Dean's Super AA

[jwc]

Dean. Couple of questions regarding your new design. I'm quite sure they will be quite popular.

I'm gonna give them a try on My dbb's.

1. Is the 2.5mH a must? Noticed you are using an existing low pass filters in AK's that are second order. The 4.0mH with a cap tightens up the woofer curve a little and I am assuming this is playing little with the integration of your high pass. The reason I am asking is because for me to use previous designs on my dbb's, I am adding a second low pass. Now this presents interesting info to my amp but I am not concerned about that. I am using a second vented K33 to "catch up" to the mid and tweet. Doing this with the Standard Type A is working well (2x 2.5mH inductors). When doing this, the two woofers presents more sounds going into the higher frequencies than the one vented woofer. I wish I had a curve for the one K33 in the scala or the khorn therefore I would know exactly what the roll off is. That way I could match em with the right inductor. Take a look at my theoretical curve of one vs two k33's in a vented enclosure. There is more sound at the higher frequencies.

At this point you are asking; "Damn Cullison, what is your question". Well do you see major flaws in my using your design with a second low pass on my dbb's.

2. Trying to make a way for the Super AA to be integrated with a standard Cornscala. Now here is the thing. The tweeter on your network isn't too far from other Klipsch designs such as the AL-3 and AK-3. Now think about the B-3 for a minute. The tweeter comes off a second order where there is a 2uF cap and a 0.125mH inductor. Now this comes from tap #4 on a T2A. So here is my question:

Say use your network and keep every thing the same except for the high pass tweeter portion. Could it have its own T2A in which the high pass input is going to tap#5. Then off tap#4 goes to your 2.2uF, .15mH, 3.9uF with diode? The sqwuaker could then be down attenuated to match the low pass and the tweeter. Now wouldn't you have a Cornscala network that could cross the K401/K55 at 400hZ and the overall dB be somewhat like a Cornwall?

I'm sorry if I sound like a mad scientist here. However, sometimes that freedom of mad thinking gets me somewhere.

3. Easier question. That Super AA will allow me to try an 8 ohm Altec 902-8T on my dbb's with Altec 511b's? Right?

I may think of some more dumb questions later.[]

[D-MAN]

Let me barge in here as this is interesting to me, too.

2 x K33E's wired in series, putting you at approx. 6.5 Ohms total VC impedance. The normal 2.5 would work, I would think. That's what the Khorn uses (2.4 mH - about 6 Ohms for a single K33E in the horn). I think your low pass inductor is too high of a value.

Try a single 2.4 or so for the inductor for both drivers in series and see.

I could be wrong - the heavy-hitters like Bob, Dean and Al will be the authorities on this.

DM

[jwc]

Let me barge in here as this is interesting to me, too.

2 x K33E's wired in series, putting you at approx. 6.5 Ohms total VC impedance. The normal 2.5 would work, I would think. That's what the Khorn uses (about 6 Ohms for a single K33E in the horn). I think your low pass inductor is too high of a value, and therefore your bandpass is too high.

Try a single 2.4 for the inductor for both drivers and see.

I could be wrong - the heavy-hitters like Bob, Dean and Al will be the authorities on this.

DM

D dawg.

I considered the wired in series option. However, theoretically, the dB of the low end wouldn't match my goal. The two K33's in parallel hit arguably 104-107dbb. In series we are talking much lower. I don't wan't to lose dbb. Just a personal goal here. I see your point in trying to have a roll off like what might be with the horn low ends like a scala. However, I bet the roll off is much different for the K33 in a vented vs horn enclosure when crossed at the same frequency.

Other thing. As you raise the impedance, you will have to raise the value of the inductor. Then the "crossing" wouldn't be like what is in the classic sound of a Cornwall. However, I know that dated Cornwall models were crossed high at 1000hZ. Not sure what that sounds like.

I = Z/(2 x pi x f)

I almost went with a solo 1.25 mH for both woofers when in parallel. This would keep the crossover frequency like the Cornwall for each woofer/enclosure. However, I had a bunch of 2.5mH's around and just ran one to each woofer. Same result.

jc

[jwc]

Here is what I have so far. Waiting for the diode and Litz from Solen. The autoformer is T2A. May experiment later with the universal ones.

