Jubilee comments/discussion continued.

[Coytee]

Why is that ? Do they need them ?

I don't know anything about them (and I mean NOTHING!!!) but...with that said, I've read some comments about different horns collapsing verticals (or something like that) and their need for electronic equalization might be different than another horn that doesn't collapse the vertical (like the K402, I think)

So, as I think I recall reading, some horns have the eq sort of "built in" the horn itself and other horns add some electronically.....either way, the sound is getting "modified" in one way or another in both situations.

The above is merely speculation as to what I bet the answer will be.

[PrestonTom]

The Altecs you are referring to are actually a fairly old design.

One of the reasons that one might need EQ is due to the use of a CD horn (eg. the k-402). These horns have the feature that there is little additional high frequency roll off as you go off axis. The usual side effect, however, is that on axis or off axis the high frequencies need to be boosted. This would be CD equalization. If you look at your Behringer or EV there is probably a "knob" that is labelled this way.

CD horns have been around for quite sometime (the Manta Ray was an early version) and the need for CD eq has also been around since then. Some have implied that EQ for horns is a new requirement. No it is not.

-Tom

[Marvel]

The model 19 uses an 811b horn, crossing over at 1200Hz. There are a number of Klipsch folks here who have used the 511b horn on their LS or Khorn, and found it to be a huge improvement. They seem to sometimes move on to one of the tractrix horns...

I think there is more than one person on here who has the Model 19. A freind of mine had a pair that I really liked, but my LS are ok by me. He did, however, always play them far to loud for the room they were in.

[mikebse2a3]

One of the reasons that one might need EQ is due to the use of a CD horn (eg. the k-402). These horns have the feature that there is little additional high frequency roll off as you go off axis. The usual side effect, however, is that on axis or off axis the high frequencies need to be boosted. This would be CD equalization. If you look at your Behringer or EV there is probably a "knob" that is labelled this way.

Actually in many cases the Horn/Driver is much more efficient than the LF section (even the Klipschorn and Jubilee) and so the lower frequency end of the Horn/Driver is reduced and not much if any boost needs to be applied to the top end.

This is a good example of how many don't understand the principles of the Jubilee Design and think EQ is doing something terrible to their sound wether it's passive or active crossovers being used.

The passive version of the Jubilee is a good example of this since you can't boost with a passive!

The active version is also the same in that the compensation is in reality acheived by a reduction in drive and not boosting for the K402/K69A.

mike tn

[Blvdre]

Good point Mike, there's no "boosting" involved, only attenuation. My knowledge is pretty minimal in these matters, but I do know that even with exponential horns, such as the 511 or 811's, a bit of contouring is still required. I had to add an RC contour to my 511/902 LaScala set-up to make things right. After the fact, I looked at the model 19 x-over contour, and my values were virtually spot-on. Whoever posted those 902 response curves here, thanks.

[PrestonTom]

One of the reasons that one might need EQ is due to the use of a CD horn (eg. the k-402). These horns have the feature that there is little additional high frequency roll off as you go off axis. The usual side effect, however, is that on axis or off axis the high frequencies need to be boosted. This would be CD equalization. If you look at your Behringer or EV there is probably a "knob" that is labelled this way.

Actually in many cases the Horn/Driver is much more efficient than the LF section (even the Klipschorn and Jubilee) and so the lower frequency end of the Horn/Driver is reduced and not much if any boost needs to be applied to the top end.

This is a good example of how many don't understand the principles of the Jubilee Design and think EQ is doing something terrible to their sound wether it's passive or active crossovers being used.

The passive version of the Jubilee is a good example of this since you can't boost with a passive!

The active version is also the same in that the compensation is in reality acheived by a reduction in drive and not boosting for the K402/K69A.

mike tn

Mike, you are of course correct. But I think folks might be getting confused. I was using the term "boost" a bit loosely . I was using it in the relative sense and you are using it in the absolute sense. Yes, LF sections do not have the sensitivity that the HF sections have and they will differ by several dB. Hence the need for an autoformer to cut some gain in the top section of a K-Horn. The Jubilees will also share this difference in sensitivity between the the two sections.

However my point was that EQ is not being used to merely compensate for differences in sensitivity, that is trivial and only requires a voltage divider or whatever. If someone considers that to be EQ, then this has been done for decades (sometimes why crossovers are called balancing networks). But the CD equalizing is a bit different since that is happening only to a single driver. Yes, you are correct the high frequencies are technically not being boosted, rather the low frequencies are being attenuated. The net effect is still the same and the response is being equalized. What is different is that the EQ is not being applied to hide some defect in the transducer. The characteristic of a CD horn is that it frequency response is approximately uniform (although not flat) as you go off axis (within reason). However the physics dictates that the response can not be both flat and also uniform as a function of angle. Hence the need for CD equalization. I am belaboring this point because CD eq is different in that equalization is not used to "compensate" for deficiencies in a driver.

