When I said 40's I was meaning upper 40's. I'd have to go back and find the SPL trace but I think it was 92dB at 48hz measured via Smaart RTA in room with music. Didn't hear anything nasty. Just to be clear this was not an anechoic measurement. I was just surprised and had smaart handy so took a look...
Fair enough.
Due to the extreme box size restrictions, your only possible solution is going to have to include a Linkwitz transform active circuit and lots of amplifier power......you won't pull this off passively with any driver and level of effectiveness......period.
To save weight, I'd suggest drivers of a neo variation or dual diaphragm BMRs such as the new micro subs from Tang Band. In your application, you'll only need clean response from the woofer to 250hz as the 10F has shown to extend that low and still have exceptional power handling and low distortion.
Honestly, this sounds more like asking the question of could rather than should.
To save weight, I'd suggest drivers of a neo variation or dual diaphragm BMRs such as the new micro subs from Tang Band. In your application, you'll only need clean response from the woofer to 250hz as the 10F has shown to extend that low and still have exceptional power handling and low distortion.
Honestly, this sounds more like asking the question of could rather than should.
The L16 is a very enticing choice. I'm seriously considering using this, even though I would need to increase the size of the speaker. My main concern is that despite being optimized for sealed enclosures, for some reason it models very poorly in a tiny airspace. The power required to reach x-max is on the order of 150 watts when other drivers only need ~50 watts. Can someone try modelling and see if they get the same thing? It seems really strange that it needs this much power.
I plugged the L16RNX Seas specs into WinISD and got this for a sealed, Qtc = 0.703 alignment: Vbox = 5.6 L, F3 = 84 Hz and 25W for 100 dB. Xmax occurs at 15 Hz at 100 dB.
The U16RCY/P sealed has the same F3, but Vbox = 7.25 L and 18 W gives 100 dB. Xmax occurs at 81 Hz at 100 dB. The pronounced cone breakup of the L16RNX is absent due to different cone material.
The L16RNX goes a lot lower in a vented alignment: Vbox = 10.L, F3 = 47 Hz and 25W gives 100 dB. Xmax occurs at 35 Hz at 100 dB.
The U16RCY/P sealed has the same F3, but Vbox = 7.25 L and 18 W gives 100 dB. Xmax occurs at 81 Hz at 100 dB. The pronounced cone breakup of the L16RNX is absent due to different cone material.
The L16RNX goes a lot lower in a vented alignment: Vbox = 10.L, F3 = 47 Hz and 25W gives 100 dB. Xmax occurs at 35 Hz at 100 dB.
This is what I get with the SEAS L16RN-SL. It doesn't seem right. The excursion level goes way down below the peak at 110Hz when most drivers stay flat as the frequency goes down. It takes way too much power (150 watts) to reach xmax at the lower frequencies. What do you guys think?
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
What exactly are you simming? 4 L16RN-SL in 6L?
Thanks
"This is what I get with the SEAS L16RN-SL. It doesn't seem right. The excursion level goes way down below the peak at 110Hz when most drivers stay flat as the frequency goes down. It takes way too much power (150 watts) to reach xmax at the lower frequencies. What do you guys think?"
Thanks
"This is what I get with the SEAS L16RN-SL. It doesn't seem right. The excursion level goes way down below the peak at 110Hz when most drivers stay flat as the frequency goes down. It takes way too much power (150 watts) to reach xmax at the lower frequencies. What do you guys think?"
Oops, I posted the wrong links.
Here is the simulation for 1 SEAS L16RN-SL in 1.5L with 250W and the excursion graph, followed the simulation for 1 SEAS ER15RLY in the same 1.5L, but only 50W. As you can see, the L16RN-SL needs a ridiculous 250W to hit xmax in the bass frequencies while the ER15RLY only needs 80 watts to hit xmax.
Now the SEAS ER15RLY
Here is the simulation for 1 SEAS L16RN-SL in 1.5L with 250W and the excursion graph, followed the simulation for 1 SEAS ER15RLY in the same 1.5L, but only 50W. As you can see, the L16RN-SL needs a ridiculous 250W to hit xmax in the bass frequencies while the ER15RLY only needs 80 watts to hit xmax.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Now the SEAS ER15RLY
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Good catch. You are right. The Sd is 25% bigger on the L16, so the amount of power needed to achieve xmax (which is almost 20% higher) makes perfect sense now
The problem I see with the L16 in 1.5L is the extremely high Qts of 1.57 that causes a 4 dB hump in its response at about 160 Hz. F3 (ignoring the hump) is about 100 Hz. The ER15 has an improved Qts of 1.02, bit it rolls off earlier with an F3 of about 122 Hz.
An alternate, but more expensive solution would be to use 4 Tangband W3-2108 mini-subwoofers in a total of 1.5L. Using a series/parallel wiring scheme would give a 4 ohm impedance. Qts is 1.04 and F3 is 90 Hz. 100W gives about 103 dB and excursion peaks at about 3.1 mm and flattens out to 2.7 mm below F3. The W3-2108 has 7mm Xmax, leaving some excursion available for a Linkwitz Transform to boost the output below F3. Just a (expensive) thought.
