WAJ on AUDIO - for truth in hifi / stereo / high-end audio



BASSic ISSUES; Poor Midbass in HiFi: Classical Music the Cause?  

Could combining different drivers be one avenue to the solution?

by W.A.J.



For those of us who listen to a variety of musical genres, and especially for virtually all forms of popular music, I believe it imperative for us to get the bass (especially midbass) of our systems correct - even more so than it is for lovers of the classics.

Before going further, I should point-out that this article is mostly comprised of excerpts of my e-mails to an audiophile-friend of mine. In this instance, we were discussing ways my speaker-systems could possibly be improved. And so, a familiarization with said DIY speaker-systems may be helpful in understanding some of the discussion. Briefly; consider a mid-woofer system comprising a combination of outstanding (especially at low-mids) KLH 12" drivers (2 per side) combined with the 7" mid-woofer from Yamaha's award-winning studio-monitor; the NS10m. (Both types functioning as mid-woofers excelling at different characteristics). Also consider that an Altec 802/811 compression-driver/horn combination provides mids and highs, from atop a 'full-range' tower standing just over 5' in total. And also consider that one each of a pair of 'Goodmans of England' 18" woofers is ensconced in a large 5' sub-woofer, per side. Both the towers and subs fluctuate between sealed-box and bass-reflex modes of operation (dependinig on whichever experiment is being conducted at any particular time). And, for the near-future, the iconic Jensen Imperial horn cabinet is currently being seriously considered (in the discussions cited above) as a 'housing-unit' for these drivers - i.e., effectively, combining fullrange and sub-woofer systems in one cab - with serious increases in efficiency and bass-potency. (This combination would be potentially similar, in principle, to the example, here, for instance, and there's a version of an original Imperial, here). Efficiency of the complete system is now already at around 95db/w/m. (Details of this system may be seen here, and here - subject  to slight changes, as always).

However, though tips on system-building may be garnered (by the brave) this article is mainly about recognizing the deplorable state of midbass-reproduction, generally, in the realms of high-end audio. We look at why this may be so, and we also look at a possible avenue to a solution.

But first, let's regognize the midbass characteristics of main speaker-enclosure types (sub-categories notwithstanding). These are depicted in a general order of quality, starting at #1 with the best of all, though it's possible that in certain instances a lesser design may outperform a theoretically-'superior' type:

1) Front-loaded horn: Pros; traditionally, the best performer of all, at midbass (and overall) especially with regard to large theatre-horn types; those with the equivalent of large cones - ideally 15"ers, or doubled 12s. Highly efficient & dynamic. Cons; though amply aided in transient-response by the virtues of horn-loading, small-coned systems (especially those with sealed-boxes) tend to be 'quicker', more detailed, and articulate, at the leading-edges of notes.

2)Rear-loaded horns: Pros; As above. Cons; As above. Additionally, unless specifically oriented to excel at midbass (whether by accident or by design) many are relatively compromised by a significant drop-out in this region, due to phase cancellations.

3) Acoustic-suspension or Sealed-box: Pros; Excellent at midbass, but... Cons; Inefficient. Also, though excellent, they're commonly used with small-coned drivers (i.e. less than the equivalent of 15"). Because of this, they're typically unsurpassed at the leading-edges of midbass notes, yet they fail to follow-thru with the realistic 'substance' and/or weight which larger drivers excel at. (Note, however, that this criticism is not applicable to large sealed-box designs which incorporate the equivalent of large drivers - as exemplified by top-of-the-line Dunlavys and A-Rs, for instance - these are still not as efficient as horns, though. Here's a review of the AR-9, page 1 & page 2). 

4) Bass-Reflex: Pros; More efficient than acoustic-suspension (still less than horns) and can be good at midbass, BUT.... Cons; this is the most abused and corrupted design known to man. Many designers corrupt this type by seeking more extension than their resouces warrant. Therefore, in so doing, they garner the illusion of bass-extension partly by compromising the midbass performance. The result is a soft, mushy, wooly, indistinct, and ill-defined midbass which mistakenly displays some of the characteristics of deeper-bass. The sad reality is that this is by far the most common type of midbass displayed by typical speaker-systems today. Arguably, a 'pretty' sound (many people like it) but totally unrealistic. Absolutely awful, in reality - and the reason for this article.

