JubScala?

[edwinr]

Hi Steve,

I've been searching for that other discussion. Is it recent? I've searched the 'net too. It seems as though this Patrician design is very old...

[SSnyder]

Hi Steve,

I've been searching for that other discussion. Is it recent? I've searched the 'net too. It seems as though this Patrician design is very old...

Hi go to 2-channels fomum and the seen add for 2 Klipschornes and 1 was Patrican thanks

[DrWho]

If your 90dB covers 50dB SPL to 140dB SPL, then chances are you're not using about 30dB of that dynamic range, and now you're down to 60dB of dynamic range (ouch)....and at 50dB SPL, you're gonna be above the noise floor of the room so it's not gonna be masked completely.

I could be totally wrong here but I'm pretty sure no one here is playing their music at 140db in their home listening room...

Let's say your speaker is 108dBspl at 1m with 2.83V input, which is 9dBV. Most amplifiers have a gain of around 26dB. If your max output level on the DSP is 20dBV, then you're looking at a max SPL of 145dB (108dBspl - 9dBV +20dBV + 26dB), which with 90dB of dynamic range will put your noise floor at 55dBspl.

I agree that people aren't listening this loud, but what they're doing is turning down the output of their preamp to adjust the volume. What happens is the SPL of the noise floor doesn't change, but the music gets quieter. So now if you want to listen at 90dBspl, now you're only talking 35dB of dynamic range, yikes!

If your amplifier has volume controls, then you can certainly trim down the input sensitivity, but this trim usually doesn't change the internal gain of the amplifier. What happens is usually you raise the effective output impedance of the device upstream from the amplifier, which means for every ~6dB of input sensitivity reduction, you lose ~3dB of dynamic range in the amplifier (these are very rough numbers to present a concept...the actual math is very non-linear). But let's say your amplifier has an output noise voltage of 25dBspl through the speaker. Attenuate the input sensitivity of the amplifier by 20dB and now its noise floor raises to 35dBspl. The noise floor from the DSP amplified by the amplifier is also now 35dBspl and your system will sum to 38dBspl. That's gonna be loud enough to hear the noise floor pumping since you're gonna need at least 20dB of masking to really hide it.

[Chris A]

Mike,

1) Are you taking into account the thresholding of human hearing, effectively reducing the ear's required S/N at very low SPLs? How old are you assuming your critical system listeners to be?

2) Also, are you assuming that most folks are setting up their gain structure in order to get 100% out of the last amplifier stage? It seems to me that if there were a problem with noise floor (...which I don't believe that I've experienced in my listening...), that would be the first response by the user--to adjust the gains to "prevent noise floor audibility".

3) All this is interesting discussion, but I'm still not aware that I'm listening to the postulated problem. I'm listening to Dx38s on a daily basis. What am I missing in detecting the problem?

4) I'd probably prefer to discuss driver FM distortion and IMD before I'd get really excited about my own electronics noise floor, FWIW. We haven't even discussed the noise floor of tube electronics, cables (common mode noise rejection), and TTs in this thread...[:-*]

Are we talking about some effect that is 20+ dB away from most currently limiting system performance issues?

Chris [:^)]

[DrWho]

1) I'm referring to my own personal experiences, so I guess 26? [A]

2) I am merely presenting a usage case to explain the analogy and set up the later discussion about input trims and their impact on dynamic range.

3) "Inner detail"

4) Driver FM distorion and driver IMD usually doesn't impact the design of the electronics...

Btw, I feel a sense of defensiveness coming from those that enjoy the Dx38. I am certainly not trying to imply that you can't enjoy good audio from the Dx38. Knowing more about the unit and some of the limitations shouldn't change one's enjoyment of the unit, but it might help those looking to take things to the next level.

Also, broad band noise really isn't the issue when you start operating closer to the noise floor. Power supply artifacts become much more pronounced (non-linear mixing and digital spurs) and the electronics won't be as linear either (so more distortion). It's hard to measure the distortion at these levels though because the noise is louder than the distortion...but that doesn't mean it isn't audible. 20dB is a very generic rule of thumb concerning masking. I personally prefer to design so that you don't have to rely on masking in the first place.

