inches = 6 inches (approx) or 15cm which is the shortest distance available in the receiver.?

In your opinion and with your experience, between the sub at the front wall (at the beginning of this thread) or at the rea wall closed to MLP which one is the less bad (or the best !) ?

There was no FL and FR individual measurements at the 100 Hz XO so delay timing analysis was not done for that XO.

I will provide it soon today in case you would have a moment to take a look .

 

[thexder]

The mains to sub timing look questionable, but this cannot be confirmed with this data.

Sorry for barging in. I, too, am investigating impulse responses.
How does this timing look? Not measured with timing reference. Also, no delays were used. Just phase/polarity settings.
Thank you.

 

[jtalden]

The delay in the receiver was set at 90cm (36 inches). I want to be sure. You suggest to set delay at 36 inches - 32 inches = 6 inches (approx) or 15cm which is the shortest distance available in the receiver.?

Yes, That is correct.

In your opinion and with your experience, between the sub at the front wall (at the beginning of this thread) or at the back wall closed to MLP which one is the less bad (or the best !) ?
[/QUOTE]

The chart below shows the SPL comparison of the settings in Post 10 option 1, Post 10 option 2, and Post 20/21. The SPL through the XO range is very similar for all 3 options. It represents the best that can be done given the main speaker and the LP positions chosen. There is a big improvement however in the low bass with the sub in the front of the room. Either of the Post 10 options is the better choice in my opinion.