Overview of subwoofer management

[Keith_W]

I wrote this post on another forum, but I thought it might benefit SNA readers, so I am reposting it here. The question that was asked was: "how do you integrate a subwoofer to the mains?". When you add a sub to a system, you need to:

 

1. Volume match the sub to the mains. Failure to do this will result in either thin or overwhelming bass.

2. Smooth out the frequency peaks and dips in the room because ALL bass does this, not just subs. Failure to do this may result in lumpy sounding bass, depending on your room. 

3. Low pass the sub and high pass the mains. Failure to do this will result in misaligned phase over the band where freqs overlap (i.e. 20Hz up to the low pass freq of the sub), producing cancellation at many points in the band where SW and mains overlap. The exception is if you are using DSP such as MSO (Multi-Sub Optimizer) which can exploit multiple subwoofers and mains speakers to smoothen the bass response. 

4. Align the phase at the XO region. Failure to do this may produce cancellation at the XO point. 

5. Time align the sub to the mains. Failure to do this will result in "flabby" sounding bass, where the upper freqs are heard first, and the bass freqs 10-20ms later. It will sound disjointed.

6. Check the reverberation of bass freqs. There is usually more bass reverb than found in higher freqs and it is more difficult to address. Adding subs usually exacerbates the problem. DSP may be able to help (e.g. with a VBA or Dirac ART) but it is better addressed with physical solutions. 

 

And all this is assuming that you have bought the right sub and chosen a proper location for your subwoofer in the first place, AND assuming you are using a single subwoofer. Using multiple subwoofers adds to problems that need to be addressed. Subwoofer integration is as important than buying an appropriate sub, if not more important. Failure to do this will produce peaks and dips that are impossible to fix. These are the options: 

 

1. Do not use any type of crossover. Use the mains full range, low pass the sub only. 

This is what many people seem to do. All they do is adjust the controls on the SW plate amp until it sounds right - i.e. rough volume and phase matching and maybe some PEQ. Most people do this subjectively, but it is extremely difficult to do without measurements, and it is impossible to achieve a satisfactory result with any degree of precision. What happens is - the entire band of the subwoofer overlaps with the mains, and unless the phase is perfectly aligned (very difficult even with DSP, and DEFINITELY impossible if the sub is ported) it will produce cancellation at multiple points where the freqs overlap, maybe even cancelling out all bass freqs altogether if improperly adjusted. Your sub is now cancelling bass instead of augmenting it. Some subwoofer manufacturers actually recommend this (e.g. REL). It is really disappointing to see subwoofer manufacturers mislead their customers this way. Surely they know better, yet they spread misinformation. The advantage of this approach is ease of use and marketing. You can sell more subwoofers if you convince your customers that adding a sub is easy. To be fair to REL, all subwoofer manufacturers have a vested interest in misleading the consumer on the difficulty of adding subs. So they usually do not mention it in their marketing, or they may promote some other half baked solution. But at least they are not actively promoting a solution that will worsen the sound, like REL. 

 

2. High pass the mains, and low pass the sub with an external analog crossover, or built-in XO in the sub. 

This is what the remaining majority users do, usually without measurements. This is better, because the overlap region is smaller, and therefore less potential for phase cancellation. But - still no time alignment nor phase alignment, and no DSP for the natural peaks and dips in the FR. Furthermore, if the subs have DSP but the mains don't (e.g. as seen in Velodyne subwoofers), the DSP itself introduces additional latency, up to 30ms. This is HUGE and it will definitely worsen the time misalignment and produce flabby sounding bass. This DSP is usually low resolution and IIR because of the requirement to keep latency as low as possible whilst running on hardware with minimal processing power, so it will correct bass freq peaks/dips in a rough fashion. 

 

Disclosure: like most of us, I used to spend money first and then wonder why my solution failed later. I attempted to do this with a DEQX for the subs and analog XO for the mains. The DEQX adds 30ms of latency, well outside the Haas fusion zone, which produced a disjointed sound when used this way. I blamed the DEQX at the time. But in hindsight, it was because I was not using the DEQX properly. I switched to a Marchand XO which immediately removed the 30ms latency and resulted in an audible improvement, something I mistakenly attributed to the superiority of analog. Beware of drawing the wrong conclusions from equipment changes, I certainly made that mistake. 

