Which crossovers should I get?

[greg928gts]

I'm ready to buy a pair of crossovers for my Khorns. I use SET tube amps, a nice CD front end, and listen to rock/jazz at as high a volume as 2 watts into Khorns will produce. So far I've preferred the A crossover to the AA.

Should I get ALK's, Dean's I dunno what you call em crossovers, or ALK ES (extreme slope)? Cost is not an object here, I don't believe there's enough of a difference in the prices of them to matter to me.

Thanks for any input.

Greg

[Deang]

LOL.

Do you have a set of A's?

Change 13uF to 40uF, and add a .20mH inductor in series with the cap and connect back to tap 5.

Change 2uF to 3.3uF, and hook this leg back to positive input.

Add a 10 ohm resistor between taps 5 and 0 on the autoformer.

Adjust squawker output to suit your taste. You can leave it on taps 4 and 0, or drop it down a little using taps 5 and 2.

Now you have a nice flat stable impedance for your Horus', very low loss due to the low part count, and a better first order filter with smoother transition points. I guess we can call it the "dunno what you call em crossovers" for now.

I've been strong arming Al into coming up with some simpler designs. I burden him with my ideas, and he's been using the recipes he whips up to teach me how he does filters. The one above is the highest bang for the buck design, and can be done for very little money on an existing Type A.

[scott0527]

----------------

On 4/21/2005 10:58:29 AM greg928s4 wrote:

I'm ready to buy a pair of crossovers for my Khorns. I use SET tube amps, a nice CD front end, and listen to rock/jazz at as high a volume as 2 watts into Khorns will produce. So far I've preferred the A crossover to the AA.

Should I get ALK's, Dean's I dunno what you call em crossovers, or ALK ES (extreme slope)? Cost is not an object here, I don't believe there's enough of a difference in the prices of them to matter to me.

Thanks for any input.

Greg

----------------

I think the ALK ES's are pretty expensive. About twice the price of an ALK A regular. They look to have twice as much "junk" in them so why not twice the price. The theory behind them makes sense.

I have ALK A crossovers in mine now. I like 'em. I bought the kits from Al. They were fun to build. They look cool and my wife gets turned on when I say I'm soldering a capacitor.

[greg928gts]

Dean, I remember just a little while ago that you were building some crossovers for Khorns that presented a very stable impedence to the amplifiers and I thought you had a name for them.

I don't want to modify an older set of crossovers. I'm ready for all brand new and upgraded components. I like any simple idea, but I want all top notch parts. Can you help or are you too busy right now?

Or do you think I should just get a set of ALK's?

I guess my hesitation with the ALK's is the quantity of components and my flea-sized amps. My amps didn't do so well with the AK-3's, I presume because the power got lost inside the componentry and couldn't find it's way out of the horns.

Greg

[cjgeraci]

Greg, those crossovers you just referred to were designed by John Albright - and yes, Dean makes them. Called DHAxovers.

I have a pair in my Belles, which are run by a Peach pre and a Marantz 8B. They are great-sounding crossovers. Very smooth at all volumes yet extremely detailed. To me, they sound as detailed as freshened "A"s, but with a smoother mid-horn. They are priced less than the standard ALKs. I cannot give you a direct-comparison audio-wise to the ALKs since I have not heard those, but knowing Al's fine work, ALKs would also be a great option.

That being said, I really like the DHAxovers and will probably put those in any future K-horns or Belles I might acquire.

Carl.

[greg928gts]

Thanks Carl.

So, Dean's DHAXovers, regular ALK's, or ESALK's?

Anyone?

Greg

[3dzapper]

ALKs have the advantage of "tunable" squaker output to match your system and room while maintaining a constant impedance load on the amplifier. They work great with all types and powers of amps from flea to arc welder. The Polyswitch tweeter protection can be in line or out simply by switching one wire on the output strip. Plus, with the removal of a jumper or two they are bi-wireable.

Rick

[greg928gts]

Thanks Rick. I have to get used to your new avatar.

Greg

[3dzapper]

It has an archery flair!

Rick

[ricktate]

While i have home built ALKs i would like to try BECs rebuilt AAs on mine to,,,,,just a thought would be nice to have both...Rick

[NOSValves]

Greg,

Did you get thise tubes for your dad?

Craig

[Deang]

I sent two people to Bob last month, and one to Al Thursday night. Every situation is different. What works well for one person, may not be the best solution for another.

