Sealed TC-3000, EP2500, BFD, Project Start

[aktiondan]

I wanted to get some feedback before I delve into sub project for a friend from work. Here's the gist of it:

Sub: 15" TC-3000 Subwoofer
Box: 90-100L Sealed
Amp: Behringer EP2500
EQ: Feedback Destroyer Pro DSP1124P

The sub has been purchased and has arrived. We're planning on picking up the amp/EQ before the weekend along with some cables, speaker wire and 1/4"-RCA adapters. The project started out as a ported box with an F3 of 21Hz, that netted 6 cu.ft. with (2) 6" ports that stretched 58" into the box. After deciding that was just shy of physically impossible, given the dimensions of the box we were trying to work with (and needing to keep the WAF on the marginally-high side of things, but after all it is a 15" sub, so not much you can really do there), the project then shifted gears to a sealed enclosure with active EQ'ing. A 21" cube with 3/4" MDF topped off with 3/4" oak yields a net volume of 90L and an F3 of 38Hz and a resultant Qtc of 0.64. That's when the BFD came into the picture. Though most people seem to use it for correcting room modes, which we will also do, we will also being using it to tailor the natural roll-off of the sealed box and to give us an F3 that's a little better than 38 Hz. At 2k watts, we'll just barely be breaking Xmax on that massive driver at 20Hz.

Anyway, I'll be posting pics of the sub and it progresses along, and anyone with comments or suggestions, please feel free to reply. This will be my first attempt at using the infamous BFD, but fortunately not my first attempt at building a sub. However, this will go on my list of the biggest and the baddest. Can't wait to get started.

[Exocer]

Sweet!

I also own a TC-3k but have it in a 110 liter sealed box, ep1500 for power and BFD for EQ. In my 1350ft^3 room it kicks some serious tail! One thing I found interesting is the in-room measured response without EQ doesn't show the large hump around 50hz shown in simulations, very good! So I pretty much ended up using the BFD to get a little more output down low ala wide Q boost of 3db or so around 20hz.

What size is your room if you don't mind my asking?

[aktiondan]

After taking a 5-month long break, this project is back in works. I didn't mean to start a thread only to have it go nowhere, but my friend for whom I'm building this sub is for came down with some serious health problems, so we had to put the project on hold. Anyway, he's back after lots of chemo and bed rest and really wants to get the sub going. I've got the wood, we're going to do 3/4" MDF and 1/4" oak all around. Still looking to do a sealed box but I'm leaning towards 3.4-3.6 cu.ft. about 100L and I'm starting to shy away from doing the EP2500 and leaning towards PE's HPSA1000 plate amp. I suppose for simplicity sake, not b/c I think it will sound better. The difference is 1kW which equates to 3dB, so how much worse can it be? I like the simplicity factor of mounting the amp in the box, with its integrated parametric EQ, high-pass/low-pass filters, and no fan. It just seems like it would be a cleaner approach, less extra cables, wires, and audio gear to hideaway.

Has anyone used a TC-3000 with the HPSA1000? I know the sub will want more, but it still seems like a kilowatt of power to this sub will still be plenty. Besides in a sealed box this size according to Unibox, the max power the sub cab actually handle is only a round 1444W which is only 1.6dB more than 1,000W. Is the EP2500 really worth the usable 1.6dB of extra power? I know I'm looking at this from a purely numbers point of view, but while in the design phase, that's all I really have to go by. I feel like if the sub were for me, it would be a no-brainer, but since this is for someone else, I kinda need to take that into account as well. It's gotta be simple but still sound great. We're not trying to set any SPL records either or create a system that's hard to maintain. Thanks for any input!

Here's a couple of pics, I know most people have seen this sub, but it truly is impressive.

[WillyD]

Has anyone used a TC-3000 with the HPSA1000? I know the sub will want more, but it still seems like a kilowatt of power to this sub will still be plenty. Besides in a sealed box this size according to Unibox, the max power the sub cab actually handle is only a round 1444W which is only 1.6dB more than 1,000W. Is the EP2500 really worth the usable 1.6dB of extra power?

Well the TC-3000 can truthfully handle more. But in reality, the EP2500 may only give you another 600-700W of output. But, its significantly cheaper than the HPSA1000. But if the savings isn't important, don't hesitate getting the HPSA. It'll work just fine.