 

3. Use hardware based DSP to high pass the mains, and low pass the subs

Examples of hardware based DSP: MiniDSP, built-in bass management of AVR's, Genelec GLM, DLBC/Dirac Art, etc. This is the most common recommendation you will see on SNA for DSP but it is still not the ultimate. This is better than all the previous solutions, because it ticks nearly all the boxes of the requirements laid out above. Nearly all hardware based DSP use automated software algorithms that removes decision from the user and these algorithms sometimes gets it wrong, usually with no way to over-ride the algorithm. For e.g. some AVR's calculate the delay by directing you to enter the distance of the sub from the listener. This is wrong, the delay should be measured with a microphone with a timing impulse (we can discuss why). Even worse, Denon and Marantz AVR's use 300m/s as the speed of sound (it should be 343m/s) - they are literally rewriting the laws of physics or assuming all their customers live at the top of Everest. This should be an absolute scandal, yet I have never seen a reviewer mention this. Also, processing is done on low powered hardware, and the low latency requirement again means IIR filters or mixed phase filters and USB microphones (or even worse - cheap uncalibrated microphones). You are usually limited in how much delay you have available, and depending on your situation, that may not be enough. So while there is some bass management, it is not the best - better precision, but not great precision. It may correct the bass up to 80-90% of what needs to be done if the algorithm gets it right. If you use this approach, you need to be aware of the shortcomings and do what you can to mitigate it. 

 

4. Use software based DSP to high pass the mains, and low pass the subs

Examples: Acourate, Audiolense, Focus Fidelity, REW/RePhase, Matlab, Octave. These software packages vary in the degree of automation (i.e. use of a software algorithm), but most let you over-ride the defaults and give you more control over what you need to do. Some (like Acourate, REW/RePhase) are completely manual and relies on you to interpret graphs and make decisions. Octave and Matlab are even more manual in that you need to enter mathematical equations. It is my belief that manual correction is superior to "one button DSP" if you know what you are doing, and most people don't. In that case, "one button DSP" is superior for you. The disadvantage of this approach is the steep learning curve which varies depending on the degree of automation offered by the software, requirement for a certain type of hardware configuration (i.e. some kind of computer must always be in the signal chain), lack of convenience, and difficulty of use. The solution also lacks robustness, because it is software running on a computer - it will never be as robust as a simple electronic circuit. One bad update from Microsoft or Apple, or download some malware - and your solution is broken. Of course there are ways to mitigate this but beyond the scope of this post. 

 

In short: Subwoofers are for advanced users, and even then, most advanced users still get it wrong. Bass impacts all of us in this hobby - do you even want bass capability if you listen to undemanding music? If you do want bass, how much inconvenience and difficulty are you willing to put up with? Better subwoofer integration comes at the cost of inconvenience and greater difficulty. The sheer difficulty of DSP may make this an insurmountable problem for many. So my recommendation is: DON'T add a subwoofer, buy main speakers that are capable of low bass instead. You still need some form of bass management, but this is greatly simplified. Don't get me wrong, adding a subwoofer still adds benefits, but unless you are prepared to manage the subs - it can range from anything from a slight upgrade, to a sidegrade, or even a downgrade. 

Edited April 10 by Keith_W
mistakes and typos corrected

[Steff]

3 hours ago, Keith_W said:

Even worse, Denon and Marantz AVR's use 300m/s as the speed of sound (it should be 343m/s) - they are literally rewriting the laws of physics or assuming all their customers live at the top of Everest.

 

Thank you for elaborating approaches & your experience.

I was reminded of this gentleman's set up (very large REL stack behind his TAD speakers):

[Whites]

@Keith_W thanks for posting that information. I was thinking on trying option 2 as I had been reading articles on the 6moons website and the main reviewer has provided quite detailed articles on his experience.
 

I didn’t realise it would that difficult. I am still going to give it a go, just to satisfy my own interest and it could be quite fun…..maybe

[Keith_W]

6 Moons are in service / or in collusion with the manufacturers and I would not trust what they say. I don't think they know what they are talking about either. Anybody who has ever picked up a microphone and looked at their sub, and know what they are looking at, will realize that fixing all the problems created by adding a sub or subs is not a trivial issue. I can tell you that I still have problems that I haven't fixed, mostly because I am struggling with the software. 