The issue about Altecs (511, 811) is entirely different. These are not CD horns. So EQ wold be needed for adjusting the relative sensitivities of the drivers (if you want to call that EQ). If it is also needed to "improve" the frequency response of the horn/driver, then that is another "kind" of equalization. I think it is this last kind of equalization that folks are afraid of (whether it is attenuation or boosting). Ideally a horn/driver should not need "that kind" of equalization As far as the Jubilee goes, EQ is used 1) to match the sensitivity of the two section, 2) to provide CD equalization of the K-402 horn and 3) to knock down some peaks in the response (mostly on the bass bin).

I think we are in agreement,

-Tom

[mikebse2a3]

The issue about Altecs (511, 811) is entirely different. These are not CD horns. So EQ wold be needed for adjusting the relative sensitivities of the drivers (if you want to call that EQ). If it is also needed to "improve" the frequency response of the horn/driver, then that is another "kind" of equalization. I think it is this last kind of equalization that folks are afraid of (whether it is attenuation or boosting). Ideally a horn/driver should not need "that kind" of equalization As far as the Jubilee goes, EQ is used 1) to match the sensitivity of the two section, 2) to provide CD equalization of the K-402 horn and 3) to knock down some peaks in the response (mostly on the bass bin).

I think we are in agreement,

-Tom

Yes we are in agreement.

I knew you actually understood Tom how the EQing was being done but as you pointed out people can be confused when we speak of boosting the response and that is what I wanted to clarify.

It is completely understandable why (History is full of EQing being misused) people are concerned and reluctant to accept EQing but when used properly EQing (passive or active) can be an appropriate tool to take a loudspeaker system to a higher level of performance than is possible without it.

So what I would like for people to come away from this discussion is EQing as applied to the Jubilee in particular is that this isn't a case of someone just randomly using EQing with no way to verify how well it is being implemented but instead realize that Roy has the knowledge, experience and tools to look at this in all the domains(frequency, phase, impedance...etc... with a thoroughly analyzed and consistent listening enviroment) to verify that any EQing will bring the total system to a higher performance level than would be acheivable without it.

mike tn

[sfogg]

Tom,

"What is different is that the EQ is not being applied to hide some defect in the transducer. The characteristic of a CD horn is that it frequency response is approximately uniform (although not flat) as you go off axis (within reason). However the physics dictates that the response can not be both flat and also uniform as a function of angle. Hence the need for CD equalization. I am belaboring this point because CD eq is different in that equalization is not used to "compensate" for deficiencies in a driver."

Gotta kind of clarify that first sentence. The whole reason we need EQ 'boost' at all is because the driver is rolling off. Look at the plane wave tube response of the driver and it rolls off. Look at the same driver on a CD horn and it will show the same rolloff as well.

When you then put the driver on a horn that basically focuses the HF into a smaller and smaller area (collapsing polar response) that focusing of the HF is effectivly boosting the output of the driver at higher frequencies. The horn is EQing the response of the driver (by trading dispersion for output) to deal with the mass rolloff of the driver. If the horn and driver are well matched this will give relatively flat response directly on axis but rolled off response off axis which in room will result in a somewhat rolled off sound even on axis since we hear the composite of on and off axis response in room. If the horn/driver aren't well matched the response won't be either.

A CD horn doesn't have that trade off of dispersion for output. So the natural response of the driver comes through... and as such needs external boost. With the external boost we get flat response on axis and off axis. The external boost can be matched directly to the driver used.

If one considers a horn as a type of amplifier a CD horn has flat response (what goes in is what comes out) while something like a 511B has a tipped up response on the top end... what goes in comes out boosted on the top end.

Shawn

[DrWho]

Good point Mike, there's no "boosting" involved, only attenuation.

For what it's worth...

I know you're referring to the age old notion that "attenuation" is better than "boosting", but that's not looking at the big picture. In other words, it is entirely acceptable to boost.

All that really matters is the final power going into the driver for the SPL you desire.

For example, let's say you have 10dB boost at 10kHz, relative to 1kHz. If you're listening at 100dB with 1W at 1kHz, then you will be pushing 10W at 10kHz. Now let's say you attenuate 1kHz by 10dB instead of boosting at 10kHz. If you're listening at 100dB at 10kHz, then you'll be turning the amp up to the point that you've got 10W at 10kHz...and due to the attenuation, you will be running 1W at 1kHz.

The only time the notion of boosting or attenuating matters is when it involves the gain structure of your signal path. The device introducing the most amount of noise will determine where the most gain in the signal path is needed. Generally speaking, you want more gain up front and less towards the end (which is kinda the opposite of what we see with power amps....and is the reason line level devices need to have such crazy low noise/distortion levels).

[Blvdre]

No, I wasn't making a judgment call, just pointing out that when contouring w/ a passive network, you're shaping via subtraction. In your example (all other considerations aside), I see that it's a wash.

Interesting point about the acceptability of attenuation v.s. boost. Maybe people are disdainful of boost because it involves active circuitry. I would think that, depending on your personal taste, either method can be satisfactory, as long as properly executed.

[Edgar]

Interesting point about the acceptability of attenuation v.s. boost. Maybe people are disdainful of boost because it involves active circuitry. I would think that, depending on your personal taste, either method can be satisfactory, as long as properly executed.

