Designing a custom bass cab - third iteration, 2x8"+15" compact neo 3-way...

[Paul K]

I'm using WINISD pro alpha for my project's design (the kick drum shell for a bass speaker cab.) Adding insulation to the inside of the cab makes the cab bahave as if it were larger, and bigger is better. So far, so good. Winalpha has no "input" for the amount of insulation (no suprise, as it'd be tough to measure) My question: does this added insulation change the size/length of the port that is required for optimum performance?

[alexclaber]

This is something I just learnt about. WinISD Pro can calculate the effect of insulation by changing the values of Ql Qa and Qp but I have no idea how you work out them!

 

Although a sealed box will produce better bass due to increased apparent volume if you insulate the cab heavily, a ported box can actually be over-insulated, which damps the resonant effect of the port and thus reduces bass response. In fact, an over-insulated ported cab will end up with similar response to a same size sealed cab but worse group delay. Can't find the article I read about this but it's somewhere out there on t'interweb...

 

Useful thread: http://billfitzmaurice.com/phpBB2/viewtopic.php?t=1688

 

My theoretical experimentation has found that the port should be tuned 2-4 semitones above the lowest frequency the cab will have to handle, the larger the cab appears to the speaker, the lower it will need to be tuned to prevent driver unloading too severely below the tuning frequency. For a low B in an undersized cab (100 litres instead of 150 litres as would give maximum bass) I'm tuning to 37Hz. Once I build the cab I'll slide the port in and out until I get minimum excursion with low D (note closest to tuning frequency). I won't be using much woofer enclosure insulation as the cab is 3-way.

 

Alex

 

P.S. Enclosures losses (due to leakage, port losses and insulation losses) affects the volume that the speaker perceives but the port sees the actual measured internal volume.

[alexclaber]

Iteration 3.5

 

In the interest of seeing if it's possible to make this a full production design, I've been experimenting with the trusty Eminence Alpha 8MR - a very good value, sensitive and high power midrange speaker whose only weakness is rather peaky response (which had really put off flat-response-obsessed moi).

 

Initially when just replacing the Precision Audio 8"s with the same crossover points the Alpha is a lot louder, with very peaky midrange response. But if I connect the Alphas in series then the sensitivity matches the Kappalite 15" and the inductance of each midrange acts as a low pass filter on the other midrange. The response still has peaks at the top and bottom of the midrange which I don't like.

 

However, if I shift the high pass filter point up from 400Hz to 1000Hz and reduce the slope from 2nd (12dB/oct) to 1st (6dB/oct) order then the big peak around 600Hz is removed. Excellent. If I then add a small series inductor to the midranges I then lower the low pass filter point to 3000Hz and reduce the slope from 2nd to 1st order and remove the big peak at 3kHz. Excellent again!

 

So the midrange section has gone from:

 

Two Precision Devices PDN.8BM31 @ £140 & 4.4kg total

Two inductors and two capacitors @ $30 & 1lb total

 

to:

 

Two Eminence Alpha 8MR @ £46 & 4kg total

One inductor and one capacitor @ $6 & 0.2lb total

 

£155 vs £50

5kg vs 4kg

 

A nice improvement all round!!!

 

And now that the midranges are crossed over at 1kHz and are running at 16 ohms total, there is no need for a low pass filter on the woofer to maintain 8 ohms total impedance. If the woofer was front-firing removing this filter would cause a hump around 1.2kHz but the side-firing arrangement means that the woofer's output is acoustically filtered and won't interfere with the midrange drivers.

 

And because the midrange drivers are sealed back they don't need sub-enclosures, which makes the build easier, increases the volume the woofer sees and reduces the weight further.

 

What a breakthrough!!!

 

Alex

[alexclaber]

One final tweak - the woofer has been moved from being in a side-firing slanted baffle to being in a down-firing slanted baffle. Why? More efficient use of cab volume, better woofer chamber outer panel damping from floor, better boundary reinforcement from floor, less plywood needed for build, therefore lower weight.

 

This requires the relocation of the ports. For reduced weight and easier tuning I'm going for rear-mounted plastic port tubes, which will be supported by the side handles.

 

I've also decided to revise the construction, from heavily braced balsa spaceframe and 1/4" ply skin, to 1/2" ply panels with 1"x1" batting and 2"x2" cross-bracing. Should weigh in at about 60lbs complete and be a cheaper and easier build

 

Frequency response plots and technical drawings to follow.