[bazzle]

There is also another case where if you are happy and get enjoyment from the way you have "setup" your system then that surely counts for something?

 

I have been happy with a sub crawl and letting my AVR (Yamaha 2085) do the majority with 80hz set for all the other speakers initially and Yamahas microphone on a stand with the full cal method.

[Keith_W]

NOBODY is happy with the way they have set up their system! Almost everyone in this hobby wonders how it might be improved. If they were happy, they wouldn't be in constant search for upgrades!  

[zydeco]

@Keith_W

 

Thanks for the well thought out post. I’ve had several cracks at subwoofers and have some additional observations:

 

1. Option 3 probably delivers 80% of the benefits of Option 4 at 20% of the effort / complexity.

 

2. Option 3 seems to be where the market, and investment, is at and I’d guess that we will see continuing improvements as seen by the emergence of Dirac ART.

 

3. Option 4 gets even more complex, but not impossible, when one wants to integrate A/V sources into the mix. 

 

4. Option 4 has, traditionally, relied on the goodwill of the developer or enthusiasts to get one started. More recently, however, we’ve seen some businesses setting up services to assist in filter creation which is a big step forward.

 

5. Option 5 (no subs) has its own problems - namely a lack of full range reproduction and the challenge of where to position speakers vs the front wall. 

[Raffinator]

Is there any benefit to using hardware based crossovers such as this one?

 

https://sublimeacoustic.com/products/k235-stereo-3-way-active-crossover

 

 

[Keith_W]

2 hours ago, Raffinator said:

Is there any benefit to using hardware based crossovers such as this one?

 

https://sublimeacoustic.com/products/k235-stereo-3-way-active-crossover

 

Take a look at the requirements listed in the first post to find your answer. 

 

1. Volume match the sub to the mains. Failure to do this will result in either thin or overwhelming bass. - Yes. 

2. Smooth out the frequency peaks and dips in the room because ALL bass does this, not just subs. Failure to do this may result in lumpy sounding bass, depending on your room.  - No

3. Low pass the sub and high pass the mains. Failure to do this will result in misaligned phase over the band where freqs overlap (i.e. 20Hz up to the low pass freq of the sub), producing cancellation at many points in the band where SW and mains overlap. The exception is if you are using DSP such as MSO (Multi-Sub Optimizer) which can exploit multiple subwoofers and mains speakers to smoothen the bass response. - Yes

4. Align the phase at the XO region. Failure to do this may produce cancellation at the XO point. - No

5. Time align the sub to the mains. Failure to do this will result in "flabby" sounding bass, where the upper freqs are heard first, and the bass freqs 10-20ms later. It will sound disjointed. - No

 

Quick answer: there will be some benefit. But it only does half of what you need it to do. It will work, it is better than nothing, but it's not a great solution. In fact it is barely better than doing nothing. The biggie is lack of time alignment and absence of proper amplitude control. 

[Raffinator]

@Keith_W in terms of your “no sub” suggestion, do you have recommendations for mains speakers that go low enough with bass that cover this?

 

Towers that can do this, without needing too much space away from a wall, would be in my planning zone. Without being grater than $10k…

Edited April 18 by Raffinator

[Keith_W]

58 minutes ago, Raffinator said:

@Keith_W in terms of your “no sub” suggestion, do you have recommendations for mains speakers that go low enough with bass that cover this?

 

Towers that can do this, without needing too much space away from a wall, would be in my planning zone. Without being grater than $10k…

 

Yeah, unfortunately I don't know the speaker market < $10k. There are WAY TOO MANY SPEAKERS in that market for me to know them all. I also think that speakers are very personal. What works for me may not work for you. And what works for me right now may not work for me any more - if I moved house and into another listening room. 

 

And BTW, lest I be misconstrued, I think subwoofers are definitely beneficial. I have a pair of subs, and I have a setting on my DSP that immediately switches between two settings - mains high passed + subs on, and mains full range with subs off. Both have been EQ'ed so that the frequency response is the same. But they don't sound the same - the version with the subs is noticeably more spacious and physical. I just think that if you add subs, you have to think about integrating them. And this requires a lot of extra trouble and knowledge. It isn't as easy as subwoofer manufacturers would have you believe. 

