Let's design a 15" sealed cab

[JimmyM]

So, are we set on the Beta 15?

I'm using the Delta, but you go right on ahead.

 

[George Mann]

I'm confused about your goals. Are you asking for design information, like in the title "
Let's design a 15" sealed cab"

or speaker driver recommendations/discussion?

If you want design information, but prefer art, you may continue to disagree with the established cab designers.

I am asking for people to weigh in on all the parameters of the design goal. I am also an engineer by the way. But do to my unorthodox way of thinking (outside the box), I will be forced limited my own contributions.

 

[George Mann]

I'm using the Delta, but you go right on ahead.

We need accurate information on this driver.

 

[JimmyM]

We need accurate information on this driver.

That lets me out! I just put them in my B-15 cabs because Bruce at fliptops.net said it works well in B-15 cabs.

 

[FerruleCat]

Is WinISD telling me right that the Beta in 3cf is excursion-limited to about 50W? Response seems to fall off pretty quick below 90Hz, too. I used WinISD's provided driver database.

 

[wisconsindead]

Science!

Why try to design a sealed cab? just for fun? Shouldn't the goal of all cab designers be a cab that has no voice and reliably reproduces the input signal (i.e. who cares if its sealed or ported)? There may be inherent advantages to designing a voiced cab, but otherwise I think people should focus on designing cabs that have nice flat response curves.

 

[seamonkey]

And I have found that good sound is as much an art as it is a science!

Science!

Why try to design a sealed cab? just for fun? Shouldn't the goal of all cab designers be a cab that has no voice and reliably reproduces the input signal (i.e. who cares if its sealed or ported)? There may be inherent advantages to designing a voiced cab, but otherwise I think people should focus on designing cabs that have nice flat response curves.

+1
Here are some musings by Sean Olive. he says they can predict listener preference to 86%, but read down in the comments where he says they can actually get this to 100% by controlling context effects.

http://seanolive.blogspot.ca/2013/04/the-relationship-between-perception-and.html

Might as well start with science. You can always degrade with the EQ knobs

 

[iiipopes]

Have you played a B-15R?

No. I played a B15 when it was still a fairly new amplifier. What makes a B15 work is a low-wattage highly compliant speaker cone that has as low a resonant frequency that could be engineered at the time. I'm not going back.

 

[wisconsindead]

+1
Here are some musings by Sean Olive. he says they can predict listener preference to 86%, but read down in the comments where he says they can actually get this to 100% by controlling context effects.

http://seanolive.blogspot.ca/2013/04/the-relationship-between-perception-and.html

Might as well start with science. You can always degrade with the EQ knobs

Thanks for sharing that article. Interesting study and results. BTW I love your avatar

I like that people want to be innovative and creative (and this does at times bring about progress) but at the end of the day I think the big picture should be addressed. You wanna F*** around, fine idc, but I will probably think you're rubbing your pocket. There is a thread going on right now about poor bass output from an E string and everyone prior to myself started suggesting brand A and B and C and no one seemed to mention that the poor output almost certainly has something else involved and probably isn't related to the strings at all. So in a similar fashion, lets address variables, whats the actual goal, how can we obtain what we want? But merely suggesting drivers without addressing the real question at hand seems to be a waste of time. Then again I'm the one sitting on the internet Cheers.

 

[Woodstockz]

The top 2 lines are a part of good science and engineering. However, listening tests are not used to verify the science. That's already verified. Instead the listening tests are used to hear what it sounds like. The science and models do not in themselves predict how we will perceive the sound - tone. It takes experience to understand what a particular modeling result is likely to sound like. Fine tuning is as much trial and error (art and intuition) as it is recalculating. The best engineers and scientist I have known were very creative and artistic. However, they did not doubt fundamentals. There is a very wide range in tonal preferences, and therefore a wide range of opinions on the significance of modeling results.

In believe that it is all good science and engineering. I have used this methodology for over 40 years.
So #3 is not good science and engineering? - Measurements with test equipment? Then how do you troubleshot? How do you refine your design? How do you refine your models? I have had to clean up after people that only went as far as #1 and 2.
#4? - Imagination? Einstein said that imagination is more important than knowledge.
#5? - Challenging dogma? Kepler...
Sorry for tangent.

Anyways, for the OP, check out the Ampeg 86-167-08, the 15" speaker used in the B-15R.

 

[Woodstockz]

No. I played a B15 when it was still a fairly new amplifier. What makes a B15 work is a low-wattage highly compliant speaker cone that has as low a resonant frequency that could be engineered at the time. I'm not going back.

B-15R has a sealed cabinet and 4x the power of the early B-15s. Totally different design.

 

[AstroSonic]

Is WinISD telling me right that the Beta in 3cf is excursion-limited to about 50W? Response seems to fall off pretty quick below 90Hz, too. I used WinISD's provided driver database.

