subs + bass horns + kicks?

[seamus]

I have been experimenting with 6th order BPH designs for subs that go low (hit 30hz with authority) to compliment a stack of short bass horns.

The subs would be a 2 x 21 from about 30 - 60hz, bass horns would 2 x 18 from 50 - 150hz and the tops will be some sort of 3 way (think F1 evo7/res5 or danley sh50).

My question is does having subs + bass + kicks/low mid horns servicing narrow bands introduce too many potential complications (horn lengths, phase etc...).

I've always had systems with bass horns and love the sound. When coupled they kick like mule and fidelity is great. The issue is they never went low enough no matter how many I brought to the venue.

I used to haul around 10 Cerwin Vega earthquakes and my friend would bring 12 TSW124 (we like horns).

Anyway i'd love to keep the horns belting hence the idea of having a narrow band sub underneath them. However I'm wondering if its a better idea to use a sub that goes up to the 100hz range and use the low mid in the tops to meet them (likely a 15" or 12" driver).

Thanks!

[Ricci]

Hello Seamus...

Personally I'm not a fan of subdividing the bass region into a 2-way for only covering roughly 30-120Hz, but a lot of people prefer this. 

The reason that I do not is because 50-150Hz bass/midbass is easy. Has been for decades. 25-50Hz octave is much harder and requires more cab size, driver displacement and power. Every octave lower in frequency requires 4X the air displacement to maintain the same SPL. Most people have some type of limit on the amount of gear they can haul, both size and weight, the amplifier power available, cost, etc...A good sub design can run up to 100-120Hz and good mains can get down to meet them with power. The resources used for the midbass/kick bins could be used in the bottom of the sub range and the mains. It's also slightly more complicated to setup a 3 or 4 way system. 

Opinions differ on this. A lot of it is personal preference and the way people like to EQ or flavor their sound. 

 

[seamus]

Hi Ricci,

Thanks for your response. Great designs by btw.

It could just be that I haven't heard enough 6th bph type subs stacked up but it seems to me that they don't kick as hard in the 80 - 150hz ish range. Hence why I was considering going the 2-way route.

Do you think that there could be any benefit of a 2-way bass system when it comes to nailing that 60-100hz  and 30hz ranges equally? Or is it a pointless exercise and just stack up more subs.

 

[seamus]

Let me rephrase.

On the basis that a narrower band equals more efficiency will a 2-way bass system be perceived as being more powerful?

[Ricci]

The 2 way may be perceived as more powerful in the midbass region. It will not be as powerful in the low range, because system resources are being used for midbass only. 

If your midbass bins are crushing powerful down to say 60Hz. You would need 2 or 3x the cabinet size and power to keep up with subs at 30Hz. 

Take this hypothetical scenario.

Assume you have 10kw of amp available and 160cu ft of cab volume. This is what'll fit in the truck space and work on the AC mains. Also assume that you have 3 different sub designs: a full bandwidth 30-120Hz sub that's 20cu ft, a 30-60Hz LF only sub that's also 20cu ft (LF extension + efficiency requires big), and midbass cabs that are 10cu ft and work 50-150Hz. 

System 1 is all the midbass "sub" cabs that don't get below 50Hz. You could fit 16 of these in the truck. Output would be chest caving high in the midbass, but notes below 50Hz are going to be way down in level and anything much below 40Hz is MIA.

System 2 is a hybrid with 6 LF cabs covering 30-60Hz and 4 midbass bins covering 60-120Hz. Assume 1Kw per cab, so 6kW for the low octave and 4kw for the upper. This is a typical ratio and would = the same amount of truck space as system 1. You now have the bottom end that was missing from system 1 but you've reduced your midbass power greatly by moving from 16 midbass cabs down to only 4. 

System 3 is 8 full range subs covering the whole 30-120Hz. This much bandwidth and more is not that difficult to cover well with one design. This has the most LF performance due to all of the power and all of the cabinet volume is available for the LF <60Hz. 8 cabs vs 6 and 10kW vs 6kw or some lesser amount. It still wouldn't touch the midbass output of system 1 because the system is now optimized to extend much lower. However it may equal or better the midbass output from system 2. 4 midbass cabs equaling a much smaller total volume and on less power vs 8 subs with all of the cab volume and power available. Also there is no crossover in the bass region. Output can be adjusted to taste with EQ.  

