Need Help Deciding Khorn X Over

[sfogg]

Dean,

"Right now, I see more gear, more wires, and a higher potential for ground loops and noise."

True, the potential is certainly there. Maximizing levels through digital crossovers takes some work and is required to get the best out of them and to keep noise floors low.

I actually kind of think the Behringer DCX2496s requiring a 'hot' signal in/out of them is kind of useful when it is setup properly. Don't get me wrong... getting it set properly takes work but once it is setup right the hot signal in/out helps lower the noise floor of the system. The reason being is in most equipment the noise floor is more or less constant. The higher you can get the signal away from the noise floor the better off you are. With the Behringer you have to feed it a pretty strong signal to drive its ADCs to full resolution. That means the signal is further away from the noise floor or your sources and of the Behringers ADC itself. The output of the Behringer is also pretty high. To be able to use its DACs full resolution with almost any amps you need attenuation on the amplifiers inputs. The attenuation on the amps inputs also attenuates the upstream noise too. So then you are basically left with just the self noise in the amps. Which with most speakers isn't an issue but can be with horns.

"I also can't for the life of me figure out how a layperson can pull it off without some decent measuring equipment."

I'd certainly agree with that too. With instruments you at least have a verifiable 'goal' for the system. Doing it by ear with music I think could lead to even more second guessing since one doesn't really know if what they are hearing is due to the crossover setup or differences in the different music.

Setting them up with instruments will get you in the ballpark much easier. Then after you get in the ballpark you can tweak it a little more from there by ear if desired.

In my case in my heavily treated room my goal was pretty simple.... I wanted to be flat through the crossover regions (which are narrow) and wanted to tame any large peaks in the response to be relatively flat overall on pink noise. (Not the same as being flat on white noise) If I needed to tweak response for a bad recording I do that in the Lexicon with its Tone and Tilt controls.

As far as time aligning the drivers I had the Behringer do its auto align function do that for me. You plug a mic into the Behringer and it plays some clicks through the drivers and measures the delay times and sets its delays accordingly. I've never really bothered to play around with that though to see how it audibly changes things. The audibility of that is probably lower in my system since it is crossed so sharply (about 210dB/octave mid to woofer) compared to a first order setup.

The audibility of time alignment between something like a center channel and the L/Rs is *very* audible though in how seamless a blend you get. One of the best tweaks on the Lex. is altering the centers time alignment 'distance' while listening to music with heavy vocals in them till you hit the spot where things just really lock into place. It is fairly dramatic.

BTW, I forgot to mention last night... if you *really* wanted to drive youself mental.... with many of the digital crossovers out there they can store multiple pre-sets. The Behringer has something like 24 of them. You can literally use that to store different crossover points, different slopes, different EQ...etc...etc so you can change between them pretty easily and even change them via RS232. If I wanted to I could literally setup mine such that as I turned the volume up the system would automatically change between pre-sets. So for example it could be setup such that at low volume levels the crossovers are first order, at higher levels it could switch to higher order slopes and even do things like at really high levels roll off the top end. Or whatever. Most systems couldn't do that level of automation though.

Shawn

P.S. Don't forget to send me your address....

[Colin]

geess Shawn, last night I was all set to try active crossover again and this morning you got me scared off, maybe I should just spend $10K on electrostatics, another $10K on an amp and call it quits, at least they would look nice!

[sfogg]

Colin,

It isn't brain surgery... but it also isn't the same as installing a new pair of passive crossovers.

IMO for success one needs to understand the steps required to set it up before starting it. If they understand what needs to be done they should be able to proceed in a logical manor to get it done. But there are multiple steps and it can be time consuming. I put in quite a bit of time into it when I did mine.

Shawn

[DrWho]

"With the right crossiver it will allow you to time-align the bass bin which would be a huge step forward."

So, you've done this with the Klipschorn and have heard the "huge step forward"?

I've heard active setups with Heritage speakers twice, and wasn't

much impressed with the sound from either. I'll be trying it myself

here shortly, so in due time I'll have a solid opinion about it based

on my own personal experience. Right now, I see more gear, more wires,

and a higher potential for ground loops and noise. I also can't for the

life of me figure out how a layperson can pull it off without some

decent measuring equipment.

Advice based on speculation is worthless.

I have heard the difference of time-alignment many times on many

different sound systems so I feel I can comment on the difference it

would make with the khorn. It is clearly audible, but maybe I've got

the advantage of being trained to hear it? But don't take my opinion on

the matter...all the engineers at klipsch feel the same way; even PWK,

once Roy was able to get the idea across (it has nothing to do with

hearing two different sounds, which is what PWK kept having in his

head).

As far as more gear is concerned, by going to an active crossover you

are running the same amount of gear. You have source>preamp>crossover>amp>speaker instead of source>preamp>amp>crossover>speaker.