[D-MAN]

As you raise the impedance, you will have to raise the value of the inductor.

But theory would state that 2 x 3.4 Ohm drivers wired in series = 6.8 Ohms or so. The series will cut the output by 1/2, for sure, but 2 drivers in series=1 driver output (96.5 db normal), ignoring other losses, of course.

The main point is, the amplifier sees 6+ Ohms, exactly what it would see if

there was a Khorn on it (i.e., a K33E in the horn itself which raises its impedance to about 6 Ohms total). Therefore, I would go with the same single inductor value as used by the standard Klipsch lowpass inductors, as they are VERY close to what you would have in series.

The Fs of the drivers will not change whether wired in parallel or series. I am confused by what you're saying about that.

If 96db sensitivity (result of series connection) isn't cutting it (sounds like that's the issue) then a special shelving crossover could be designed to lower the mid/tweet output to match.

*****

But I'm sort of getting what your talking about - what about using more efficient drivers? I'm sure you considered that, too.

OK parallel K33E's - 1.6 Ohms total - that's a problem. Shouldn't the low pass inductor be lowered rather than raised? It would go down to 1.2 mH for a 400Hz point?

I'm going to shut up now, this is going over my head. I'm hoping to learn the answer from the other guys. I want to experiment with dual woofer setups, too, as you know.

DM

[jwc]

As you raise the impedance, you will have to raise the value of the inductor.

But theory would state that 2 x 3.4 Ohm drivers wired in series = 6.8 Ohms or so. The series will cut the output by 1/2, for sure, but 2 drivers in series=1 driver output (96.5 db normal), ignoring other losses, of course.

The main point is, the amplifier sees 6+ Ohms, exactly what it would see if

there was a Khorn on it (i.e., a K33E in the horn itself which raises its impedance to about 6 Ohms total). Therefore, I would go with the same single inductor value as used by the standard Klipsch lowpass inductors, as they are VERY close to what you would have in series.

The Fs of the drivers will not change whether wired in parallel or series. I am confused by what you're saying about that.

If 96db sensitivity (result of series connection) isn't cutting it (sounds like that's the issue) then a special shelving crossover could be designed to lower the mid/tweet output to match.

*****

But I'm sort of getting what your talking about - what about using more efficient drivers? I'm sure you considered that, too.

OK parallel K33E's - 1.6 Ohms total - that's a problem. Shouldn't the low pass inductor be lowered rather than raised? It would go down to 1.2 mH for a 400Hz point?

I'm going to shut up now, this is going over my head. I'm hoping to learn the answer from the other guys. I want to experiment with dual woofer setups, too, as you know.

DM

D-Man. Don't shut up. You are right. Your last sentence said:

"OK parallel K33E's - 1.6 Ohms total - that's a problem. Shouldn't the low pass inductor be lowered rather than raised? It would go down to 1.2 mH for a 400Hz point?"

Yes man. I lowered the impedance so therefore I did lower the inductor value. Each K33 with its own 2.5 inductor wired parallel is the same thing as a parallel setup using just one 1.25 inductor for both drivers..

Yes the Fs of the driver would be the same if the crossover frequency changed but the Roll off might be different.

Yes I could attenuate the mid and tweet. This would then compromise my goal with this setup. I'm not worried about the ~2 ohm presentation to my amp. The amp is a beast and probably doesn't hiccup.

It would be hard to find a more efficient driver for these cabinets. Ran a ton through WinISD and none was more efficient....more lower dB with lower Hz if I changed the tuning of the cabinet.

[Deang]

Hey, you guys are asking the same questions I would ask -- and I have others too. Let's start with below:

On 7/13/2005 1:05:22 AM djk wrote:

A) One 4 ohm woofer and a 2.5mH inductor 600hz (Cornwall).

Two 4 ohm woofers and a 1.25mH inductor 600hz (2 ohm parallel connection).

