Driver size and complexity of music

0 Members and 1 Guest are viewing this topic. Read 4462 times.

chadh

Driver size and complexity of music
« on: 24 May 2009, 02:43 pm »

In another thread in this circle, there's a discussion about the perceived weaknesses of a small (4.5") single driver with regard to its ability to play various types of music.  In particular, claims were made that suggested the inability of small drivers to cope with "complex" music.

Now, it's clear to me that I don't really understand how speakers work.  But it seems to make sense to me that small things (like drivers) have a hard time make deep sounds.  But it's not obvious to me why small drivers would have a hard time dealing with "complex" sounds.  For example, when dealing with orchestral work, and needing to produce many different sounds at the one time, why would a small driver face greater problems than a large driver?  They both work the same way.

Maybe I've just misinterpreted the claim.  I can see why it would be easier for multiple drivers to produce different sounds if those sounds are being produced by the different drivers. 

Chad

rjbond3rd

Re: Driver size and complexity of music
« Reply #1 on: 24 May 2009, 03:31 pm »
In my limited experience, there are a couple related issues.  First, excursion can be an issue with many fullrangers, and as you exceed xmax, the music gets compressed.  The second issue is Doppler distortion (the low frequencies modulate the high frequencies due to cone motion).  However, these issues go away in proportion to how you high-pass them.

Everything has limitations, and fullrangers have to be babied a little.  But in my mind, it's still better to have speakers that can do an outstanding job on some music than to have speakers that can do a mediocre job on everything.

Scott F.

Re: Driver size and complexity of music
« Reply #2 on: 24 May 2009, 03:36 pm »
Chad,

This is a common misconception perpetuated by people who have heard single driver systems that may not be of the best design, that or using inferior drivers (somebody doing a system on the "cheap" rather than investing serious money on quality drivers).

Granted, single drivers have absolute limitations when it comes to the frequency extremes. No debate there at all. Augment your favorite driver on whether the high side or low side with an active (or passive) crossover in the first two or last octave and you will find single drivers are FAR more coherent across the frequency bands they cover than a multi-way speaker.

You have to keep in mind that though while a speaker is moving forwards or backwards along its travels, it still is receiving tons of signal information that equates to additional micro-movements of the cone. Every speaker does the same thing whether a full range or a tweeter.

Some single drivers seem to do a better job at transcribing that information than others. For me and and my experiences, the higher the Gaussian strength of the magnet (read=Alnico), the better the coherency of the driver.

All that said, the thought that a single driver can't do complex music is simply absurd. In my system I can play anything from the audiophile "little girl with guitar", Mahler to NIN, Tool, to the Postal Service and it all is extremely coherent.

As always, if you ever find yourself in St Louis, drop me a PM and I'll be happy to prove this to any disbelievers first hand.

chadh

Re: Driver size and complexity of music
« Reply #3 on: 24 May 2009, 07:18 pm »

Scott,

I myself have 4.5" (Fostex) drivers in my single driver (Omega) speakers.  And the more I hear other speakers, the clearer it is to me that "coherence" is the huge strength of my speakers (and, I presume, most other single drivers).   

I don't listen to much orchestral music, though, so I'm not about to deny the claim that others might make that their little 4.5" drivers don't do that style of music justice.  On the other hand, I don't really understand why a 6" or 8" single driver would would be less likely to lose its coherence when dealing with that kind of music than a 4.5" driver.  Now, if it's simply a failure to deliver the bottom octaves well, then I appreciate that a bigger driver would help.  But the ability of a driver to respond accurately to lots of information in a signal isn't obviously a function of its diameter.  But then again, lots of things about audio is far from obvious to me.

If anything, I would have imagined that a light, nimble driver would be able to deliver more information coherently than a bigger, heavier, clumsy one.  But I might well be wrong.

The one drawback that I really notice with my speakers is a very specific one.  For some reason, I've got a real weakness for polyphonic choral work.  There are times when I'd happily listen to the Tallis Scholars sing Palestrina all day long.  My speakers sound beautiful when I play this music, but just lack some weight.  I want to be totally drenched in this music, engulfed in the swirling maelstrom of vocal lines.  But, no matter how loud I turn it up, it doesn't ever get there.  I imagine that this kind of issue would likely be solved with a bigger driver, but I'd hate to lose any of the ability that my teeny-weeny drivers have to respond swiftly and faithfully to the music.