I wasn't really ready to talk about this yet, but I guess I need to say something.

I've never really been convinced that Al's way is the best way, in fact, after my experience with the Heresy ALKs, I'd pretty much decided that something was out of place with the design. To put it bluntly, I just didn't like it. Speaking with Al about it, he seemed to think it might be related to the fact that he didn't have a K-22 on hand to run complex impedance on, and extrapolated from available data he had on the driver. I suggested something else -- that maybe I just didn't like the sound with the network. So, is it the network, or my ears? How about a combination of factors? Say, the relationship between the size/shape of the room, seating distance, power/volume levels, gear, and the ear/brain -- which when all jumbled together creates the overall perception of sound. Complicating the whole thing even further is the fact that you can have two people standing side by side, listening to the exact same thing -- and come to completely different conclusions about the sound. So, I say there is no "best" network from a circuit perspective (assuming proper design), but only a network that provides the interaction and voicing best suited for the situation -- which when you get right down to it, is going to be dictated and based on simple preference.

While working through the books I have, and trying to apply and integrate what I was learning with the autotransformer - I realized that I didn't really understand the autotransformer very well. So, emails went off to John Albright and Al K in the hopes that they could clear the rubble out of the road for me. Feeling a little frustrated, I decided a phone call might help, and since my last two calls had went to John, I decided to give him a break and call Al instead. Al is extremely patient, and didn't seem to have a problem with the fact that I didn't have a clue as to what he was talking about. Instead of giving up on me, he took me down to the basics and started working me through it. Four phone calls and many hours later things were finally sinking in and making a lot more sense. It wasn't easy for me, since I first had to learn how the thing actually works, then how it behaves on both sides of the filter(s), the impact on the values as different taps are used, how to integrate its use with a bandpass on the squawker, and finally -- how to use it properly with different filter slopes. Suffice it to say at this point that if one wants to maintain the use of the autotransformer for its low loss properties, and take advantage of the flexibility it offers in adjusting attenuation to suit individual taste -- then constant impedance is the way to go.

Though Al doesn't agree with my conclusion that the larger the room and/or the louder one plays, the steeper the slope between squawker and tweeter they need -- he agrees that people just plain like different things. His response to me when I told him I really preferred the sound of a 1st order network (Type A/ DHAxover) -- was that I must like the sound that comes from drivers interfering with each other! LOL, well yeah - I guess I do!! So, even though he doesn't completely agree with me, he has let me convince him to help me come up with some solid, yet simpler designs. I sent my ideas in schematic form, which I offered up as "homework". Al reviewed and corrected them, and then used them to teach me what I'd done wrong -- with why and how the fix needed to be done. The result is a nice set of schematics which are the property of Al (ALK Engineering) and myself. Once we finish working out the details of our professional relationship, the schematics will be posted on Al's site for download, and I will start offering builds. Once my site is completed, the schematics will be there as well.

Someone has to be thinking, Man, don't we have enough of these things already? Well, yes and no. There are too many that could have been made better, and a only a few that are really good. I felt we needed a few more good ones that were both affordable and offered some flexibility to the end user in the form of attenuation adjustment.

[Erik Mandaville]

"I felt we needed a few more good ones that were both affordable and offered some flexibility to the end user in the form of attenuation adjustment."

I agree. I think this is where the approach of using a constant impedance variable L-pad has an advantage over the auditioning of various combinations of input and output taps on autoformers. An idea I proposed in an earlier thread, which I remember some thought could be useful, would be to use a shorting, multi-poled rotary switch, that could be used to switch between various settings on the autoformer -- which in a sense creates a variable L-pad from the autoformer. What I think is convenient about both of these approaches to squawker attenuation is the fact that the output levels can be listened to on the fly -- and quickly. Moreover, variable L-pads can also be used on tweeters in the same way. Some people have experimented with different HF drivers, which often have greater efficiency than the stock Klipsch tweeters, and an L-pad can be a very simple and perfectly good sounding solution to balancing between the drivers. I have been contemplating the inclusion of an easy to follow tutorial on how to install a simple L-pad volume control on squawkers -- ANYONE can do this! Quite a few people have brought up this concern, and some have found that even the lowest settings on the autoformer are not enough. A variable L-pad, matched to the impedance of the driver in question, works perfectly for this. When I listened to the Altec 511B (thanks again, Shawn) for the first time with our borrowed center channel La Scala, I found it more efficient than the stock Klipsch midhorn. Because I have L-pads on all of the networks I've built, it was very easy to simply turn it down, while listening, to the level that blended best.

