Zilch's AK Design Collaborative - Econowave Speaker

[Russellc]

EWG: Gentlemen, THEY'RE ALIVE.
I shall reserve my judgement until I complete more auditioning.

So what's the verdict? Dont be shy, search this thread under my moniker....I've posted all sorts of half cobbled up messes that look like a 3 yr. old did....come on and pics please!

Russellc

 

[Russellc]

Find the frequency where the wavelength of the sound wave is equal to the smallest dimension (usually the width) of the baffle. That will be the point where the center of the woofer (if centered on the baffle) will be 1/2 wave away from the outside edge of the cabinet. At any frequency lower than this, the wavelength will be BIGGER than the baffle dimension, and the sound waves will no longer be "supported" by the baffle... they will 'wrap around' the cabinet, which will reduce the effective output to the front of the speaker.

In scientific terms, we call this the "2-pi to 4-pi radiation transition frequency". In layman's terms, it the frequency where the bass drops in level because of the loss of baffle support. In short, it's known as the "baffle step frequency"... because, if you graph it, it looks like a "step" in the response curve at that frequency.

Fortunately, it's about a 6 dB step, in theory... pretty much equivalent to the output of a second woofer added in parallel to the first. So, if you pick a crossover frequency for the second woofer EQUIVALENT to the frequency where the baffle is 1 wavelength wide... you will be in the ballpark.

IME, I've had good luck with a sort of modified approach. Tune both woofers to an EBS vented (long-tail rolloff vented box) or low-Q (near Qtc of .5 or so) sealed alignment, and use a lower frequency lowpass crossover on the .5 woofer (about 1 octave below baffle-step). This makes for a more subtle baffle-step compensation... more like 2-3 dB rather than the full 6 dB. I find this works better for speakers in real rooms... where the speaker is still somewhat "supported" below the baffle step by proximity to a wall behind the speaker cabinet (which "restores" to an extent, the 2-pi baffling effect below the nominal baffle step frequency of the cabinet)...

In short... there's no hard-and-fast rules on what will sound best in your particular ".5 woofer" setup... but these will give some good guidelines on where to start...

Regards,
Gordon.

Great explaination gordon, and thanks for explaining it not only in technical terms, but then giving it a "real world" explanation for those like myself, who are a little more technically challenged.

I bet the smaller systems Zilch and Skywaverider are doing may be great candidates for this, and still maintain their small footprint.

Zilch and Skywave, how low do the smaller guides support the Selenium? or do they require a driver with lower response?

Thanks in advance,

Russellc

 

[GordonW]

Great explaination gordon, and thanks for explaining it not only in technical terms, but then giving it a "real world" explanation for those like myself, who are a little more technically challenged.

I bet the smaller systems Zilch and Skywaverider are doing may be great candidates for this, and still maintain their small footprint.

The nice thing about ".5 woofer" alignments, in addition, is that these allow the use of efficient woofers- with lightweight cones, and improved HF behavior- instead of one woofer, with a heavier cone, in order to achieve the same low-frequency response behavior. This allows the woofer that is doing the midrange to perform better into the upper part of its range.

This is why many people like the JBL 4435 studio monitors so much... they were one of the first ".5 woofer" systems to come to market. In those, they took similar cabinet volume to what would be required for a single woofer to play that kind of bottom end (a single 2245 18", in essence, as seen in the 4345 studio monitor) and substituted two smaller woofers with lighter cone mass (the 2234- basically, a 2235 WITHOUT the additional bass-extending and efficiency-reducing mass ring required for low end response from a single woofer). This allowed the 15" woofer to perform pretty admirably well, crossed over at about 1KHz to the bi-radial "butt-cheeks" horn... where as, in other systems, the 2235 is usually used at 500 Hz and lower... at least in the higher-end studio monitors...

Regards,
Gordon.

 

[skywave-rider]

...

I bet the smaller systems Zilch and Skywaverider are doing may be great candidates for this, and still maintain their small footprint.

Zilch and Skywave, how low do the smaller guides support the Selenium? or do they require a driver with lower response?

Thanks in advance,

Russellc

Zilch has said he believes working out the compensation for the 8x8 90x50 waveguide is worth pursuing.

I am using one set with standard Econowave x-overs and D220ti, and another active with BMS 4552nds.

I put the Econowave versions in a college radio production control room as nearfield monitors. Right now the x-over is breadboarded, but I intend to build circuit boards for the x-overs as Jack shows, and hopefully update the compensation at that time. They sound very good, and should hold up to student abuse as well.:nono::drunk::nerd: We'll see.

The BMS versions stay at home as nearfields. Those are the ones with the JBL LE8T-Hs.

 

[RayW]

Yeah, what Gordon said. I've never tuned one 2 octaves below baffle step, but it's room/placement dependent and if Gordon says that works for him then I'll check it out next time I do one.

Ray

 

[Zilch]

Zilch and Skywave, how low do the smaller guides support the Selenium? or do they require a driver with lower response?

In theory, it's the opposite, actually. The smaller the waveguide, the less pattern control it exhibits at lower frequencies. At some dimension, it becomes more suitable for use as a "tweeter," and any lower frequency capability of the compression driver goes unused with the requisite higher crossover frequency.

At these preliminary design stages, I look to how JBL uses the new waveguides in actual product; it's a good bet they know what they're doing. I am posting links to the dual-woofer versions this time so that those with interest in pursuing the "0.5" approach can also see how JBL has implemented it in these recent designs:

http://www.jblproservice.com/pdf/AE Series Compact Models/AC26-WH.pdf

http://www.jblproservice.com/pdf/AE Series Compact Models/AC2895-WH.pdf

In the single-woofer versions, they have the 6" square waveguide crossed lower:

http://www.jblproservice.com/pdf/AE Series Compact Models/AC16-WH.pdf

http://www.jblproservice.com/pdf/AE Series Compact Models/AC1895-WH.pdf

These products define a range in which the new waveguides might be used successfully, in the context of the particular drivers and filter topologies JBL employed.

