My own "Twins"...

[Jack Gilvey]

Only fraternal, perhaps, but they do share a ~1ft^3 Vb, and are decidedly more modest than Ryan's. I wanted to try out the various Eq functions (LT-style) of my new Behringer DEQ2496 as well as experimenting with positioning of dual subs of a size amenable to moving around freely. So, I threw together a new box (the one in front) to match the Vb of an old one from scrap MDF and Baltic Birch, put in a couple dowels as cross-braces, and sprayed it black. The "old beater" is in the background, having gotten itself a fresh coat also. The drivers are Adire DPL12's, and the boxes are pretty well stuffed.
Now I just need to get my TrueRTA setup finalized...

[steve nn]

various Eq functions (LT-style) of my new Behringer DEQ2496

Hey Jack,
I'll be very interested in what you come up with as many others will be. Lets see what the DEQ2496 can actually do standing in for a LT.

[ThomasW]

Those are really cute, bet they're heavy little buggers.....:wink:

As usual under construction. Created with lots of help from Dennis H.

http://home.comcast.net/~thomasw-2/SubwooferSetupandEQpageDEQ2496-LT.html

[opt-e]

Nice, reminds me of the Sunfire True Subwoofers. :T

[Jack Gilvey]

I'll be very interested in what you come up with as many others will be. Lets see what the DEQ2496 can actually do standing in for a LT.

As it provides the same correction/extension filters as an LT circuit, it should be able to achieve comparable in-room results, which are more important than a perfect spreadsheet. As I understand LT, you're basically using one BP filter to get rid of the Qtc peak, a shelf to extend response, and another BP to "soften the knee" at the new Fc to your desired Qtc. As stated in that awesome link Thomas posted (love it):

If practice, you don't really need to worry too much about calculating Q and all that stuff. A perfect theoretical LT will work perfectly in an anechoic chamber but we use our subs in real rooms with nasty peaks and valleys, room gain, etc. We'll also be using the PEQ to correct for the room so we can forget the two parametric filters of the LT to start with. Just use the 12dB high shelving filter at your in-room F3 and cut the highs as much as you'd like to extend the bass. Once you have the extension you'd like, you can start using the band filters to flatten out the peaks and dips.

And that's the way I'll approach it, treating the final in-room FR as a whole instead of Eq-ing the sub's anechoic response (via a dedicated LT) then again flattening it once in place. The natural Qtc of the subs is ~.95, so it's a fairly high starting point.
Won't get to much dirty work until I get my laptop and USB interface, though, I don't feel like lugging this pc upstairs.

[Jack Gilvey]

My on-the-cheap painting station, complete with exhaust fan. Only somewhat effective, but beats the pollen-laden wind in the driveway (my black car has been yellow for the past month):



Even to a workhorse, one can add a bit of class :

[steve nn]

As stated in that awesome link Thomas posted (love it):

Yes, I just checked it out and it will be most valuable and shed much light on the decision process in the direction a guy goes. Nice job Thomas and Paul.

The natural Qtc of the subs is ~.95, so it's a fairly high starting point.

Perfect! This is what interests me so much about this Thread. I had planned on building a 2.1 enclosure and seeing what I could get out of it (SPL-SQ) with the BASSIS and eq, but Thankfully you've taken it on. :T

[---k---]

Jack,

Looks great. I'm very grateful that you undertook this project. I'm looking forward to hearing your results with the DEQ. I'm starting to really want a little more low end from my Twins. I think the DEQ is the way to get there.

[ThomasW]

I 'volunteered' Jack to do a step-by-step writeup of how he creates his LT. That way the rest of us can learn along with him...:wink:

[---k---]

Seconded!

I'm most interested in how much effect the shelving filter has below 20hz. I've read Thomas say that someone measured the BFD down to 5hz, depending on filter Q. I just need to see before and after graphs.

[Jack Gilvey]

Alrighty, then. Probably be about a week before I get the measuring rig going. Of course, I'll be playing with it on the IB, too.

[ThomasW]

I've read Thomas say that someone measured the BFD down to 5hz, depending on filter Q. I just need to see before and after graphs.

That was done with test equipment on a bench, not in-room measurements. Very few people are equipped to do accurate infra-sonic measurements of operating loudspeakers.

[Jack Gilvey]

Very few people are equipped to do accurate infra-sonic measurements of operating loudspeakers.