You are discovering one of the potential problems with sealed cabs: the excursion limited power handling plummets just below the system resonance, then recovers to some extent. My modeling runs for the beta 15A in 3 cf sealed show: F3 56 Hz, F5 52 Hz, a peak thais is up 1 db at 70 and 210 Hz, and maxes out at +2.8 db @ 92 Hz, and has a power handling sag (below the system resonance) of 85w@64Hz. Given that the fundamental is typically something like 20% of a notes total energy, the 85w excursion limit would be reached when the total note energy was around 400w. This may also explain why the Delta 15a (with a 2.7mm Xmax) seems to have no problems handling bass. Some playing styles do produce sub-bass transients with a lot of energy. A high pass filter should solve that issue. The 2.8 db peak may seem excessive to some and too little to others. It is something I have found that I prefer (a little peaking), as it produces a rich bass that is still articulate. It also allows the use of a reasonable size cab volume (3 cf). As peaking goes higher than around 3 db, the note definition begins to suffer. Fat bass territory seems to be in the 3-5 db range. Above 5 db of peaking the bass becomes soft and pillowy, eventually exhibiting audible boom - a resonance with a peaking and decay characteristic that is audibly different from the instrument (the notes can blur into each other), and the harmonics are smothered.

I have modeled a bunch of drivers for sealed cab use and the Beat 15a is one of the best regardless of cost (so it's a great buy as well). If you are going for maximum deep bass extension and power handling, a ported cab is a much better approach. However, for a tight, punchy, responsive tone a sealed cab can deliver the goods, as long as a player doesn't get carried away with boosting the bass, or doesn't try to play a stadium...

For those desiring a flat (hifi) response and more power handling in the deeper bass, the CB15-8 will meet the need (2.5 cf: F3 72 Hz, F5 62 Hz, and a power handling minimum of 180w@40 Hz), although many will find it lacking in highs. The Delta 15LFAs also a good alternative between the Beta 15a's 2.8 db peaking, and the CB15-8's flat response: 3 cf, F3 62 Hz, F5 53 Hz, 1 db of peaking at 100 & 130 Hz with maximum peaking of 1.2 db @ 115 Hz, and a power handling minimum of 165w@ 50Hz. In terms of sound, these have less than 4 db more output capability in the low bass than the Beta 15a. There are some 18 inch drivers that look very promising, but need a mid.

Lot of choices. Some generalizations: peaking can be increased by using a smaller box, or decreased by using a larger box. With a given driver F3 is higher in smaller boxes and lower in larger boxes. You have to juggle peaking and F3, by changing the box volume.

 

[FerruleCat]

Thanks very much for the informative post. Much to learn!

 

[Buzz E]

I can't believe Linnen hasn't popped in with his insights.

 

[AstroSonic]

So in a similar fashion, lets address variables, whats the actual goal, how can we obtain what we want? But merely suggesting drivers without addressing the real question at hand seems to be a waste of time. Then again I'm the one sitting on the internet Cheers.

You're a little late, but spot on - from post #12:

People need to decide what they want performance-wise in a cab and design accordingly. That would include the tonal quality of the bass (articulate, well defined, rich, fat, etc.), bass extension (do you want the deep fundamentals, or how deep do you want fundamentals to be reproduced), and how much power must the cab take to produce the volume you need (how big is the venue, and how loudly do you play). You also need to consider the mids and highs (will a mid or tweeter be needed), and their impact on the tone of the cab.

Until the above questions are answered there is little point in working on a design (except maybe to learn, and that is a good one!).

These questions really are important.

 

[Midak]

Using any suitable (Qts and power handling), readily obtainable off-the-shelf driver! It should be balanced as much as possible between reasonable size and weight vs. performance.

Any current or seasoned designer/builder is encouraged to participate.

Since I am currently designing a 15'' (ported) cabinet, you may want to have a look here.

 

[kringle77]

I have personally used the 3015lf, Basslite 2515, delta 15lf, delta pro15 and a few others. Delta pro wasn't bad but alittle lean in the lows, delta 15lf has good power handling but kind of blah sounding, the basslite was really pretty good but needs some top end help and only comes in 4 ohms (other wise I would say to add a 4" mid) and there have been some others. My opinion is that you put two beta 15's in a 4 cf box and call it a day. It seems like the best overall in most ways.

 

[gumtown]

Any unusual/radical physics defying high power/small size ideas?
Can you do a sealed isobaric box + mids, and passive filters to smooth peak responses?

 

[iiipopes]

B-15R has a sealed cabinet and 4x the power of the early B-15s. Totally different design.

Still not going back. Unless you want to carry it to the gig and set it up for me.

 

[George Mann]

Thanks for your input AstroSonic. BTW, I do respect the contributions that you and Duke have contributed to this board. My only argument with you was over how to handle the rather unruly behavior of the Alpha 15, and had once again failed to realize that it was beyond the abilities of my fellow TBer's (not meant as an insult BTW).

So once again, the Beta 15 in a 3-cu.ft. box (damping?) with high-pass filter (where to set?), and for those wanting a flat hi-frequency response, add 2 notch filters of 4dB @ 1KHz, and 9dB @ 1.9KHz which should provide for a smooth roll-off.