La Scala - REW Measurement Help

[Robbie010]

I have just taken the very first REW measurement of my clone La Scala which shows a substantial peak at 140Hz.

 

 

This is L & R channel overlaid with 1/6th octave smoothing. 

 

Initial reading suggest that this is a room mode caused by the low ceilings in my room. Has anyone come across this before? 

One thread suggested this could be corrcted by raising the speaker off the floor?

[Chris A]

What's the Q_ts of your woofers?  A high Q_ts will give you that big rise in response just before you get to the resonant frequency F_s, which is about 100 Hz in a La Scala bass bin. 

 

You could use a passive notch filter centered on 146.6 Hz, with an attenuation of about 18-20 dB and a "Q" of about 6-8. 

 

I actually recommend using to good DSP crossover to flatten the response using its PEQs.

 

Chris

[Panelhead]

  Try moving the speaker. Also raise it to see if the low ceiling is affecting it. 

  EQ can flatten the 160 Hz peak and the 200 -500 Hz dip.

  Some feel bass bin flex creates the 140 Hz peak. What did you use to construct the bass bins? I doubt this, it would be output level driven. Seems the peak is common at all levels.

[Chris A]

21 minutes ago, Panelhead said:

Some feel bass bin flex creates the 140 Hz peak.

Not an 18 dB rise, however, as shown above.  That magnitude of midbass rise likely won't be attenuated by varying its boundary gain.  In addition, if you look at the response below 100 Hz, it tells you that the bass bin is presently either using boundary gain (i.e., a room corner) or there is a bass reflex chamber attached to the back chamber of the woofers--like djk, et al. advocated. 

 

There is a much milder anechoic midbass rise at about that frequency (oscillating response) due to the undersized mouths on the bass bin, which creates response ripples nearer the F_s of the bass bin/woofer.

 

Chris

[Robbie010]

39 minutes ago, Panelhead said:

  Try moving the speaker. Also raise it to see if the low ceiling is affecting it. 

  EQ can flatten the 160 Hz peak and the 200 -500 Hz dip.

  Some feel bass bin flex creates the 140 Hz peak. What did you use to construct the bass bins? I doubt this, it would be output level driven. Seems the peak is common at all levels.

 

Thanks.

 

I used 3/4 ply and installed braces.

 

I have an RME ADi-2 DAC that I have borrowed, so I will have a play around with the EQ setting on that and see if I can correct the spike that way.

 

Definately think I'll end up gong down the DSP route though.

[Panelhead]

  I am using Room Perfect DRC to do correction on LS II. The effect is very positive. Previous home speakers were all changed. Just it is more obvious on the LS.

  But it makes small amplitude and phase corrections.  Not 18 dB. 

  Once above 500 Hz your clone looks very good to above 10 kHz. . The high end is rolling at what looks like 12K. Is that tweeter or measurement related?

[Panelhead]

1 hour ago, Chris A said:

Not an 18 dB rise, however, as shown above.  That magnitude of midbass rise likely won't be attenuated by varying its boundary gain.  In addition, if you look at the response below 100 Hz, it tells you that the bass bin is presently either using boundary gain (i.e., a room corner) or there is a bass reflex chamber attached to the back chamber of the woofers--like djk, et al. advocated. 

 

There is a much milder anechoic midbass rise at about that frequency (oscillating response) due to the undersized mouths on the bass bin, which creates response ripples nearer the F_s of the bass bin/woofer.

 

Chris

  Chris,

    Your knowledge and input is impressive.

 Looking at the peak around 70 Hz, it is clear there is either boundary or other reinforcement. Now.

[Robbie010]

39 minutes ago, Panelhead said:

  Chris,

    Your knowledge and input is impressive.

 Looking at the peak around 70 Hz, it is clear there is either boundary or other reinforcement. Now.

I’m in awe of these responses...... I really don’t have the technical knowledge to understand whats going on or to reply to some of the responses above.
 

Reading between the lines of your reply above, I’m assuming that the fact that my la scala are very close to the back wall is not helping with the 140/150Hz peak i.e. the boundary effect??

[Chris A]

To get the bass bins mostly out of boundary effect from any room wall at 146 Hz, you would need to move them anywhere from 2 to 4 feet away from all of the walls/floor/ceiling.

 

That might be difficult to do, and you'll probably wind up with an additional ~3 dB dip in response somewhere above that frequency, in addition to that which you currently have between about 200-600 Hz.  I think the best way to deal with the 100-200 Hz midbass peak is to apply an attenuating EQ filter until you get the response flat across that band, and leave the bass bins where they are. 

