Double Bass One for the speaker experts: Hi-Q drivers, small boxes, DSP-EQ & phase response

[swingingoodtime]

Hi all,

I'm looking at building a PA-sub that will double suitably as a DB amp; the electronics will include a Behringer DEQ2496 digital EQ, so I'll be able to dial in some fine-granularity EQ & compression settings that suit the bass to boot (ain't technology great?).

As always, small cabinet size & weight is desirable. I'm happy to go a sealed cabinet as it won't really need to go down flat to 40Hz; more like 60-70Hz, and the transient capability of a sealed cabinet is better than that of a ported box. I can therefore consider drivers with a slightly higher Q (ie, 0.5 - 0.8) but in my research I've come across the occasional driver with a very high Q - up to 2 (!) which I’m assuming are dedicated ‘infinite baffle’ drivers.

Although I'd likely not use one of these, the speaker-geek in me got to thinking in general: in a small enclosure (ie with a high enclosure Q) these high-Q drivers would therefore peak quite high as expected (generally around Fo), but this can be flattened out using DSP-EQ to produce a reasonably favourable curve with the roll-off around 70Hz (not a great stretch for many drivers) and this would be far more desirable than boosting.

However, in terms of phase response: given the high Q of the box this would be less than ideal to begin with, but would flattening the response improve the phase response any? Or would it stay the same (or even get worse?)

Thoughts, anyone?

Thanks heaps!

Paul

 

[drurb]

Why are you concerned about the phase response at all in this application?

 

[anonymous02282011]

" PA-sub that will double suitably as a DB amp"

seems to be a contradiction there because generally PA-subs deliver the really low stuff and when amplifying the DB those frequencies are generally avoided because they cause feedback and make the DB sound boomy and muddy.

Not saying it can't be done; just something to consider as you think through this.

 

[swingingoodtime]

Why are you concerned about the phase response at all in this application?

The relevance is in the intended design: the 'sub' part of the sat/sub combination/PA-application is supposed to function as a standalone DB amp/cab when the sats are removed. I can set this up using the Behringer & a switchable electronic XO (or better still, go an all in one dbx Driverack box which has the DSP-EQ & XO all-in one). The idea is that I can create a (recallable) EQ/XO profile in the Beh/dbx that suits the sat/sub combination and another one that suits just the 'sub' on its own for the DB amp application, ie:

The sat/sub (PA) profile would roll the sub off from 200Hz down (ish), with the highs being xo'd from 200Hz up for the satellites.

I could still run a DB through this, but if I need just a DB amp, I'll leave the Sats at home & dial in the DB amp profile, which would have just one bandpass filter (eg, 70Hz - 8000(?)Hz) applied to the sub. The phase-response concerns would be relevant here as the sub would now be handling more than just lo-bass and phase/time smear would impact transients & clarity; I hate hearing the upper-mids kick in some time after the low end as it throws the rhythm into chaos (a bad room can produce this effect too, I've found).

Wondering why I'm bothering with all of this? I primarily sing & am looking to get my PA gear down to as little & portable as possible; a PA sub/DB amp combo would eliminate one box around the house. The L1 approach (ie, line array satellites) removes the need to lug speakers AND stands (not to mention a fair few cables if I locate the sats on top of the sub cabinet). The DSP-EQ will provide a much finer level of EQ control than a std amp (both the Beh & dbx have General EQ, Dynamic EQ, Parametric EQ, compander & limiter functions on-board) for both DB & PA applications.

Being the singer, I tend to lug far more gear than anyone else; the entire PA probably weighs about 80-90 kg over about 6-7 loads (FOH, foldback, cables, stands, blah blah blah....) and the recent NYE gig with the 50 stairs up to the venue (no elevator!!!) was just a killer . The running joke is that I'm looking forward to doing just DB so I can actually be more portable heading to a gig...

" PA-sub that will double suitably as a DB amp"

seems to be a contradiction there because generally PA-subs deliver the really low stuff and when amplifying the DB those frequencies are generally avoided because they cause feedback and make the DB sound boomy and muddy.

Not saying it can't be done; just something to consider as you think through this.