C) Two 4 ohm woofers and a 5mH inductor 600hz (8 ohm series connection).

2.83V sensitivity around 100hz

A)101dB

B)107dB

C)101dB

Dennis based his post on 600Hz, but let's put that aside for a second. I don't have any problem understanding the 2.5mH value in the horn loaded types, but it doesn't seem to make the most sense in the Heresy and Cornwall. Complicating it further is the different ways people are using the K-33 specs with the different software and calculators. We also now see Al including the voice coil inductance as part of his calculations.

If we just look at a single K-33 with a first order Butterworth -- the straight up math says 2.5mH gives us a 250Hz crossover "point", and the "correct" value for 400Hz would be 1.6mH. So I think the first order of business would be to determine whether or not to stay with the Klipsch way of doing it, or to simply start from scratch and begin working with some 2nd order alignments. If it were me, I would be using the math to get close and keep me out of trouble, but then trying some different things on that low pass section until it hit my ears right.

I would avoid using the behavior of K-33 in a horn loaded bottom to determine anything it might do in a bass reflex design. When you horn load the K-33 -- it's not a 4 ohm driver anymore!

The parallel connection bugs me. No amp, regardless of quality or design likes hovering around down there. Why are you shooting for maximum sensitivity? It's not like you're running a 2A3 SET amp -- you have gobs of quality power. You have enough power that even at 101dB/w you'll be able to drive yourself out of the room. Hey, when you hit 125db+ the top is going to start compressing, and in this case -- what does it matter if it starts doing it at 60 watts or 200? Paralleling those drivers just means you'll have max SPL at 60 or 70 watts, and your amp will be so hot you'll be able to cook a steak on it.

Edited November 8, 2015 by Deang

[jwc]

Dean, yes those statements by Dennis were helpful back in my dbb thread.

"Two 4 ohm woofers and a 1.25mH inductor 600hz (2 ohm parallel connection)." This is exactly what I did except I used two 2.5 inductors for each woofer in parallel. He agreed that is the same thing.

When trying to rationalize exactly what PWK/Klipsch's intention were with that 2.5 inductor for the K33 in a bass reflex enclosure was a mistery to me. What is more a mistery is then using the 4.0mH with a parallel cap for a second order low pass filter for the B-3.

As far as going with what the ears want...Is a great idea. I need to master the whole autoformer thing a little more before i get better with that.

As far as the 2 ohm load and my amp. Yes this sounds unconventional and I knew it would raise eyebrows. Eventually, I will tinker with this cabinet with two different 8 ohm woofer like D-man wants. That would be ideal.

My amp doesn't even get warm. And I really don't turn it up too loud. The plan in the end was to get a tube amp, have a efficient speaker. Put that big MAC for use with another speaker.

All points well taken.

I'm seriously thinking about that mod for the Super AA to be used with a Cornscala

jc

[BS Button]

I'm going out to rake some leaves.

[mark1101]

IMHO, what I think would be a great idea is if you guys came up with a modern design for the AL-3.

2nd order woofer

2nd order squawker IN (add a 1st or 2nd order Squawker OUT, this would be the major improvement)

extreme slope tweeter

Great sounding network stock, but definitely gets harsh once pushed. Need to get the squawker outta there on top, and have the ability to attenuate the squawker like you guys already do now.

Sounds like you guys are rethinking everything anyhow now.

[]

[Deang]

IMHO, what I think would be a great idea is if you guys came up with a modern design for the AL-3.

Well, for all intents and purposes the AL-3 is a "modern design". If you build it up with better caps it takes a pretty good leap forward in sound -- but still has a bit of a hollow sound through the midrange.

2nd order woofer

2nd order squawker IN (add a 1st or 2nd order Squawker OUT, this would be the major improvement)

extreme slope tweeter

The elliptical tweeter section of the AL-3 is an "extreme slope" of sorts. As for a network with a second order squawker section -- it already exists -- it's called the Universal Type A Replacement, better known as the ALK. I also like the sound of a 2nd order on the woofer. I'm seriously considering a hybrid type network incorporating the low pass of the later networks with the squawker and tweeter sections of the SAA. Tempting, but it sounds so good the way it is I hate to mess with it.

