JBL310s subwoofer polarity

[witwald]

@A Surfer Yesterday I received some additional information from SVS. They mentioned that I needed to be aware that the DSP latency of the SB-2000 Pro is about 5.5ms, and that this alone would present time alignment and phase integration challenges with full-range loudspeakers. Of course, that begs to question as to whether or not there is any time delay between the signals output from PRE OUT 1 and PRE OUT 2 on the M3. The high-pass filtered signal from PRE AMP 2 does go through an extra step of DSP, so maybe there is some extra time delay added there. To actually determine whether or not there is would require an oscilloscope or a two-channel FFT analyzer.

Knowing the value of the DSP latency, I added its time delay to my simulation, and I got the following result. As you can see, there is a large dip in the frequency response through much of the bass region, which is due to the phase cancellation effects.

4th-order Butterworth LPF –3dB at 62 Hz, +ve polarity, PEQ F=58 Hz, Q=2, Boost=–1.8 dB, delay 5.5 ms

Inverting the polarity of the subwoofer produces the following result, which is clearly much better. Please give the change in polarity on the subwoofer a try to see what it sounds like. My expectation would be that you can raise the level of your subwoofer and still get a good sounding response, as the –14dB setting could be a little low (although I am sort of flying blind on that point!).

4th-order Butterworth LPF –3dB at 62 Hz, –ve polarity, PEQ F=58 Hz, Q=2, Boost=–1.8 dB, delay 5.5 ms

 

[witwald]

@A Surfer You mentioned earlier that you wanted to try the 60Hz high-pass filter setting on the M3. I've gone and prepared a simulation of that combination. This includes the DSP processing latency in the SB-2000 pro, and so the set up also required a polarity inversion on the subwoofer to obtain the best integration. The plot of the results is shown below. Note that I've adjusted the low-frequency bass response of the subwoofer near 18Hz in order to get it to better match with the published measured data. This involved increasing the roll off rate by about 6dB/octave.

Subwoofer (DSP delay 5.5 ms): 4th-order Butterworth LPF –3dB at 45 Hz, –ve polarity, PEQ F=46 Hz, Boost=–1.5 dB, Q=2
NAD M3: 2nd-order Butterworth HPF "60Hz setting", –3dB at 58Hz

 

[A Surfer]

@witwald Thank you for obtaining that information from SVS. I am more than happy to contact technical support at NAD. I have in the past and they were very forthcoming and helpful. If I was to contact NAD the information that would benefit your simulations would be about any delay stemming from DSP correction from the Pre out to the mains correct?

I am still not very skilled at interpreting what I am seeing in the above two scenarios that you have graphed. Based on what you are seeing do you feel that the M3 passed at 60Hz reasonably offers a little more low frequency energy into the mains, but in such a way as to not result in any noticeable blurring? Sadly where this experimenting is concerned I will be out all day and into the evening with a girlfriend so the next phase of trials will have to wait until I return home. I am typically up late so I expect I will be having at least a few hours of listening time and will be able to report back to you the results.

I love the current nimble integration, just love it. I feel the drivers are tracking very well, and as per your new simulation where you incorporated the additional information from SVS I can absolutely believe that dip is there, hence the perceived drop in low frequency quantity and for lack of a better term, some drop in impact.

I think that I need to split the difference in terms of presentation styles versus my old, meaty, but somewhat blurred low frequency integration and where your modelling has taken my system thus far. I will say again that without a doubt, your predictions were very good and yielding quite favourable results and I know that with your guidance I will be able to obtain the best from this modest, but capable system.

I will also turn the subs up to -10dB for all future testing as if I remember a conversation with SVS support I had some time ago, I believe they had felt that the subs should be playing at that level. I think I will grab a picture for you now. The next phase, probably in a few months would ne to get a good microphone and do some REW analysis. The room is likely not very destructive, but absolutely it isn't ideal either. I imagine the low, reflective drop ceiling will be involved and it would be interesting to do an analysis at the listening position although I don't feel that there is likely to be any major voids or peaks. Still, assumptions need to be tested.

 

[A Surfer]

Here are three images. One shows the configuration as seen from just behind the listening position. The M3 and Gustard X16 are raised off the floor (temporary solution) using several very dense landscaping wall bricks which hopefully yields a fairly stable base. Ideally I would not have the electronics so close because of vibration, but not much that I could do.