[crtexcnndrm99]

These informational posts and a few similar threads recently have been superbly helpful, thank you. 
 

In terms of the DSP option, which minidsp would be needed for a two-way horn system with external crossover (I.e., the possibility to biamp if needed) and two subs? If the aim is to DSP for the active subs and at least the LF in the speaker system… 

[Keith_W]

In your case, I see: 

 

- minimum of 4 channels required, preferably 6 if you wish to bypass the external XO and use the MiniDSP instead. 

- a turntable, meaning you need an ADC if you wish to keep using it in your system 

 

The problem is, there is no MiniDSP that ticks all the boxes. You can get a MiniDSP with an ADC (the SHD series) but those have a max of 4 channels out. Or you can get one of the 8 channel ones, but you will need to add an ADC to your system - Flex Eight, Flex HT (if you need HDMI input), or DDRC-88D (you will have to BYO 8 channel DAC). 

 

MiniDSP's real advantage is low cost, ease of use, and robustness. I say "ease of use" with a pinch of salt because ALL DSP products are not easy to use, but some are easier than others. The disadvantages are limited processing power and therefore limited potential for correction. They also recommend USB mics, which will prevent you from obtaining time alignment.

 

 

I know there is a procedure to recover the clock from USB mics with REW and most other major software but it is complex and inconvenient. And I don't know if you can implement it in a MiniDSP even if you recover the clock - I am far from a MiniDSP expert. Maybe some other MiniDSP users can chime in and help you. AFAIK it's amplitude correction only, but that is enough for most people. The effect of time alignment can vary from subtle to massive depending on the delay on your subs - i.e. it is system dependent. 

 

Another alternative is to look for an older DEQX unit. They have the same SHARC processor as the MiniDSP, and come with built-in mic preamps and have and ADC with 6 DAC channels - everything you need. Of course, the new DEQX Premate is on another level with an ARM processor and it uses lin phase FIR's and has 32768 taps per channel. But it is astoundingly expensive. 

[Satanica]

1 hour ago, Keith_W said:

 They also recommend USB mics, which will prevent you from obtaining time alignment.

 

I know there is a procedure to recover the clock from USB mics with REW and most other major software but it is complex and inconvenient. And I don't know if you can implement it in a MiniDSP even if you recover the clock - I am far from a MiniDSP expert. Maybe some other MiniDSP users can chime in and help you. AFAIK it's amplitude correction only, but that is enough for most people. The effect of time alignment can vary from subtle to massive depending on the delay on your subs - i.e. it is system dependent.

 

Maybe I'm missing something, but I don't understand what you're critiquing here, because one can use the acoustic time alignment function in REW for time alignment.

And what does it matter if a mic adds some delay when time alignment is relative between speakers and subwoofers?

[Keith_W]

11 minutes ago, Satanica said:

 

Maybe I'm missing something, but I don't understand what you're critiquing here, because one can use the acoustic time alignment function in REW for time alignment.

And what does it matter if a mic adds some delay when time alignment is relative between speakers and subwoofers?

 

USB mics aren't clock sync'ed to the DAC, and do not have a loopback function so it is reliant on the clock of the USB interface. It turns out that I was wrong, I should not have said that about ALL USB mics, it appears as if the UMIK-2 can provide acceptable clocking because it uses an internal clock. See John Mulcahy's post in the REW forum. The issue is inconsistent timing measurements. 

 

The other issue is that you need a calibrated mic for DSP, otherwise an improper measurement will result in improper correction. Fortunately calibrated USB mics are available. 

 

And BTW there are additional steps when using a USB mic. It has to be run in Java mode (not ASIO), which means that Windows might resample REW or the mic's output (or both) and again screw up the timing. To get around this, set the default sample rate in Windows to match the sample rate you are using with REW. 

 

All of this depends on whether timing is important to the user in the first place. I don't know if MiniDSP can adjust the timing. Anyone? If it doesn't, then this concern is irrelevant. 

