Jump to content

BioBrian

In at the deep end - SB 15 inch for sensitive bass

Recommended Posts

I've had the feeling something's been missing, so have decided to do something about it.

 

I like to play orchestral and pipe organ music, and of course I'm listening to the gear, not just the music  :love.  Also, I've mentioned in other threads the wish to hear soft bass in proper proportion to the rest of the sound, eg quiet strokes with a soft stick on a large bass drum.

 

Sometimes in organ music it's as though a run of notes is missing one or two, and my sense is that it's the very low notes that just run out of puff with 10" drivers. I've spent ages developing a 3-way, using SEAS's lovely W26FX002, but even in vented 92 litre boxes, it's not giving me that solid, sometimes cheeky bass that I know is in the recording.

 

I do have a sub, based on my home-made Playmaster 300W amp and a 12" XLS driver, but prefer the sound to come from a properly balanced L + R pair. Proper speakers, in my book, shouldn't need subs!

 

The thinking is that what I don't like about subs is their insensitivity, in that they tend to be heavy rubbery monsters, usually in a box far too small, requiring huge power to get moving, so these delicate LF sounds might not be true to life. Why not look at Sensitive LF drivers, like this SB Acoustics 15", at 93 dB?

 

So these creatures arrived, and on opening the boxes I got quite a surprise. For 15" woofers, they actually measure 16 1/2", or 417 mm across. And at 10.2 kg, quite a lift.

 

post-135890-0-21792200-1437538229_thumb.

 

Question is: what's the best way to go from here? Infinite possibilities are tripping me up (out) a bit, so I thought I'd put it out here...

 

Edit: oh yeah, forgot to say - it's   SB42FHC75-6

Edited by BioBrian

Share this post


Link to post
Share on other sites

Question is: what's the best way to go from here?

 

Decide what cabinet size you can work with (bigger is better) ..... and from there choose a driver to match your cabinet choice.

 

Candidates are sealed box or (front) horn .... or if you're ok with 'vented', then some type of ported box, or a tapped horn.

Share this post


Link to post
Share on other sites

SB Acoustics T&S parameters are usually spot on and given that this model fits the conditions for a classic B4 alignment the internal volume of the enclosure (Vb) should be 300 litres.

Share this post


Link to post
Share on other sites

Thanks for quick replies, gents. The opening post was getting a bit long, so I didn't add any thoughts of my own.

 

I've put it through WinISD and come out with 320 l vented/22.5 Hz, but have whittled it down to 260 l/24 Hz minimum. This gives F3 = 24 Hz, F6 = 21 Hz, using double 103 x 257 mm vents.

 

As an ultimate dream, it's in a 5-way using Scan Speak drivers above, something like:

 

post-135890-0-75232300-1437541693_thumb.

Edited by BioBrian

Share this post


Link to post
Share on other sites

Thanks for quick replies, gents. The opening post was getting a bit long, so I didn't add any thoughts of my own.

 

I've put it through WinISD and come out with 320 l vented/22.5 Hz, but have whittled it down to 260 l/24 Hz minimum. This gives F3 = 24 Hz, F6 = 21 Hz, using double 103 x 257 mm vents.

 

As an ultimate dream, it's in a 5-way using Scan Speak drivers above, something like:

 

attachicon.gifSB 15 inch - possible 5-way mock-up.jpg

 

Looks the goods but what happens when someone opens the door...? :thumb:

Share this post


Link to post
Share on other sites

Hell glad you pointed that out, Hensa! There's 2 downward steps outside, onto concrete :(.

 

You can tell when a bloke lives alone:

 

- Puts his new toys in the doorway

 

- Loves the idea of 320+ litre cabs

Share this post


Link to post
Share on other sites

I don't want to waste anybody's time - here's a bit more stuff to emerge over the last couple of weeks:

 

SB42 for bottom 2 octaves. 20-40 & 40-80 Hz. Maybe a bit more. What size inductor? Thinking at least 10mH C-coil, but not sure how to calculate.

 

SS 26W/8861T00 for next 2 octaves (Upper bass & Lower mid): 80-160 & 160-320 Hz. Closed box, 46 l is about as small as I would risk without more learned advice. Hopefully low-distortion, punchy, dry.

 

Illuminators above that: heaps of excursion, if needed, for better power tolerance/lower distortion than SB Satori drivers.

 

D3004/620000 tweeter seems to be the unloved kid in the family. It has the 3-petal flower thing on the front, made of black foam rubber. It has the best off-axis response I've seen in a tweeter, but everybody only talks about it's sister, the R3004, with the cute pointy nose.

 

Here's a back view:

 

post-135890-0-96217000-1437545006_thumb.