An alternate, but more expensive solution would be to use 4 Tangband W3-2108 mini-subwoofers in a total of 1.5L. Using a series/parallel wiring scheme would give a 4 ohm impedance. Qts is 1.04 and F3 is 90 Hz. 100W gives about 103 dB and excursion peaks at about 3.1 mm and flattens out to 2.7 mm below F3. The W3-2108 has 7mm Xmax, leaving some excursion available for a Linkwitz Transform to boost the output below F3. Just a (expensive
) thought.Attachments
Last edited:
An alternate, but more expensive solution would be to use 4 Tangband W3-2108 mini-subwoofers in a total of 1.5L. Using a series/parallel wiring scheme would give a 4 ohm impedance. Qts is 1.04 and F3 is 90 Hz. 100W gives about 103 dB and excursion peaks at about 3.1 mm and flattens out to 2.7 mm below F3. The W3-2108 has 7mm Xmax, leaving some excursion available for a Linkwitz Transform to boost the output below F3. Just a (expensive
Haha there's definitely no such thing as bang for the buck for this build. One speaker will require around $1200 in parts and electronics!
Unfortunately there is no way to fit 16 of those W3-2108's in a 14'' x 6'' x 6'' box. Fitting 4 of those L16's and a midrange and tweeter in a box of that size is already incredibly challenging. Plus, I'm already severely power limited with 4 L16's in a cabinet. I need ~1000 watts to reach xmax on all 4 woofers in that cabinet size.
Regarding the hump, I can EQ that hump away and then apply the boost. Do you see any problems with that?
Also, I'm thinking I should use the L16RNX instead of the L16RN-SL. Even though the RN-SL is designed for sealed use while the RNX is for ported, the RNX actually models a lot better and the lower Qts makes it 1dB more efficient for bass frequencies.
Last edited:
The box is way too small, especially for the L16.
You're running it at >1,5, which gives little or no bass extension, and a boomblaster style hump in the midbass. The compliance stiffness of the air in the box protects the woofer against high excursion.
Johan-Kr
Maybe I misunderstood - do you have 1.5L for each of the 4 speakers or a total of 1.5L for all 4? I thought the latter and modeled four W3's in a total of 1.5L or 0.375L per speaker. I see your sim for the L16 and ER15 is for 1.5L per speaker.
Thanks.
Thanks.
Haha there's definitely no such thing as bang for the buck for this build. One speaker will require around $1200 in parts and electronics!
Unfortunately there is no way to fit 16 of those W3-2108's in a 14'' x 6'' x 6'' box. Fitting 4 of those L16's and a midrange and tweeter in a box of that size is already incredibly challenging. Plus, I'm already severely power limited with 4 L16's in a cabinet. I need ~1000 watts to reach xmax on all 4 woofers in that cabinet size.
Regarding the hump, I can EQ that hump away and then apply the boost. Do you see any problems with that?
Also, I'm thinking I should use the L16RNX instead of the L16RN-SL. Even though the RN-SL is designed for sealed use while the RNX is for ported, the RNX actually models a lot better and the lower Qts makes it 1dB more efficient for bass frequencies.
Yes it has. The rear woofers are low passed at 150Hz and the front woofers are low passed at 300Hz.
It is 1.5L per driver for a total of 6L for 4 drivers. The spreadsheet I have does not work well (It refuses to work on Windows, gives a "Compile error in hidden module: ThisWorkbook, Pulseio" error, and I tried on Excel 2003, 2007, and 2013). While it does work on OS X, I cannot change the driver configuration, so I'm forced to model one driver.
There will be EQ to boost bass and correct the shape of the response. The DSP tuning will be much more sophisticated than just a Linkwitz transform, which is required to make the speaker usable with the amount of boost I'm adding to the woofers.
Last edited:
The performance is exceptional.
A test will never tell you if you like something or not.
Most of the problems with a 5 inch in the bass region don't come from what Zaph measures.
You have to know what a test tells you and what it does not and then if it is significant and then for whom it is significant.
Example:
When I make vented box, I make it as large as possible (but I don't like large boxes) and the vent as long as possible.
Going to get **** for this but: you don't have to box plot or anything (maybe to see GD or something I don't know); or maybe if you're listening to the speaker outdoors or want a lot of midbass or peaky bass.
I've never made a speaker and thought: I made the tuning too low.
I mean, that's why you make your own speakers because you know what you like and you build things better than others.
I like sealed woofers (large, ~8inch) with cone tweeter or ring radiator or 30mm dome.
If you like the way graphs look and think they sound good, build that.
Build all the 1" - 5-7" woofer 2-way vented speakers you want. I build them sometimes, but I digital EQ the crap out of them.
Smartest thing I ever read on the forums written by a cheeky guy I met 15 years later: "you can't listen to a graph!"
That TB woofer sucks all the performance in the bass region you can out of the cone area with good mids also (not really a great idea for longevity), build quality sucks *** (some one said this already), so I build SB acoustics maybe or nothing. It's really a cheap illuminator woofer, with a ferrite motor, which I kind of prefer for longevity/heat dissipation.
Doesn't make sense to build speakers out of cheap drivers unless you are hard core hobbiest and have time to burn or like to stay busy (I don't know
) and have tons of time to waste, in my humble opinion of course
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.