To say that a BSC circuit adds group delay and is therefore bad is simply wrong, because the final response is what matters and the flatness of frequency response and the group delay are both better with the BSC. Call it nitpicking if you like, I call it correcting an error or misunderstanding.
I'm well aware that the sharper cutoff of a maximally flat bass reflex has a lot higher group delay than a closed box. Basic filter theory. (Nobody says you must use a maximally flat response by the way, an early gradual rolloff alignment has much less group delay and still has many of the benefits of a bass reflex like reduced excursion)
However it's pointless to say that the group delay is higher if you don't know what the thresholds of audibility are. The reality is that the thresholds of audibility for group delay at low frequencies <~50Hz are VERY high. It takes a LOT of group delay to even be audible at all, (many 10's of milliseconds, and increasing rapidly as frequency goes down) likewise a high Q resonance at low bass frequencies is virtually inaudible on normal music content, (provided it doesn't introduce a peak in the amplitude response) particularly if the bass notes have overtones, as its the harmonic overtones which provide the "speed" and definition of the note.
If you don't believe me, have a look at the following research paper, which I did manage to find after all:
http://www.genelec.com/documents/publications/aes116th_2.pdf
Here is the original thread where it was discussed:
http://www.diyaudio.com/forums/multi-way/172806-flat-not-correct-stereo-system-35.html#post2570014
Bottom line is that significant group delay and time domain ringing of low frequency resonant poles are both virtually inaudible and don't contribute to "slow" bass.
All speakers are a set of balanced compromises, if obtaining a lower group delay means that you must sacrifice other important performance criteria, you had better be pretty sure that what you're trying to optimize is even audible in the first place...
Last edited:
"Slow bass" and "lack of punch" of BR comes IME mostly from two factors:
1) Too high tuning freq > 30Hz
2) Driver and port running on the edge (compression, chuffing etc)
3) Front ports which leak MF (in 2-ways)
All of this often comes together in smaller floorstanders and bookshelf speakers.
A well executed and properly tuned BR with a strong driver and rear port does have zero issues with that. It tends to be a big box, though. BR must be big to work properly.
I did a lot of experiments with independantly adjustable magnitude and phase response (via FIR-filters), BR's with 50Hz tunings or so did improve with phase correction.
1) Too high tuning freq > 30Hz
2) Driver and port running on the edge (compression, chuffing etc)
3) Front ports which leak MF (in 2-ways)
All of this often comes together in smaller floorstanders and bookshelf speakers.
A well executed and properly tuned BR with a strong driver and rear port does have zero issues with that. It tends to be a big box, though. BR must be big to work properly.
I did a lot of experiments with independantly adjustable magnitude and phase response (via FIR-filters), BR's with 50Hz tunings or so did improve with phase correction.
Last edited:
The audibility of group delay / phase in the bass region is effectively the same topic as absolute polarity because the ear is sensitive to the actual waveform.
Secondly, a rather long group delay in the bass gives less masking of higher freq content (of, say, a kick drum signal), which is perceived less punchy / sharper even when freq response is exactly identical.
Secondly, a rather long group delay in the bass gives less masking of higher freq content (of, say, a kick drum signal), which is perceived less punchy / sharper even when freq response is exactly identical.
just to point out, i love closed box speakers, and some of the best designs came from the 70s, like the custom bbc monitors 'moonlit' by my late father during his time at Goodmans in portsmouth during that era. Sadly i didnt get them in the will. Grills made from seives and T27 with ally dome pressed from a pie dish. Mint. But with my living space, somehow a reflex works despite the L shaped listening room. My main speakers are a total sham at the moment, major tinkering, unrelenting tinkering. The closed box al130 for example SHOULD be reasonable, but it doesnt have weight, and the smaller box actually honks up the midrange, where the larger reflex doesnt. Another reason im going off narrow but deep enclosures.
Great link!
GREAT LINK! Good to see an old Genelec doc, im may have to use my Imagination At Work tomorrow and print that!
GREAT LINK! Good to see an old Genelec doc, im may have to use my Imagination At Work tomorrow and print that!
just a short note
though I happen to like closed design, a vented BR design does not have to result in sharp rolloff
with the right choice of woofer you can calculate it to have round and smooth rolloff
ofcourse less deep, yes
though I happen to like closed design, a vented BR design does not have to result in sharp rolloff
with the right choice of woofer you can calculate it to have round and smooth rolloff
ofcourse less deep, yes
Ah yes, the old fast/slow. One thing I have looked for is easy of integration. I guess I find low Q sealed boxes easier to integrate and thus a better overall system result. I have only built subs for music, not HT ultra low gut wrenching special effects.