I am also providing a full length shot of the room taken standing on the stairs at the far end of the basement. Again, just to show the room dimensions. The third picture was taken standing behind the electronics and shows the listening position that the speakers fire towards. The mains are placed far enough from the front wall and also with adequate distance from the side walls. They do need to be leveled more as the floor slopes down towards the walls on both sides there so I am sure the imaging will become even better once the mains are fully leveled.

 

[Jdunk54nl]

First, polarity flips can be your friend. Sometimes you might be in time, but out of phase by 180 degrees. Sometimes you can be in phase but out of time. You want both!

What does this look like when measured in a room?

Pictures from here: https://www.merlijnvanveen.nl/en/st...gnment-the-foolproof-relative-absolute-method


This is in phase at the arrow (phase lines overlap), but out of time (slopes of the lines are different). Note* This one isn't that bad of an out of time, but still out of time.

This is in time (slopes of phase are the same) but out of phase by about 180 degrees. Every line separation going up is 30 degrees.

This is in phase and in time, phase lines overlap AND slope are the same. This was achieved by a polarity flip. If you are 180 degrees out of phase, this would be an ideal spot for it.

 

[Jdunk54nl]

Ideally you want to have the same slopes of the lines and be within 60 degrees of phase. 60 degrees gives ~+5db summation and being perfectly in time and phase should give ~+6db summation. So not a huge difference.

But that is using programs like SMAART or Open Sound Meter to measure phase directly and adjust accordingly. I use both of them and usually SMAART and use the method from the link above.

This was one of my latest integrations and not an "ideal" like the pictures above (phase graph in the middle):
Phase was lined up pretty well. But if you look at like 800-1000hz you can see the green line right speaker does something weird (reflection off of table right next to it) and then if you look at the pink summation at the bottom, you can see it does not sum well there either.

 

[Jdunk54nl]

So, if you do not have a microphone to measure how can you do this?

Actually pretty easy and here are some good steps.

1) Download REW and open it.
2) Connect your computer to your system and open REW's generator and set it up similarly to the picture below: Make sure your low cut and high cut are covering your sub to main crossover.
Pink Periodic Noise
Custom frequency that covers your crossover between mains and subs


3) Listen to this and flip the polarity to see which one sounds the QUIETEST! We listen for the quietest because that is easier to tell compared to the loudest. Adjust the main speaker delay by increasing it (add distance if AVR to main speakers or subtract distance from the subwoofers or remove delay from them). Listen to when this gets the absolute quietest. There will be a range of values where it sounds the quietest. Choose the one in the middle of this range.

We add delay to mains due to the nature of subs and they are always naturally delayed, it will never just be the distance to them. We just had a thread on this with articles explaining why. I will try to find it.

4) Now that you found the quietest setting, flip the sub polarity back to the other one and it should now be the loudest and best integration you can get by ear.

 

[witwald]

We add delay to mains due to the nature of subs and they are always naturally delayed, it will never just be the distance to them.

Why are subwoofers naturally delayed? Are you referring to the DSP-related processing delay that is introduced by the digital filtering in many of today's suwboofers? If a bandpass response is used to model a subwoofer, it all integrates well with high-pass filtered main speakers. There is no need to add extra delay to the main speakers to get correct integration.

 

[witwald]

My M3 was not attenuating the signal to the mains at all, so full range on the biamp setting there. Now as per your modelling I moved the biamp high pass to 80Hz. The subs were already configured to positive polarity and 24dB slope. The previous crossover point was actually 80Hz and the volume was (and remains -14dB).

That is a quite unusual setting for your system. I'm assuming that your PL200 speakers have a –3dB point at about 45Hz with a 4th-order roll off from the vented enclosure. If your subwoofer crossover was set to 80Hz, then you would have had a very large region of overlap. The interactions are further complicated by the DSP processing delay on the subwoofer.

Below is a model that I have created of the abovementioned configuration. The red curve is the subwoofer's output with a 4th-order Butterworth 80Hz LP filter applied. Due to the heavy attenuation of the subwoofer's output, which is needed otherwise the bass boost around 50Hz will be very large, it's not providing a lot of assistance at frequencies below 35Hz or so. This configuration is less than ideal.