 

Alex

[alexclaber]

An update! Since getting the design to this point I have:

 

1. Shelved it account of it being too big compared to my Acmes.

2. Designed a more compact 2x6"+12" horizonal version designed for stacking as a pair at gigs and standaline at rehearsals.

3. Shelved that on account of it being too inefficient - I played a seriously loud gig recently and decided I wanted something that could stay very clean at high SPL.

4. Bought ALL the Bill Fitzmaurice speaker plans, gone through them in great detail and decided to build a DR250 and Titan 39.

5. Then played a quieter gig where we sounded so much better and resolved that low stage volume is the key do good vocals - thus the DR250 and Titan 39 is complete overkill.

6. Therefore decided to build an OmniTop 12 and some direct radiator 15" subs.

7. Then realised that I might as well fit midranges and tweeters to those subs and do away with the OmniTop 12.

8. Ended up back a design which is remarkably close to the previous 2x8" plus 15" design!

 

Here's how it stacks up: For the lows an Eminence Kappalite 3015LF in a ported box, 80 litres, tuned to 40Hz. For the mids a pair of Eminence Alpha 8MR sealed back drivers (therefore appearing to share the same enclosure but not in acoustic terms). And for the highs a short line array made of a series of cut-and-glued GT-1016 piezo tweeters.

 

The box itself will be quite wide, fairly short and shallower than it is tall - 25" wide, 18.5" high but only 15.75" deep. The port will be a slot port, firing out of the side of the cabinet, with the rear baffle forming one side of the port and a secondary baffle running parallel to that. Because of the cab's width this allows me to fit an unusually large port into a small box, giving both correct tuning and high power handling without port noise or compression.

 

The 15" woofer will be mounted forward firing on the front baffle, on the opposite side of the box to where the port exits from the side panel. Beside the 15" woofer will be the two 8" midrange speakers, vertically aligned for good dispersion and control without negative comb filtering. Because stages can be quite confined I feel it's best to have the port (which needs to fire into free air) and the midrange speakers on the same side of the cab so that are away from any reflective boundaries. The cabinet can therefore be turned upside down (though I'll make sure it looks right both ways up) to suit the venue.

 

The front baffle will actually be stepped - the part of the baffle where the 8" midranges are mounted will be ~2" recessed, and the tweeter array will be mounted in a semi-circular pipe, which will be fixed in a vertical line directly in front of the two 8" midrange speakers. This will improve the horizontal dispersion of the midrange speakers and prevent beaming in their higher ranges, whilst also saving front baffle space. I don't know if any benefits from the coherent source will be nullified by the lack of time alignment, we shall see.

 

The midrange speakers will be bandpassed at 1kHz and 3kHz with a first order filter. They will be connected in series to get the right sensitivity and improve the voltage handling. The tweeters do not require a crossover as their capacitive nature acts like a first order high pass filter. The cabinet will be biamped from my PLX 3002 - one channel providing up to 550W to the woofer, the other channel for the midranges and tweeters - and there will be a first order lowpass passive line level filter set at 500Hz for the woofer and maybe a first order highpass passive line level filter at ~400Hz for the mids/tweeters. I may make the woofer low pass switchable between a number of different values to provide a midrange boost as the woofer is lowpassed progressively higher. There will also be a tweeter on/off switch.

 

I'll be making a matched pair of cabinets, which can be stacked so that the midrange speakers act as a short line array, or used separately as small PA speakers. When using two speakers, they will run in parallel, still biamped, so that the woofers receive up to 900W from my amp. The midranges can handle approximately half the power of the woofer but their impedance is twice as high - they will be all but impossible to blow and should play very loud without power compression.

 

I'm going to take a gamble and build the cab from 9mm okoumé ply with extensive bracing and prestressing. If this proves insufficiently stiff I'll use 12mm ply for a 6lb weigh penalty.

 

Here's the basic specs I expect to achieve:

 

Average sensivity = 97dB

-3dB @ 60Hz, -6dB @ 46Hz, -10dB @ 38Hz

+/-2dB from 70Hz to 3.5kHz, +/- 4dB to 20kHz

25" wide, 18.5" high, 15.75" deep

~ 40lbs weight

450W RMS woofer section; 250W RMS midrange

800W RMS excursion limited

 

ISP Technologies make a slightly similar but 2-way cab, though this didn't dawn on me until now, just to give you a feel for my plans:

 

http://www.isptechnologies.com/BV115Sglow.jpg

 

http://www.isptechnologies.com/bass_vector115.htm

 

Mine will be shallower and narrower, with the vertical slot port firing out of the back of the side you see in that picture, behind the handle. And the midranges will be recessed to fit the tweeter array in front of them. And it'll be a passive cab and far far lighter.