Erik

[Al Klappenberger]

Guys,

Since Dean is telling everybody about the long list of possible crossover

upgrades variations that he can do, here's another one that's in the works.

This one will be a very simple modification that can be made the Klipsch "AA"

network that will actually correct most of the deficiencies in it's design

and cost far less that buying new AA networks with all their deficiencies

intact. It's kind of a "half step" to a true constant impedance network. All

the caps will be replaced with new ones of better quality and corrected

values. An extra inductor will also be added. The paste-up attachment shows a

direct computer simulation comparison between the original AA and the

improved "AA". I call it the "super AA". If you're interested in the idea,

you need to twist Dean's arm a bit! The idea is to send Dean your AA and he

will send you a upgraded one.

Al K.

[Deang]

Al, you let the cat out of the bag.

Erik, I always liked the idea of being able to attenuate, what I don't like is variable L-pads. Even if they were sonically the best approach, from a purely business perspective I wouldn't use them. Dust gets inside of them, and then they get scratchy. As dirt accumulates, it just goes from bad to worse. What I sell I build to last half a lifetime with no hassles, and I don't want people calling me 5 years down the road with problems. Fixed L-pads are better, but neither compares favorably to what the autoformer delivers sonically. Yes I know, you say you can't hear a difference, but I can.

In case anyone is wondering about the tweeter, I have a design that adds another, smaller autoformer to the tweeter leg. It's one that Al had UT make for his ESNs, and goes in 2db increments. It didn't take long to see other uses for it, and I'm going to add one to the ALKs I'm building for my Klipschorns.

[Al Klappenberger]

Dean,

Yep, that cat is out of the bag! The mods haven't been made to a real AA yet to test it out, but I'm sure it will be a big improvement at a very modest cost.

Al K.

[Erik Mandaville]

Dean:

Alright, I understand that. I can't help but mention the fact that I have a pair of speakers I built in 1990 that use poorer quality L-pads than what I've used in my Klipschorns, as well as that of a friend. They still work perfectly, and are free of oxidation after all that time -- even in the hot, humid climate of SE Texas. An L-pad is really just a volume control, and steps can be taken, just as in a preamp or integrated amplifier, to prevent or at least reduce contamination of dust. Tube sockets on amps with exposed tubes are actually far more likely to have problems with dust than would an L-pad contained in a protective metal housing.

It's true, I hear no difference whatsoever between an variable L-pad and the autoformer. The switch mechanism is something you might find useful, although that too has contacts that may need to be cleaned from time to time -- as would the contacts of quick release connectors used on the various taps of an autoformer. Corrosion effects static parts, as well. The sound is of course the primary concern, and you know what you like.

Last a lifetime?! Dean, we both know nothing in this hobby seems to last longer than a couple of years at most before we start looking for the next upgrade or 'thing' to try. I know that feeling well! I thought I was set until an Altec 511B from sfogg hit our doorstep. Man, that is one nice sounding horn, but Marie thinks it's not particularly helpful (I'm being diplomatic here!)in an interior design sense. We are conferencing about that from time-to-time, but I'm going to get a pair and hide them in the closet until I can think of a subtle way of installing them in the Klipschorns without her knowing. I have to say, I very much respect the fact that she wants the horns to remain true to PWK's work -- and not end up looking like the experiment of some crazy audio scientist (I love the extreme look of the Altecs though).

.....and as usual, I'm getting way away from the topic of this thread -- sorry about that.

Erik

[Erik Mandaville]

See? even well-known speaker companies use them!

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=50597&item=5769180459&rd=1&ssPageName=WDVW

[sfogg]

Dean,

"Fixed L-pads are better, but neither compares favorably to what the autoformer delivers sonically."

An L-Pad doesn't have to be resistor based, there are also transformer based l-pads on the market.

Shawn

[sfogg]

Rick,

"The Polyswitch tweeter protection can be in line or out simply by switching one wire on the output strip."

That isn't in AL's Type As. That was something I added to the ones I built. Very easy for anyone to add if they wanted to but it isn't a standard function of Al's network.

Shawn