Yeah, I've thought about that a fair bit. As I'm custom building this to my specific needs (which generally never needs uber-lo bass; mainly vocals and keys), I'm deliberately not making it a true 'sub'; it'll only go down to about 60-70Hz at most. If I ever use a BG'ist, he (or she) will probably bring their own amp, so the system won't need to deal with those frequencies. A keyboard might, but not thunderously or frequently so, and most rooms will reinforce the lo-end some what anyway. And if I ever get good enough to sing & DB at the same time, at least I could use the PA setup knowing it'll still sound ok for the DB. But thanks for the heads-up.

Phew! After all of that; any thoughts on the phase-reponse issue, please...?

 

[Mike Arnopol]

Problem is that the High Q drivers that are suitable for pro sound generally have limited high end. (above 1K) If you're not concerned about uber-bass, the prosound drivers that are good for DB and EB are lower QTS and designed for reflex. Have you gone through the US speaker catalog? I've talked to a guy there who was pretty knowledgeable. They have a huge catalog. I'm not a reflex fan, but if you tune the cab low enough the range above 60hz will all be coming from the speaker, not the port. As long as you have EQ, who cares if you have to bump it up a bit below 100 hz?

 

[Happy Steve]

I just mentioned in the speaker size thread about my Bose 802 speakers, and how the 802C controller has 13dB boost at around 60Hz. If it is good enough for Bose...

Interestingly too in the context of this thread, I have a Behringer DSP1124P "feedback destroyer" which I have used 1 of the presets in '12 channel parametric' mode to emulate the EQ of the Bose controller box very successfully...

And sorry, but I have no idea about the phase response.

Steve

 

[swingingoodtime]

Problem is that the High Q drivers that are suitable for pro sound generally have limited high end. (above 1K) If you're not concerned about uber-bass, the prosound drivers that are good for DB and EB are lower QTS and designed for reflex. Have you gone through the US speaker catalog? I've talked to a guy there who was pretty knowledgeable. They have a huge catalog. I'm not a reflex fan, but if you tune the cab low enough the range above 60hz will all be coming from the speaker, not the port. As long as you have EQ, who cares if you have to bump it up a bit below 100 hz?

I'm not so much of a reflex fan myself; has it's uses, but for DB it seems pointless; no need to drive deep & not as good for transients. I've a few low-Q options but am not considering them for this reason; would rather go sealed cab.

As mentioned in my OP, this was more of a general extension of thought than a definite intention to use mega-high-Q drivers per se; having said that, the knowledge gathered in these discussions may well also be translated into the design of the satellite columns too; hi-sensitivity/hi-Q drivers are far more common in the smaller sizes (ie, 4") that I'm considering for that application too (I think they're often ceiling speakers).

Re EQ; I'm particularly interest in the concept of EQ'ing down rather than up, as this avoids excursion issues. With the right driver, the natural roll-off, even in a hi-Q box, may fit neatly into the 60-70Hz range, so no need to boost the lower end at all. Still concerned about the phase impacts, though.

I have only one thing to say re the US Speaker Catalog (& Parts-Express, etc): Damn the Tyranny of Distance!

(Shipping charges, mutter, grumble, grumble...!)

 

[swingingoodtime]

I just mentioned in the speaker size thread about my Bose 802 speakers, and how the 802C controller has 13dB boost at around 60Hz. If it is good enough for Bose...

Yikes; what's the excursion like below this frequency under heavy drive conditions?

Interestingly too in the context of this thread, I have a Behringer DSP1124P "feedback destroyer" which I have used 1 of the presets in '12 channel parametric' mode to emulate the EQ of the Bose controller box very successfully...

Hah; cool. Clever.

And sorry, but I have no idea about the phase response.

Steve

No worries; thanks for the input!