 

Alex

[_Sweet Willie_]

I haven't re-read the early part of the thread, so forgive me if I suggest something that has already been suggested...

 

Coax 8" driver + regular 8" driver. Really. Instead of the tweeter array.

 

 

[alexclaber]

Coax 8" driver + regular 8" driver. Really. Instead of the tweeter array.

 

Good point, though something I have considered. Unfortunately there aren't a lot of pro audio 8" drivers, especially lightweight ones that can produce a lot of midrange.

 

BMS make a nice 8" neo coax and matching non-coax speaker. The coax costs £165 and the non-coax £90; total cost ~£255. Also due to the drivers involved, particularly the coax horn, there are a LOT more crossover components needed and therefore the design gets more complicated and the phase and transient response less good. Also the BMS drivers need their own sub-enclosure which add weight and more significantly size. Also the tweeter would need protection circuitry whilst the piezos are basically indestructible.

 

On the other hand each Alpha 8MR costs £23.11 and six piezos total $8.64; total cost ~£50. The Alpha 8MR is also significantly more sensitive than the BMS (which go a lot lower and flatter at the expense of sensitivity but I don't need them to go low, I have a 15" for that) though the power handling isn't quite as high.

 

One of my earlier designs used the BMS 8" neos - it was a revelation to come across a crossover design that allowed me to get sufficiently flat response from the far cheaper and louder Alpha 8MRs and I couldn't believe how I could get these amazingly good value speakers to outperform the seriously high-end BMS speakers.

 

Alex

 

P.S. It seems that the £ (GBP sterling) sign is not appearing in the preview - I presume this is incorrect, if not I shall be grumpy...

[_Sweet Willie_]

Fair enough, Alex. My brief perusal of various speaker outlets/manufacturers suggested to me that the better coax set-ups tended not to be neodymium.

 

Some of my thoughts on this are influenced by the joy that is the Bag End coax loaded cabinet...

 

Peace.

--SW

 

[alexclaber]

I don't believe that the magnet material has any bearing on how the speaker sounds/performs - magnetism is magnetism is magnetism. However due to their being a bit of a shortage of neodymium and only finite R&D resources it is taking some time for all pro audio drivers to change over (but I believe 99% of them will).

 

That's why you don't see neo magnets in Acme or Bergantino cabs - because no-one sells a high excursion neo magnet 10" or 12" woofer suitable for bass guitar, and although both companies use OEM speakers they use models closely based on production models, therefore for Eminence you only have to order 50 speakers minimum. If they wanted a neo magnet high excursion woofer making it would certainly be easily possible - unfortunately due to the number of differences to Eminence's existing speaker line it would require a minimum order of 500 woofers which isn't financially viable for any company in this market. But give it time and the shipping cost of all these speakers from China will preclude ceramic magnets in all except for those wanted by descendants of alnico lovers...

 

Bag End certainly make some fine cabs - their compact 15" does a good job of disproving those that believe 15" speakers are slow!

 

Alex

[alexclaber]

Here's the predicted frequency response plot for this design with 2.83V input (1W nominal into 15" woofer but only 0.5W nominal into midranges and even less into tweeters):

 

http://www.thehumanzoo.com/alex/Speaker%20Plots/2815SLA_sensitivity.JPG

 

Below 200Hz the woofer response was modelled in WinISD Pro, all other data is based on manufacturer's measured data with crossover points/slopes factored in.

 

Alex

[_Sweet Willie_]

I don't believe that the magnet material has any bearing on how the speaker sounds/performs - magnetism is magnetism is magnetism.

 

I didn't mean to imply this. To clarify, I meant more that the selections for coax options that I've seen in the on-line marketplace seem to be more extensive for non-neo than neo -- and the quality difference lies in availability and which manufacturers are making what coax drivers at what quality levels that are neo vs non-neo (rather than some kind of inherent physics advantage of one kind over the other).

 

Also, I find really interesting reading in your thought process leading thru this cab design.

 

Peace.

--SW

[alexclaber]

Right, I've shelved that design too!