[Owen Bartley]

Dan, this is awesome, there aren't enough sealed projects going on lately! I hope your friend is doing well, and he'll love his new sub. One more thing to consider about going plate amp vs pro amp is that with the plate you'll have to run a sub cable to the final location, whereas you'll be running speaker cable with a pro amp. This might have an impact if the sub isn't going to be near the equipment rack.

Other than that, keep at it, I really want to see how this turns out!

[aktiondan]

Thanks for the responses. I was leaning towards the PE plate amp, but it seems like every other day I keeping going back to the Behringer. I'm just drawn to the thought of all that power. As well as getting to try out the BFD. A few years ago I built a sub out of one of the older TC Sounds drivers and powered it with a 1,200W QSC amp and it was absolutely insane. Until I got carried away with the 20Hz sine waves and cooked the thing. So there is a bit of hesitation to go down that path again, I'd rather stay on the conservative side of things just in case. Especially since how the sub will be used/treated/played/abused is sort of out of my hands since it won't be in my own home. I mean, you just never know.

At any rate, I threw together the cut sheet for the pieces and drew up a side view of the internals. Should be a decent sub, hopefully I'll get to keep it for a few days after its finished before my friend wants it. I'll be sure to run some REW numbers on it too. I've got a brand new uncal'd EMC8000 mic that needs to get used for something.

[aktiondan]

Wood has officially been cut. The box size is now set in stone. This is all I was able to get done today, I cut (9) pieces of wood using the ole' poor man's table saw jig. This almost works easier than trying to slide a huge 4 x 8 foot sheet of MDF onto a less-than-ideal $99 table saw from Sears. The basic pieces are: qty (5) 19-1/4" x 19-1/4" pieces for front x2, bracing x2 and rear panel x1. Qty (2) sides are 19-1/14" x 20-3/4". And the top and bottom are both 20-3/4" x 20-3/4". There's plenty of pieces left over for adding 2" x 3/4" x ?" bracing throughout the enclosure. Though the dimensions were optimized for a 4' x 8', the (5) main pieces that were cut across the length of the board left 1/8" to spare! But even that was no accident. Having the cut sheet drawn out in CAD beforehand really helps get the most out of each cut. I squeezed in that extra 1/4" all the way around which bought me an other 4.2L without really making the box noticeably any bigger. The final predicted net volume is going to be 97.6L or 3.45 cu.ft. Here's just a few pics, nothing exciting yet, but at least some of the wood is cut, that's a start, and sometimes just getting started is the hardest part.

[Owen Bartley]

Dan, way to squeeze out a little extra volume by tweaking at the cut-list stage! I'm too anal about all the math I would have already done to even think about maximizing volume at that point, great idea. I hope you were still able to measure twice, and cut once!

[aktiondan]

Yes, I measured several times and managed to just have to cut once. I've made that mistake all too many times. So the HPSA-1000 amp arrived on Friday. I pulled it out (this thing is pretty beefy, it weighs some 17 lbs.) and made a few measurements using REW of the parametric EQ. That EQ is actually pretty cool, I just wish there were more than just one of them inside there. You can basically boost the low end of sealed sub and get some better extension, or you can tame one room mode, but not both. I've been trying to figure out how to import the exported .txt file into UniBox so I can add the response to the modeled data but keep getting some "string too long error". Anybody know anything about that?

Here's a few screen shots of just what you can do with the parametric EQ. I suppose for the BFD gurus, this is small beans, but at least it's something, and better than most plate amps. The 18Hz high-pass rumble filter cannot be defeated and can be easily seen in each plot. It causes the response to be a couple dB down at 20Hz with the EQ flat. But if you dial up the BW to 0.2 and the gain to 2.5 right a 18Hz, you can get the response flat to 18Hz before it starts to roll off. I wish the BW could be adjusted tighter, even 0.1 isn't narrow enough to tame a nasty room mode without attenuating too many surround frequencies, but again, better than nothing. We'll most likely be using it to yield a flatter in-room response and lower f3 since it's a small sealed box with a natural f3 of ~37Hz. The amp will run of headroom before excursion is ever reached, so we'll bump up the boost. I want to figure out a way to tap out of the crossover into an RCA on the back of the amp so once the sub is done, I can still get into the crossover part and make measurements with REW. The dial indicators on the back of the amp are hardly an indication of the what the dial is actually doing, let alone how its altering the FR.

Anyway, here's some plots, as well as a pic of my setup for measuring, on the floor in my living room, where else?

[WillyD]

The 18Hz high-pass rumble filter cannot be defeated and can be easily seen in each plot. It causes the response to be a couple dB down at 20Hz with the EQ flat.