Edited by BioBrian

Share this post


Link to post
Share on other sites

but not sure how to calculate.

 

Have you designed a crossover before?    The more "ways" the more difficult.....    drivers will only produce the sound they're told to by the crossover filter   ;)

Share this post


Link to post
Share on other sites
Moderator

\One can never have enough subs, alas only a 4 way :D

post-104378-0-22773200-1437548346_thumb.

Edited by joz

Share this post


Link to post
Share on other sites

henry218 is correct in questioning your initial plan.The sensitivity and impedance variations,will not allow  for a basic passive crossover network to be effective in a five way set-up.A three way system would be a more 'do-able'  option providing you can find a mid-range driver and tweeter with sensitivities of 93dB /1w/1m or better.Otherwise,you may have to look at an active crossover with additional amplifiers

Share this post


Link to post
Share on other sites

So we see what I meant by "infinite possibilities", and all that that throws up. There's a lot of tradition to be questioned (and drawn upon) when looking at such a different set of goals.

 

Dave, "designed a crossover" is not quite how I'd describe it, but I've spent the last year and a half taking the basic plan of a LR2 crossover, and making it work for the drivers I had. Change this piece, measure, change that one, measure, listen, then start it all again. I've done runs of tests, eg with a 0.27 mH and 10 different caps, and then 4.7 uF and 6 different coils. The worse part is the taxing of my memory, even when taking careful notes! The pattern is pretty basic, and I've learned a few tricks, and discovered meaning in other people's tricks. As you know, it's no easy ride, and I did only start this after 2 careers bit the dust. I'm not a welder or hydraulics engineer, but I built a front end loader. I'm not a carpenter/plumber etc, but I built my house. I just get in and try to learn what's most needed to do the job.

 

@ Henry218 - 5-way is possible. It's been done, including very recently. In the scenario above it would be really 4 1/2-way, as the 26W 10" woofer has no Hi Pass filter, other than the sealed enclosure. Band-pass filters for the following 1 or 2 mids (could be 4-way if I can find the right drivers) should follow the same principles as in a 3-way. One problem I see is that the system impedance can drop with every additional driver and filter element.

 

The point of the thread, however, is to find a good application for this driver. It was chosen for its high sensitivity, highish Qms (5.4), low (22.5 Hz) Fs, low inductance (2.1 mH), etc. The only drawback I found, and you guys can probably find more, is the Mms at 160 g, but hey it's Sd of 850 sq cm is good reason for it to have a good stiff cone.

 

This driver, in the baffle I've been playing with, has a woofer width of 575 mm, so calculated Baffle Step is(?) 115 over 0.575 = 200Hz. This gives me the option of bi-amping, and using the difference in amplifier gain to correct for both BS, and the difference in sensitivity between this woofer and the drivers above.

 

@ VanArn - thanks for the warning, and this is one reason I've exposed myself so blatantly - having LspCAD and the brains/experience to use it would be a great thing. But I don't have health or patience to sit at this machine long enough for that. But I've seen this concept work. It is a bit "baffling" to see that most mid drivers are of lower sensitivity than most tweeters and a lot of woofers, but they still end up getting attenuated in the final circuit. The SPL curves for mids never reach the level of the SPL of the whole system, so I've come to conclude that they are probably designed to have lower sensitivity.

 

93 dB is already reduced by a series coil, and the parallel capacitor. Trial and error for the size of these is quite expensive, so maybe somebody can work out if 10 or 18 mH or whatever would give a cutoff frequency around 80 Hz? The cap is easy to play with; I have several, and they're easy to use as multiples. I've made graphs of resistance according to various inductors, but the sands are shifting all the time with other influences. This bit should be the easiest.

 

An LCR for the impedance peak is needed too - any clues on calculation for this?

 

The calculators I've seen for impedance only affect the slow rise above resonance, and in this case, there is no rise in the working band!

 

Of course the biggest question for me is matching of phase, and all that that entails. I'm making that aspect somewhat easier by having the upper drivers in their own boxes, for shifting fore-and-aft. Apparently this makes it simpler to bring off a simple LR2 filter. MiniDSP or DEQX are not out of the question, but I prefer to keep my Bach before my Byte and give passive a decent shot.

 

By the way, I've not been at all happy with the on-line calculators that have shown up so far. They don't even specify whether you enter Nominal impedance or Re. All calculators should come with a step-by-step guide in what they need you to input, otherwise they're useless! (Showing some restraint here  :) ).

 

I've already had satisfaction from these cats - bi-amped on the floor in front of my 3-ways (driven through 5.6 and 6.8 mH series coils, the biggest I have ATM). Even with no baffles, they are filling in the space I was lacking. There's a grip to the room and a sense of fullfilment that's not been there. Just don't want to waste time building baffles I'll be throwing away. Need to explore, so I'm hoping to hear something inspiring!