Hopefully I can get some reasonably accurate readings, at least for comparative purposes if not on an absolute basis. I've got the ECM8000 mic, just picked up the M-Audio MobilePre USB interface (bus-powered, phantom power, nice for laptop use for outdoor measuring) at GC recommended in Ilkka's TrueRTA guide , and I'll use Ed's ECM8000 correction factors from that thread, too.

[steve nn]

Hopefully I can get some reasonably accurate readings, at least for comparative purposes if not on an absolute basis. I've got the ECM8000 mic, just picked up the M-Audio MobilePre USB interface (bus-powered, phantom power, nice for laptop use for outdoor measuring) at GC recommended in Ilkka's TrueRTA guide , and I'll use Ed's ECM8000 correction factors from that thread, too.

Good! Your locked and loaded Jack.

[Exocer]

Jack, im really interested in your measurements. I'll definitely be checking back from time to time to see your progress.

[Exocer]

Any progress? :B

[Jack Gilvey]

Nahh, nut'n 'til I get a laptop set up upstairs in the HT. I have to admit, I've been more than a little distracted with getting into rudimentary personal multi-tracking using that Mobilepre and ECM 8000 mic in conjunction with Cubase SE. Here's an experiment with a G 'n R song I haven't played in a really long time...all first takes, so go easy ('specially on the vocals, I was having fun. ).

http://home.comcast.net/~tgilvey/Patience-01.mp3

Anyway, I'll get serious soon. :T

[SteveCallas]

Damn Jack, that was very good.....well aside from the final set of lyrics , but I realize his voice is near impossible to replicate when he gets in that somewhat whiny tone. But really, that was awesome, pretty spot on. I have to wonder - and be honest - at any point during the recording, did you do the Axle shuffle? :B

[Dennis H]

Nice, Jack! How many tracks?

[mazurek]

I've done fake LT circuit on my DEQ before. Its really really easy if your sub is 0.707, then you can just use one shelving filter. Where the system can run into problems for me is equalizing midwoofers and woofers, I ran into thermal compression in some cases. But if the boost stayed low, everything was ok. Also, if you want to avoid clipping, you need to attenuate the rest of your signal (my LT circuit had 11dB boost, therefore everything else gets attenuated). My main experience is that this works really well for some low freq boost, but can cause problems otherwise.

[Dennis H]

Yeah, with the DEQ, you don't wan't your total to go above 0dBFS. So, if your filters boost 10dB at some frequency (you need to add them all up), you need to cut the input at least 10dB to keep from clipping the output.

[Jack Gilvey]

Damn Jack, that was very good.....well aside from the final set of lyrics , but I realize his voice is near impossible to replicate when he gets in that somewhat whiny tone.

Heh...thanks much, Steve. I used to do a passable Axle ("Knockin' on Heaven's Door" was part of our set), but that was when I was in my last band ~15 years ago....when I was about your age.

But really, that was awesome, pretty spot on. I have to wonder - and be honest - at any point during the recording, did you do the Axle shuffle?

See previous answer.

Nice, Jack! How many tracks?

Thanks! I only used four, but a couple morphed into something else. Lead guitar became rhythm for the last part, and the whistle track was lead vocal. Background vocals were stuck in there somewhere. Makes it a little tough to mix, I'll get more professional as I play with the program.

I've not even scratched the surface of this program, yet. It's incredible what you can get these days for recording. For $200 a few companies offer not only a full USB/Firewire interface with inputs for mic/ guitar (with amp-simulator, effects, etc.)/bass/synth, but a packaged 24-48 track recording/mastering suite with full 24/96 resolution and all the effects/processing of a studio rack.

Yeah, with the DEQ, you don't wan't your total to go above 0dBFS. So, if your filters boost 10dB at some frequency (you need to add them all up), you need to cut the input at least 10dB to keep from clipping the output.

Noted.

[---k---]

Jack,
Sounded good to me. On my laptop speakers, I'm not sure I could tell what I was supposed to be listening for. Sounds like what I remember it sounding like. I may have to see if I can find the original somewhere to compare.

[---k---]

Yeah, with the DEQ, you don't wan't your total to go above 0dBFS. So, if your filters boost 10dB at some frequency (you need to add them all up), you need to cut the input at least 10dB to keep from clipping the output.

Really? Is it that easy? I figured the Q would have an effect also. So, I would have to sum the area under the curve to equal 0. I've thought about it in the past, but I've never worried about it -- could explain some things.