It actually can be defeated, with a mod. Not sure if you are into that kind of thing, but it's doable.

[Sonnie]

Very interesting!

[imbeaujp]

Verry nice project ! I will follow your progress and final results !

JP

[aktiondan]

I did read about that mod from this guy's site earlier this week, http://maxhawk.bravepages.com/projects/subwooferamp/ (beware, popups galore) but I just meant that there isn't simply a switch you can flip to defeat it easily. I'd be into it doing the mod, however for this sealed box, I'm not sure I would benefit a whole lot by pushing it down much further. An 18Hz high-pass should fine for this project. Besides, it shouldn't tax the amp as much if/when there is content below 18Hz pounding away, after all, it is only a thousand watts with zero active cooling. So we'll leave it as-is. I didn't get a chance to actually work on building or cutting anything else today, just messed around with the amp.

[WillyD]

I agree with ya.

[aktiondan]

I sent an email to Kristian, the author of Unibox and he made an update to Unibox to allow direct imports of .txt files from REW straight into Unibox for use in the active EQ section. There was an error before due to the some ASCII variation in how the .txt files look from REW and what Unibox actually wanted to see. But with the fix, it works great. So just to try it out, I modeled the TC-3000 in a 99L sealed box with and without the measured active EQ from the HPSA-1000 to show the difference. This is with the boost at +6, the BW at 0.2 and the Freq at 22Hz. This actually looks like it should yield a really nice flat response from about 24 to 100 Hz. In-room response should be a bit higher in the low end, but otherwise, much flatter than if used without the EQ. It's no Linkwitz-Transform, but at this setting, it's essentially doing the same thing. Kristian said he should be updating his site with version 4.08 soon.

[Exocer]

I too am interested in having the Unibox update, is it available online somewhere?

The FR in that simulation looks very good. It would be interesting to see exactly how the room effects things though. Its a bit difficult to estimate what the FR would look like before having some in-room plots to work with.

[aktiondan]

You can download it here http://home20.inet.tele.dk/kou/ubmodel.html

You're right, without any in-room measurements, it's hard to tell exactly what to set the parametric EQ to. I'm hoping I don't need +6dB of boost to yield a nice flat response. It's just fun to be able to model this kind of stuff, and do it with measured data, and be able to see how the modified response will look. In the end, I'm sure we'll just measure the in-room response with REW and then tweak the EQ to give a suitable FR. So how it models may not be the bottom line, but it gives a good starting point.

What would be really fun to model some Linkwitz-Transform circuits, build them up, measure them with REW, import the responses into Unibox, model the predicted modified response, build the box, and finally measure the near-field and in-room responses and then compare all the data. That would be pretty cool.

[Owen Bartley]

aktiondan wrote:

What would be really fun to model some Linkwitz-Transform circuits...

Sounds good to me... keep us updated on this one!

First of all, good call about the amp mod for now. You can always go in and make the change later if you feel like you're missing some low content and your amp isn't getting too hot. Second, I think it's unlikely you'll need to put much boost on this, mostly because it's usually the really low end that needs a lift. So you should be able to tame out one nasty room mode if it exists!

[aktiondan]

Owen Bartley wrote:

You can always go in and make the change later if you feel like you're missing some low content and your amp isn't getting too hot.

Of course. I was actually quite surprised when I ran that plot and saw how quickly the combined rolloff is. Output at 10Hz is a whopping 20dB lower. I suppose it makes sense though, you end up with a 30dB/octave rolloff, 12dB for the natural rolloff of the sealed box, and another 18dB for the filter in the amp (at 20Hz output power is about 106dB w/o the filter and an octave lower it's about 76dB with it, so the math checks out). I bet if you pitted two of these subs against each other, one with and one without the highpass filter, you'd have a handful of people who could hear the extended low frequency content and like that sub better. So contrary to what I had initially believed, a highpass filter or rumble filter on a sealed sub may actually do more harm than good, other than for the reasons already listed (power consumption, heat, etc.). I do not think that sealed subs by design need rumble filters since the box itself is a highpass filter with a 12dB/octave rolloff and the sealed box itself offers greater control over the excursion of the driver at low frequencies. The need stems from ported boxes whose drivers unload below the tuning frequency of the enclosure. I think most of the guys here know that, why else do we all want tuning frequencies into the teens? Because excursion skyrockets and output go to pot below the tuning frequency. Hmmm, I may have to rethink messing around with that filter after all...I guess we'll just wait and see.