Edited by BioBrian

Share this post


Link to post
Share on other sites

Like your 15", very nice.

Honey comb carbon driver cone like the JAMO Concert series woofers, huge chassis and engine, should be a lively performer.

Great project

 

Cheers

Share this post


Link to post
Share on other sites

i didnt say its not possible.. just i wonder why you want to build 5way passive... :D

Share this post


Link to post
Share on other sites

i didnt say its not possible.. just i wonder why you want to build 5way passive... :D

Yeah just put an active xover/amp on them. Dayton audio 150 comes to mind.

Share this post


Link to post
Share on other sites

The point of the thread, however, is to find a good application for this driver. It was chosen for its high sensitivity, highish Qms (5.4), low (22.5 Hz) Fs, low inductance (2.1 mH), etc. The only drawback I found, and you guys can probably find more, is the Mms at 160 g

 

The driver is fine for a hifi subwoofer.

Share this post


Link to post
Share on other sites

By the way, I've not been at all happy with the on-line calculators that have shown up so far. They don't even specify whether you enter Nominal impedance or Re. \

 

You want neither.  The impedance of real drivers vary with frequency.

 

Without something like lspcad, it is often even difficult to design a good two-way speaker.

 

My advice is to find a well documented, high performing design.... and build that.

Share this post


Link to post
Share on other sites

there's 2 way of doing this,

 

1. doing it by trial ($$$ as you need to stock up the parts)

2. get a really good measurements (the first 3 drivers need to be measured on the roof... like Paul Spencer did... :P ) and do simulation.

 

i found that 4 way is the maximum, i don't really see any benefit going 5way, unless its 4.1/2/3/4 :D

Edited by henry218

Share this post


Link to post
Share on other sites

\One can never have enough subs, alas only a 4 way :D

attachicon.gifIlluminator boxes.jpg

Respect,

I got into trouble today playing Interstellar on my ht aka (the whole neighbourhood heard it) with 2x12" & 2x15"

And I've got 2x12" shivas there doing nothing......

Share this post


Link to post
Share on other sites

there's 2 way of doing this

 

There's quite a lot of overlap.

 

 

Near field measurements can be used up to Hz = 10950 / width(cm)       (but this rule does not hold for a horn)

 

Impedance measurements, can be used along with 'flat' responses .... in order take some 'trial' out of 'trial'

Share this post


Link to post
Share on other sites

For inspiration Troels has just released a 5 way......................

 

For driver measurements I've just bought the dayton dats from lsk. Don't ask me how well it works as I'm just too busy to even look at it yet.

 

One thought runs through my mind.

Is it not possible to use that bass driver as a passive or active supplement to your existing speakers ?

Thus cutting your possible work load down buy a large chunk.

Share this post


Link to post
Share on other sites

@ Henry218 - I have a great roof overhang, but the river noise would be a problem. I have a hay shed, which is really anechoic with the right amount of hay, but I'd need to run the computer & amp from an inverter, and I'm not sure how they would like that. Have thought about making an anechoic 'igloo' kind of thing though, just with hay and enough wire or something to hold it together - could be fun. Or I could wait for the next time it snows - that's an incredibly quiet environment!

 

@ HiFiplus - I'll revisit Bagby's software - I didn't even have Omnimic last time, and found it very clunky to use - I need somebody over my shoulder for that stuff. Gets very depressing not knowing WTH to do next. TL is very tempting, but I can't see it working if using offset, like the TQWT. Does anybody know how to design a TL with the woofer having about 1/3rd box volume and then tapering to a small vent? I've looked at Martin King, but gave up B4 downloading MathCAD sheets... Vented is easy...

 

@ Davewantsmoore - Not sure why you point out "The impedance of real drivers vary with frequency". Dayton DATS gives a curve that does just that. Usual big spike at resonance, then gradual climb up. This driver shows about 75 Ohm at 22.5 Hz, then takes a very slow ride, not crossing 15 Ohm until 1500 Hz - due to its low inductance? I expect it will behave differently once in a box, and later when it's broken in a bit. The spike can be almost flattened by an LCR, but my experience with the SEAS woofer was that it hardly affected the SPL curve, so I cut it out.

 

@ 125dBmonster - thanks, all positivity appreciated!

 

@ joz - I used to have violin lessons in Elwood - glad you didn't live next door then...

Share this post


Link to post
Share on other sites

Just had sideways thought.

What about open baffle or H frame ?

 

Also, their are some velly smart people here, and some more on diyaudio.

maybe also post there and see if you can get more ideas or options...

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Recently Browsing   0 members

    No registered users viewing this page.

×