Chad

Scott F.

Re: Driver size and complexity of music
« Reply #4 on: 25 May 2009, 12:30 am »
Chad,

Though I don't have much choral music, John Blaine our late GAS recording engineer shared a stellar recording of a local group who called themselves Armonia Early Music Ensemble. John recorded them in the chapel at Prior High School here in town. John was a minimalist when it came to recording. A simple pair of mics, a mixing board with no EQing then straight into his digital recorder. His post processing consisted of a minor leveling of the signal on the computer then saving the file as a wav.

After listening to this recording, I can imagine what you are missing not having that "heft" and "body" to the music. Even though the human voice doesn't extend down very far, the resonances of this chapel with a male voice are well into the upper part of the last octave, maybe just a bit above. If your speakers don't reach that far down solidly (and they don't as I remember then I heard them some time ago), you should probably consider a sub.

I'm lucky, I use a pair of 15" Altecs actively crossed over at 125Hz then above that my Lowther PM2As. My Altec's add tons of weight to the music in the right proportions. If you don't have some sort of sub rolled under your Omegas, you will definitely be missing that room filling bass they leave out.

Tell you what, you may try heading to Best Buy or Costco and trying one of there little 8" or 10" sealed subs (do the sealed, they are tighter sounding than the ported ones). You'll probably have to crank the crossover down to near or below 40Hz to get a decent blend between it and your Omegas. I think then you will find that you are getting closer to the sound you want.

Those little subs aren't overly expensive. You can get them for about $150-$200 or so. Who knows, maybe you can find something on Craigslist a little cheaper just to experiment. Once you get a better fell for the advantages a sub has in your system, you can go for something a little better and resell the cheap one.

If you like the sound of your 4.5" Fostex, don't look any farther. Stick with what you've got an go the sub route. As you walk up the Fostex line the bigger drivers can sound a little weird at certain frequencies (just my opinion, to each their own, not looking for a battle by any means). It may have something to do with what you mentioned about the heavier paper cones and not being able to articulate the music like a lighter, stiffer cone, not sure but that's my guess anyway.

Not sure I really gave much of an answer, just something to consider that I think might help. :scratch:

Docere

Re: Driver size and complexity of music
« Reply #5 on: 25 May 2009, 01:23 am »
I don't have heaps of time:

Chad, generally speaking, your 4.5"-ers are going to have a greater linear cone motion - in and out (excursion) - to excite the same amount of air as a larger driver; it will need to move farther in order to reproduce a given frequency at a given volume than will the larger driver. That need to move a greater distance brings excursion limits and the Doppler Effect more into play. A driver can move only so far and designing a driver to have higher excursion introduces some design challenges and tradeoffs. The Doppler Effect becomes more pronounced as a driver has to move further and a greater amount of (higher) frequencies are riding on the larger (bass) excursions. Regarding Doppler Effect, some claim it is audible/problematic, others not. I am guessing one's opinion comes down to the type of music one listens to, at what volumes, and whether they are sensitive to the effect.

I would anticipate that polyphonic choral music at decent volumes, with its sustained power requirements accented with subtle nuances at frequencies our hearing is most sensitive, would be challenging for many systems. I recall playing Lisa Gerrard at decent levels through a nice two-way (6.5" mid-bass) and reasonably powerful amp and hearing obvious compression. As for orchestral, I think it has most of its energy in the 200 to 500hz range? I suspect this would be a challenge for most any small driver.

Although larger coned drivers generally have lower excursion requirements and therefore Doppler Effects they have their own challenges. Higher mass cones, more scope for resonances and colourations, reduced high frequency dispersion, more limited/lower quality upper frequencies... again, these are generalisations. When run genuine full-range without bass support, I think the larger drivers can provide a greater sense of macro-dynamics, effortlessness, palpability, and balance; smaller drivers, when run within their limitations, can seem more nuanced, subtle, and direct.