 

Here's the latest - again the Kappalite 3015LF in an 80 litre / 40 Hz enclosure for the bottom. But instead of a pair of 8" mids and an array of piezos for the top, a single 1" compression driver on a large constant directivity horn. So a shift from a 3-way design to a 2-way design.

 

The driver I plan to use is the BMS 4552ND - an 80W RMS, 450W peak, neodymium magnet, 1" throat, 1.75" voice coil driver, with 113dB sensitivity on a typical horn.

 

http://www.bmspro.info/photos/bmspro_info/4/8/8/659089.jpg

 

The horn I plan to use is the BMS 2221 - 90deg x40deg short throw, constant directivity horn, cut-off frequency of 730 Hz, dimensions: 300H x 170W x 115D mm; i.e. way bigger than anything you see on a bass cab!

 

Here's its smaller sister the 2119 which cuts off at 900Hz (haven't found a pic of the 2221 yet):

 

http://www.bmspro.info/photos/bmspro_info/9/4/2/766522.jpg

 

This means I can cross the woofer over to the horn at 800Hz, maintaining a good polar pattern across the frequency spectrum. These BMS drivers are really really nice - they get used in high-end home audio and studio monitoring as well as in very nice PA speakers. Should sound really clear and clean and go plenty loud enough. It does throw up some other challenges - like I can't use anything like the ultra simple 1st order crossover I had planned - more on that later...

 

Of course I'll only make one cab with the horn - the other will be a simple 15" cab.

 

Alex

[Rocky McDougall]

Alex, In you opinion, what confirguation has the most potential for a Bass cabinet that the average person could carry in the trunk of a car. Loud enough for small to medium venues. I have leaned toward the 1x15 with a mid driver as compared to the 2x10 simply because of the weight difference. (two heavy magnets compared to one heavy magnet) I know the Neo's are probably the way to go for weight savings. Do you think two 8" speakers could compare to one 12". The use of multiple small drivers seem to be gaining in popularity.

Rocky

[Paul K]

As far as weight savings of neo speakers, it's astounding. I've purchased an Eminence Basslite for installation in my kick drum speaker cab. THe difference in weight between it and the EV it is replacing is enough to make my wife comment that now even she would roadie for me. I haven't installed it yet, as I'm going to make a removable shelf for the Superfly amp and will do it all in one shot (gotta finish my taxes first). But damn. The weight savings is way more than I expected; well worth the money. Sound might even be minimally improved since the new speaker takes up less room in the cab. A significant amount? I dunno. But I'll post when it's done.

 

Peace

Paul K

[alexclaber]

Alex, In you opinion, what confirguation has the most potential for a Bass cabinet that the average person could carry in the trunk of a car. Loud enough for small to medium venues.

 

Rocky, as this design process has progressed I've ended up trying to design exactly that - a cab that is similarly small and light compared to my Acme Low-B2, manages a similarly flat clean response and goes louder overall. I've had to give up some low end extension to get more SPL but I think the trade-off will work well.

 

I have leaned toward the 1x15 with a mid driver as compared to the 2x10 simply because of the weight difference. (two heavy magnets compared to one heavy magnet) I know the Neo's are probably the way to go for weight savings. Do you think two 8" speakers could compare to one 12". The use of multiple small drivers seem to be gaining in popularity.

Rocky

 

I don't believe smaller speakers than 10" work for live bass guitar except for very low SPL or in very large numbers (I'm still not quite sure how the PJB cabs manage it...)

 

The sensitivity of a speaker is tied to area, moving mass, magnetic motor strength, and if you try to get a low enough resonant frequency from a smaller speaker you have to raise the moving mass too far thus lowering the sensitivity. A larger speaker can have the same moving mass and suspension stiffness to get the same low resonant frequency but because the cone area is larger the sensitivity is greater.

 

Of course as you increase the speaker size the more difficult it is to get a similar in-cab resonant frequency from the speaker to the free air resonant frequency, as the air within the enclosure acts as a spring thus raising the resonance. That's why the Whappo Grande with its 21" speaker has far less bass extension than an Acme with its 10" speakers.

 

If you want a cheap, light and reasonably compact bass cab I'd recommend a few things:

 

1. Bill Fitzmaurice Omni 10 - a horn-loaded 2x10" with either Deltalite II 2510 or Basslite S2010 speakers depending on the power you intend to use. Cheap, light, easy build, compact, loud.