Note: Popularly employed passive crossovers also degrade the performance of this and all speaker-types, and may be much more of a factor in the degradation of midbas transients & dynamics than most audiophiles realize - as I've recently discovered. (More to come, on that).  

With the backdrop now set, perhaps we should now repeat the opening line and continue from there. So, here goes:

The BASSic Issues: For those of us who listen to a variety of musical genres, and especially for virtually all forms of popular music, I believe it imperative for us to get the bass (especially midbass) of our systems correct - even more so than it is for lovers of the classics.

Why would I say that? After all, the fan of the classics would like to have correct midbass too, for sure. But it's also a fact that the classics do not engage the midbass frequencies (for large drums, for instance) with anywhere near the regularity as it would apply to popular music, and especially with regard to the percussive, transient, and dynamic aspects of bass reproduction. Indeed, one could spend an entire listening-session without encountering the fore-mentioned elements of bass - countless sessions, in fact.

Not so with popular music. Midbass (specifically; of a percussive nature, and especially regarding the ubiquitous kick-drum) is the back-bone, the very foundation, of popular music.

Very nearly every single track is dominated by it. This is why I say it's extremely important for a system to correctly reproduce the midbass, and the transients there-in, if it is to properly reproduce significant amounts of popular music. This is less of an imperative for a system  playing mostly classics, not because correct midbass would not be appreciated here, but because the irregularity with which it is encountered would cause deviant reproduction to be less bothersome - it could even go un-noticed. Not so with popular music since this deviant midbass would be noticed on every single beat, more or less. This is why I say there is much more of an imperative for a system playing significant amounts of popular music to be exemplary at midbass reproduction, along with all else.

But how many speaker-systems are there with the ability to replicate a well-recorded kick-drum in a manner virtually indistinguishable from the real thing? Only a pathetic few, I assure you!

Incidentally, I also believe most designers of typical high-end speakers are staunch classical listeners. They must be! Just look around at the multitudes of high-end speaker-systems which are deplorable at midbass and you'll see what I mean. These must have been designed by those who are not exposed to much in the way of midbass-transients, by any means. Yet these speakers are highly rated (reviewers must be biased toward the classics too) I can't understand it.

I'm sure you're aware of the wonderful reputation of the Spendor BC-1, for example. Well, based on my experience with it over too many years, I can tell you; it absolutely STINKS at midbass. [By the way, this is no reflection on the designer, Spencer Hughes, or on those who commissioned the design, the BBC, as both recognized it for what it was; a compromised medium-sized monitor for confined spaces. It could only have been considered 'very good' in the context of its limitations - and it was. But this reputation was blown out of proportion, in my opinion. It was never meant to exemplify the state of the art, as too many seem to believe. Others in the BBC line-up carried that mantle; the large-coned LS5/1A, LSU/10, LS3/1, and LS5/8 (aka Harbeth) are only some examples of BBC's very best, thru the years, all utilizing large mid-woofers from 12" thru 15" to a whopping 18 inches in diameter.]

Only the so-called 'gurus' listening to the classics exclusively could have rated the BC-1 speaker-system (with its 'soft & light' midbass, among other things) as highly as it is, even today. And, based on the similarly atrocious quality of the majority of high-end speakers, in the midbass region, one must conclude that the classics are too heavily depended upon, with regard to source-material in assessing speaker-systems. That's the point I'm making, here.

But it seems these 'gurus' are the ones who set the standards for all. I say this because even speaker-systems you'd have expected to be geared towards popular music display the same deplorable characteristics, with regard to midbass-transients. For instance; a JBL 4430 studio-monitor, which I tested against my cheap KLH, displayed very similar bass/midbass to that of the unmodified KLH, with its cheap crossover 'mucking-up' the response in a monumental manner, as I've recently discovered. And the handicapped KLH, itself, displayed very similar bass/midbass to the Spendor (similarly 'soft') which epitomizes the midbass of most popular speakers. The JBL L-100 was a domesticated studio-monitor, a pair of which I'd owned (often forgotten) during my tenure with the BC1. And though I suppose one could say its bass was significantly more substantial than that of the BC1, it certainly wasn't much more realistic - with similar characteristics, basically. (The BC1 beats it everywhere-else - as it does most others - except in dynamics, incidentally).