[Jack Gilvey]

Sounded good to me. On my laptop speakers, I'm not sure I could tell what I was supposed to be listening for.

Nothing in particular, just posted it since I did it using the same mic and pre I'll use for room measurements.

Sounds like what I remember it sounding like. I may have to see if I can find the original somewhere to compare.

I don't think it'd hold up too well...just take it for what it is.

[---k---]

Jack,

Have you applied a shelving filter yet? I see a DEQ on everyone's faviorite auction site for a decent price ending tonight. I'm tempted. I would really like to see how it preforms at less than 20hz before spending more money. I told the wife the other day that I was "done" for a while. oooops.

[Jack Gilvey]

Not yet, sorry...no computer on that floor. I was hoping to avoid this, but later in the week I'm going to move this whole shindig (such a friggin' nest) upstairs and set up for a few hours of tests. This'll be just to gauge the effect of the shelving filters and stuff.

[---k---]

Well, that is just not good enough!!!



I went ahead an ordered one with the Behringer mic. I got next week off, so hopefully it will get here soon and I'll be able to play.

[Jack Gilvey]

Cool...we'll compare notes. I'm thinking it'll be easier just to bring a sub/amp down here and just test the nearfied FR of the effect of various filters. I promised to paint the nursery today, but look for some prelim stuff tomorrow.

[---k---]

Sounds like a plan. Let me know of any tips/mistakes you come across.

[Hdale85]

Nursery huh? Baby on the way or already there?

[Jack Gilvey]

Yeah, August...I'm covered in lavender.

I did take a quick look (sub sitting in the middle of the room with the mic on the floor in front 6" away and an old plate amp running it) at the effect of a high-shelving filter doing ~15dB cut set at ~40Hz. Adding a boost got tricky, as I kept running into clipping/distortion, but cutting the highs worked ok (as pointed out in Thomas' FAQ/Dennis' LT writeup). Room gain from proper corner placement might flatten that, I'll check it upstairs at a later date.

[---k---]

I got my DEQ today. After running to the local music store and getting robbed for the XLR cables to hook it up, I got it up and running. It is quiet the swiss army knife. I found it pretty easy to setup a shelving filter. It seemed to have the effect I was looking for. I'll have to break out RoomEQWiz tomorrow.

But, there is so much I haven't figured out yet. I guess I'm going to have to read the manual.

[Jack Gilvey]

Yeah, it's a full-featured unit. I ran a sweep on the L12dB filter, the low-shelving. This is the filter set for 12dB of gain at 30Hz:

[Jack Gilvey]

I amended the above post. The original graph was inaccurate as it seems my TrueRTA Sound System calibration using the 1/8" jacks of my MobilePre resulted in a boosted response down low, confirmed with a quicksweep. I checked it this morning using the 1/4" instead and it calibrated correctly, so that FR should be correct for L12dB with 12dB of gain at 30.3 Hz.

[---k---]

Jack,
You're too slow!!! :P

I'm in the middle of playing with my DEQ, I'll try and upload some photos of the setup, camera shots of the DEQ in action, and other stuff later today or tomorrow to maybe help get started on the guide.

Anyway, just some real quick results:


Red is the base line, taken at my seat with no correction.
Green is the result of the DEQ Auto EQ feature. It applied a +7 boost at 20hz, then some cuts; 25hz(-5), 31(-3), 50(-13), 639(-13), 80(-13).
Blue is the result of a single +12@44hz shelving filter.

I think I need to dial the boost back a bit and apply some cuts - then listen of course. But, the DEQ is a fun unit once you get the hang of it.

Since the AutoEQ function manipulates only the Graphical EQ, and not the Parametric EQ, I'm not sure how good it is.

[Jack Gilvey]

Jack,
You're too slow!!!

Story of my life. Hopefully the laptop shows up one of these days so I can play. I shall eagerly follow your progress in the meantime. Just playing with it and comparing the curves to what TrueRTA reports, it does seem accurate as far as what it says it's doing correlating to what actually happens to the signal. If you get some clipping in the unit, try rolling off with the high shelf instead.

Since the AutoEQ function manipulates only the Graphical EQ, and not the Parametric EQ, I'm not sure how good it is.

Not much, and it's apparently something of a crapshoot down low. It did well with a full-range signal, but I'd stick with manual PEQ for below 100Hz.