As with all things, it's in the execution. Part of this is related to the concept of gain structure -- cut followed by boost can raise the noise floor and reduce dynamic range. Boost can also easily be abused, as in attempts to use narrow PEQ filters to "fill in" room nulls (it is impossible). Cut is perhaps less easily abused, but ultimately there is nothing intrinsically wrong with either cut or boost, properly applied.

Greg

[whatever55]

What-a-burger... you have a PM....

[Islander]

With the JubScala, the HF and LF are balanced by boosting the LF by 3dB. With my setup in my room, it seemed a bit bright, so I attenuated the HF by 1dB and haven't altered Roy's recommended settings in any other way, other than adjusting the delay for the HF section to suit the tweeters' positions on top of the bass bin.

[DrWho]

Do we not have mid horn designs that are relatively flat in response?

Are you talking specifically just the horn? Or a horn + driver response? And are you talking on-axis frequency response, or frequency power response?

Are we not able to combine a relatively flat-response midrange horn with a relatively flat-response tweeter and a simple network that all improves on the Khorn design?

For sure, but there is more to be improved upon than merely copying newer horn techniques into the framework of the Khorn...

I would be much more interested in a design that included the Jub bass bin, a tractrix midrange horn with a high quality 2" mid driver, a lush sounding tweeter like the Beyma CP-25 and a simple network like the "A" network, but of course made to balance these particular drivers. This is not my expertise, but it just seems to me that compared to the Khorn, this combination would offer higher efficiency, a relatively flat frequency response, a deeper voicing, smoother response through the midrange, and sweeter highs. I could run a speaker like this (and achieve superior sound quality) with the same 2 watts per channel that I'm feeding into my Khorns.

Ok, now let's compare that to the Jubilee...

You would have an extra crossover point, which brings along all sorts of sounds characteristic to xovers. Lower slope xovers with no time-alignment is guaranteed to cause comb-filtering, which again has its own characteristic sound. Efficiency with a passive xover is actually lower than using an active xover. Active xovers also allow for better damping of the drivers. The lack of EQ (or a balancing network) is going to result in a rougher frequency response. Also, using horns that EQ up the driver are going to result in a very non-linear power response.

Using a flat response driver on a flat power response horn will avoid needing to use any CD EQ, but they don't exist.

I think the only thing you gain by collapsing vertical polars is less cone-excursion from the active driver - which should result in less driver distortion. Going 3-way might allow one to collapse the verticals and not sacrifice the power response as much.

That said, I think some people prefer the sound of highs with exagerated power response...and projecting from higher in the air too. There needs to be a willingness to retrain your ears...

[sfogg]

Mike,

"Lower slope xovers with no time-alignment is guaranteed to cause comb-filtering, which again has its own characteristic sound."

Lower slope crossovers for non-coincident sources are guaranteed to comb filter with or without delay. Delay just lets you alter where the combing occurs.

Shawn

[Coytee]

Delay just lets you alter where the combing occurs

I've been altering where my combing occurs... I dont need to comb in the back anymore and my wife is trying to keep me from growing a 'comb over' to compensate...[:$]

[Don Richard]

Delay just lets you alter where the combing occurs.

And it also aims the dominant lobe directly at the position to which the drivers were synchronized.

[greg928gts]

That said, I think some people prefer the sound of highs with exagerated power response...and projecting from higher in the air too. There needs to be a willingness to retrain your ears...

I appreciate your technical knowledge, it's impressive. But if I have to retrain my ears to like something that just doesn't sound as good to me, then all that technical knowledge is of no benefit.

All speakers sound different to everybody, that's why there are so many different kinds out there. The subjective nature of it all means it isn't about being technically correct. In the end they have to sound good to the end user. It's not up to the end user to convince themselves that the most technically correct speaker is better sounding.

Greg

[DrWho]

Mike,

"Lower slope xovers with no time-alignment is guaranteed to cause comb-filtering, which again has its own characteristic sound."

Lower slope crossovers for non-coincident sources are guaranteed to comb filter with or without delay. Delay just lets you alter where the combing occurs.

Shawn

Ah, very good point. I should probably rephrase that to say comb-filtering "on-axis".

[DrWho]

It's not up to the end user to convince themselves that the most technically correct speaker is better sounding.

My problem is the end user then claims to be an audiophile.

I absolutely enjoy music, but I also greatly enjoy the audiophile pursuit (the playback of music with minimal distortion). I have found that my greatest listening enjoyment has been the result of really good music played back on really good systems. That doesn't mean I can't enjoy good music on a crappy system (like my car) or that I can't enjoy really stupid music on a really good system (as long as the recording quality is exceptional).

To me, enjoying music and being an audiophile are two different things. It's a double benefit when I get to do both at the same time.

However, I do get rather bored when I hear the same sounds imparted onto every song in order to make them more palatable. It's like watching home-made movies where someone overuses the same effect over and over....and over....and over... [|-)]

Btw, how do you know what you like if you don't train your ears? I have found that I have blaimed a lot of things incorrectly until taking the time to learn a bit more....I can't imagine how skewed some people's perspectives are when they've never tried to learn anything about what they hear. Of course, they also seem to be the most vocal about knowing what they like to hear...