[Chris A]

4) Driver FM distortion and driver IMD usually doesn't impact the design of the electronics...

I thought this was a thread on the JubScala...[A]

3) "Inner detail"

Okay, I guess you have me on that. If you're ever through D/FW, drop in for a critical listening test. My ears aren't 26 years old anymore [] []

Btw, I feel a sense of defensiveness coming from those that enjoy the Dx38. I am certainly not trying to imply that you can't enjoy good audio from the Dx38. Knowing more about the unit and some of the limitations shouldn't change one's enjoyment of the unit, but it might help those looking to take things to the next level.

Fair enough - note that some folks read only a portion of what is written, then run off and say things like "the Dx38 is a piece of [bs] "Just trying to understand your points, and I believe that I do now.

Recently, I've been reading on the demagnetization of typical permanent magnet materials (ferrite, Alnico, ceramic, and neodymium-ferrite-boron) and "flux modulation distortion" of real drivers. This is an area of real interest, since I believe it was that PWK guy who said the following:

"...But a small percent of modulation distortion can raise hell with the way it sounds."

Taken from 4/89 Speaker Builder Magazine, "A Visit to the Klipsch Kingdom" by Bruce Edgar

I much prefer horn-loaded speakers, tweeters through subwoofers (not direct radiators) because of their ~25 dBspl freedom from distortion and their ~15 dBspl increase in output over using the same drivers as direct radiators [...PWK's performance numbers quoted from the same article].

[DrWho]

Recently, I've been reading on the demagnetization of typical permanent magnet materials (ferrite, Alnico, ceramic, and neodymium-ferrite-boron) and "flux modulation distortion" of real drivers. This is an area of real interest, since I believe it was that PWK guy who said the following:

Have you come across anything on shorting rings, or audio specific literature? I'd love to read up a bit more on it...

[Chris A]

http://www.audioasylum.com/cgi/vt.mpl?f=hug&m=152594

http://www.linkwitzlab.com/frontiers_3.htm

[mikebse2a3]

"Let's say your speaker is 108dBspl at 1m with 2.83V input, which is 9dBV. Most amplifiers have a gain of around 26dB. If your max output level on the DSP is 20dBV, then you're looking at a max SPL of 145dB (108dBspl - 9dBV +20dBV + 26dB), which with 90dB of dynamic range will put your noise floor at 55dBspl."

Again using max SPL of 145db is a crazy reference for a home listening enviroment.

"I agree that people aren't listening this loud, but what they're doing is turning down the output of their preamp to adjust the volume. What happens is the SPL of the noise floor doesn't change, but the music gets quieter. So now if you want to listen at 90dBspl, now you're only talking 35dB of dynamic range, yikes! "

And this is why we need to properly adjust the gain balance of each stage of the system for maximum signal to noise ratio and thus maximum dynamic range.

"If your amplifier has volume controls, then you can certainly trim down the input sensitivity, but this trim usually doesn't change the internal gain of the amplifier. What happens is usually you raise the effective output impedance of the device upstream from the amplifier, which means for every ~6dB of input sensitivity reduction, you lose ~3dB of dynamic range in the amplifier (these are very rough numbers to present a concept...the actual math is very non-linear). But let's say your amplifier has an output noise voltage of 25dBspl through the speaker. Attenuate the input sensitivity of the amplifier by 20dB and now its noise floor raises to 35dBspl. The noise floor from the DSP amplified by the amplifier is also now 35dBspl and your system will sum to 38dBspl. That's gonna be loud enough to hear the noise floor pumping since you're gonna need at least 20dB of masking to really hide it."

The typical amplifiers sensitivity control's basic function works as a variable voltage divider with the ability to reduce the output voltage from the source(this includes any noise floor signal as well as musical signals) that the amplifier is receiving it's signal from.