I believe a combination of factors; featuring cobbled designs (mostly bass-reflex) compromised driver-quality, and the ills of passive crossovers, account for the poor quality of midbass reproduction in high-end audio, generally. Those are the real causes, so I wouldn't blame the classics for that. But if those that test these systems are relying mostly on the classics as the medium with which they assess performance, then this would explain the poor midbass performance of hifi speakers, in general.

A trend toward a tighter (though light) mid bass seems now to be in process amongst popular speakers (though a lack of realistic 'weight' detracts from realism, obviously). But the general reality (i.e. since the ousting of theatre-type horns and acoustic-suspension in the 70s) has long been the soft, wooly, pretty mid-bass of those like the Spendor - copied, to a 'T', by virtually all and sundry. Add to that scenario; latter-day overtures toward a tighter but lighter presentation.

Which is 'better'; tight & light, or soft & wooly? These seem to be our choices, amongst the popular brands of speakers (with 'bloated and boomy' lurking on the outskirts, in botched prosound-type designs).

Blended Sonic Characteristics: Have you ever heard a soft kick-drum, by the way - or Tom-Tom, or Timpani, or Kettle? (Perhaps there are 'light' versions of these too - as with light-beer). I just wondered.

Yet this soft, pretty, midbass is so dominant, and so accepted by audiophiles, that many don't even recognize flab-free, accurately-toned, and superior midbass - or the general characteristics there-of - when they hear it. I like relating this next story (there's another version of it in another article, here) bear with me: One of the main drummers of  the 60s' Motown's studio-band recently revealed to an interviewer, in a documentary, that the best kick-drum sound he'd ever managed to produce is once when he substituted a cardboard beer-box carton for a real kick-drum. (That documentary was; "Standing in the Shadows of Motown". The drummer was Uriel Jones.  Refer to his comments starting at around 53 minutes into the documentary). On hearing his remarks, I went out of my way to take stock of what such drums really sound like, in real life. I discovered that the guy's implication is absolutely true - this is what many kick-drums (not all) really sound like; a thick-cardboard beer-carton. Pound on one with the ball of your hand (not the knuckles) and you'll see.   

Now let's switch, for a moment to that little white-coned monitor found in most studios around the world - the monitor many love, and many love to hate; the Yamaha NS10m. (Here's a research paper on it, by Dr. Keith Holland & associates - and here's another review). I once read an account of one user's vehement criticism of this speaker, and was galvanized into action. Why? This user complained that he didn't like the Yamaha NS10m because its 'bass' sounded 'cardboardy' (his wording). He complained that the NS10 sounded different from 'normal' speakers; it lacked deep-bass (really now - a  7"er lacks deep-bass - amazing) and most annoying of all is that its midbass, with kick-drums for instance, sounded 'cardboardy'.

Needless to say, on reading that, I immediately launched into a search, and wouldn't have stopped until I'd found a few of these 'awful' NS10s, which did what most other speakers simply couldn't - i.e. mimic the characteristics and tone of the typical kick-drum. And, yes, it's absolutely true. With some drums NS10s do sound like they're reproducing drums that sound similar to cardboard boxes (as they absolutely should) and furthermore, they highlight and match the similar, but less distinct, tonal-trait of my own Goodmans 18" subs - together they're inimitable. [Oh sure, the NS10 is from the 'tight & light' camp, but without the NS10 shaping the response and tone at the crucial beginning of notes, the Goodmans subs become nearly ordinary - this region ceases to be outstanding in its quality, detail, tone, and general accuracy. It's in the (upper-bass, upper-midbass) regions, at which small mid-woofers like the NS10s' excel, where the majority of the tone, detail, and sonic-characteristics of bass instruments are most evident. Allowing an outstandingly accurate small driver to span its normal range and, additionally, to delve deeply into the sub-woofers' territory (to its natural limits at around 60 hz, perhaps) imbues the fore-mentioned region (upper-bass/midbass, especially) with much of the small driver's desirable characteristics. It is also the combination of the characteristics of outstanding small and large drivers (with the latter conveying realistic 'weight') that garners unprecedented realism, overall, in this case.]

