Understanding subwoofers

[3ll3d00d]

Not that this noob can necessarily grok that, but could you link to that? I musta lost track and can't find it searching.

Understanding subwoofers

From my more noobish POV Why is a DSP considered absolutely necessary for proper subwoofer integration, and what specific room acoustics or phase alignment issues make it indispensable If phase issues need tweaking beyond basic inversion, or delays required to align arrival times at MLP...

www.audiosciencereview.com

Say my stack per L/R side is

it's extraordinarily complex and I imagine quite unlikely to be necessary but have fun fwiw there was a craze for MBMs on avsforum ~7-8 yrs ago so you'll find plenty of real life projects if you search over there

 

[EJ3]

I said that the subwoofers are usually delayed with respect to the main speakers. What happens when you delay them even more?

If the DIY subs (which the the pre-amp feds the crossover that separates the sub amp signals from the main amp signals to send to the propor amps, which then send their power to the respective subs and mains:
and the mains are using the floor firing (including the port) subs as their stands,
do the subs still need to be delayed?

 

[Keith_W]

If the DIY subs (which the the pre-amp feds the crossover that separates the sub amp signals from the main amp signals to send to the propor amps, which then send their power to the respective subs and mains:
and the mains are using the floor firing (including the port) subs as their stands,
do the subs still need to be delayed?

I am having difficulty understanding what you mean.

In any case, subs almost never need to be delayed, because they are ALREADY delayed. It's just the minimum-phase nature of the beast. What needs to be delayed are the mains.

 

[john61ct]

Understanding subwoofers

From my more noobish POV Why is a DSP considered absolutely necessary for proper subwoofer integration, and what specific room acoustics or phase alignment issues make it indispensable If phase issues need tweaking beyond basic inversion, or delays required to align arrival times at MLP...

www.audiosciencereview.com

Aha, I saw no indication those screenshots were wrt Overdrive10

And as I stated I do not (yet) have enough understanding yet to extract any meaning from them.

 

[3ll3d00d]

Aha, I saw no indication those screenshots were wrt Overdrive10

And as I stated I do not (yet) have enough understanding yet to extract any meaning from them.

peaks in frequency response = ringing in time
peaks in group delay = some abrupt change in frequency response is usually around (which often means ringing of some sort or a port or similar)
usual fix for such peaks = dsp

but then that can get swamped by the room anyway so back to square 1

(NB: for that particular driver, it just seems like it needs a bigger box to me)

 

[EJ3]

I am having difficulty understanding what you mean.

In any case, subs almost never need to be delayed, because they are ALREADY delayed. It's just the minimum-phase nature of the beast. What needs to be delayed are the mains.

Thank you for your response.
I'll try to clarify (hopefully) a bit: the subs are used as stands for the main left and right speakers and the subs speakers and port are directed at the floor (how the subs were designed). Each sub abuts the wall behind them and the main speakers (which have a forward firing port) are about 6" off the wall, sitting on top of the subs.
The PreAmp line out (L & R) sends the signals to the crossover (LP 55Hz, HP 70Hz), which then sends the respective signals to the sub amps and the main amps (all in the amp rack) which then sends the (via speaker wire) to the subs and the mains.
I'm curious, (in a pure analog system) if the sub and main amps are receiving the signal simultaneously from the sub/mains crossover to their respective amps (in the amp rack), then via the speaker wire) to the subs and the mains, how is it that the subs are delayed when there is no other electronic gear than what I have already mentioned, in between the preamp and the speakers/subs?
I have plans for DSP, etc. But it will be a while before I can get to that point.

 

[Flaesh]

I said that the subwoofers are usually delayed with respect to the main speakers.

I simply compensate this by delaying the mains. Right when I build the system.

What happens when you delay them even more?

Without any use of DSP. Move the subwoofer(s) to another room! And listen to what happens

 

[kyuu]

I simply compensate this by delaying the mains. Right when I build the system.

I feel like I'm going a little crazy following this exchange. Delaying the mains is exactly what Keith said one needs to do, right? But that can't be done with DSP integrated into the subwoofer, unless the signal for the mains is passing through it (which is unusual except for some nearfield monitoring systems it seems like).

 

[Flaesh]

Delaying the mains is exactly what ... one needs to do, right?

Exactly. In most cases (except for a few specific ones). An external processor for the whole system is what is appropriate.
I don't buy those mass-market boxes with plate amplifiers and obscure limiters myself, and I don't recommend them to anyone.

 

[EJ3]

Exactly. In most cases (except for a few specific ones). An external processor for the whole system is what is appropriate.
I don't buy those mass-market boxes with plate amplifiers and obscure limiters myself, and I don't recommend them to anyone.