Agreed the typical amplifier's inherent design gain and noise floor itself is pratically unchanged by the action of the sensitivity control. To test any amplifiers design for incompatability (due to it's inherent design's noise floor) with especially high efficiency loudspeakers one can use a shorting plug on it's input and the resulting noise floor heard will be the best that this combination can acheive.

As far as lowering the input sensitivity control and the noise floor of the amplifier going up I will have to disagree and I've never observed this in any properly working amplifier.

Anyway I believe I'll go listen to some music..........and maybe just for the heck of it I'll try to hit 145db with my noisy tube amplifiers...[]

mike tn

[Coytee]

and maybe just for the heck of it I'll try to hit 145db

C'mon over to my place and we'll rock out with some Marie..... [H]

[Chris A]

and maybe just for the heck of it I'll try to hit 145db with my noisy tube amplifiers...

Do tube amps have 100+ dB of SNR? (This is a straight-up question.) Is there a difference between P-P and SETs in this aspect? Maybe this question should go into another thread...

Chris

[DrWho]

As far as lowering the input sensitivity control and the noise floor of the amplifier going up I will have to disagree and I've never observed this in any properly working amplifier.

Then you disagree with having never measured. But if you want to talk theory, then let's have a real technical discussion on noise figure.

Again using max SPL of 145db is a crazy reference for a home listening enviroment.

Are you trying to show ignorance? Do you ever recall listening to a Dx38 with QSC amps cranked up all the way? Or listening to high sensitivity speakers with Crown XTi amplifiers? Could you even hear the hiss? How loud do you think it was? And do you believe the dynamic range specs on those amplifiers?

Anyway I believe I'll go listen to some music..........and maybe just for the heck of it I'll try to hit 145db with my noisy tube amplifiers...

I think it'd be more productive if ya took your head outta the sand, cuz then you wouldn't need 145dBspl... []

[ClaudeJ1]

As far as lowering the input sensitivity control and the noise floor of the amplifier going up I will have to disagree and I've never observed this in any properly working amplifier.

Then you disagree with having never measured. But if you want to talk theory, then let's have a real technical discussion on noise figure.

Again using max SPL of 145db is a crazy reference for a home listening enviroment.

Are you trying to show ignorance? Do you ever recall listening to a Dx38 with QSC amps cranked up all the way? Or listening to high sensitivity speakers with Crown XTi amplifiers? Could you even hear the hiss? How loud do you think it was? And do you believe the dynamic range specs on those amplifiers?

Anyway I believe I'll go listen to some music..........and maybe just for the heck of it I'll try to hit 145db with my noisy tube amplifiers...

I think it'd be more productive if ya took your head outta the sand, cuz then you wouldn't need 145dBspl...

Now boys, we need to play nice.

Mikes,

I this is why I have always like dbWatts to describe amplifiers and not just watts..................but doesn't having a potentionmeter turned down at the front end of a VOLTAGE gain device (not power, in the discussion, eh?.........hmmmmm not sure about that) also drop down the noise coming from the source in a linear proportion?

[ClaudeJ1]

Realistically, though, so much music uses peak limiting (some properly, some not) in the recording phase, that to encounter dynamics on most music beyone 60 db is a bit of a rare to non-existent occurence, is it not?

It still can be handled by any Jube or JubeScala (not really a proper term without at least a 402 horn with a TAD or Klipsch driver IMHO) with ease with any good amplifier, so this too much "theoretical" stuff is a waste of time in the real world.

[DrWho]

doesn't having a potentionmeter turned down at the front end of a VOLTAGE gain device (not power, in the discussion, eh?.........hmmmmm not sure about that) also drop down the noise coming from the source in a linear proportion?

Nope, it's not linear at all...the noise reduction is always less than the signal reduction, which means your dynamic range will shrink. Eventually, you get to a point where there is no noise reduction at all and you only reduce your signal.