If we proceed to use the SB-2000 Pro to boost the bass using the subwoofer integration methodology of REL, then we can set the low pass filter on the SB-2000 Pro subwoofer to 4th-order and 40Hz in order to blend with the natural roll off of the PL200 mains. As you can see, the red curve (subwoofer output) now adds a lot of extra bass extension, which is the desired result. The subwoofer is also connected with positive polarity, which I tend to believe is a preferred option. The NAD M3 amplifier is supplying a full range signal to the main speakers, so we don't get the benefit of reduced excursion levels and power input into the main speakers. In this configuration, the integrated output is –3dB at 21Hz and –6dB at 18Hz.

Subwoofer (DSP delay 5.5 ms): 4th-order Butterworth LPF –3dB at 40 Hz, +ve polarity, PEQ F=40 Hz, Boost=–2.7 dB, Q=2
NAD M3: HPF inactive (main speakers full range, with assumed 4th-order Butterworth response)

 

[witwald]

My M3 was not attenuating the signal to the mains at all, so full range on the biamp setting there. Now as per your modelling I moved the biamp high pass to 80Hz. The subs were already configured to positive polarity and 24dB slope. The previous crossover point was actually 80Hz and the volume was (and remains –14dB).

Another possibility that comes to mind is to use the 40Hz HPF on the NAD M3, and then adjust the LPF on the SB-2000 Pro accordingly to achieve the best possible integration. Here are the results of that simulation, which appear to be very good. It also has the benefit that you will utilise more of the natural upper bass response of the PL200 speakers above 44Hz (the approximate acoustic crossover frequency). You may like to try this set up to see how it performs for you. The modelling indicates that the bass response will be –3dB at 20Hz, and –6 dB at 17Hz.

Subwoofer (DSP delay 5.5 ms): 4th-order Butterworth LPF –3dB at 36 Hz, –ve polarity, PEQ None
NAD M3: 2nd-order Butterworth HPF "40Hz setting", –3dB at 41Hz

This configuration is also very good when it comes to tuning the output of the subwoofer to set its response at the preferred level. You can see from the above plot that the output at 30Hz is more or less due entirely to the subwoofer. Hence, the 32Hz bass peak of "Negative Girl" can be used to adjust the output of the subwoofer to get a good balance between the relative outputs of the subwoofers and the main speakers.

The following two plots below demonstrate this behaviour. There I have adjusted the output level of the subwoofer by +3dB relative to the "flat" level, and in the other plot by –3dB. It can be seen that the bass response below about 50Hz is what is predominantly affected by these changes, although there is some interaction up to about 90Hz or so of course.

Subwoofer set to +3dB re "flat" setting in the first plot

Subwoofer set to –3dB re "flat" setting in the first plot

 

[Jdunk54nl]

Why are subwoofers naturally delayed? Are you referring to the DSP-related processing delay that is introduced by the digital filtering in many of today's suwboofers? If a bandpass response is used to model a subwoofer, it all integrates well with high-pass filtered main speakers. There is no need to add extra delay to the main speakers to get correct integration.

Yes,
This article touches on it:
https://www.soundoctor.com/whitepapers/subs.htm

I would disagree about the last statement.
Every home/office/etc that I have measured (or friends) using SMAART needs extra delay (compared to just distance) on the mains to get the best phase relationship (in the MLP) between the subs and mains. This isn't models, this is actual measurements in a room. Every video online that I have watched on how to get proper phase integration (from people measuring phase directly) has had the same thing, some of these are huge concert halls to big meeting rooms, etc.

Have you measured phase in a room directly using software like Smaart or Open sound meter? If not try it and then we can continue these discussions.

 

[witwald]

I am more than happy to contact technical support at NAD. I have in the past and they were very forthcoming and helpful. If I was to contact NAD the information that would benefit your simulations would be about any delay stemming from DSP correction from the Pre out to the mains correct?

Exactly. As the M3's PRE OUT 1 is always full range, as the crossover settings don't affect this output, I would expect that it requires fewer computations (time) than does the HP filtered PRE OUT 2 that is then connected to MAIN IN when external subwoofers are being used. It would be very handy to have that DSP delay information from NAD.

I think that the question to ask of NAD is "What is the individual DSP processing delay associated with PRE OUT 1 and PRE OUT 2?". This will then allow us to determine what the difference is.