But how remarkable is this? The user was exposed to the accuracy of the NS10 and never recognized it as such. He lamented not having the type of midbass most speakers produce (as described above, re; Spendor, etc). Now, how nonsensical is that? I suppose this gives new credence to the old saying; 'Some wouldn't know good sound even if it were to come up and bite them on the... ear'. Too true!

[Perhaps most significantly, as previously hinted, I've discovered that having such a light, tight, and accurate driver working in conjunction (and largely overlapping) with the large woofer causes the response to be just as 'quick' as that of a small woofer by itself - except that the small/large-driver combo is more realistically weighty. I've further discovered that neither the small nor the large driver, by itself, is anywhere near as good or as realistic as when both are combined in this manner. I also believe that Wilson Audio's employment of a similar strategy (if not as radical) accounts for the renowned alacrity of the bass-response in their flag-ship model, despite issues elsewhere. (See page 2 of the X-1 review, and refer to 'System & Crossover', paragraph 2).

But the concept of blending the sonic characteristics of different drivers is a recurrent feature in several aspects of my DIY speaker-system. (The cheap KLHs' only reason for existence in my system is for their unparalleld low-mids performance, for instance - nothing else - other strengths are garnered from other drivers and strategies. The lower-midrange of the KLH is more realistic than that of ANY other driver or  speaker-system I've EVER encountered, regardless of cost - and I've encountered quite a few. Midbass of the KLH is crap, tho, so it's purposely subdued in order to be over-rided and handled outstandingly in my system by the fore-mentioned NS10/Goodmans combo. And deep-bass from the latter driver can also be awesome).

This concept of combining drivers with contrasting/complementary strengths tends to elict performance beyond the capabilities of any one driver. I call it; Blended Sonic Characteristics (BSC) for now. And where 'BSC' specifically applies to my typical configuration for bass, I call this the 'double-bass' system - i.e. 2 separate systems optimised for upper-bass/midbass and lower-bass reproduction. Decware, in their modification of the Jensen Imperial speaker-enclosure design, have also recognized the benefits of a similar strategy (which, perhaps, explains why I'm seriously considering a 'mildly-adjusted' version of it, for my own use).] 

The Ball and the Murky-Pool: Returning to the midbass-response of speakers, in general, here's one way I'd describe it: Let's imagine bass, including deep-bass, as a body of water, a pond. Let's also imagine midbass-transients (the sound of a kick-drum, for example) as a ball, a bowling-ball, held just beneath the surface and charged with the task of quickly rising above the surface in the instance of each midbass-transient.  

In my view, the way most speakers depict the midbass-transient/bowling-ball is in such a way that virtually the whole ball remains submerged in the murky waters of the pond (which represents other bass frequencies). Concentrating on the typical type of response offered by my Spendor, I'd say seven-eighths of that ball is generally submerged in that murky pool. Midbass-transients are feeble and indistinct with such speakers, with perhaps a faint depiction of the initial transient in evidence (i.e. the exposed one-eighth of that ball) above the murk of surrounding bass. The rest of that kick-drum note, the body and the end of the note, are hidden beneath the murk of the accompanying bass frequencies, and even modulated by them. (And the same kick-drum note alone, 'naked', without the other accompanying bass frequencies, is not much more convincing).

Better speakers like the folded Klipschorn would display as much as a half of that ball, above the surface of the pool, above the accompanying bass. And a good front-loaded straight-horn might have done somewhat better. More of the body of that kick-drum note is exposed, in both cases. Very good, in my opinion, and more convincing. But, despite the vagaries of the recording process, is it possible to reproduce nearly all of that note? Could we garner much more of the leading-edges and ending of that note, for instance, more detail, and more of the defining tone of the instrument's note, as generated at the initial stages of said note?

I believe so. And, in my view, this involves the combination of the large driver working in unison with the small - specifically, in my case; working with a light, tight, low-compliance, paper-coned, small-driver, which is outstanding at the the leading edges (in addition to the initial substance and tone, and even the actual ending of notes) at midbass - something like the NS10's mid-woofer. I'd suggest that such a combo would expose three-quarters to the whole of that kick-bass note (recordings permitting) regardless of any other accompanying bass frequencies. (Note that I'm not referring to latter-day digital recordings, which are better, BASSically, than recordings of the 50s to 70s, for instance - I refer to the playback of the more difficult recordings of that vintage-era).

