Labhorns finshed, some initial measurements

[RobWells]

Yes I did eq the curve. The 80Hz dip is a proper null though - I tried adding 15dB gain at 80Hz and it didn't change the dip.

If you look at the 'in room' plots of the labs before eq (page 2 of this thread) then the purple line shows the dip, the other 2 don't.

The measurements were taken in 3 places across my 2 seater couch - left seat, centre, and right seat. I then took the measurements again 1 foot in front as the couch moves back and forwards a bit day to day. Thing is, if I sit in the seat with the 80Hz dip and listen to music the sound doesn't noticably change if I move 18" left to the place without a dip.

__________

10Hz..

As you know, below a certain frequency the room goes to 'pressure mode' In my case this is at ~30Hz. As room gain this low gives me approx 12dB/oct boost, and my subs roll off at 12dB / oct they cancel each other out and give me a pretty flat response below 30Hz (down to 15Hz at least - then they drop at 6dB/oct according to the plot). I've attatched a plot of my 3 tempests in room, no eq to show you where the pressure mode kicks in. Also note no room modes below that point .

At the moment I'm using 2 tempests, but am waiting on 1 to be delivered so I'll be using 4 below 30Hz eventually. I've estimated my room to give me a free 10dB at 15Hz, which can't be bad.

I'm using a dcx2496 crossover to eq the labs and the tempests. Just eq the tempests flat then add a 24dB lowpass at 30Hz,. The labs only need a 12dB high pass as they roll off naturally at 12dB/oct in room anyway.

Cheers,

Rob.

[BassAwdyO]

Wow I am suprised by the amount of gain you're getting down low. I was under the impression that most construction wasnt structurally rigid enough to support this much gain, although I've never actually measured the gain in my room. Is your room brick/concrete?

If you cant hear the dip... it cant be that bad Try some sine sweeps though and it might show up.

How low are you going with your mains? If they're reaching down past 80hz they could perhaps even out that dip a bit. At least that's what I've experienced about using mutiple units for a modal frequency with non symmetrical placement.

[paulspencer]

Rob,

Another way to tame room modes is to construct the envelope to absorb bass. Plasterboard (drywall) mounted so that it is allowed to move can apparently do this quite well - two layers mounted on furring channels. This is what Dr Earl Geddes recommends in his home theatre book.

[RobWells]

Hi Bass,

The room is single layer drywall. I know it sounds a lot of gain but thats what it measures. My spl meter gives similar readings.

I designed the mains to roll off @80Hz to go with my processors xo. I have a deq2496 on them at the moment so could boost the mains a bit to lose the dip. I want to try speaker placement / seating position first though. I think we get a bit too uptight on the graphs though. It doesn't sound like I'm missing bass moving in and out of the 'bad' area. Its just that I know its there


Paul,

I've been reading up on bass traps and room treatments a bit recently, and am looking at trying some corner traps if I can make them disappear in the room(not easy as it's my living room really). I rent my house so can't really go into more 'structural' solutions. I was going to buy the Geddes book, but have just bought the Handbook for Sound Engineers instead. Arrived today - lots of reading for the winter evenings.

Cheers,

Rob.

[paulspencer]

Rob, I've heard many criticisms of bass traps being ineffective. Perhaps you can make some panel bass absorbers which won't take up much space - portable ones

[RobWells]

Hi Paul,

Yes - I've read some stuff saying bass traps don't work at all, but other stuff saying they do well. The ones where theres a 4 / 5" bit of rigid fibreglass spanning a corner seem to be regarded better than the foam wedge type things. I guess as with all room treatments its a case of trying.

Rob.

[Tom Danley]

Hi Rob

Your 80Hz notch is (most likely) a floor to ceiling mode, it cannot be eq’d away as the more energy you put in, the greater the cancellation energy is also.
With an RTA (which is very limited resolution) it is often possible to seemingly “fix” nulls with EQ, however, this only appears to be as a result of the limited resolution.
What’s worse, attempting to fix a notch with EQ totally screws with the phase while being unable to actually fix the problem.
Peaks on the other hand or the speakers low frequency response can be fixed with EQ and in so doing, one also fixes the speaker’s phase.
Eq has an equal capacity to fix or mess up, depending what you try to do with it. Minimum phase things can be fixed in both amplitude and phase, time related issues (such as a reflected wave at 80Hz) cannot.

Given you probably have “enough” horse power with a pair of Labs, you could try changing the orientation to see if the mode can be driven less severely.
If the horns are in the corners now, try them mid wall with the mouths facing up, in other words experiment.

One I have found to be very useful and see no problem with is that you can also use a thing called reciprocity to your advantage.
Take your lab sub (or better /easier a smaller sub with flat response) and place it in the main listening position.
Then, take your measurement mic (much easier to move around than a woofer) and measure the response in the various positions you could put the sub.
Do to the transfer functions being identical (or very nearly) going either direction, one can directly snoop around for a “best spot” for the woofer, with a microphone.

One last thought, when a horn is too small, its length has a direct and powerful effect on the low corner F. Since you have these indoors and can do what ever you want, you realize it is then that adding to the path length with an external horn will lower the corner F.
An additional 2 feet of path should lower it to where the expansion takes over being the limit (this should be around 25-28Hz).
An experiment you could try would be to place one lab, against a wall and pointed into a corner. If you place the mouth about 16 - 18 inches from the corner (facing into the corner), the external shape of that air path is close enough to load the horn for a foot and a half or more. This also makes the effective mouth larger so you may get a better response shape too.
Anyway, its cool to hear about your adventures with Lab’s indoors.
Best,

Tom Danley

Danley sound Labs

[slowmotion]

Tom,

Could this 80Hz notch be "worked around" by
setting the hp crossover at that frequency ?


cheers

[RobWells]

Thanks for the pointers Tom, and thanks for the free design too!

I've a couple of weeks off over Christmas to have a play with positioning. My room is a bit small to squeeze in 2 labs, but I will be able to try the upfiring suggestion. (They'll only fit behind my screen. The room is 20' long x 10' wide x 7'7" - 9' high. Ceiling goes __/^^^^^\__ ) Screen is across the 10' wall. 2 doors and a chimney stack add to positioning probs, as does the limited positions for couches / normal TV etc... Its kind of fitting a pint in a quart pot. I'm also sqeezing 600L of sealed subs behind the screen too.

I think the main thing is to take more measurements in available listening positions first before moving the labs - it took enough effort to lift one on top of the other (I built with mdf as they're for home use)

At the moment I've got one lab on top of the other, against the back wall (floor to ceiling) with the mouth 29" from the side wall.


Cheers,

Rob.