 

[jon3673]

Paul,
When I first started playing UB, I found that the best sound I could get was by using a pair of stacked Bose 802 speakers driven with a Walter Woods amp. I even got great response in a small room by using just one Bose 802. My son also set this up for his EB in a rock band and absolutely loved them.
I walked away from them because I hated to move around that much equipment, plus my UB. However, one of the better UB players in this town still swears by this arrangement.
I recently heard another UB player using the new Bose L1 system e/w 2 bass sub speakers. He feels that it is a better response that I used to get using the Bose 802s.
I also now own a Bose L1 system that I was using for a PA on a trio gig. After hearing my friend play thru his system, I tried using it for my UB. Unfortunately, I did not get a good response until I added the 2nd woofer. I now agree that it is the best thing that I’ve ever used for amplification of an UB. But I hate to carry around this much equipment! Especially when I’m just playing in a small group!
Therefore, I normally just use my AI Contra. Infact, I’ve even stopped playing thru the Bose when I use it for a PA, because the AI is about the same size and weight as the extra Bose bass speaker that is needed, and this way my sound is completely separated. This feels better for me.
Bose does make a speaker called the Panaray MB4 Subwoofer. I ran across it when I was thinking about adding the second bass speaker to my L1 system. It appears physically to be two L1 bass speakers mounted into one cabinet. However, when I attempted to use one with my Bose L1 system, instead of the 2 sub woofers, I found that it is not compatible with that system.
I’ve seen some of them on the used market going for around $300. You may want to investigate this possibility.
Good luck on your search!
Jon

 

[drurb]

...The phase-response concerns would be relevant here as the sub would now be handling more than just lo-bass and phase/time smear would impact transients & clarity.

Considering what you are amplifying and reproducing, from the point of view of physics and psychoacoustics, this simply is not true. Want to prove it to yourself? Take long samples of the sounds you wish to reproduce and make versions of them with various non-linear phase shifts intentionally applied. I'll bet you a cookie you can't reliably hear the difference. I'm cheating though, because I'm familiar with published studies on the subject*​. Suffice it to say that the acoustic conditions under which you could not only discern the difference but, more relevantly, state a preference are not even close to met by your conditions. In other words, forget the phase response for what you're doing.

*​Please don't ask me to list them-- they are too numerous and can be found in JASA and JAES.

 

[drurb]

I just mentioned in the speaker size thread about my Bose 802 speakers, and how the 802C controller has 13dB boost at around 60Hz. If it is good enough for Bose...

Here's why it's not always a good idea, especially for a ported design:

...I'm particularly interest in the concept of EQ'ing down rather than up, as this avoids excursion issues.

 

[swingingoodtime]

In other words, forget the phase response for what you're doing.

Hey, cool, if I'm wrong & it's one less thing to think about, then great! Thanks heaps for sharing your knowledge.

Out of interest, though, would this still apply for the upper (non-DB realm) frequencies? ie, for the PA satellites? I'm wondering if the shorter frequency wavelengths are more susceptible to phase issues, or whether in a PA setting/live sound environment it's largely irrelevant... (I'll admit to be thinking from a hi-fi design perspective, where subtle detail is more easily detectable than at a gig).

 

[swingingoodtime]

Paul,
When I first started playing UB, I found that the best sound I could get was by using a pair of stacked Bose 802 speakers driven with a Walter Woods amp. I even got great response in a small room by using just one Bose 802. My son also set this up for his EB in a rock band and absolutely loved them.
I walked away from them because I hated to move around that much equipment, plus my UB. However, one of the better UB players in this town still swears by this arrangement.
I recently heard another UB player using the new Bose L1 system e/w 2 bass sub speakers. He feels that it is a better response that I used to get using the Bose 802s.
I also now own a Bose L1 system that I was using for a PA on a trio gig. After hearing my friend play thru his system, I tried using it for my UB. Unfortunately, I did not get a good response until I added the 2nd woofer. I now agree that it is the best thing that I’ve ever used for amplification of an UB. But I hate to carry around this much equipment! Especially when I’m just playing in a small group!
Therefore, I normally just use my AI Contra. Infact, I’ve even stopped playing thru the Bose when I use it for a PA, because the AI is about the same size and weight as the extra Bose bass speaker that is needed, and this way my sound is completely separated. This feels better for me.
Bose does make a speaker called the Panaray MB4 Subwoofer. I ran across it when I was thinking about adding the second bass speaker to my L1 system. It appears physically to be two L1 bass speakers mounted into one cabinet. However, when I attempted to use one with my Bose L1 system, instead of the 2 sub woofers, I found that it is not compatible with that system.
I’ve seen some of them on the used market going for around $300. You may want to investigate this possibility.
Good luck on your search!
Jon

Hi Jon,

Thanks for that (although why does it not surprise me that the compatibility issues would push a consumer to have to spend more...?)

