Triple driver - sealed. Pros/cons, and is it too much for a newbie

[bear]

Originally Posted by MedPed

Thanks moonfly.
That is sort of what I was thinking. Based on what I have heard, sealed is better to my ears. (I havent heard THs). Question is, at what cost. Am I delusional in thinking I can make a sealed sub for under one grand which will get me usable 18-20 Hz output in my room. I am not sure in spl terms, but would hazard a guess and set about 115-120 db as my goal - after eq.
<snip>



BTW, in looking at drivers I couldn't help noticing how sensitivity goes own as drivers LF range begins to extend beyond (lower than) 35hz. From above 90 to the mid 80s. Likely from increased reinforcement of the cone. How relevant is that? It seemed fairly significant.
<snip>

115-120db below ~35Hz is very difficult to achieve via the speakers alone, with some room lift, possible.

You start to run into excursion and power limiting for any given driver.

One of the main reasons that speakers start to drop in sensitivity as their size, mass and Fs goes down is that the gap around the VC has to be made larger, assuming a high Xmax driver (which it almost always is these days, and almost always is with a low Fs) because otherwise the VC will rub on the pole pieces. So there is a balancing act between the mechanical aspects which results in a lessening of the magnetic/motor efficiency. Also, for equal output, as the frequency drops, the excursion has to increase substantially. This sets another limitation.

Back to the finished speaker, the limitation is going to be a function of the Xmax vs. frequency vs. power handling. This sets a pretty immutable limitation. So if you drop 10 dB in sensitivity from 90 to 80dB SPL/1w/1m (for example) you need to make that up in amplifier power. BUT you now have "wasted" 10dB of amplifier power just getting back up in level. If you used 1 watt for the 80dB, then to get to 90 you will use about >8watts. Not so bad? Well if it took 10 watts to do the 80 in the room, you'd need 80watts just to get to 90, and >640watts to get to 100dB! That's assuming the driver could *handle* 640watts, and that did not exceed the xmax at the frequency of interest.

Now, what if you had a 12db/oct EQ slope from say 100Hz down? You'd have +12dB of EQ at 50Hz and 24dB of EQ at 25Hz. Figure out what happens to the power levels and the power handling of the driver then!

Obviously, there are subwoofers that do use EQ, so it is *possible* to design a usable system. But that's where the design comes into play. There are practical limits and tradeoffs that must be taken into account no matter what.

So, when you say "115-120dB SPL" it is critical to understand what that means, and how that is measured, and if it is realistic.

My advice is to read a whole lot more on the subject, and probably follow an existing project that you determine meets your specific requirements. Check some of the build and design threads here too...

And, the other thing is that there isn't likely to be one single design or idea that reaches "perfection", I think you'll find that different designs do different things better or worse than others, and that you may or may not be satisfied fully with any or all of them! Thus, "DIY" lives!

_-_-bear

[tb46]

Hi MedPed,

Post #54: "...usable 18-20 Hz output in my room. I am not sure in spl terms, but would hazard a guess and set about 115-120 db as my goal..."

Example:

m R g S r has just put together a thread on the Dayton Audio RSS460HO-4 18" Reference HO Subwoofer 4 ohm 295-472 New HT subwoofer. Dayton Ref 18".

I tried this one in a sealed V_net=140L box in Hornresp. At the low frequency end of your goal: it takes three of those boxes (@ 250W ea.) to get to your SPL goal, assuming some room gain, and keeping the excursion below Xmax. It will take an LTC or some form of equalization to flatten the response.

Regards,

[turbodawg]

Originally Posted by tb46

Hi MedPed,

Post #54: "...usable 18-20 Hz output in my room. I am not sure in spl terms, but would hazard a guess and set about 115-120 db as my goal..."

Example:

m R g S r has just put together a thread on the Dayton Audio RSS460HO-4 18" Reference HO Subwoofer 4 ohm 295-472 New HT subwoofer. Dayton Ref 18".

I tried this one in a sealed V_net=140L box in Hornresp. At the low frequency end of your goal: it takes three of those boxes (@ 250W ea.) to get to your SPL goal, assuming some room gain, and keeping the excursion below Xmax. It will take an LTC or some form of equalization to flatten the response.

Regards,

That model is at x-max?

[MedPed]

Man this a lot of information.
@bear. I stand corrected. Higher freq diffract more.

The 115-120 db sounds too ambitious that low. I like the suggestions I have received so far. The cabinet sizes seem horrifically large though.

Let's assume I decide to go with twin 18" in sealed enclosures each. Powered by the inuke 6000dsp - or other comparable amp. Is the inuke capable to handle all the eq required? And realistically what is the minimum cabinet size I am looking at without sacrificing too much. I don't want to go trough the trouble and end up cutting corners in the wrong places.
Lastly, dedicated circuit for the amps?

I feel I am getting somewhere!!!

[Moonfly]

Originally Posted by bear

Try this outside with high freqs and call me back... you hear a speaker behind it because of reflections in the room, for the most part. Yes, the bass likely does radiate to the rear somewhat, but again outside you'd be surprised at what does not get there at all...

also you won't realize how beamy most speakers are until you listen outside (or in an anechoic chamber), because inside the "space" gets "filled up" by reflections galore.

