Zilch's AK Design Collaborative - Econowave Speaker
- Thread starter Zilch
- Start date
I presume you saw where Jack's reworking Seville baffles to fit E'Wave in the cabs:
http://audiokarma.org/forums/showthread.php?p=2126901#post2126901
Nice layout on the crossovers! :thmbsp:
I read somewhere that somebody's making them. In the AK K forum, perhaps, or the factory main one?
[We don't use bad words here....
]
Zen Mod
New Member
I read somewhere that somebody's making them. In the AK K forum, perhaps, or the factory main one?
[We don't use bad words here....
sorry Zilch - you lost me at least twice .....
"factory main one " ?
"bad words" ?
hehe - I'm pretty well known over at DiyA , about my passion for OT everywhere ....... and I hope that's my lonely sin :yes:
:banana
, but - besides some Yidish word ( one crazy preamp there is named exactly with that word ) , I don't use any other "bad " word on net ..... 
so - I'm willing to learn how to make my own matching autoformers ; in day of my youth , I concluded that they're superior to plain resistive attenuator section....
Thanks In Advance , again .....
It may have been on DIY Audio I read it actually. Otherwise:
http://audiokarma.org/forums/forumdisplay.php?f=131
http://forums.klipsch.com/
[They use the "K" word shamelessly both places there.... ]
Autotransformers for attenuation are not so simple to use or all that superior, it turns out. JBL used to employ them commonly, but abandoned them 25 or 30 years ago. Some info over on LHF about that.
http://audiokarma.org/forums/forumdisplay.php?f=131
http://forums.klipsch.com/
[They use the "K" word shamelessly both places there....
]Autotransformers for attenuation are not so simple to use or all that superior, it turns out. JBL used to employ them commonly, but abandoned them 25 or 30 years ago. Some info over on LHF about that.
Zen Mod
New Member
WOW do these sound GREAT!!! I replaced the Passive radiators, mounted the new crossovers, and fired them up. They are fast, detailed, dynamic, and have bottom end that won't quite. The only problem is that the left channel is set at full volume on my 300B amp and the right channel is set and 3/4 volume just the get them balanced. I switched everything up and the only thing I know now to cause the reduced volume is the left crossover. I built them identical. So what would cause this problem??? P.S., I am running the horns with a 300B amp and the woofs with an Adcom GCA 510 amp. I will refinish the cabs and the grill frames in the weeks to come.
Last edited:
gogofast
Addicted Member
nice! looks like the horns fit right in.
so you said "heil no!", huh?
oh, and maybe i missed earlier posts about this, but you're bi-amping? how's the crossover configured for this? can you possibly direct me to the posts that discuss this matter? thanks.
Wayne Parham
Active Member
Attenuation verses compensation
A word about autoformers for padding. It's a great idea, but it's an impedance modifier. That's not so bad, in fact, it's a good thing as long as you design for it. But the real problem in this application is what we're looking for isn't straight attenuation - we're looking for top-octave compensation. That makes autoformers and L-Pads ineffective, or only partially effective, because equalization is required.
The equalization required is the conjugate of driver power response, so you can't just use a bypassed resistor either. A capacitor bypassed resistor would give basically a diagonal line of rising response with no initial flat shelf, and that only if the load presented by the driver were purely resistive, which it's not, a different can of worms. The ideal curve has a shelf of flat response from crossover up to about 5kHz or so, where the driver response begins to droop 6dB/octave, so augmentation is added in at that point. There are a handful of characteristics that the right circuit has to provide and deal with.
My approach has always been to slightly underdamp the crossover filter, which causes it to provide a tiny bit of peaking. Then, adding a bypassed attenuator afterward yields an overall curve that's just what we want. I designed it with Spice, so I was able to get the curve aligned just where I wanted it. I think this is the most useful circuit topology for equalizing a CD horn with a compression driver, using a slightly oversized load that provides a slightly underdamped curve followed by a capacitor bypassed attenuator. Works perfect for padding with top-octave compensation as required for CD equalization.
A word about autoformers for padding. It's a great idea, but it's an impedance modifier. That's not so bad, in fact, it's a good thing as long as you design for it. But the real problem in this application is what we're looking for isn't straight attenuation - we're looking for top-octave compensation. That makes autoformers and L-Pads ineffective, or only partially effective, because equalization is required.
The equalization required is the conjugate of driver power response, so you can't just use a bypassed resistor either. A capacitor bypassed resistor would give basically a diagonal line of rising response with no initial flat shelf, and that only if the load presented by the driver were purely resistive, which it's not, a different can of worms. The ideal curve has a shelf of flat response from crossover up to about 5kHz or so, where the driver response begins to droop 6dB/octave, so augmentation is added in at that point. There are a handful of characteristics that the right circuit has to provide and deal with.
My approach has always been to slightly underdamp the crossover filter, which causes it to provide a tiny bit of peaking. Then, adding a bypassed attenuator afterward yields an overall curve that's just what we want. I designed it with Spice, so I was able to get the curve aligned just where I wanted it. I think this is the most useful circuit topology for equalizing a CD horn with a compression driver, using a slightly oversized load that provides a slightly underdamped curve followed by a capacitor bypassed attenuator. Works perfect for padding with top-octave compensation as required for CD equalization.
You switched them and the problem followed the crossover?
Maybe one of the group can come over and help you troubleshoot.
I'm not seeing the L-pads there?
Zilch, yes, the problem is in the crossover. The only piece in the crossover that is not new is the .47uf cap. every thing else is new. I am not using L-pads because I am bi-amping and both amps have volume controls. Other than the volume issues, they sound great. The ESS go down very low and the whole sound is very balanced and flat, yet dynamic. I have a big room and these fill it as good as any system I've had.
Dnewman, I switched the channels on the amp and the left is still down in volume, so it must be the crossover. I also switched the waveguides/drivers and the left is still down in volume. Cheers.
Dnewman, I switched the channels on the amp and the left is still down in volume, so it must be the crossover. I also switched the waveguides/drivers and the left is still down in volume. Cheers.
Thank you again, Wayne!
Here's one of JBL's autoformer attenuation circuits, just before they discontinued using them. The compensaton is tagged on after the attenuation.
R2 & R3 are, let's say, "dissipative."
http://www.jblproservice.com/pdf/Network Schematics/3120A Network.pdf
Here's one of JBL's autoformer attenuation circuits, just before they discontinued using them. The compensaton is tagged on after the attenuation.
R2 & R3 are, let's say, "dissipative."
http://www.jblproservice.com/pdf/Network Schematics/3120A Network.pdf
[The Zilchster scores a big ol' "DUH!"
:]Edit: As penance, I have traced your crossover wiring from the pics, and I don't see anything amiss. I assume the black resistor in series with the 3.9 Ohm green one is 12 Ohms, and the two white ones across the white 0.47 uF are all 62 Ohms. It all kinda stacks in proximity there where the common (ground) connection is made; you may want to check that none of the component leads are touching in that area.
It's possible that the louder one is actually the defective, i.e., the attenuation isn't operating as it should....
FYI: the JBL PRO shop is out of the $9.90 waveguide for the next 2 weeks.
There are alternate dealers in US but they declined to give me contact info as I am in
Canada. They also will not ship to Canada.
There is a distributor Soundcraft in Quebec (514-595 3966). It is Canadian Thanksgiving, so will be able to phone tomorrow to find out about availability and price.
There are alternate dealers in US but they declined to give me contact info as I am in
Canada. They also will not ship to Canada.
There is a distributor Soundcraft in Quebec (514-595 3966). It is Canadian Thanksgiving, so will be able to phone tomorrow to find out about availability and price.