 

Chris

[Robbie010]

16 minutes ago, Chris A said:

To get the bass bins mostly out of boundary effect from any room wall at 146 Hz, you would need to move them anywhere from 2 to 4 feet away from all of the walls/floor/ceiling.

 

That might be difficult to do, and you'll probably wind up with an additional ~3 dB dip in response somewhere above that frequency, in addition to that which you currently have between about 200-600 Hz.  I think the best way to deal with the 100-200 Hz midbass peak is to apply an attenuating EQ filter until you get the response flat across that band, and leave the bass bins where they are. 

 

Chris

DSP it is. 

[PrestonTom]

Using DSP is a good decision

Good luck, -Tom

[Robbie010]

Thread resurrection!

 

I have been researching the various DSP options and came across these Hypex NCORE plate amps with intergrated DSP:

 

https://www.hypex.nl/product/fusionamp-fa123/154

 

https://www.hypex.nl/documenten/download/1711 

 

2x 125W amplification, plus 100W tweeter channel, analogue and digital inputs, 15 biquads per channel.

 

Is there any reason why I shouldn't use something like this with my DIY La Scala?

 

I really like the idea of a neat and tidy package which allows direct connection of the source and even has the add-on for volume and source selection via IR remote control.

[Chris A]

If you're doing two-way, then it seems that a two-way Fusion Amp (FA122) would be a little less expensive than the three-way FA123.  If the DSP section of that plate amp/DSP crossover is quiet enough, then this is a pretty big deal, because the NCore amplifiers are basically the reference standard for amplifier performance.  I would make a decision soon if I were in your shoes (Brexit vs. non-Brexit-member taxes and export/import duties).

 

Chris

[Robbie010]

36 minutes ago, Chris A said:

If you're doing two-way, then it seems that a two-way Fusion Amp (FA122) would be a little less expensive than the three-way FA123.  If the DSP section of that plate amp/DSP crossover is quiet enough, then this is a pretty big deal, because the NCore amplifiers are basically the reference standard for amplifier performance.  I would make a decision soon if I were in your shoes (Brexit vs. non-Brexit-member taxes and export/import duties).

 

Chris

Thanks, Chris. It does look promising. 
 

I suppose I could initially buy 1 and do a quick, dirty setup to see what the noise levels are like. 

[Chris A]

Here's a similar configuration FA series amplifier with performance measurements:

 

https://www.avsforum.com/threads/measuring-amplifiers.1780914/page-72#post-56439070. 

 

It looks really good.  If I were to replace any of my present amplifiers, this is what I'd probably use. I presently bi-amp/tri-amp all loudspeakers using DSP crossovers in my setup--eight stereo amplifiers. The FA series amplifiers have been out on the market since at least April 2017, I find.  I don't know how I missed this, but looks like it really is a game changer for what I'd call the "new audiophile" crowd (a phrase that I coined to describe those that embrace DSP crossovers and clean class-D amplification).

 

I'd actually recommend using these FA 122s or FA 123s on Jubilees and the K-402-MEH, for instance if everything performs as advertised.  It certainly will clean up the clutter of my rack mounted D-75A amplifiers, eliminate the need for a First Watt amplifier (a class-A room heater) driving the K-402/TADs, the Xilica XP crossover, the miniDSP 2x4 HD, and allow me to finally eliminate my AVP in favor of a PCIe-to-AES  digital-to-digital conversion card (like the RME HDSPe AES card running in a desktop PC).

 

Chris

[Robbie010]

4 minutes ago, Chris A said:

I'd actually recommend using these FA 122s or FA 123s on Jubilees and the K-402-MEH, for instance if everything performs as advertised.  It certainly will clean up the clutter of my rack mounted D-75A amplifiers, eliminate the need for a First Watt amplifier (a class-A room heater) driving the K-402/TADs, the Xilica XP crossover, the miniDSP 2x4 HD, and allow me to finally eliminate my AVP in favor of a PCIe-to-AES  digital-to-digital conversion card (like the RME HDSPe AES card running in a desktop PC).

 

I'm thinking I'll go with the FA123, as this can do both 2 way and 3 way and gives a little more flexability.

 

One of the main things that has put me off going down the DSP route thus far is the thought of all the extra boxes - pre-amp, DSP box, multple power amplifiers etc etc. This unit seems to eliminate all that hastle.

[ClaudeJ1]

13 minutes ago, Chris A said:

Here's a similar configuration FA series amplifier with performance measurements:

 

https://www.avsforum.com/threads/measuring-amplifiers.1780914/page-72#post-56439070. 