I am still not very skilled at interpreting what I am seeing in the above two scenarios that you have graphed. Based on what you are seeing do you feel that the M3 passed at 60Hz reasonably offers a little more low frequency energy into the mains, but in such a way as to not result in any noticeable blurring?

I think that would be the case. Reproducing 60Hz using two 6.5" drivers isn't that difficult. However, if the system is being pushed to very high SPLs, then of course those two drivers will be called upon to have quite large excursions. It should still work reasonably well, though, and I'd expect good overall sound quality.

I think that I need to split the difference in terms of presentation styles versus my old, meaty, but somewhat blurred low frequency integration and where your modelling has taken my system thus far.

With some of the configurations that I've covered in previous posts, I think that you will still get a very solid bass response, with the benefit that it is quite a bit more extended. Knowing that there is a 5.5ms processing delay in the SB-2000 Pro, I saw in the simulations that I've conducted that using the 40Hz HPF setting on the NAD M3, and then adjust the LPF on the SB-2000 Pro for 4th-order 36Hz operation (with no PEQ) seems to offer very good integration between the subwoofers and the main speakers. This configuration really works well to allow for simple up/down level adjustments of the subwoofer's output to get an acceptable integration simply by ear.

I will also turn the subs up to -10dB for all future testing as if I remember a conversation with SVS support I had some time ago, I believe they had felt that the subs should be playing at that level.

I would tend to agree with the SVS suggestion. When I read that you used a –14dB setting, I thought that this was unusually low in value. However, knowing that you were driving your main speakers full range, that low subwoofer level is necessary to try and achieve some sort of balance to the bass output, as the integration was not of a standard nature. I expect that using a higher subwoofer level would have resulted in excessive combined bass output, and that's why you selected the lower level based on your listening tests.

The next phase, probably in a few months would be to get a good microphone and do some REW analysis.

That's an excellent idea, as a microphone and some REW measurements will certainly help to better understand what's happening with regard to the integration.

 

[witwald]

Every home/office/etc that I have measured (or friends) using SMAART needs extra delay (compared to just distance) on the mains to get the best phase relationship (in the MLP) between the subs and mains. This isn't models, this is actual measurements in a room. Every video online that I have watched on how to get proper phase integration (from people measuring phase directly) has had the same thing, some of these are huge concert halls to big meeting rooms, etc.

I fully recognise that analog filters have group delay associated with their responses. Maybe you could have expanded on the important features to make things clearer.

When trying to achieve smooth integration between a subwoofer and the main speakers, there is a need for complementary filter responses in order to get good results. That's a basic tenet of loudspeaker crossover design. Any group delay in the LPF and HPF sections needs to be complementary in nature in for smooth blending between the responses to occur. Keep in mind that my models have included the natural responses of the subwoofers and the main speakers to some level of fidelity, with the phase responses being embodied in those models.

The same principles apply to the blending between woofers and midranges, or bass-mids and tweeters, or midranges and tweeters. There is nothing special about the response of a subwoofer. It is just like a bandpass filtered midrange unit, albeit with the frequency response of the midrange being suitably scaled. The subwoofer-to-woofer integration behaves substantially like that of a midrange-to-tweeter integration problem, albeit the subwoofer often has a natural bandwidth measured in octaves that is likely to be less than that of a typical midrange unit.

Have you measured phase in a room directly using software like Smaart or Open sound meter? If not try it and then we can continue these discussions.

I've measured magnitude and phase responses of loudspeakers when designing and building crossover networks for two way loudspeaker systems using CLIO and IMP. I am aware of concepts such as phase cancellation, polarity reversal, group delay, and time delay. VituixCAD has many features for including all of those effects in the simulations, and I have made use of them over and over again. In any measurements, just like in models, it is important to identify the contributing factors. So, please, continue the discussion if you would care to do so.

The following is a simple model created in VituixCAD. It has a subwoofer with a 4th-order Butterworth HP response that is –3dB at 20 Hz. The subwoofer's natural response at high frequencies is assumed to be a 2nd-order Butterworth LP response with a –3dB point at 180Hz. These seem quite typical behaviour of a good subwoofer.

In this particular example, the main speaker is assumed to have a 4th-order HP response that is –6dB at 80Hz, with the response shape being that of a 4th-order HP Linkwitz-Riley filter. It should be possible to build a speaker such as this. After all, it's just another filter shape, and the Thiele-Small vented box theory spent a lot of effort analysing maximally-flat vented systems as these are usually desirable.