 

I have seen complaints in Audiolense's forum about USB microphones. I know if Bernt has gone as far as issuing a recommendation against using them though. Uli from Acourate firmly says not to use them, but he does provide a workaround to recover the timing information from USB mics. He sent me this message: 
 

“The logsweep recorder allows you to add a Dirac pulse before, between and behind the sweep (checkbox Dirac). The user can load the sweep signal (\Documents\Acourate\LogSweep\Logsweep48.wav) and inspect it. The measured sweep contains (more or less) the speaker reaction on the Dirac pulses and you can measure the actual distance between the pulses and compare it to the nominal distance. This leads to a correction factor which allows to re-sample the recording to a new sample rate. And then the resulting pulse response fits much better regarding the time accuracy.

All in all a complex procedure. It will quickly convince you that mics with 48V phantom power are a better solution.

 

BTW for a USB mic (typically not supported by Asio driver) you have to use Asio4All which then uses Windows sound. The user has to live with all the effects of Windows sound.”

[Satanica]

11 minutes ago, Keith_W said:

 

USB mics aren't clock sync'ed to the DAC, and do not have a loopback function so it is reliant on the clock of the USB interface. It turns out that I was wrong, I should not have said that about ALL USB mics, it appears as if the UMIK-2 can provide acceptable clocking because it uses an internal clock. See John Mulcahy's post in the REW forum. The issue is inconsistent timing measurements. 

 

The other issue is that you need a calibrated mic for DSP, otherwise an improper measurement will result in improper correction. Fortunately calibrated USB mics are available. 

 

And BTW there are additional steps when using a USB mic. It has to be run in Java mode (not ASIO), which means that Windows might resample REW or the mic's output (or both) and again screw up the timing. To get around this, set the default sample rate in Windows to match the sample rate you are using with REW. 

 

All of this depends on whether timing is important to the user in the first place. I don't know if MiniDSP can adjust the timing. Anyone? If it doesn't, then this concern is irrelevant.

 

You're talking about timing for subwoofers to main speakers, those inaccuracies of a UMIK-1 (fractions of a millisecond) are inconsequential at the frequencies speakers and subs are going to be crossed over. Yes you can use a miniDSP for time alignment as you can delay channels individually.

[Keith_W]

19 minutes ago, Satanica said:

 

You're talking about timing for subwoofers to main speakers, those inaccuracies of a UMIK-1 (fractions of a millisecond) are inconsequential at the frequencies speakers and subs are going to be crossed over. Yes you can use a miniDSP for time alignment as you can delay channels individually.

 

What is the granularity of the delay? Dependent on the sampling rate? 

FWIW 48kHz sampling rate = the minimum delay is 20.8us. For 192kHz, that is 5.2us. That is a fraction of a millisecond - 0.0208ms and 0.0052ms respectively. You will notice John Mulcahy documented up to a 100us variance with USB mics. Granted, you don't need that kind of precision when it comes to aligning subwoofers. It's more important for high freqs. 

[Satanica]

2 hours ago, Keith_W said:

What is the granularity of the delay? Dependent on the sampling rate?

 

In the two units I'm currently running, it appears to be down to the hundredth (0.01) of a millisecond at 96kHz (maximum sampling rate).

[zydeco]

@Keith_W

Any insights onto stereo vs mono bass (for stereo systems)? 

[Keith_W]

Umm, that is a bit of a can of worms @zydeco. Most people agree that multiple subwoofers are better, and that every subwoofer in the system should ideally be individually controlled. But "stereo subs" is not just multiple subwoofers, the subs have to be specifically set up for stereo. David Griesinger and Todd Welti advocate setting the subs up wide apart, preferably on either side of the listener, or if that is not possible - on both corners of the long wall of the listening room. It produces what they call "bassiousness". You can read more about it here.

 

Others, notably Toole, say that the idea has no merit. Most recordings have mono bass due to the nature of microphone placement - the wavelengths are long, the distance between mics is short, you simply can not record stereo bass unless you specifically set out to do so. And if we are talking LP, 100% of them have mono bass due to the limitations of vinyl. It takes several cycles for our ears to even hear the bass, and that low frequencies are omnidirectional anyway.