2408H-1 is a tweeter driver. How low will the 8" square waveguide play nicely with a more general-purpose driver like E'Wave's Selenium?

Measure, then listen.... :yes:

 

[GordonW]

Yeah, what Gordon said. I've never tuned one 2 octaves below baffle step, but it's room/placement dependent and if Gordon says that works for him then I'll check it out next time I do one.

Ray

The last one was a special case... huge box, very wide, with short legs. So, it probably was "floor baffled" better than most... and therefore, needed less boost.

I also had a low-Q lowpass on that woofer... which tended, probably, to "droop" the top octave (500-1K) a tad.

It does depend on a number of factors...

Ray, do you use first or second order filters on the ".5 woofer"? IIRC, I used a first-order filter on that driver at the low frequency (90 Hz or so) and then paralleled that whole setup with the woofer that was doing more "full range", on the same low pass filter as the "full range" woofer (so that the ".5 woofer was 6dB/oct. lowpass at about 90 Hz, transitioning to 18dB/oct. at 1.1KHz or so)... that might also make for a few differences...

Regards,
Gordon.

 

[RayW]

I use a 1st order ".5" because it the shallower slope is closer to the slope of the baffle step. I also don't use a full 6dB step. Doing in-room measurements shows that, in the cases I've measured, something like 4dB is generally about right.

Ray

 

[GordonW]

I use a 1st order ".5" because it the shallower slope is closer to the slope of the baffle step. I also don't use a full 6dB step. Doing in-room measurements shows that, in the cases I've measured, something like 4dB is generally about right.

Ray

Unless you're outside, with the speaker on a pole... that's about the only time I've ever felt that full 6dB step was warranted. Any sort of boundary reinforcement, and somewhere between 2-4 dB has been my "comfort zone" too.

Regards,
Gordon.

 

[skywave-rider]

I can't do polars or anything fancy, but here are RTAs of 8" square waveguide on BMS 4552nd. First is with active crossover at 1600 Hz, where I'm using it, second with crossover dialed down to 500 Hz, for comparo. The bump centered at 10kHz is shown to be a problem with my mic. Taking that into consideration, RTA is pretty flat.

This is only to help show low end limit, compensation and driver is not Econowave. And I don't have my Seleniums (Selenia?) here right now.
I had comparable RTA of the Econowaveguide, but have not located it yet. But there's tons of it posted here anyway.

 

[Zilch]

That says the waveguide's good down to E'Wave's nominal 1.6 kHz, on-axis, at least.

Do you have the 6" square ones for a similar test, Skywave?

 

[skywave-rider]

Unless you're outside, with the speaker on a pole... that's about the only time I've ever felt that full 6dB step was warranted. Any sort of boundary reinforcement, and somewhere between 2-4 dB has been my "comfort zone" too.

Regards,
Gordon.

Gordon,
If you're using a first order filter for a woofer running in .5 mode, how does one control the level of step if the full 6dB is not warranted? I would guess it's where you set the .5 frequency, or is that too simplistic?

 

[skywave-rider]

That says the waveguide's good down to E'Wave's nominal 1.6 kHz, on-axis, at least.

Do you have the 6" square ones for a similar test, Skywave?

Yes, but they are the 90x90s, are those the ones you mean? Regardless I'll post a similar for those.

I tried them in an active 3 way and found the HF to rise like crazy. Took no pictures.

For the 3 way test I switched to the 8x8. Had no mid horn drivers other than what you see in the pic.

 

[Zilch]

Yes, but they are the 90x90s, are those the ones you mean?

Yup, 90° x 90°, jus' like what's in 4425, 4430, and 4435....

 

[skywave-rider]

6x6

Here's RTA, 1 meter on axis for 6x6, 90x90 waveguide -- better than I remembered. It's not raw, same compensation to compare to 8x8, but I can post raw if u want. First 1.6kHz, then 500Hz.

 

[Zilch]

Here's RTA, 1 meter on axis for 6x6, 90x90 waveguide -- better than I remembered.

COOL!

Looks like we're good to go with those, too.

I'll run them with Seleniums and E'Wave crossover for comparision.... :thmbsp:

********

Of note: Skywave's in New York City, I'm in California, and Jack's in Arizona. We all have the same test gear, and can compare measurement results via AK.

This is a "No Wank Zone."

[Well, Wank can play, too, when he gets with the program.... ]

 

[skywave-rider]

COOL!

Looks like we're good to go with those, too.

I'll run them with Seleniums and E'Wave crossover for comparision.... :thmbsp:

:beer:

 

[Kegger]

Ray, Gordon, anyone else who may have experience,

Anyone try or know the ramafications of a .5 woofer situation using 2 different woofers?

Maybe like say a 10" high efficency woofer that goes up high nicely then use say a 12"
or 15" "real" bass driver that digs deep maybe not efficent as the 10" for the .5 woofer.

Some kinda situation like that at least?

 

[jackgiff]

Looks like I am missing lot's of fun. I am caught up replacing the electrolytic caps in a Sansui G-8000 for my nephew in Washington. Big heavy tub, but a very well made piece of equipment.

Maybe the design of the 6 and 8 inch square waveguides will be done when it is time to finish my HT setup. That would be COOOOL!

I can't imagine the 6 inch would play down to 1.6 KHz. Have you seen any similar sized waveguides or "horns" play that low, Zilch?

 

[Bauhausler]

Altec Seville??

Quick question because I know someone will know this answer. Will the EWave waveguide fit in the area above the woofer in a Seville baffle or does a new baffle board have to be made?

Thx