From the plot below, it is evident that, even without any delay on the main speakers, we get almost textbook-quality smooth integration between the subwoofer and the main speaker. Why might that be? There doesn't appear to be any "inherent delay" in the subwoofer's response that causes grief for the correct and smooth integration of the two response functions. Or am I missing something here? Of course, the implementation of the Linkwitz-Riley filters needs to be in the analog domain, so that there is no DSP processing delay added to our system. Or if there is DSP filtering at work, then the processing delays on the subwoofer and main speaker should be identical for this simulation to be accurate. That much we can understand from this simple model, let alone accurate measurements. Furthermore, the slight dip around 200Hz is caused by a mismatch in the phase response between the acoustic filtered outputs of the subwoofer and the main speaker. This is caused by two factors: 1) the natural LP response of the main speakers and 2) the natural bandpass response of the subwoofer, in the frequency range of interest.

Now, if the output of the subwoofer is delayed by 5.5ms relative to that of the main speaker, then we get the following response. The response has a pronounced dip just above the crossover frequency. Note that this time delay is not inherent in the subwoofer, but could be due to some DSP-associated processing delay. The 5.5ms was chosen as it seems to be representative of DSP-assisted subwoofers, so might be often encountered in practice.

Why is the dip centred on a frequency a little higher in value than the crossover frequency? It's not due to any inherent delay in the subwoofer, but simply due to the fact that the LP and HP responses are not exactly complementary in the strict Linkwitz-Riley sense. They are close, but not perfectly matched in this particular case.

How do we fix this problem that has been introduced by delaying the output of the subwoofer relative to the output from the main speaker? The deep null suggests, to those who may have simulated and/or measured such behaviour, that a polarity inversion could help. The following plot shows the result obtained when the polarity of the subwoofer is inverted (many subwoofers have this ability). However, although we have inverted the polarity of the subwoofer, the summed response is nowhere near as flat as that which was shown on the first plot. The reason for this is that the time delay added to the subwoofer has a linear phase response.

As the phase response of acoustic outputs of both the subwoofer and woofer is nonlinear, trying to correct the large dip via a simple polarity inversion is not quite sophisticated enough, as the results above clearly show. In this instance, the best approach would be to add a delay of 5.5ms to the main speakers, and we then would get the good integration shown in the very first plot. As is seen above, this is another example of the Linkwitz-Riley filter topology's remarkable tolerance to phase errors in the acoustic HP and LP responses, which was pointed out by Linkwitz in his journal paper all those years ago, and hence the popularity of Linkwitz-Riley filters.

This example hopefully helps to illustrate that the subwoofer has no inherent delay, and it can be treated just like any other filter response function. If there are delays that show up in the measurements, then they are coming from other sources.

 

[witwald]

Yes, This article touches on it: https://www.soundoctor.com/whitepapers/subs.htm

From that article I quote: "Simply connecting a sub to existing mains speaker (or amp) terminals is the WORST POSSIBLE WAY to do this. EVERYTHING scientific and acoustic about this method is wrong, from the additive delay issues to the back EMF of the mains affecting the LF signal. However there are plenty of people who simply do not understand correctly integrated bass...".

Apart from the general tone, if I have interpreted the above correctly, it seems that the general concepts associated with crossover network design don't apply to subwoofers. Yet, subwoofers are just like any other driver: they have a magnitude and phase response. How those parameters and behaviours are understood and used in a crossover network will serve to allow correct integration of subwoofers and main speakers. Just like any other drivers in a multi-way loudspeaker system with its associated crossover networks. What makes a subwoofer so special that it needs it's own crossover network theory? Nothing springs immediately to mind, as a subwoofer is just another loudspeaker subsystem.

Another quote: "Impulse response (NOT frequency response) really is the holy grail of all of audio."

But isn't the frequency response an embodiment of the impulse response. That's how MLS (Maximum Length Measurement) systems work. So what is so special about an impulse response that is not present in the frequency response? Of course, some important characteristics of a loudspeaker system can be more easily identified on an impulse response time history, but that's not what the statement quoted even remotely attempts to touch on.