[---k---]

Well, a bit more playing. Graphs are starting to look pretty good (as long as I keep my head in a vice )

Lots of Shelving filters. Everyone: I wouldn't call myself an expert on all this, so if I've done something wronge, please don't hesitate to tell me a better way. Also, don't follow my example until Thomas says so.



I got the following filters:
L12db +12 @44.8 (roughly fs of my box)
H6db -15 @55.7
L6db +3 @ 22.4

Red = Base Line
Green = L12db +12 @44.8 only (top line)
Blue = H6db -15 @55.7 (bottom line)
Magenta = L6db +3 @ 22.4 (middle)
Teal = All three (flattest line)

Now, I need to do some listening.

[Dennis H]

L12db +12 @44.8 (roughly fs of my box)

I think you'd get better results with your first filter if you used a lower frequency, say 20 Hz to start. Using a high cut, you can set it at Fb. Using a low boost, you need to set it lower, like an octave or so. I marked up Jack's pic to show how it works. Both settings would give more or less the same shape curve.

[Dennis H]

Okay, here's how I'd approach it, starting with your red line. Use a shelving filter to flatten it out below 30 Hz -- the old LT thing. It's really good from the high 20s up to 45 or so, so don't touch it there. Then use parametric band filters to pull down the room peaks. You should be able to get it pretty flat up to 100 or so instead of rolling off sharply above 70 like it does now. I'm assuming you are bypassing the receiver's crossover for now?

[Jack Gilvey]

I think you'd get better results with your first filter if you used a lower frequency, say 20 Hz to start. Using a high cut, you can set it at Fb. Using a low boost, you need to set it lower, like an octave or so.

As you see in my graph, setting the 12dB Low Shelf to 30Hz means the chosen boost is already fully present at 30Hz, and begins ramping up much higher.
With the high cut, you also gain some headroom in the unit itself, correct?

Well, a bit more playing. Graphs are starting to look pretty good (as long as I keep my head in a vice )

That's why room treatments are by far more effective than even the best eq.

[---k---]

Yeah, I just spent some time doing some listening and staring at my graphs. I came to the same conclusion you did: Why the heck am I using that big cut? and I should be using a parametic to fix the room at 60 -70, and not the selving.

Upon listening, it was clear that I had more low end. The selving filter in the DEQ definitely extends down to 10hz Movies were much more impressive. (Except for Black Hawk Down - I saw the cone moving like crazy during the Irene scene, but didn't feel or hear much. Hummmm. I'll have to invite myself over to someone's house with a sub that does that scene well Chris? )

But, with music, the kick drum sounded a little flatter. I listened to some Blue Man Group and Bella Fleck. I pulled the cut back, and got Planet Drum in now, it sounds better.

I was going through the receiver with the crossover enabled. I've done it both ways in the past, but this makes more sense to me. Do you have reasoning to do it the other way? If I don't use the receiver's cross over, those spikes you see at 90 and 100 are HUGE.

Thanks for your suggestions Dennis. My room is very ugly. So, the parametics for the room crap is tricky. I'll play more tomorrow.

[Dennis H]

As you see in my graph, setting the 12dB Low Shelf to 30Hz means the chosen boost is already fully present at 30Hz, and begins ramping up much higher.
With the high cut, you also gain some headroom in the unit itself, correct?

Theoretically, it's 3dB short of full boost or cut at Fc. It does look like it overshoots 1/2dB or so down below 20Hz.

Headroom should be about the same either way with an analog signal. There might be an advantage running the ADC hotter but I wouldn't sweat it for a sub. Just don't turn the preamp up so loud it clips the EQ box anywhere -- ADC, DSP, DAC. With a digital signal, the input is basically at max volume all the time so you need to be more careful to keep it below 0dBFS internally to avoid clipping the DACs.

[---k---]

That's why room treatments are by far more effective than even the best eq.

Got some. On my list of projects for this week is to build a couple more. Biggest problems is that I only have one usuable corner. The rest have doors.

[Dennis H]

I was going through the receiver with the crossover enabled. I've done it both ways in the past, but this makes more sense to me. Do you have reasoning to do it the other way? If I don't use the receiver's cross over, those spikes you see at 90 and 100 are HUGE.

Ah, okay that's fine. In that case you do want it to roll off above the XO freq so maybe your shelving filters are doing just what they should.