That's great feedback on the L1's sound in this application; I'll try to design the sub with plenty of SPL headroom.

Cheers,

Paul

 

[drurb]

Hey, cool, if I'm wrong & it's one less thing to think about, then great! Thanks heaps for sharing your knowledge.

Out of interest, though, would this still apply for the upper (non-DB realm) frequencies? ie, for the PA satellites? I'm wondering if the shorter frequency wavelengths are more susceptible to phase issues, or whether in a PA setting/live sound environment it's largely irrelevant... (I'll admit to be thinking from a hi-fi design perspective, where subtle detail is more easily detectable than at a gig).

Indeed, it's another matter when we are talking about broadband signals (in English, signals that have a broad range of frequencies) such as music. I also tend to think from a hi-fi design perspective. What you are asking really comes down to what is technically referred to as sensitivity to "group delay." It's one thing, of course, to ask what are the conditions under which people can detect differences in controlled conditions. It's quite another to ask if the differences can be detected or appreciated in uncontrolled conditions-- for example, listening to running music in your friend's house. Finally, and the condition that really counts here, is whether even if you can detect a difference, you can reliably state a preference. Now, I must admit that I'd have to go back to the literature to sensibly talk about the precise numbers and conditions but, frankly, I wouldn't worry much at all about the PA satellites in a real-world gig situation. One reason for not doing so is that you'd be hard pressed to control what you need to control anyway. Imagine that you went out of your way to "time align" the PA satellites. Now, someone moves them a few feet or someone applies a bit of EQ here and there using a standard mixing board with electronic filters. Bang! Out the door goes all your hard work re phase-response, yet the audience is still having a great time.

P.S. Would you be kind enough to fill out your profile? I like to get to know folks here and appreciate reading where they're from, what equipment/instruments they use, what experience they have, and what types of music they play. Thanks.

 

[fdeck]

I'm usually more worried about time-aligning the drummer.

 

[swingingoodtime]

Indeed, it's another matter when we are talking about broadband signals (in English, signals that have a broad range of frequencies) such as music. I also tend to think from a hi-fi design perspective. What you are asking really comes down to what is technically referred to as sensitivity to "group delay." It's one thing, of course, to ask what are the conditions under which people can detect differences in controlled conditions. It's quite another to ask if the differences can be detected or appreciated in uncontrolled conditions-- for example, listening to running music in your friend's house. Finally, and the condition that really counts here, is whether even if you can detect a difference, you can reliably state a preference. Now, I must admit that I'd have to go back to the literature to sensibly talk about the precise numbers and conditions but, frankly, I wouldn't worry much at all about the PA satellites in a real-world gig situation. One reason for not doing so is that you'd be hard pressed to control what you need to control anyway. Imagine that you went out of your way to "time align" the PA satellites. Now, someone moves them a few feet or someone applies a bit of EQ here and there using a standard mixing board with electronic filters. Bang! Out the door goes all your hard work re phase-response, yet the audience is still having a great time.

P.S. Would you be kind enough to fill out your profile? I like to get to know folks here and appreciate reading where they're from, what equipment/instruments they use, what experience they have, and what types of music they play. Thanks.

Cool, thanks for that. I'll push on with the project (slowly! The usual obstacles: time & $$ ); my main concern now is the generally low power handling of the satellite drivers; anywhere between 15-30W max each. Although I'm looking at a stick of 16 drivers which should yield about 103-105dB/W/m sensitivity in total (driving 8 drivers in parallel/series config per channel), the limited max power rating doesn't leave a lot of room to move from there (ie, dealing with peaks) compared to a good pair of, say, 8" drivers which would be far more robust (and have comparable sensitivity;a pair of Beta 8A's, each driver driven per channel, would produce 98dB/W/m. 4W would bring this up to 104dB, and the Beta's have a power handling of 225/450W (RMS/Max), so plenty of room to move from there. Not that I'd ever use all of that, but I'd hate for a peak to blow the smaller drivers. Fortunately, the limiter on the Behringer may provide some protection on this front, but still...) *sigh* swings & roundabouts: size & portability vs cost & power handling...

I've updated my profile now; sorry about that. It's been on the list of things to do for ages now...

 

[swingingoodtime]

I'm usually more worried about time-aligning the drummer.

LOL!