Originally Posted by tuxedocivic
Again, sorry for being a noob. I didn't see any reason for BSC so thanks for confirming that.
The more I think about this though, the more I think having dual opposed subs will create wave length issues when the rear sub bounces off the back wall.
For instance, if the distance from the rear sub to the rear wall and back to the front sub is say 1m and quartre wavelength of 80hz is 1m and you're sitting 2 or 4m back from the front of the sub, then you will be cancelling out part of the 80hz note, not creating MORE output. Rather decreasing output. At other frequencies however you'll reinforce the wavelength for more output. Thus you'll end up with a jaggy response at the listening positing.
This is a very astute observation. I've tried to talk about this phenomenon in the past, but it never gained traction. Nevertheless, I've experimented with the idea and found it to be one of those things that should be obvious but somehow gets overlooked, not just in this particular scenario, but in many other subwoofer designs as well.
Here's an illustration that shows why I used up/down firing dual-opposed vs front/back (or side-side, which is the same difference):
As I've covered many times over the years, phase and boundary reflection anomalies are typically mistaken for room mode problems and invariably, PEQ is used to tame them.
The problem goes to this question:
Originally Posted by pbc
While we're looking at FRs, taking a look at this ...
Ignoring the fact that I seem to have a rising response from about 28 to 48hz (which I didn't bother taming more as the two responses matched), is it just me, or does it appears my mains are almost 10db down versus my subs? Could it be the test tone in my AVR is 10db hot on the sub channel? When I get home I'll take a look at the response up to 500hz (which is where I think I stopped it) but I think it had the same issue.
Think I need to pull out my old Avia disc and recalibrate the levels. If I was in fact 10db hot, that could explain a lot of things! Thought I doubt the PB13 would have handled Pulse at +6db to reference without some serious "farting" noises that I experienced back when we blind tested the PB13 at Jakeman's place (where we hit those noises at even -10db in his 4000 treated theater.
I believe that if pbc would expand his graph to a wider bandwidth, say up to 2kHz, he will see that the sub calibration is much closer to level and that there is just a trough in response from 50-200Hz.
The trough is fairly typical IMO, and is caused by the interaction of boundaries and sub driver/mains driver placement in the room.
Originally Posted by stgdz
so what does a dual opposed setup have over a quad single woofer setup?
Seems a quad would have a better room response as its multiples.
IMO, a dual single driver system, when stacked in a corner, will usually yield a better crossover region response than a side-side or front-back dual opposed sub in the same corner, for the reason picture in the illustration I posted above.
The final tweak is to adjust the phase (or digital delay, as the case may be) to get the sub properly tuned to the mains.
In any rectangular room, placement of 4 subs, one at the center of each wall, places the subs at different distances to the LP, which I've found to be a disaster below 20Hz. I've found the harmon paper referred to as being very contrived with grossly misleading graphs and having no information or data regarding response below 20Hz.
I also have not seen any real room graphs from sub owners that directly compares different sub placements that shows the Harmon suggested placement to be the best possible placement.
Originally Posted by pbc
Lastly, any comments regarding what software I can try to run some "better" RTA graphs with at reference levels? I don't mind paying for TrueRTA if it will give me better info vs REW, but I think I'd need to then invest in a ECM mic and preamp at the least which are expensive to find up here. Or am I pretty much limited given the laptop sound card and CM140? I.e., will I clip the sound card before being able to take readings at reference?
SpectrumLab is free and I've found it to be an amazing tool. I was testing new stuff last night and I ran this comparo of the 1st chapter of star trek where the Romulan ship emerges from the black hole.
One graph is with a mic at the LP and the other graph is straight off the players SW output jack. The player has all speakers set to small so the SW out gets LFE + RB. The subwoofer system has all speakers set to small as well:
This comparison will show EVERY tiny difference. For example, in my system, I have reference response to just below 5Hz, after which it rolls off fairly quickly vs the digital copy, which shows content to 3Hz.
SL also shows the raised noise floor below 10Hz due to the recent installation of a new attic fan, which I will now be able to address instead of being oblivious to the constant infrasonic drone.
SL shows a slight dip in response at 73Hz as well as every other variation from 0dB in FR at the LP.
SL would also show any added harmonics. For example, in this scene, there is a constant ULF at around 6Hz and its 2H of around 12Hz. If the subs were adding 3HD at 18Hz, this would be painfully obvious in the graph. The same holds true throughout the subs bandwidth.
This is the best way to incrementally increase the SW output level as you repeat a scene like the one used to see where your system is weakest at what playback level. You can use other similar scenes to verify the problem and either limit playback level or upgrade.
Again, I have found no better way to 'see' what your sub system is doing with various source at the playback that suits you in your room.