As for smaller full-range drivers being unsuitable for complex music - especially orchestral - in general, I think there are a range of reasons people believe this to be the case. A worst case scenario would be a person who likes to listen to music at high volumes trying out a system built around small medium to highish Qts drivers loaded into in resonant enclosures driven full range by poorly designed high output impedance amps of marginal (or low) power for the job. Such systems are quite common and may reproduce simple music to many people's liking; they won't do complex music well at most any volume, let alone volumes some people demand.

There are exceptions to these generalisations, of sorts. Individual driver designers can stretch the paradigm, or achieve a more positive balance of the benefits and constraints.

To my current way of thinking, a speaker loading that minimises cone excursion whilst getting appropriate energy in the upper-bass/lower midrange is important; from <~ 80hz, support with a well implemented sub. The amp (SET, please) and speaker would best be high-passed around the 80hz point, to reduce the demands on both. Tune the system so that it sounds balanced at a volume level you like to listen? or a little lower. You may just end up enjoying music played back at lower levels, whilst preserving your hearing, requiring fewer system compromises, and likely saving a little money as well.

This was a bit of a ramble! I hope it makes some sense.

Cheers
Raymond

Edit: clarified one point; fixed some typos.
« Last Edit: 25 May 2009, 02:41 am by Docere »

chadh

Re: Driver size and complexity of music
« Reply #6 on: 25 May 2009, 01:10 pm »

Thanks guys. I appreciate your efforts to enhance my musical experiences.  I've toyed with the idea of a sub before, but never seriously.  My amp is handily equipped, though, with one set of full range outputs and one set with a high-pass filter.

Chad

chrisby

  • Full Member
  • Posts: 772
Re: Driver size and complexity of music
« Reply #7 on: 25 May 2009, 06:20 pm »

Thanks guys. I appreciate your efforts to enhance my musical experiences.  I've toyed with the idea of a sub before, but never seriously.  My amp is handily equipped, though, with one set of full range outputs and one set with a high-pass filter.

Chad

Chad, excuse if this has been answered before, but exactly what is the amp that is "handily equipped"?   It's been my experience that many of the HT class amps or receivers my have "flexible" filters or XO's, but their quality can be inferior to dedicated separates or even low parts count DIY PLLXO. 

Using the full-range output to drive a sub can't harm anything physically, but could cause major sonic issues, so you'd still want to implement a low pass filter.
 

chadh

Re: Driver size and complexity of music
« Reply #8 on: 25 May 2009, 06:27 pm »
My "handily equipped" amp is the 5w/channel version of Roger Modjeski's Music Reference EM7.  It has two pairs of binding posts on it.  One has a high-pass filter at around 100Hz, the other has no high-pass filter and gives a little bass boost below around 400Hz (I think).

I suspect that Roger has employed something very similar to the PLLXO idea you suggested, but has simply built it into his amp just before one of the sets of binding posts.

My pre-amp also has two sets of outputs.  If I were to employ a sub, I imagine I'd take one set of pre-amp outputs to the sub, the other to the amp, and then run my speakers off the amplifier outputs that utilize the high-pass filter.

Chad
 

doug s.

  • Full Member
  • Posts: 6572
  • makin' music
Re: Driver size and complexity of music
« Reply #9 on: 25 May 2009, 07:24 pm »
chad, i would heartily recommend you crossing your speakers at 100hz or so, and adding a pair of subwoofers for the frequencies below that.  besides getting the low end, crossing your mains at 100hz will allow them to do everything above that frequency even better than they do now, being relieved of seeing the low frequency info that they surely struggle with...  while i prefer using an active outboard x-over, your amp as descried may work out yust perfectly...

also, while i always prefer using stereo subs, unless you are going to be crossing over much lower than 100hz, and are using a steep slope x-over, you will almost definitely need a pair of subs.  otherwise, imaging will suffer.  if you insist on trying yust one sub, i would place it for proper imaging, not best bass response, and set it centered exactly between your mains...

doug s.

chrisby

  • Full Member
  • Posts: 772
Re: Driver size and complexity of music
« Reply #10 on: 25 May 2009, 09:01 pm »
To echo what Doug mentioned, when properly implemented ( i.e. XO point and slopes) the most noticeable improvements to adding (stereo) woofers should be in midrange coherency and overall clarity at higher SPL's.  The woofers/subs should be undetectable.