 

That article states clearly that you have a binary choice to make - you can either have smooth bass (ideally positioned subwoofers) or stereo bass.  I don't agree, you can set up your subs for stereo bass and then use DSP to get smooth bass. Up to a limit though, how well this works is dependent on your room. 

 

I have no business arguing against Griesinger, Welti, or Toole. All I can do is try it in my own system and see if I like the effect. Some months ago, I reconfigured the system and one of the goals was to try stereo subs, so my system is configured against the long wall with the subs in the front corners. I then made two sets of filters, one with the subs on, the other with the subs off, with both equalized to exactly the same frequency response and all the other variables controlled as much as possible. The drivers in my mains are very good, they can actually go flat down to 25Hz, so I made my subs roll off at 25Hz as well. The difference was pretty dramatic, the version with the subs has an incredible sense of envelopment, as if the sound is coming from all around you. 

 

Now this is far from a controlled experiment. You could argue that this experiment is a test between speakers alone vs. speakers + subs (albeit with the same freq response), in which case the version with subs is the clear winner. It is not an experiment that proves the validity of stereo bass. A real controlled experiment would have another pair of identical subwoofers placed in "ideal" locations in the room, then everything equalized to the same FR, time alignment, etc. and filters made so that I can switch between them (or preferably, have someone else switch them for me and I am blinded to what I am listening to). But I can't do that - not enough subs. I don't even have complete freedom to place the subs anywhere I like given the limitations of furniture, doors, etc. 

 

All this is a long-winded way of saying: I don't have a strong opinion on this one. The experts disagree, and I have not performed a proper experiment to validate one view or another. May I ask you what you think about stereo subs? 

Edited April 25 by Keith_W

[andyr]

9 hours ago, zydeco said:

@Keith_W

Any insights onto stereo vs mono bass (for stereo systems)? 

 

21 minutes ago, Keith_W said:

All this is a long-winded way of saying: I don't have a strong opinion on this one. The experts disagree, and I have not performed a proper experiment to validate one view or another. 

 

Surely it depends what freq you 'cross over' to the subs at?  (And by that, I mean the frequ which the 'mains' are rolled off at - since rolling off the mains delivers the optimal sub-integration AFAIAC.)

 

If you roll off the mains at 110Hz (like I do, due to the LF limitations of my 6" Satori mid/woofers) ... then you need symmetrically placed stereo subs; if you roll off your mains at, say, 50Hz - then you can get away with one sub (placed just about anywhere).

 

[Keith_W]

5 minutes ago, andyr said:

Surely it depends what freq you 'cross over' to the subs at?  (And by that, I mean the frequ which the 'mains' are rolled off at - since rolling off the mains delivers the optimal sub-integration AFAIAC.)

 

If you roll off the mains at 110Hz (like I do, due to the LF limitations of my 6" Satori mid/woofers) ... then you need symmetrically placed stereo subs; if you roll off your mains at, say, 50Hz - then you can get away with one sub (placed just about anywhere).

 

 

For sure! If you roll them off that high, time alignment becomes more important. 

[davm]

I'm playing around with a pair of DIY ripole subs that I built to work with some Quad ESL63 speakers.

 

Integration is using a DEQX HDP Express II.

 

The ripole subs have a similar dipole-style radiation pattern to the ESL63s and I've chosen to place each ESL63 on top of a subwoofer.

 

Ripole subs don't go nearly as deep as more conventional subs -- I'm seeing in-room response starting to fall off from flat at 28Hz. But in principle they should work well with the Quads as:

1. they're dipole like the Quads, and (hopefully) the bass radiation pattern is close to null where each Quad's electrostatic film is inside the vertical panel -- maybe reducing the impact of lower bass notes on the Quads
2. they use two drivers per sub aligned to "push" and "pull" against each other as they operate, making them  faster/more agile.

The DEQX has been configured to apply high-pass filters to the ESL63s and low-pass filters to the subs. I've got the crossover set at 120Hz to remove bass energy from the Quads hopefully allowing higher volume levels overall.

 

I must have made at least 20 configurations and room measurements so far, partly driven by needing to get better at using the DEQX and, more importantly, understanding the way changes impact the sound.