The real moral of the story is to not use an amplifier with a crap load of gain, but you can't find amplifiers without a lot of gain, which is why you can't get anywhere close to ideal with the products on the market today. Trying to play the attenuation game is only a band-aid and not a real solution to the problem...

[mikebse2a3]

mikebse2a3 said: As far as lowering the input sensitivity control and the noise floor of the amplifier going up I will have to disagree and I've never observed this in any properly working amplifier.

Then you disagree with having never measured. But if you want to talk theory, then let's have a real technical discussion on noise figure.

Mike have you ever shorted(ie: equilivant to turning a sensitivity control all the way down on many amplifiers) the input of an amplifier to measure (or listen to) it's noise floor? If you have and measured (or heard) an increase in the noise floor you have a faulty amplifier or a crappy design..!

mikebse2a3 said: Again using max SPL of 145db is a crazy reference for a home listening enviroment.

Are you trying to show ignorance? Do you ever recall listening to a Dx38 with QSC amps cranked up all the way? Or listening to high sensitivity speakers with Crown XTi amplifiers? Could you even hear the hiss? How loud do you think it was? And do you believe the dynamic range specs on those amplifiers?

I'll just leave your "ignorance comment" alone Mike.

Believe me Mike I have never been in a room with the DX38, QSC amps cranked up all the way to insane levels...!!!

Let me make this simple, match your amplifiers input sensitivity correctly to the EV DX38 and the noise floor of the EV is inaudible when listening in a typical home listening room even if listening at what I believe a majority of people would call very loud levels.

And lest you accuse me of being defensive of the EV DX38 again I will remind you I have already stated that I'm now using the EV DC-One because of it's increased features and I have no noise floor issues from either unit when either unit is placed in my system. Simply put the EV DC-One allows me the ability do what I consider essential adjustments to my system because of it's increased number of filters in it's inpu/output channels and that is the only reason it has replaced the EV DX38 in my system at this time.

I think it'd be more productive if ya took your head outta the sand, cuz then you wouldn't need 145dBspl...

Mike, I've simply been trying to get you to understand that your reference levels of 145db to acheive your 55db noise floor has nothing to do in the real world were I believe most listeners(exception made here for Coytee) will not even come close to those kinds of levels and it only serves to scare people off of an excellent processor that they might want to consider for their systems if it's price/features would meet their needs.

Now I think I will try to get the "sand outta my head" and do something more productive than debate you farther with this.

mike tn

[Chris A]

exception made here for Coytee

The nail-gun king's reputation preceeds him--once again. []

By the way - I'll repeat myself: while I do see the theoretical advantage of avoiding gain-staging the input or output of the Dx38, I have yet to hear anything limiting from this unit in real life (although there has been a lot of angst and subsequent discussions over assumed-but-as-yet-unproven shortcomings).

Any other comments that I''ve made here or anywhere else that might appear to detract from personal or product reputation (with the possible exception of SET amplification [6] ) were not intended.

I will say that the Jubscala is probably a miraculous improvement over the basic La Scala's sound, I'll also say that there is at least a couple of more tweaks that can improve its performance:

1) stiffening the La Scala bass bin to decrease its resonances, and

2) turning the bass bins around to face a corner (assuming the you have two good corners to put them into)...

...will significantly improve their sound - dramatically so.

Chris

[SSnyder]

Realistically, though, so much music uses peak limiting (some properly, some not) in the recording phase, that to encounter dynamics on most music beyone 60 db is a bit of a rare to non-existent occurence, is it not?

It still can be handled by any Jube or JubeScala (not really a proper term without at least a 402 horn with a TAD or Klipsch driver IMHO) with ease with any good amplifier, so this too much "theoretical" stuff is a waste of time in the real world.

60dB no flipn' way modern erra CD more like 3dB dinamic range

[Coytee]

(exception made here for Coytee)

Well, I'll have you know that little snide remark just got your Marie Osmond Box Set [G], returned to the store. So much for your Christmas present!

[Coytee]

The nail-gun king's reputation preceeds him--once again.

Yours too!