This is over simplified but you get the gist. Again not saying this is correct or the only way to do things. Lot's of people love their 2 and 3 way bass systems. I think it's partly due to the flexibility to turn up one section or another as an easy way to EQ the sound to taste. 

The takeaway is that going for lower frequencies has tradeoffs.

[seamus]

Thanks Ricci, really appreciate your insight.

What you've said clarifies what I have been thinking.

I understand that 2-way will sacrifice low end output.

Thanks again

[SME]

I very much lean toward fewer crossovers being better.  Crossovers are rarely precise enough to not degrade sound quality.  I would also add that SPL isn't everything.  Better sound quality can provide more apparent loudness (and more listener enjoyment in general) at lower SPL.  So while you may be sacrificing low-end on paper, if fewer crossover resonances means you're "hearing more bass", you might still come out ahead.

[seamus]

Hey SME, that was something I was worried about having crossovers in that fun zone causing phases issues between the cabs defeating the whole purpose.

Would you say that a 21" sub covering 30 - 100hz ish necessarily sounds better than splitting that range between sub and bass cabs?

I have some 15" fane drivers laying about so I'm going to design a low mid/upper bass horn (i'd assume to be in the range of 70 - 250hz). I'll post on the forum.

Bellow that I'll try and come up with a sub design down to 30 or just build some skrams.

Is that still too 2-way bass do you think? could try and tune the 15s a bit higher.

[Ricci]

Seamus.

Also have a look at the studies for preferred response curves. There are many. JBL, Harman, D&B and some others have published target curves... A trend emerges. Most people prefer a bit of a rise towards the bass but it doesn't start truly ramping up until below 150Hz and below 70Hz is where the true boost is. The majority of the heavy lifting is below 80Hz. 

[SME]

On 3/2/2022 at 5:15 AM, seamus said:

Hey SME, that was something I was worried about having crossovers in that fun zone causing phases issues between the cabs defeating the whole purpose.

Would you say that a 21" sub covering 30 - 100hz ish necessarily sounds better than splitting that range between sub and bass cabs?

I have some 15" fane drivers laying about so I'm going to design a low mid/upper bass horn (i'd assume to be in the range of 70 - 250hz). I'll post on the forum.

Bellow that I'll try and come up with a sub design down to 30 or just build some skrams.

Is that still too 2-way bass do you think? could try and tune the 15s a bit higher.

I'm not sure what you mean by the "whole purpose" here.  As noted above, it's not hard to cover 30-100ish Hz solidly with a BP6 like SKRAM, and many mid-woofer sections get down to 100 Hz no problem.  From there, it's a matter of tuning things to get the best bass sound, and it's a lot easier to tune with only one crossover in the bass area.  When crossovers are close together, the problems are significantly worse.  With that said, I think your idea of using bass horns to cover 2+ octaves from 70-250 Hz is likely to work out a lot better than trying to use dedicated "kick bins" to cover only ~1 octave worth.

Really, the "sub sound" lives in the frequencies below 250 Hz or so.  That's because 2nd and 3rd harmonics contribute substantially to bass sounds.  A 65 Hz fundamental has 2nd and 3rd harmonics at 130 Hz and 195 Hz.  All of these are crucial to the most impressive, most tactile bass sound.  (Higher harmonics matter too but are of diminishing importance.)To be clear, this doesn't mean you want "flat response" across the bass.  A power response that rises toward the bass sounds most natural. Power response concerns total sound output in all directions rather than SPL measured at some point.  Almost all speakers which measure flat on-axis in an anechoic chamber exhibit a downward sloped power response, starting at a point below which the radiation is omni-directional.  Below this point, in-room SPL response typically rises if the trend in power response is being maintained (hence the "house curve").  This tends to happen naturally in small rooms, but in a live/PA setting where bass-boosting boundaries are fewer and farther away, this can require a lot of boost for the lowest frequencies.

Hence, I'd say that if you're looking for more mid-bass "kick", you probably don't need more mid-bass SPL unless you're already working with a whole stack of SKRAMs.  What you probably need is smoother response across the mid/upper bass range, which is best served with better tuning rather than more brute.  It doesn't take much SPL at all to slam and throb if things are really tight.  Though of course it's all relative, so if you need your mids and highs at ear-blistering levels, you need some serious bass to keep up, and (except in small rooms indoors) the last octave is almost always the most demanding of SPL.