I built my pair of subs, the bare speaker had an advertised free air FR of 20Hz-80Hz, the KAB (box,) whatever you want to call it (the speaker and port present their faces to the floor), the port tuning is 29Hz, the line out coming down from my preamp to the mains & sub crossovers is LP 55Hz for the mains and HP 70Hz for the subs.
I guess that the 15Hz overlap is "fuzzy". But my mains are measured as: (with a very small bass-resonance peak. At the system resonance of 60 Hz, the output is only about 2 dB above it's average level in the upper part of the woofer’s range, and even that minor output variation was spread over almost two octaves. When the bass curve was spliced to the room-response measurement, the resulting composite frequency response is flat within about ±2 dB from 26 to 20,000 Hz. The horizontal directivity of the tweeter was only discernible in the room measurement above 10,000 Hz.
So I do not feel that anything is missing (of course, that part is subjective and remains to one day be proven [or unproven]).
My Pre-Amp has a processor loop, which would be a great place for an external processor to be inserted.

 

[rf1938]

I understand everyone is sensitive to cost, but after reading 18 pages of this of this, I can't help but suggest that a quicker and simpler way to get good sub sound is to buy a decent sub with DSP - not a cheap one if you want it to sound good - and get one of the many options of devices that offer Dirac ART or DBLC. From the complexity of their algorithm it should become obvious that integrating subs requires a great deal of measuring and a very sophisticated algorithm to fit in a sub properly. I'm sure there are DIYers who say they have done this themselves. but I am skeptical that it's possible to get even close to what Dirac can do seamlessly in a very short time. There really isn't any simple or cheap way to duplicate this. Google the "sub crawl" as an example of a time-consuming method that may give decent results with a great deal of work, but still not approach what Dirac can do for your system.

 

[john61ct]

No, the consensus among the DSP tinkerer types is that Dirac is often relatively "meh", if you are willing to learn and spend thevtime it gets very easy to do a lot better manually.

The hundreds of dollars premium you are paying is for the shortcut, avoiding the learning curve and time "wasted" tinkering.

Which is a perfectly valid choice if you can afford it.

But for those who enjoy the process and are poor, or just prefer to spend their funds on other stuff, all the automated tools are a waste

 

[Flaesh]

buy a decent sub with DSP - not a cheap one if you want it to sound good - and get one of the many options of devices that offer Dirac ART or DBLC

Ummm. If the system has a processor and Dirac is so smart, why buy a subwoofer with DSP?

 

[john61ct]

DSP is just an enabling technology

used for many many different functions. Plain Dirac Live does DRC only.

A different system is required for crossovers, bass management.

And that still leaves dozens more DSP tool types.

 

[kyuu]

DSP is just an enabling technology

used for many many different functions. Plain Dirac Live does DRC only.

A different system is required for crossovers, bass management.

Most of the devices that one runs Dirac on (AVRs, miniDSP) can handle bass management duties.

 

[EJ3]

Most of the devices that one runs Dirac on (AVRs, miniDSP) can handle bass management duties.

But some of us are not looking for AVR's (and there now seems to be some good options that are not AVRs, such as miniDSP). 2.2 channel, here.

 

[Keith_W]

Google the "sub crawl" as an example of a time-consuming method that may give decent results with a great deal of work, but still not approach what Dirac can do for your system.

I am sorry, but this is wrong. UNLESS you have a large number of subs, you STILL need to do the sub crawl before applying DSP. No amount of DSP can fix a poorly placed sub.

 

[rf1938]

DSP isn't the same as Dirac - in the case of my two subs, its only role is to protect the drivers. Dirac and the processor do all the work. And the sub crawl is not on the menu for me. Subs too heavy, I'm too old. Tough problem solved by some very clever people in Sweden. If it was easy, Dirac would be cheap. It's not.

 

[Oddball]

Ummm. If the system has a processor and Dirac is so smart, why buy a subwoofer with DSP?

Because many great subs nowadays come only with DSP? And it certainly won't really hurt you if the processor has smart EQ system. There are some posts in the ART thread with graphs that compare pre-EQd subs and then ART, vs. non-EQd subs and then ART. The result is inconclusive as small sample, but it seems that graphs do look a bit better with pre-EQd subs. While one would not want to mess with delays or phase in advanced system as ART, the basic choice of the sub mode that DSP will follow is useful by itself. My SVS subs have 2 ported modes, Arendal has 3. It is tons of fun choosing the right one as it could take days or weeks of tweaking (being sarcastic - but still useful feature).

Another thing that seems to be misunderstood in this thread is the system that one is integrating. I can definitely understand that 2.2 system can be well integrated with tools that don't require significant cash outlay but will require knowledge and time, with the later being dependent on the level of knowledge as well as experience. But on the other hand, trying to integrate 9.4.6 system with the same tools will be a bit of a pickle. So its apples to beef type of comparison. Subs actually need to be understood with the reference to the system they are serving.

 

[EJ3]

DSP isn't the same as Dirac - in the case of my two subs, its only role is to protect the drivers. Dirac and the processor do all the work. And the sub crawl is not on the menu for me. Subs too heavy, I'm too old. Tough problem solved by some very clever people in Sweden. If it was easy, Dirac would be cheap. It's not.

Casters work for me, as does putting the sub on a 'movers platform". I have a rolling chair that the level of where one's butt is, adjusts from 14"-23". I roll the sub (s) and myself along with very little effort, no "crawling" involved.