 

I feel like I've got the volume matching between the subs and the Quads dialled in. I'm working on time alignment, and there may be room for improvement here. I find the impulse plots for the subs a little challenging to interpret.

 

The DEQX offers a lot of options with xover filters and this is something I haven't really explored yet. For example, Linear Phase vs Linkwitz-Reily and the steepness of the slopes. The room measurements show a bit of a dip at the xover frequency which offers the opportunity for improvement.

 

Nonetheless, I'm pretty happy with where I am now and I'm acutely aware of the old maxim, "better is the worst enemy of good".

 

Edited April 25 by davm

[andyr]

1 hour ago, davm said:

The DEQX has been configured to apply high-pass filters to the ESL63s and low-pass filters to the subs.

 

Excellente!  

 

1 hour ago, davm said:

I've got the crossover set at 120Hz to remove bass energy from the Quads hopefully allowing higher volume levels overall.

 

Yes, you should be able to achieve that.

 

1 hour ago, davm said:

I'm working on time alignment, and there may be room for improvement here.

 

I suggest as your Quads are sitting on top of the subs ... no delay is required - they are physically time aligned.

 

1 hour ago, davm said:

I find the impulse plots for the subs a little challenging to interpret.

 

 

Hah - so do I!!  

 

1 hour ago, davm said:

The DEQX offers a lot of options with xover filters and this is something I haven't really explored yet. For example, Linear Phase vs Linkwitz-Reilly and the steepness of the slopes.

 

If you select linear phase ('FIR' filters) instead of L-R (IIR filters) then that would be a good choice.  I would suggest as steep a slope as you can get ... I use 48dB L-R (as my miniDSP unit doesn't offer FIR filters).

 

[Keith_W]

1 hour ago, davm said:

The ripole subs have a similar dipole-style radiation pattern to the ESL63s and I've chosen to place each ESL63 on top of a subwoofer.

 

Not really, Ripole subs are supposed to have a cardioid radiation pattern. But never mind, it should still work well

 

1 hour ago, davm said:

The DEQX has been configured to apply high-pass filters to the ESL63s and low-pass filters to the subs. I've got the crossover set at 120Hz to remove bass energy from the Quads hopefully allowing higher volume levels overall.

 

I suggest you measure the native response of your Quads and what distortion products appear at what frequency. 120Hz may be the right XO point but it would be nice to see if you have made the right decision

 

1 hour ago, davm said:

I feel like I've got the volume matching between the subs and the Quads dialled in. I'm working on time alignment, and there may be room for improvement here. I find the impulse plots for the subs a little challenging to interpret.

 

What procedure does DEQX use for time alignment? 

 

1 hour ago, davm said:

The DEQX offers a lot of options with xover filters and this is something I haven't really explored yet. For example, Linear Phase vs Linkwitz-Reily and the steepness of the slopes. The room measurements show a bit of a dip at the xover frequency which offers the opportunity for improvement.

 

1. If you see a dip at the XO frequency don't assume that it's due to the XO. Do a quick room sim in REW to make sure there is no null predicted. Sometimes you can get unlucky and you get a null exactly at the XO position and you end up chasing your tail wondering WTF is going on. Don't ask me how I know. 

 

2. Linear phase filters do not rotate the phase at the XO point. Use them if you can. Then the difference between Linkwitz-Riley, Butterworth, etc. is only the shape and steepness of the slopes. If using Min Phase filters, you also have to worry about phase rotation. 

 

3. If the dip is due to your XO (and not a null), that indicates a phase misalignment between your mains and your subs at the XO point. Study the phase graph closely, you should see a discontinuity at the XO point. It looks like a stair step. If you look at the Group Delay display, you will see a sharp peak at the XO point. Possible remedies include inverting the polarity of both subs, adjusting the phase of the sub with the plate amp, or adjusting the phase with DSP using a tool like RePhase. Not sure if DEQX software is able to do that. 

 

Good luck! 

 

11 minutes ago, andyr said:

I suggest as your Quads are sitting on top of the subs ... no delay is required - they are physically time aligned.

 

I wouldn't bet on it  It's not time aligned until you have a measurement proving time alignment. All subs are "slow", even if they are coplanar with the main speakers.