There are certain advantages to having the speaker directly coupled to

the ampifier: less power compression, more efficient, better damping,

no back-emf, etc etc.... The fact that you have more amplifiers doesn't

matter because there is only ever one amplifier between your ears and

the source for a given passband.

As far as 'more wires and more noise' - I don't think another foot here

or there is going to matter when your signal has already passed through

a thousand feet in the studio. And if there is noise in an active

setup, then there should be noise in the passive setup as well. Noise

is an indication of a poor power source anyway and is a good sign that

it needs to be cleaned up - which will result in an even smoother

sound. Ground loops are a sign of poor grounding, which means there is

a safety hazard - not a huge deal I don't think, but a ground loop

indicates that the 3rd prong is pretty much useless in the event of

large voltages.

And I believe a layperson should have no trouble dialing in an active

crossover system without measuring equipment....maybe an SPL meter if

they don't trust their ears. You should get very good results by

mimicking the original passive crossover - and I would argue even

optimum results that way. And you can just measure the time-delay

requirements in feet: ~1ms per foot. It certainly isn't rocket science

and I'm sure the layperson would enjoy learning along the way.

[Deang]

I just think maybe the approach isn't the best for someone new to the upgrade path, and I can't help but feel that it takes the best of what the approach has to offer to sound more satisfying than a good set of passives -- and like you said, a good deal of work as well. I'm going to do it, but I'm NOT looking foward to it. It will be interesting to see how much "better" it is for the amount of addtional money and work it takes. It may not be brain surgery, but it probably helps if you were born with a pocket protector in your hand.

[sfogg]

"And I believe a layperson should have no trouble dialing in an active crossover system without measuring equipment....maybe an SPL meter if they don't trust their ears."

Try it sometime that way......

"And you can just measure the time-delay requirements in feet: ~1ms per foot. "

That is no assurance of a proper setting. You don't know what the latency through the crossover(s) themselves are. If it isn't consistant between all outputs, and it isn't always, then a simple 1ms per foot isn't going to get the system into alignment.

Shawn

[Deang]

PWK's thoughts on the subject. The first two pages are from 1972, the last page is from 1976. So, it's been 30 years, and with the quality of the modern units certainly being superior over the ones they used back then -- it's more than possible that the old experiments and the opinions that resulted from them are no longer applicable. OTOH, I don't think everything written in the old sheets should be discounted. It's fun to disagree with PK, and I've done it plenty -- but as time marches on, I find myself going backwards and agreeing with him on things I used to disagree with him about.

[DrWho]

"And I believe a layperson should have no trouble

dialing in an active crossover system without measuring

equipment....maybe an SPL meter if they don't trust their ears."

Try it sometime that way......

I have [] It only gets way more difficult when you start deviating from the recommended/original crossover.

"And you can just measure the time-delay requirements in feet: ~1ms per foot. "

That is no assurance of a proper setting. You don't know what the

latency through the crossover(s) themselves are. If it isn't consistant

between all outputs, and it isn't always, then a simple 1ms per foot

isn't going to get the system into alignment.

Well that is going to be an issue with the piece of gear you decide to

go with. If is is of any quality the manufacturer should provide the

delays associated with different signal paths - and if they don't I'm

sure a phone call would remedy that. Or if you're really lazy you can

just get a unit with the auto-setup feature to do it all for you - like

your behringer and all the dbx units.

[Daddy Dee]

Not to fuss with anything in this thread... I just add that I'm among

the folks who wisely (for me) decided not to jump into triamping.

Though I have to say I was truly fascinated listening to Roy Delgado

talk about delay on Khorns and Jubilee. It could be reasonably attained

as a`taster with a reasonably priced piece of Behringer gear and a pair

of Teac digital amps. Now that was a fairly tempting possibility, even

for a technically challenged geek like me.

The idea of triamping makes so much sense, but I think the advice in

this thread to approach with caution of understanding and counting the

cost... this is good counsel.

[Deang]

!

[Deang]

*

[sfogg]

"It only gets way more difficult when you start deviating from the recommended/original crossover."

In other words it gets way more difficult when you try to take advantage of some of the abilities tri-amping gives you.

If one is just trying to duplicate passives they are missing out on numerous other benefits that can be had with tri-amping.

Sure you gain a few benefits doing it this way but it isn't optimized.

"If is is of any quality the manufacturer should provide the delays associated with different signal paths -"

It isn't the different signal paths. It is the DSP latency from different processing occuring on the different outputs. A manufacturer can't just put out a chart that shows this considering the thousands and thousands of possibile combinations within a unit.

"and if they don't I'm sure a phone call would remedy that. "

Not likely considering all the variables involved..

Shawn

[DrWho]

PWK's thoughts on the subject. The first two pages

are from 1972, the last page is from 1976. So, it's been 30 years, and

with the quality of the modern units certainly being superior over the

ones they used back then -- it's more than possible that the old

experiments and the opinions that resulted from them are no longer

applicable. OTOH, I don't think everything written in the old sheets

should be discounted. It's fun to disagree with PK, and I've done it

plenty -- but as time marches on, I find myself going backwards and

agreeing with him on things I used to disagree with him about.