I recently linked us to a 6moons  review of a Jean Hiraga speaker utilizing one from amongst my favorite type, and size of driver; Altec (604, in this instance). At page-2 of that review, it was pointed out that; good though the 15" Altec was at conveying the weight of instrumental notes, it failed to articulate the 'quicker' components of those notes. The review further highlighted that another system on hand, featuring small-coned-drivers, was much better at the attributes the Altec lacked, but that the small-coned system could not convey the 15" Altecs' realistic weight.

Recently, in an article entitled; 'Why the Small Driver Can Never Be As Realistic As the Large..' I made the very same point I'm making in this segment: I believe the way forward is a combination of the old with the new, a combination of the large-driver with the small - the best of both worlds, so to speak. I could be wrong, but from what I've witnessed, this does seem to be a viable course of action in the quest for higher levels of accuracy at bass/midbass reproduction. 

Conclusion: Returning to my speaker-building dilemma; under normal circumstance I would jump at the prospect of building a horn, again; the Jensen Imperial, as mentioned. (My first DIY speaker at age 13, or so, was a scaled-down slot-loaded 'corner-horn', with several other types built since then). Normally, I'd advise anyone to build bass-horns, especially for their dynamism and transient alacrity. But now I'd be inclined to add more, if possible. Specifically, I'd investigate whether a similar small-coned scheme could be incorporated with the large driver(s) of the main horn, somehow - since I'm now convinced that no horn can induce a large and heavy cone (of 15", for instance) to be as 'quick' as the small. (Perhaps doubled 12s equipped with light cones, used without passive crossovers, could be an alternative to this scheme - thus; obviating the need and encumbrance of adding smaller drivers. I'll elaborate later, as this is another recent discovery).

But, in assessing the efficacy of these theories, I implore you to stay away from classical music, in general. Stick with popular music, especially those with lots of midbass transients since this is where most speaker-systems fall apart.

Please don't get me wrong (I doubt you have, but just to be sure). My rant should in no way be construed as an attack on classical music, or on fans of it – I too am a fan, to some degree. It's an attack on the poor performance of most speaker-systems in reproducing midbass-transients, in general.

Apart from dynamics and low-midrange tone (or the lack there-of) I believe mid-bass is the region where most systems fail dismally. Let's face it, many otherwise-good systems can render a fair facsimile of the blat of a trumpet, the strike of the stick on a snare, the sonority of a violin, or even the tunefulness of a vocal note or two.  In listening to a recording on such a system, we may even begin to accept that the reproduction may be closely approaching some semblance of what the real thing could conceivably have been like. In extreme cases, we may momentarily even be fooled into thinking it's real. 

But then it happens.

Yep! The moment, the very instant, a poorly reproduced kick-drum assaults our consciousness, the illusion crumbles. That's annoying, to say the least. And, with popular music, most annoying is the fact that this soft, mushy, 'pretty', and disgustingly unrealistic depiction of the kick-drum is inevitably repetitive throughout the track - track after track - it's sickening. And the 'tight & light' presentation of some latter-day examples may be somewhat less-disgusting, but still also far from ideal - they're incomplete, with an obvious lack of realistic 'weight'.

One would have hoped that, with so much of music (apart from the classics) based on mid-bass in general (and on the kick-drum in particular) speaker-manufacturers would have gotten this region right, at least.

Not so!

As it stands, this region is absolutely, indisputably, the worst-reproduced region of the spectrum, and most noticeable as such, generally speaking. In religious circles, this would be deemed; a 'sin'. Perhaps you'd want to look into this - or not - perhaps performance, here, is already satisfactory; congrats. The preceding incorporates some of my own ways of addressing this 'sin' (hopefully, a point or two will be helpful to some).

What's more? It's a sin of commission; in most cases; the worst sort. (It's actually worse than the sin of omission most speakers commit at low-mids).

But then, these are only my opinions - BASSically!


Continue to Part 2.


Copyright 2013