Great sounding network stock, but definitely gets harsh once pushed. Need to get the squawker outta there on top, and have the ability to attenuate the squawker like you guys already do now.

All the stock networks cause the squawker to fall apart when pushed. I think Klipsch should turn the listening rooms over to those who still have all of their hearing intact -- I use my wife and daughter.

[joshnich]

Hello Dean

I know that this post will certainly illustrate my complete and total lack of technical knowlege....but what do you call the new networks that you built me. I know they are different from your Jensen based AAs that they are replacing but I get confused when i read these posts not being quite sure if you guys are talking about my new networks!

Josh

[Deang]

Hi Josh -- yes, you have the 'Super' AA. Yours do differ from my standard build in that I put some very nice 40uF AudioCap PPMF's on your boards.

[joshnich]

OK then. I cannot comment on the techincal aspects but I can certainly comment on the sound.

When I replaced the original AAs with Deans AAs ( I believe they have the stock specs but were built with improved parts) I could not believe the improvement. In my miind this was one of the most significant improvements in sound that I have recieved with any upgrade. Way more significant then tube rolling or even cartidge upgrades.

I almost didnt believe Dean when he told me that his super AAs would be a noticable improvement. He was so right. These are amazing. If you have heritage speakers and you have not updated the networks you are misssing a great deal.

If you have updated your vintage amps (scott,dynaco etc) you know what replacing old caps can do. The same is true with these networks.

Josh

[Audible Nectar]

Question, Josh:

When you replaced the original AA with the Super AA, was the original AA ORIGINAL, or had they been updated by BEC or otherwise?

[joshnich]

The super AA was in place of a dean built AA that featured Jensen caps.

J

[Deang]

...and before that?

[NOSValves]

OK then. I cannot comment on the techincal aspects but I can certainly comment on the sound. When I replaced the original AAs with Deans AAs ( I believe they have the stock specs but were built with improved parts) I could not believe the improvement. In my miind this was one of the most significant improvements in sound that I have recieved with any upgrade. Way more significant then tube rolling or even cartidge upgrades. I almost didnt believe Dean when he told me that his super AAs would be a noticable improvement. He was so right. These are amazing. If you have heritage speakers and you have not updated the networks you are misssing a great deal. If you have updated your vintage amps (scott,dynaco etc) you know what replacing old caps can do. The same is true with these networks. Josh

I'll second this post 100%. I went from original type AL(horrible), original parts Type AA better, Wired Type AA as A with stock parts, Bob's type A's (increased clarity but horribly sharp when the SPL went up), Type AA's with Auricaps/Hovlands from Dean (noticeable improvement) and finally Dean latest greatest with all top quality parts this is a real hit! Sounds wonderful at any volume level.

I also briefly had some <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" />ALK Universal networks but that was so long ago my system and I myself were so different I really can not comment one way or the other.

I'm currently trying to make a modern Klipsch Type AL3 compete with Deans latest greatest but no luck so far. But I would put it over all above except Dean's.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Craig

[jwc]

Well, I put this network together last night. I used a big board for starters so I know the look isn't that good. I also didn't use the terminal posts at all the connections because this is a trial. On these photos, the squawker taps are off. Granted, I didn't use the two Hovland 20uF. I used the solens instead. I also used a 2uF cap instead of a 2.2uF cap on the first cap for the tweeter. Will add the .2 later.

I ran them on the dbb's of mine which have the K55/511b and JBL 2404's. I also added a second low pass filter (not shown) for the second K33.

I must have done something wrong. The tweeter and woofers sound right but the squawker is "muffled" or "recessed" sounding. I used the taps from -6.7dB all the way to -2.6dB. The -2.6dB sounds the best but something still isn't right. I must have done something wrong here.

I also have never messed with the diode before. I hope that is hooked up right... It is possible that this just won't work with my dbb setup.

[jwc]

.