[seamus]

I meant the whole purpose in terms of 2-way bass and processing negatively effecting the sound.

I havent had a chance to check out the studies on preferred response curves that Ric mentioned (I'm assuming they are in the aes library).

I have been semi wittingly targeting something similar to the house curve when deploying my old system.

Is it a good idea to target a curve (that isn't flat) when designing a cabinet? Or is it a matter of the ratio of cabinets deployed and dsp. Our sweet spot for the old system was 8 x bass horns and 4 x 3-way horn loaded tops.

Thanks for the detailed response its appreciated.

[peniku8]

7 hours ago, seamus said:

Is it a good idea to target a curve (that isn't flat) when designing a cabinet?

Target maximum efficiency everywhere (within your design goals of course) and correct the response in dsp. This results in lowest distortion levels and best headroom.

[SME]

19 hours ago, seamus said:

I meant the whole purpose in terms of 2-way bass and processing negatively effecting the sound.

Well it's not always true that 2-way bass will sound worse than 1-way bass or whatever.  The end result depends on the whole system.  Nevertheless, crossovers in the bass are rarely ideal and introduce flaws which are difficult to eliminate completely with additional EQ.  Multiple crossovers make this much more difficult.

19 hours ago, seamus said:

I havent had a chance to check out the studies on preferred response curves that Ric mentioned (I'm assuming they are in the aes library).

I have been semi wittingly targeting something similar to the house curve when deploying my old system.

Is it a good idea to target a curve (that isn't flat) when designing a cabinet? Or is it a matter of the ratio of cabinets deployed and dsp. Our sweet spot for the old system was 8 x bass horns and 4 x 3-way horn loaded tops.

Attempting to design a cabinet to a particular target curve is likely to compromise overall efficiency.  The BP6 designs like the SKRAM and SKHORN have a lot more output at 100 Hz than 30 Hz, even though most applications need more output at 30 Hz than at 100 Hz.  That's not a flaw though.  It's fundamentally harder to get more bass at 30 Hz, and the higher efficiency at 100 Hz makes more amp power available for 30 Hz while keeping the voice coils a lot cooler.

As far as target curves, I think they are best treated as a sanity check rather than a strict guideline.  I've seen a lot of different "preferred curves" (including multiple allegedly recommended by Harman over the years).  They all tend to suggest a gentle downward slope and an SPL difference between bottom and top of 6-10 dB.  They vary with respect to which frequency areas are steeper, but I suspect these differences are not that meaningful, not because such differences are not audible (they very much are) but because different systems and rooms will probably sound their best with different target curves.  And don't forget that measurement location has a big impact on the broad shape of the response.  The other thing is that actual content can vary a lot in terms of broad spectral balance, even between different tracks in a given mix.

I realize this doesn't answer the question of what ratio of subs vs. tops to provision for.  The answer is probably a matter of preference in part, and may be affected by budget as well.  For my home system using experimental DSP, I rely on an estimate of in-room *acoustic power* response to optimize my bass.  I use a -3 dB/octave target for ~1 kHz and below because that yields a response for my speakers that's closest to measuring flat under anechoic conditions.  This also gets the SPL response at my seats pretty close to the "preferred curves", though I measure particularly hot from 20-70 Hz, which probably has to do with most of my seats being near a partial wall (a high pressure zone).  Subjectively, this seems to be a good reference for a "performance" system, but I know a few people would want even more bass, especially for a bass-centric genre.

For a large room / outdoor PA system, there is little boundary support, and so it'll take a lot of sub to match tops while maintaining that -3 dB/octave power curve.  I would want at least +6 dB more output per channel at 30 Hz than 120 Hz, and indeed +6 dB in terms of capability may not be enough for the content that has a steep power curve of its own.  (This is very much the case with movie tracks!)  And if a -3 dB/octave power curve isn't enough?  For -4 dB/octave from 1 kHz down one needs an additional 5 dB at 30 Hz, which requires almost twice as many subs.  On the flip side, if budget is tight and loud bass is not essential, one can use about half as many subs to target -2 dB/octave, and so on.

To be clear, most of my dB figures above are in terms of *acoustic power* output, which is not something one typically measures directly, but on the other hand Hornresp simulations (for sub/speaker design) actually give results in terms of acoustic power, not SPL.  SPL response only loosely tracks power response because of directivity effects and environmental acoustic factors.