Speaking of crossover slopes, I tend to use 4th order on my subs as I seem to be too subjective in identifying sub location which I find I can mitigate with steep crossovers. If the subs are being used as speaker stands, then I find second can work just fine so long as you don't use a metal cone with breakup problems. ( Hint, the paper Peerless work just fine, thank you)
Speaking of crossover slopes, I tend to use 4th order on my subs as I seem to be too subjective in identifying sub location which I find I can mitigate with steep crossovers. If the subs are being used as speaker stands, then I find second can work just fine so long as you don't use a metal cone with breakup problems. ( Hint, the paper Peerless work just fine, thank you)
err No ...🙄
As suspected , you read such in a book...... 🙂
Any BR speaker sounding slower than sealed is poorly designed . correct box volume, port area ( rear firing ) and low cut off will produce far superior bass response than a sealed Box...
PS: Thats a poorly designed BR graph posted ....
+10 and bingo .....
Last edited:
I asked a similar question in another thread a while ago: Whats the group delay effect of a TQWT? id imagine, using 'some' BAF to damp the cavity means that the response is somewhere between closed and reflex.
Hi Wayne. Why is it so important for BR to be on the back side of the box ? Leakage ?
I've put mine in front because rear firing BR has much more problems with room setup. Box requires too much space behind it to work properly without booming and by the time i make them sound just right they are in my face 😀
Last edited:
If you have pipe resonances you can't solve and which are audible, it is best to have the port on the rear. For maximizing bass output, it is best to have it on the front, close to the bass driver.
By flaring a port and toying around with diameter/length, it is usually possible to get very little pipe resonance, so I build mine with the port firing to the front.
vac
Me to. I have flared ports.
I've stuffed them (ports) in trial box with polyfill to try to hear the difference and if there is any audible distortion. I just ran out of bass 🙂
I've stuffed them (ports) in trial box with polyfill to try to hear the difference and if there is any audible distortion. I just ran out of bass 🙂
Last edited:
Very nice to see the forum warm to a subject. Closed Box versus Reflex is clearly a good topic. 🙂
I've fiddled about with all the reflex designs. Port on front and back. Different tunings. For sure reflex has an impressive sound. It's like it adds a natural reverb and big acoustic if you know what those popular studio devices do.
The popularity of reflex has a lot to do with getting a lot of bass out of a smallish box cheaply IMO. Closed box is a different animal which uses different driver Qts and works best with big boxes. Acoustic Suspension, which is closed box+, is capable of some stunning bass with powerful magnets too.
Putting a sock in a reflex is not quite doing what we want to turn it into closed box, so it's not a totally valid experiment. But it's my observation that closed box produces a more accurate bass where you can follow the bass guitarist's work better, even if it falls away faster. Very retro, I know. We'll probably have to agree to disagree here. 🙂
I've fiddled about with all the reflex designs. Port on front and back. Different tunings. For sure reflex has an impressive sound. It's like it adds a natural reverb and big acoustic if you know what those popular studio devices do.
The popularity of reflex has a lot to do with getting a lot of bass out of a smallish box cheaply IMO. Closed box is a different animal which uses different driver Qts and works best with big boxes. Acoustic Suspension, which is closed box+, is capable of some stunning bass with powerful magnets too.
Putting a sock in a reflex is not quite doing what we want to turn it into closed box, so it's not a totally valid experiment. But it's my observation that closed box produces a more accurate bass where you can follow the bass guitarist's work better, even if it falls away faster. Very retro, I know. We'll probably have to agree to disagree here. 🙂
I didn't want to turn it into closed box. I just wanted to see how much port resonances (if they are audible in this particular case - 114L box tuned to 38Hz with 4" port) influence overhaul sound of box. When port is stuffed, port resonances should be minimised (not claiming, just guessing and operating on a wild guess 🙂 )
Last edited:
If its a 2 way I'd say it's close to impossible to eliminate port resonances in the midrange to the point where they're inaudible. Not just that but general leakage of standing wave cavity resonances as well. Just close mic a port to see how much higher frequency crud escapes despite the port theoretically forming a band pass filter...putting the port on the back makes both the pipe resonances and the leakage from inside the box inaudible with no extra effort.
Another advantage of a rear mounted port is less than the theoretical 24dB/oct rolloff slope below box tuning due to incomplete cancellation of the out of phase woofer and port output. If the port is on the front you have a short acoustic short circuit path, if it's on the back it becomes a dipole below the box tuning frequency.
I've measured increases in bass response on the order of 2-3dB a 1/3rd octave or so below the box tuning by having the port on the back instead of the front...(on a box with a moderately large baffle)
Last edited:
@simon ,
Very true sir , very similar to having open baffle without the negatives ...
Sounds like a bigger room problem and not speakers ..🙂
The only time forward ports are necessary is for near field monitoring , apart rear is superior in every respect ....
Very true sir , very similar to having open baffle without the negatives ...
Sounds like a bigger room problem and not speakers ..🙂
The only time forward ports are necessary is for near field monitoring , apart rear is superior in every respect ....
I witnessed this quality in a reflex most in a design with 1/4" panels and close to zero damping. Currently, with room augmentation (subs) I can switch my mains to vented and back without such issues.
- Status
- Not open for further replies.