Two other quotes: 1) "so that IF we were playing back a correctly recorded IMPULSE, for example a well recorded kick drum" and 2) "So the impulse of a kick drum is nearly a square wave".

I'm not sure that a kick drum could really be called an impulse in any shape or form. It might be a relatively short-lived transient signal, but it really is not that much like an impulse in real life. Below is a waveform associated with a kick drum.

Does the above really look anything like an impulse response? Nor does it look like a square wave at the onset. In comparison, some of the band-limited impulse responses published in reviews of CD players look like the following. There appears to be a significant difference between these two waveforms, let alone their frequency response curves.

And another quote: "One part of this assumption is that the instrument is correctly recorded in the first place, ideally with a stereo pair of microphones which therefore ARE picking up the 3-dimensional phase and harmonic structure of the instrument in space." By virtue of their small size, don't microphones simply sense the pressure fluctuations at nominally a single point in space? How does a microphone know where that pressure has originated from in 3D space. The quote I referred to is all a bit confusing.

More confusion arises from this quote: "You CANNOT have multiple low frequency sources of differing phase relationships in a living room-sized room." That seems to be entirely possible, as we can have a subwoofer with some delay (a differing phase relationship) relative to the main speakers in the same room placed next to each other, or separated by large distances in terms of a wavelength. That's one of the very issues that I have been modelling with VituixCAD.

And even more from this quote: "If you have 2 LF sources of differing phase relationships (that means timing relationships) they will cancel. Period. " How can it be that a differing phase relationship of say, 1 degree, causes cancellation? What is trying to be conveyed here to the reader?

And it continues with this quote: "A port is ALWAYS nothing more than a cheap way to attempt to get free bass out of an enclosure and/or driver that's too small." I expect that JBL, who over many years have implemented 15" drivers in ported systems, might beg to differ!

And another quote: "Lets examine ported speakers. We'll start with the worst case, the port in the front. At mid bass frequencies, say 50-80 Hz, the LF driver moves IN the cabinet, the air in the cabinet is elastic, and the port air moves out of the cabinet. Because of the frequency at which the cone is moving, by the time the cone moves out (forward) again, the port air is now moving out, so in front of the cabinet the two air pressure sources sum together and you get a fake bass "bump" or "boost"."

In a ported system, isn't the driver motion at a minimum at the port resonance frequency? The driver is essentially at a null in its displacement-vs-frequency response curve. Therefore the driver adds very little to the total output at the port resonance, whereupon the total system output is dominated by the output from the port. I believe that Thiele, Small, Benson and Keele covered the driver–port interaction back in the 1960s, 1970s, and 1980s. And why is the "worst case" the one with "the port in front"? From a low-frequency perspective, the port is a monopole; it doesn't matter all that much where it is placed at the typically long wavelengths of operation.

 

[A Surfer]

@witwald Very thorough and useful discussion of your application approach to integration with DSP delay factors. I can see why the notion that the subwoofer needs to be considered as inherently delayed has great surface validity, but ultimately that isn't true. My laymans belief was that unless you had measured your room correctly that you would only be guessing where nulls and peaks were, so given this, orienting the subwoofers in as close to linear arrangement with the mains as possibility makes sense.

Now reading your explanation of the behaviour of a subwoofer in a crossover relationship with mains seems to give credence to my assumption about placing the subs as I have. I did play around with the most recent suggestions: using the Highpass of 40Hz on the M3, no PEQ on subs, subs at 36Hz, negative polarity, 0 degrees phase shift, 24dB slope and finally sub output set at -10dB.

It was so late last night and my brain was fatigued so I started listening and just realized I was too tired and I prefer experiencing a new system presentation when I am alert enough to really listen to it. I will give another listen with these settings in about an hour I hope. I'll cue up Negative Girl and a few of my favourite electronica pieces. I very much enjoy well crafted electronica music I have to say. There are some absolutely brilliant artists that I have followed for about 5 years now or thereabouts.

So given that I listen to a genre where complex low frequency content abounds, this work aligning the drivers excursion response is so crucial and there is no way on earth without the years of experience I know you possess, and the modelling software to go along with the applied experience, I would be absolutely blind. For example, I had no idea that changing polarity would introduce such different interaction effects, effects it seems you can model very accurately.