I doubt PWK's later thoughts were put down on paper...the next time you

talk with Trey have him tell you about PWK's stubborness and then how

quickly he changed his views once he saw the measurements. Back in 1976

they didn't have the equipment to measure things in the time-domain and

PWK was strongly against believing in anything that couldn't be

measured.

The problem with conducting those experiments in a big hall is that the

reverb totally masks the differences. I would argue that in a room with

good acoustics that the active crossover begins to shine way more.

I'd also argue that the old active crossovers back then were not too

great, but maybe I've just not messed around with the right ones.

[DrWho]

"It only gets way more difficult when you start deviating from the recommended/original crossover."

In other words it gets way more difficult when you try to take advantage of some of the abilities tri-amping gives you.

If one is just trying to duplicate passives they are missing out on numerous other benefits that can be had with tri-amping.

Sure you gain a few benefits doing it this way but it isn't optimized.

Well I doubt the layperson is interested in making their setup way more complicated []

"If is is of any quality the manufacturer should provide the delays

associated with different signal paths -"

It isn't the different signal paths. It is the DSP latency from

different processing occuring on the different outputs. A manufacturer

can't just put out a chart that shows this considering the thousands

and thousands of possibile combinations within a unit.

"and if they don't I'm sure a phone call would remedy that. "

Not likely considering all the variables involved..

Shawn

I was referring to the signal path of the digital signal inside the

unit - if you told the manufacturer every setting they should be able

to provide you with the latency...or heck, find someone with the same

equipment to measure it for you. The point being that it's something a

layperson can overcome without fancy measuring.

[sfogg]

"Well I doubt the layperson is interested in making their setup way more complicated"

Then by that logic they wouldn't be trying tri-amping at all as it makes their setup way more complicated to setup.

"- if you told the manufacturer every setting they should be able to provide you with the latency.."

Fine, call Germany or Japan and actually try to get these numbers....

And remember if you change anything in the unit those numbers are no longer accurate.

" with the same equipment to measure it for you. The point being that it's something a layperson can overcome without fancy measuring."

If you had someone measure it for you you didn't overcome the problem without measuring.....

IME... either do it right or don't waste your time and money.

Shawn

[Deang]

I have heard the difference of time-alignment many times on many different sound systems...

I grew up during a time when speakers like the DCM Time Window and Dahlquist DQ-10 were the rage. They sounded nice, but there were plenty of speakers that weren't time aligned that sounded subjectively better. We once spent a weekend with DQ-10's moving the panels back and forth, and the difference was barely noticeable, so much so that most of the time we weren't even sure if they'd been moved or not unless we looked. It certainly wasn't as dramatic as the difference I heard when I started upgrading the crossover components, or replacing the Philips tweeter and Motorola Piezo with a Scanspeak D2010. O.K., sure, the path length differences in the K-horn are greater -- but it's a moot point. What should really be considered here is the overall sound, and the things that can be had that improve some other performance parameters that enhance the listening experience -- for about the same amount of money -- with a lot less trouble.

"...so I feel I can comment on the difference it would make with the khorn. It is clearly audible, but maybe I've got the advantage of being trained to hear it?"

I'm not disagreeing that it isn't audible. What I took issue with was your "huge step forward" comment. Just because you say something authoritatively doesn't mean it's authoritative. There are people who have been doing this for 50 years who say the difference is so small that it's not worth bothering with.

"But don't take my opinion on the matter...all the engineers at klipsch feel the same way; even PWK, once Roy was able to get the idea across (it has nothing to do with hearing two different sounds, which is what PWK kept having in his head)."

Of course, this must explain why all the engineers at Klipsch opt for passive networks in their personal loudspeakers.

[Me Loves Khorns]

How do you figure out which passives to get. Roughly, how much would it cost to get new passives for 2 Khorns and 2 LaScalas?

Also, after reading so much on the forum, I repossitioned my Khorns last night into the corner, rather than lined up to look good with the other furniture. Wow, that alone made a nice improvement.

Once again, thanks to all who are helping me out. I am going to try to go to the Pilgrimidge in Hope. Do people discuss these issues some there. Is it possible to hear some speakers with some of these changes there?

[Colin]

Do not go to the pilgrimage! What you hear there will change your opinion of big ole horn and your idea of what and how your stereo can and should sound it will cost you big money to attend.

[Me Loves Khorns]

Do the new speakers sound better or something? The only store with klipsch here is Best Buy, and that doesn't count. What do people end up spending money on there? We have a WalMart in Waco already!

[Deang]

How do you figure out which passives to get. Roughly, how much would it cost to get new passives for 2 Khorns and 2 LaScalas?

I gave recommendations and provided links in my second to last post on page one of the thread. Anywhere from $200 to $1100 -- it's up to you.