Regarding bass at low levels:
Smaller speakers, "hifi" speakers if you will, often have more baffle step compensation than large high-efficiency systems. Part of the challenge is that they're not capable of the same high output- so while their optimum FR is tuned for more modest levels with tilted up bass (and sometimes treble), where the bigger, more efficient speaker using pro drivers is able to play significantly louder, cleanly. Accordingly people run them louder, and less BSC is used so they don't sound "Boomy" due to the higher perceived bass at the increased volume level.
I hope that's clearer than mud.
Smaller speakers, "hifi" speakers if you will, often have more baffle step compensation than large high-efficiency systems. Part of the challenge is that they're not capable of the same high output- so while their optimum FR is tuned for more modest levels with tilted up bass (and sometimes treble), where the bigger, more efficient speaker using pro drivers is able to play significantly louder, cleanly. Accordingly people run them louder, and less BSC is used so they don't sound "Boomy" due to the higher perceived bass at the increased volume level.
I hope that's clearer than mud.
Earl,
I would be interested to hear your take on what JBL did with LSR6332? how do you think their approach compares to yours, since they also feature a 12incher at the bottom but mid and top are obviously a no-horn approach.
p.s. btw as for Lynn's design in this thread I consider it similar to yours only one size bigger (and for that matter too big for what I could practice in my home).
It just looks like a standard 3-way design - with all the problems associated with this approach. I don't see anything unique, despite the verbiage.
Based on the specs, the JBL LSR6332 looks like a well-designed 3-way system. On and off-axis responses are very smooth, and impulse response is consistent with a 3-way LR4 crossover.
It's not a particularly high-efficiency speaker at 90 dB/1 watt/meter and 4-ohm impedance, but can accept 200W average, and 800W peak. Definitely intended for use with a high-power solid-state amplifier in a professional monitoring context.
If you used a tube amplifier, 60W/channel would be a rock-bottom minimum, and more would be better. The passive LR4 crossover (and vented bass alignment) would demand an amplifier damping factor higher than 40 (medium to high feedback).
Diffraction reduction is modest, but the impulse response looks pretty good, so the overall approach looks valid. After subtracting the usual marketing puff, it looks like a well-thought-out design, which puts it ahead of many speakers sold into the audiophile market.
What does it actually sound like? I have no idea. I'm not a JBL fan, but I can't argue with the measurements that JBL has posted.
It's not a particularly high-efficiency speaker at 90 dB/1 watt/meter and 4-ohm impedance, but can accept 200W average, and 800W peak. Definitely intended for use with a high-power solid-state amplifier in a professional monitoring context.
If you used a tube amplifier, 60W/channel would be a rock-bottom minimum, and more would be better. The passive LR4 crossover (and vented bass alignment) would demand an amplifier damping factor higher than 40 (medium to high feedback).
Diffraction reduction is modest, but the impulse response looks pretty good, so the overall approach looks valid. After subtracting the usual marketing puff, it looks like a well-thought-out design, which puts it ahead of many speakers sold into the audiophile market.
What does it actually sound like? I have no idea. I'm not a JBL fan, but I can't argue with the measurements that JBL has posted.
Last edited:
Sorry. Alnico magnets just ain't cheap.
Unless you're looking to replicate the BTA's driver complement with a high degree of precision, you're just in the 15" (or two), subwoofer (or four), big horn, and optional super tweeter vein that's been mined since Fred Flinstone was in diapers. It may share some similarities with what the folks here are building, but it won't be the same thing. Based on recent experience, I can say with some confidence that there is just no direct substitute for an alnico Altec.
You do have three things going for you if you want to spend less money. The first is that the GPA 416-8b isn't the only alnico option (ancient Altec and early JBL are nearly as competent). The second thing you've got going is that Lynn and g3dahl haven't yet publicized their crossovers or cabinets, so you're on the hook for a ton of design work no matter what you buy. The third thing you've got going is either ebay or the estate sales in your local newspaper's classified ads.
Frankly, the driver prices pale in comparison to the cabinet work, the crossover requirements (autoformers ain't cheap), and the parts you'll eat through while dialing in your design. This simply isn't a project for folks on a super tight budget. You'd almost certainly be better off checking out the work from the EconoWave folks, the SEOS guys, or Wayne Parham at Pi Speakers, all of whom have fielded proven designs with a much lower price point than the (as yet) unfinished BTA.