No-one wants to second guess the filters in Roger's amp, but regardless of the quality of parts, etc., they might not accommodate a seamless XO for this application, particularly if there is a boost in the LF band above the HP to the "mains" .  Unless the Omegas are weak in the 200-800Hz range (which is doubtful), you could end with some lumpy gravy.

planet10

  • Industry Participant
  • Posts: 1915
  • Frugal-phile (tm)
    • planet10-hifi
Re: Driver size and complexity of music
« Reply #11 on: 25 May 2009, 10:20 pm »

I suspect that Roger has employed something very similar to the PLLXO idea you suggested, but has simply built it into his amp just before one of the sets of binding posts.

Anything just before the binding posts is not going to be a PLLXO -- it would effectively have to be a passive loudspeaker XO.

The filtering should ideally happen between stages which means it would really need a switch.

dave

chadh

Re: Driver size and complexity of music
« Reply #12 on: 25 May 2009, 11:06 pm »

Hmmm...I was thinking about a pair of these.  But I don't think I can get them in the same finish as my Omegas.



Chad

chadh

Re: Driver size and complexity of music
« Reply #13 on: 25 May 2009, 11:26 pm »

I suspect that Roger has employed something very similar to the PLLXO idea you suggested, but has simply built it into his amp just before one of the sets of binding posts.

Anything just before the binding posts is not going to be a PLLXO -- it would effectively have to be a passive loudspeaker XO.

The filtering should ideally happen between stages which means it would really need a switch.

dave

To be honest, I don't know exactly how the amp achieves its filtering.  When I spoke, I was pretty much guessing...

I do know that Roger built a passive resistor network into the amp for me, to boost the bass output below around 400Hz.  This filtering may well happen prior to the gain stage.  He sometimes builds it into his pre-amps, sometimes into his amps.  This suggests it might sit after the gain stage in the pre-amp and before the gain stage in the amp.  It was possible to have it switchable, but I chose not to configure things this way.

Of course, this would imply that the signal that hits the other set of binding posts (which I believe has a simple high-pass filter) is also subject to the bass-boosting network.  Since I've not really contemplated the idea of using a sub before, I've never really considered the implications for this.  But it would suggest that, should I use these binding posts to hook up the Omegas, I'd be getting the bass boost in the 100-400Hz range.  I guess this is the source of Chrisby's "lumpy gravy" concern.

Focusing on the high-pass filter, though:  I'm wondering why one cares whether this filtering occurs prior to amplification (in which case it might qualify for description as a "PLLXO") or after amplification (in which case it becomes a lowly passive speaker crossover)?  I guess I should probably ask Roger this question as well.  I should understand my amplifier better.

Chad

chadh

Re: Driver size and complexity of music
« Reply #14 on: 26 May 2009, 12:19 am »

OK.  Please disregard almost everything I've said in my most recent post.  I just went and checked, and realized that almost everything I said earlier was wrong.

My MR EM7 amp has only one set of binding posts.  However, it has two sets of RCA input jacks!  One set allows me to use the passive filter that boosts the bass.  The other allows me to pass only frequencies above 100Hz. 

Hopefully that satisfies everybody's concerns:  all filtering doubtlessly occurs prior to the gain stage; the high-pass filter is almost certainly some PLLXO; and importantly there will be no bass boost network in play if I use the inputs that access the high-pass filter.

So...I should be able to use a pair of subs from my pre-amp outs and run the Omegas from 100Hz up through the amp without any serious fear of lumpy gravy, or chunky custard, or any other dismal saucing options.


Whew!

Sorry for the screw-up.

Now, if only I had a pair of nice subwoofers lying around...

Chad

nicksgem10s

Re: Driver size and complexity of music
« Reply #15 on: 26 May 2009, 12:35 am »
Hi Chad,

I don't know if you are going to try some different arrangements for subs but I would give you a recommendation to try a pair of Hawthorne Audio Augie open baffle subs.

I have used these with several speakers including when I was running full range Fostex F200A drivers last year.