_-_-bear

_-_-bear

High frequencies are not what subs are about, and I regularly listen to speakers outside Ive played with speakers with the tweeter turned off and mids turned off, so Im aware of how beamy they can be, but as far as low frequencies are concerned, sub just arent beamy at all.

[Moonfly]

Originally Posted by MedPed

Man this a lot of information.
@bear. I stand corrected. Higher freq diffract more.

The 115-120 db sounds too ambitious that low. I like the suggestions I have received so far. The cabinet sizes seem horrifically large though.

Let's assume I decide to go with twin 18" in sealed enclosures each. Powered by the inuke 6000dsp - or other comparable amp. Is the inuke capable to handle all the eq required? And realistically what is the minimum cabinet size I am looking at without sacrificing too much. I don't want to go trough the trouble and end up cutting corners in the wrong places.
Lastly, dedicated circuit for the amps?

I feel I am getting somewhere!!!

Cabinet size will depend on the speakers, the Fi model I put forward before (just because Im pretty familiar with it) for example needs a minimum of 150 litres. The driver has 28mm of x-max though and can handle 1500 watts, so combine that with a decent sized cabinet (larger means more efficient), and have two of them, and your going to get 20hz performance without issue. I take them down to at least 15hz in most rooms and these drivers never seem to get pushed that hard at normal decent listening volumes (-10).

[Moonfly]

Originally Posted by bear

The jet is a poor example.

First off most of the sound a jet makes is due to supersonic production of sound (inside the jet engine), secondly, the jet sounds noticeably different depending on your position relative to it's position WRT coming or going. Thirdly, due to its distance it is effectively a point source.

And the reason that a jet at a long distance, but over the horizon is difficult at times to localize is due to reflections of the audible sound, causing the ear to determine that the jet is where the source of the reflection is not where it actually is.

_-_-bear

The example was to illustrate non locatable sound. A jet high up in the sky is pretty much all deep rumble heard from the ground, and I'm not talking beyond horizon, fast jets roaring past in a valley are a completely different thing. Its a perfect example of how low rumble sounds are near impossible to locate as the omnidirectional sound is output from a point source.

[tb46]

Hi turbodwg,

Post #63: "...model is at x-max?"

It is at the very low end MedPed seems to be looking for, e.g.: ~~18Hz. Xmax listed as 12.75mm. So any power past 250W/4_Ohm in this relatively small sealed enclosure (140L) will exceed Xmax (power compression not taken into account). On the other hand, you might be able to do structural damage to a normal home even with two (one?) of these "beasts", and 500W.

Regards,

[weltersys]

Originally Posted by MedPed

The 115-120 db sounds too ambitious that low. I like the suggestions I have received so far. The cabinet sizes seem horrifically large though.

Let's assume I decide to go with twin 18" in sealed enclosures each. Powered by the inuke 6000dsp - or other comparable amp. Is the inuke capable to handle all the eq required? And realistically what is the minimum cabinet size I am looking at without sacrificing too much. I don't want to go trough the trouble and end up cutting corners in the wrong places.
Lastly, dedicated circuit for the amps?

I feel I am getting somewhere!!!

An amp rated for a given amount of power into a given impedance should be able to deliver that power.
The duration of how long it can deliver that power is a different story, LF waves are by definition of longer duration than HF, and some amps don't do well on sustained LF output, often an "old school" big capacitor/transformer amp can outperform a modern switching amp rated for double the power.

Lots of bass power requires high peaks to be delivered, a separate circuit is not required, but heavy wire is required to prevent brown outs (low AC line voltage) on peaks.
Even 12AWG, standard for 20 amp circuits, will brown out on peaks if the distance to the main box is more than 20 feet.
Here again, a switching amp often suffers more from brown out than the old school amps.

Still, many are selling off the heavy iron, there are good deals to be had on used amps.

By the way, a pair of Eminence Lab 12s in a 5 cube BR with an Fb of 19 Hz can do 112 dB down to below 20 Hz outdoors with 800 watts, indoors you might see 3-12 dB more.

[weltersys]

Originally Posted by Moonfly

The example was to illustrate non locatable sound. A jet high up in the sky is pretty much all deep rumble heard from the ground, and I'm not talking beyond horizon, fast jets roaring past in a valley are a completely different thing. Its a perfect example of how low rumble sounds are near impossible to locate as the omnidirectional sound is output from a point source.

I live in the hot dry desert of New Mexico.
Hot and dry result in the most HF air attenuation.

I can close my eyes and point to where a jet, 30,000 feet above was.

I say "was", as the jet will have traveled several miles in the time it takes to hear it.

Your "perfect example of how low rumble sounds are near impossible to locate" does not work for me, I can point directly to hot air balloons (which don't fly fast, and use burners that sound like a jet) or idling trucks far enough away to be barely visible.

Perhaps some people are less sensitive to localization down to a low frequency, I do find it interesting that many often say they don't "hear" but "feel" low frequencies.
I have no problem hearing undistorted LF below 20 Hz.
In fact, I have changed tables in restaurants because of air handling ducts resonating in the 15-25 Hz range.

My girlfriend, who hears to 20K, and about 50 dB better at 4K than I, won't notice such a sound unless I "point" it out.

Art