I keep seeing where the builders crossover too low for the midrange horn. If they measure out a couple of meters they should see the hole in the response. Maybe that's why it's not done yet? Maybe the Ariel actually sounds better and Lynn gave up and is enjoying the old speakers? Need a new thread Back to the Ariel?
Just out of curiosity, what is the basis for your alleged hole? IIR, you've mentioned the highpass (as opposed to the upper xover) being too low, but I don't see how that would, by nature, account for the alleged hole.
Don't measure the low end of your midrange horns nearfield to bild your crossover, measure them where you listen to them at. In the dozens i have and measured the response typically drops buy an octave or more when your out three or for meters compared to up close. When you crossover based on the nearfield measurement you end up with a hole that gives the illusion of presence and detail but becomes irratating and fatiguing over time. There are posts within this thread by builders of this BTA that clearly show this but as far as i can tell it has gone unaddressed. The horn/driver combination I would cross at 1000 or higher and work on getting the appropriate midbass horn to match below that. Stretching a horn too far above or below it's linear operating range as measured at the listening position can really set you back in wonder.
I agree with POOH's sentiments. Going below the the boundry reinforcement (h,w) causes a loss farfield. With an Le C'leach horn JM specified an octive above this cutoff region.
Typically you'll notice that a horns h, w dimension corresponds to a physical wavelength e.g. 8" waveguide = ~1700hz as you approach this frequency the reinforcement cuts off and creates an antiphase artifact that is cast into the far field and unseen in nearfield measurements. Many go as low as the spec calls for and the resaultent pinching of the far field results. Proper use would allow a cross point at least 2/3rds of an octave above or =>2700 as the new minimum for this example.
Typically you'll notice that a horns h, w dimension corresponds to a physical wavelength e.g. 8" waveguide = ~1700hz as you approach this frequency the reinforcement cuts off and creates an antiphase artifact that is cast into the far field and unseen in nearfield measurements. Many go as low as the spec calls for and the resaultent pinching of the far field results. Proper use would allow a cross point at least 2/3rds of an octave above or =>2700 as the new minimum for this example.
I'm sorry guys, but I have done my share of measurements as well, and I know the theory as well as anyone else too. I don't buy the argument being proposed here, not as I understand it being stated. There is no "anti-phase effect" below the mouths dimensional polar control limit. It narrows and then widens that's all. And this does not change with distance either. I will need either some clarification here or some supporting data.
In your Gedlee speaker you use a baffle (should help the low response of your waveguide?) and compensate for the midrange peak (I image it's rather large) with the crossover to extend the response beyond the linear loading range of driver. Sure you can do that because you use a direct radiator bass system and have a lot of room for eq. I am not sure how the BTA people handle this because i really haven't seen the crossover. I can be done the same way I suppose since they use direct radiation for bass too - might have missed it. Is the crossover for the gedlee speaker designed using nearfield measurements? Have you measured your waveguide alone at different distances?
No, I never use near-field, and yes I have measured at different distances. It changes somewhat very close but beyond about 1 meter there is no change.
So I am guessing from your response that you are not saying that ALL systems like this will have a hole, just that some could. That I agree with because it all depends on the relative phases at the crossover and how that is handled. When dealing with waveguides one MUST use real measurements for the crossover design, with accurate phase representation - classic approaches to crossover design do NOT work for a waveguide to direct radiator and if not done properly there can be a hole. But that is simply a poor design and not a requirement.
What I am saying is the horn used in this BTA system is actually pretty small and is mostly lip. I am saying using similar size horns (minus the lip) and other types including waveguides, radial, tractrix, hypex, biradial I have found measuring in near field indicates the horn can be operate at a lower usable frequency then what you get when you get back a to a normal listening distance. The size horn used here probably should be crossed over higher than the 700 cycles based on what I have observed here and I wonder if the people building the systems in this thread have tried a higher crossover point and what they have found doing it. Just making an observation and being curious.
That's fine, I can't say that I know how well the design shown here works - its all in the details - wish I had some good measurements of it. I just objected to the "hole" comment as I didn't see why that would be true. A too small horn usually just affects the polar aspects as on-axis problems can almost always be corrected - but the polar ones cannot.