These drivers combined with a Rythmik Audio A370PEQ open baffle sub amp and you will be on your way.

These Augies and this amp together may be the best value in home audio today.  The Hawthorne Audio website/forum is a great place to learn more with the nicest people around.

I love the flexibility the Augies and Rythmik provide in my system.  The Augies will keep up with your other drivers.

I use a two way speaker in a good sized room and am not satisfied with the sound unless the Augies are playing along.

Good luck on your search.

-Nick

chadh

Re: Driver size and complexity of music
« Reply #16 on: 26 May 2009, 02:04 am »

Nick,

Many thanks for the input.

I remember hearing Hawthorne drivers on an open baffle at the Audio Karma show last year.  It was such a relaxed, pleasant sounding experience - seriously, nothing like any other speakers at the show that year.  Unfortunately, a pair of 15" sub-woofer drivers on open baffles would leave me nowhere to sit, let alone walk around, in my living room.  So this is not likely to be a direction I follow in the very near future.  But maybe once I move...

Eric (aka "Launche") suggested I try the Audiokenesis "swarm" that JDubs has listed for sale.  Alas, this option also must be overlooked due to the serious space constraints in my living room.

If I come into some money, I suspect that my best bet would be to stay with Omega, and try a couple of the deep hemp cubes (which are only 13" cubes).

But we'll see.  Nothing changes very quickly in my audio world.  And most of the material I listen to really sounds lovely right now without the subs.

Chad

SET Man

Re: Driver size and complexity of music
« Reply #17 on: 26 May 2009, 02:33 am »

Hmmm...I was thinking about a pair of these.  But I don't think I can get them in the same finish as my Omegas.



Chad

Hey!

   That is a Fostex 31" super woofer I think. :D

http://www.madisound.com/catalog/index.php?cPath=45_228_311

   I've never heard or seen one but I'm sure it would be a sight to behold. :o

Take care,
Buddy :thumb:

doug s.

  • Full Member
  • Posts: 6572
  • makin' music
Re: Driver size and complexity of music
« Reply #18 on: 26 May 2009, 03:16 am »
chad, it sounds like you must have a small room.  if so, i'd strongly consider a pair of these yamaha yst-sw315 powered servo subs.  $200 each, delivered.  all you need to add is an outboard x-over.  set the subs at their highest (140hz) x-over point, and feed 'em from the x-over's low-pass outs at 50-70hz.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=400051706633

i own a pair of their predecessors, which contain two 8" drivers instead of the 315's single down-firing 10" driver.  while mine do not equal my pair of larger vmps subs, they substituted nicely, run as described above, when, due to living situation, i was unable to use the vmps.  this was in a room about 16x24.  these yamaha subs are hard to beat at anywhere the price.  in a smaller room, they may be all you need.

or you could find a pair of yst-sw1500's; but they are harder to find, cuz they have been discontinued, and there's no old stock left - they offer a bit more performance, w/12" drivers.  (i think the sw315's are also discontinued, but there's still quite a few around.)

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=120423170106

tho that seller says they listed for $3k, most sold for ~$500, in spite of what he said he paid for his...

doug s.

Docere

Re: Driver size and complexity of music
« Reply #19 on: 26 May 2009, 06:11 am »
Quote
Focusing on the high-pass filter, though:  I'm wondering why one cares whether this filtering occurs prior to amplification (in which case it might qualify for description as a "PLLXO") or after amplification (in which case it becomes a lowly passive speaker crossover)?

In short: a PLLXO (high pass), at the input of an amp, filters out the lower frequencies. The amp therefore does not need to reproduce the lower frequencies. A few random thoughts as to why this is beneficial for SET amps include: lower frequencies require considerable amplifier power to reproduce; available power is relatively limited (though this is not a problem in all cases); bass is not generally not a SET strength (distortion tends to rise as frequency falls); I do not understand the technicalities, but it may reduce amp intermodulation distortion; an amp has a relatively constant (WRT refequency) input impedance - a driver does not at the frequencies we are discussing; and a PLXXO requires smaller component values than speaker level high-pass filters...

Cheers
Raymond
« Last Edit: 26 May 2009, 08:30 am by Docere »