I'm not surprised, these are physical, purpose built machines that are subject to the laws of physics which I have to think suggests that the outside variables are at least reasonable, that the modelling predictions are then very good and translate well during application. So far my experience is that these adjustments have changed the system presentation so much, and in favourable ways that the effects are profound and complex. It will take time for my brain to really get used to the new signature. The tautness of the low end is phenomenal and the accurate tracking sounds superlative to my ears so far. Should be fun.

I will hopefully get time and reach out to NAD this week, although if I do it will be towards weeks end. Many thanks again. Cheers.

 

[A Surfer]

@witwald I had a modest, but excellent listening session using the settings that I described in my last post. I can't believe how the subwoofers have come alive, in addition to the previously stated benefits. All I can surmise is that I had terrible cancellations happening as the subs just never established themselves. With my original settings they were doing very well enough from I'll guess 50Hz up to the cross at 79Hz, but I found them very weak below probably 40-50Hz. They never seemed to sustain any true sub bass, and they just sounded subdued anywhere where you really needed them to kick in for low bass.

Now they are just so different I do not even know where to begin. That is what leads me to believe that there had to have had been a fairly impactful cancellation. I am wondering what possible role the polarity inversion may have played in this integration work?

I did experiment and only adjust the cross on the subs moving them up to 45Hz. I think I liked that better, and subjectively I did not notice any audible effect on integration, so more heft in the low end, but no blurring. At least that I could notice. If everything else holds true from the last configuration I reported where I used 36Hz on the subs, but now the cross is at 45Hz. May I ask what the modelling predicts the effect to be?

 

[witwald]

@witwaldAll I can surmise is that I had terrible cancellations happening as the subs just never established themselves. With my original settings they were doing very well enough from I'll guess 50Hz up to the cross at 79Hz, but I found them very weak below probably 40-50Hz. They never seemed to sustain any true sub bass, and they just sounded subdued anywhere where you really needed them to kick in for low bass.

I believe that your listening experience ties in quite well with the results indicated by the simulations. The lack of bass below 40Hz or so is a result of having to reduce the level of the subwoofer so that output at frequencies above 40Hz didn't become excessive.

Now they are just so different I do not even know where to begin. That is what leads me to believe that there had to have had been a fairly impactful cancellation.

I am hopeful that what we have managed to achieve is quite good subwoofer integration, albeit possibly not as ideal as it could be due to the inherent DSP latency in the SB-2000 Pro. If the position of the subwoofers is to be maintained close to the main speakers the way they are set up at the moment, and I like that layout, then to fix that would require some additional DSP work.

Keep in mind that a 1ms delay on the subwoofer corresponds to 0.345 metres (13.6 inches or 1.13 feet) of physical offset, due to the speed of sound being about 345 metres per second at sea level under standard conditions. In your case you seem to have moved your subwoofers forward by about 1 foot relative to the woofers in each PL200. Hence, about 1ms of delay has been removed. Of course, that will alter my simulations a little, but it is just another one of those variables that we can't quite pin down exactly.

I am wondering what possible role the polarity inversion may have played in this integration work?

In any subwoofer integration work, the polarity inversion is there to adjust the phase by 180° in the crossover region, if phase cancellation is occurring. It's not an ideal thing to do, but it does mean that the two responses that were previously cancelling at and around the crossover frequency are now going to sum together more constructively. I used the work "more" here as the phase matching is unlikely to be particularly precise. However, having said that, it does serve a particular purpose as far as obtaining a relatively notch-free summed response goes.

I did experiment and only adjust the cross on the subs moving them up to 45Hz. I think I liked that better, and subjectively I did not notice any audible effect on integration, so more heft in the low end, but no blurring. At least that I could notice. If everything else holds true from the last configuration I reported where I used 36Hz on the subs, but now the cross is at 45Hz. May I ask what the modelling predicts the effect to be?

I'm pleased that you are experimenting with some adjustments around a nominal solution point. It is very easy for me to model the new response that would occur with the low pass filter on the subwoofer set to 45 Hz. The response plot is shown below. I'd just to clarify one point, and I think that it is a quite important one.