 

2. ~100 litre ported cab tuned to ~45Hz with Eminence neo 15" speaker. Cheaper, lighter, less top and deep bottom, more classic bass tone.

 

3. Similar to above but with added midrange speaker and more bass oriented woofer - i.e. 3015LF instead of 3015. More clarity, more bottom, less midrange. Bit heavier, much more power handling.

 

4. The design I'm working on. More expensive, louder, slightly heavier, much clearer flatter sound.

 

5. Bill Fitzmaurice OmniTop 15 - louder than all the above but less deep bottom, still pretty light and cheap.

 

With all these designs you could add a second matching or woofer only cab for larger venues.

 

Alex

[_Sweet Willie_]

A friend of mine worked with Low Down Sound and had them build him a 3x10 w/ horn. The cab is relatively shallow (14" deep) and then has the 10s somewhat vertically aligned in the cab (21" wide, 28" tall). He had them put in higher x-max drivers, and reports that he's getting very giggable tone and volume from this set-up. Plus, it fits easily in the trunk (boot) of his car.

 

He was not going for the same even response as Claberonious Funk seems to be for this project, but did end up with something that he's quite happy with -- although if he were to get another one he might tweak the design a bit.

 

Peace.

--s-uu

 

[alexclaber]

Latest bright idea - 1/4" ply with extensive bracing, to give almost sandwich panel construction. Once I have a build space I shall buy a 3015LF and some plywood and see if I can get enough stiffness to make it viable.

 

Alex

[Paul K]

Your original idea of foam core was, I think, perfect. Much better than 1/4" with pounds of bracing. I know the cost of a sheet of it is a budget buster, but I did OK making my own foam core with both luan and 1/4" ply and a bunch of Gorilla Glue. Granted, I only neede two 22" round pieces. But it was easy.

Peace

Paul K

[alexclaber]

What kind of foam did you use for the core? (Downside of foamcore vs braced 1/4" is the extra external size - but it would definitely be lighter).

 

Alex

[Paul K]

I simply bought at a craft store this foam core that was marketed as a "story board". You could write on it with an erasable marker. It's very light, very stiff, and 1/4 inch thick. I'll stop in to the store this week, as it's in the mall I work at for more details on the exact product and dimensions. It was very, very cheap. Under $3.00 US for a two square foot piece.

 

I don't think that external size would be affected very much at all when you consider that the internal bracing also decreases interior volume. You still have to make the inside of the box as big as you want the inside of the box to be.

Peace

Paul K

[barticus]

I've also been toying with winISD to build a ported speaker cab out of a 16 or 18 inch dia. floor tom and a neo speaker. Still can't find a 4 ohm 12" neo to do it with, and I don't want to use two 8 ohm speakers. Very interested in how yours turns out.

 

Hey, I replied to your thread about the drum speaker, but didn't realize you're in Ithaca. I also toy around with drums (making electronic kits), and I'm a speaker builder. It also happens that I'll be in Newark Valley (halfway between you and Binghampton) in June. We should get together and talk shop.

[Paul K]

Baricus:

Mapquest says that Newark Valley would make a good motorcycle ride. Shoot me and e-mail when you're in town. I've got Sun. and Wed. off from work.

Peace

Paul K

[Paul K]

The board is used is "Ghostline Foam Board". 5mm thick, coated paper surface with a polystyrene core. 22 inches by 28 inches for $3.39 US (sorry to mix english and metric measurments; what kind of a scientist am I?) 32 inch by 40 inch board is $6.98 US. But this place has coupons in the paper all the time....

 

A similar product is viewable at http://www.elmers.com/products/product/prod_list.asp?prodCat=9. A related product on that page that caught my eye is the corregated plastic board, at a measly 4 mm thick. Rough up the glossy surface before you glue.

 

Build it, and they will come.

Peace

Paul K

[alexclaber]

I've been looking into this today and found that there are house building products which use expanded polystyrene sandwiched in between oriented strandboard for structural walls.

 

With sandwich construction the further apart the load bearing layers are, the stiffer the structure. So I'm thinking of something along the lines of 4mm ply for the inner and outer layers with 10-15mm expanded polystyrene between. I'll build it with square section wood along all the edges, build the outer layers first, cut the foam to fit inside and glue that up, then cut the inner panels and glue them in.

 

Also wondering about using some kind of phenolic or HDPE outer layer to give me a nice tough finish and then use right angle aluminium strip to cover the outer edges. Fiddly but tough and light.

 

Alex