As you can see below, when the LP filter on the subwoofer is set to 45Hz, the actual acoustic crossover may or may not be at 45Hz. In our case it isn't, as it is located at about 48.5Hz. That's not terrible in and of itself, as there are many interactions happening within your listening room, and I would be the last person to say that I have accurately captured all of those behaviours/interactions in my model. Still, the models are helpful to visualise what is actually happening, so that's why I've tried to provide lots of simulations of various cases. The model is more or less treating your 3-way loudspeaker system as if it had come with a subwoofer built into the bottom of the enclosure (i.e., a 4-way loudspeaker system).

As you can see in the plot below, the 45Hz LP filter now interacts with the high passed main speakers somewhat differently than before. We now potentially have a boost at 53Hz of about 1.5dB. This would serve to explain why your listening session indicated that there was "more heft in the low end, but no blurring". That's good, and what I would have expected, as the 50Hz region is usually associated with a bit of heft to the kick drums.

One of the interesting things about the choice of 45Hz LP filter is that as you drop the level of the subwoofer, the boost at 50Hz tends to remain relatively unaffected. This is illustrated below, where I have deliberately reduced the output level of the subwoofer by 3dB, yet a 0.7dB peak at 58Hz still remains. This gives a little bit of emphasis around that frequency, whereas at 30Hz the level has dropped the full 3dB.

Just for the sake of interest, the following simulation shows what happens when we increase the subwoofer output level by 3dB. As you can see, we now have a lot of extra bass boost over a very wide frequency range. How that comes across during listening will, as you well know, depend greatly on the low-frequency content of the track that is being played. In any case, it's probably a bit too much bass overall, but even then it is only a boost of 3dB or so.

Feel free to keep the questions coming! Now that I've set up the main blocks in VituixCAD, I've had a bit of practice adjusting them to suit the requirements of each simulation. I'm happy to share them if you or anyone else would like to have a copy.

 

[A Surfer]

@witwald Thank you, I am sure that I will have questions. The reason I moved the mains back was based on reading about speaker placement they were previously too far forward and in a area to avoid, if the advice I was reading was to be believed and the source was quite credible I felt. I will link to that information when I remember. In the ideal world I would have the subs and mains oriented perfectly with one another. It may happen yet. For now they are wonderfully integrated thanks to your modelling and patience.

I thought that the change of 45Hz was doing exactly what the model showed, and I like the persistence of the effect even if I were to adjust the subs lower in amplitude (which I suspect I will not). This has been very eye opening for me and to say helpful would be an epic understatement. I look forward to seeking your guidance a little further as needed. Thank you once again and I Look forward to chatting elsewhere in the forums as well. Cheers.

 

[A Surfer]

@witwald I just wanted to thank you again. My system sounds fantastic thanks to your modelling. I know there is improvement yet to be had as I haven't done REW and corrections yet, but your modelling has the integration issue solved. It sounds unbelievably unified. It is almost hard to understand how actually dropping the subs so far down (45Hz vrs 80Hz) has actually resulted in what seems like even more bass. The low bass is quite a bit more present sounding. And this is with the M3 limiting the mains by a 40Hz cross.

Is this evidence of how impactful cancelation really can be if the integration is poor? I may try moving the subs crossover point again just to see. I am wondering if the subs are dialed in at 52Hz does the integration weaken in any meaningful way? I would also subjectively be curious if the low frequencies would begin to sound smeared at all versus the extremely tight integration I hear now at 45Hz.

 

[witwald]

@A Surfer I've done a simulation with the subwoofer crossover set to 52Hz. As you can see, setting the volume on the subwoofer to maintain its contribution to low-frequency output results in a broad +2dB peak centred on 60Hz. This shows the interaction between the high-pass and low-pass acoustic roll-offs, which in this case only have about 25degrees of phase difference at 60Hz.

Subwoofer (DSP delay 5.5 ms): 4th-order Butterworth LPF –3dB at 52 Hz, –ve polarity, PEQ None
NAD M3: 2nd-order Butterworth HPF "40Hz setting", –3dB at 41Hz

If I now switch the polarity of the subwoofer back to +ve, we get the following. Notice that a dip that is very broad and 4.5dB deep has developed. I expect things like drums will sound quite anaemic with this setting, but the really low frequencies (e.g., 30Hz) will still come through quite powerfully

Subwoofer (DSP delay 5.5 ms): 4th-order Butterworth LPF –3dB at 52 Hz, +ve polarity, PEQ None
NAD M3: 2nd-order Butterworth HPF "40Hz setting", –3dB at 41Hz