 

It looks really good.  If I were to replace any of my present amplifiers, this is what I'd probably use. I presently bi-amp/tri-amp all loudspeakers using DSP crossovers in my setup--eight stereo amplifiers. The FA series amplifiers have been out on the market since at least April 2017, I find.  I don't know how I missed this, but looks like it really is a game changer for what I'd call the "new audiophile" crowd (a phrase that I coined to describe those that embrace DSP crossovers and clean class-D amplification).

 

I'd actually recommend using these FA 122s or FA 123s on Jubilees and the K-402-MEH, for instance if everything performs as advertised.  It certainly will clean up the clutter of my rack mounted D-75A amplifiers, eliminate the need for a First Watt amplifier (a class-A room heater) driving the K-402/TADs, the Xilica XP crossover, the miniDSP 2x4 HD, and allow me to finally eliminate my AVP in favor of a PCIe-to-AES  digital-to-digital conversion card (like the RME HDSPe AES card running in a desktop PC).

 

Chris

I concur with your assessment of the S.O.T.A. 2020 Audio landscape here, Chris. I'm using NC400's on my SH-50 mains for Stereo, and Behringer A800 Class D for the 2 Subs, then TI-class D chip amp based little boxes for  all the rest of the full Atmos channels from my Yamaha. Upgrading to the top of the line TI stuff 'cause it's cheaper for the less critical channels. I may use the Fusion Amps for my basement system which will be all horns also.......FH1's or Quarter Pie bass with Celestion Axi 2050's on K-402's and single OThorn Sub below 90 Hz. against the short wall. The newer Bruno Putzey designs of his new company are just a tad better (no pun intended) than the NCseries and perform just a shade less good than the THX-AAA design of the Benchmark 100Wpc Power Amp. Keeping in mind that most music does not tax the power output capabilities of most amps on horns, it's still the first Watt that counts the most, and with Hypex Tech. on Horns, there's still an overkill amount of headroom between normal volumes of 83 db at the sweet spot vs. Stupid Loud without ever clipping.

[Chris A]

32 minutes ago, Robbie010 said:

One of the main things that has put me off going down the DSP route thus far is the thought of all the extra boxes - pre-amp, DSP box, multiple power amplifiers etc etc. This unit seems to eliminate all that hassle.

Yes.

 

I'm sure that Klipsch consumer products engineers have probably been eyeing this for a while.  If I were in their shoes, this is how I'd rethink the replacement for the now defunct Palladium series, and perhaps gain additional market share in the process.  This takes this last issue out of the higher-end horn-loaded loudspeaker marketplace i.e., the time offsets that are built into the loudspeakers from the different length horns for each "way" of the loudspeakers. These plate DSP/amplifier modules could basically eliminate that issue. 

 

It would probably revitalize sales of Khorns, La Scalas, and even perhaps a powered version of the Cornwall IV, with time alignment...and user selectable EQ presets in the crossovers to adjust for different room boundary conditions. Perhaps this would also dramatically speed up the process of developing new loudspeaker models (including the Heritage series).  It's a lot simpler and faster to create DSP presets in the anechoic chamber (which they are likely doing anyway nowadays in order to design the passive networks) than to develop custom passive crossovers for each loudspeaker model.

 

It would also likely inject a burst of potential interest in the newer home-version Jubilee to have the ability to field-replace/upgrade compression drivers on the K-402s, including using dual-diaphragm drivers like the BMS 4592ND and the new B&C dual diaphragm driver by using a three-way FA123 plate amplifier in each loudspeaker.  Voila! All problems addressed at once.  In the DIY arena, the K-402-MEH would also be a cinch with this.

 

16 minutes ago, ClaudeJ1 said:

Keeping in mind that most music does not tax the power output capabilities of most amps on horns, it's still the first Watt that counts the most, and with Hypex Tech. on Horns, there's still an overkill amount of headroom between normal volumes of 83 db at the sweet spot vs. Stupid Loud without ever clipping.

A lot of people forget that these DSP crossovers offer something that passives cannot: absolute and instantaneous over-power/over-current protection from input transients and careless owners who use too much power and passive crossovers.  This could eliminate a lot of driver failures. 

 

The problem in any company is internal marketing and upper management saying "we can't sell that".  Getting past the "we've never done it that way before" stage is the real issue.  The delta production costs here I would think would be minimal--mostly the cost of the DSP/amplifier, subtracting the touch labor costs to build passive crossovers and test them.  You could even absorb some level of driver supplier performance drift with this. 

 

This is the day that I believe that John Dunlavy was hoping would arrive (RIP) to eliminate all the passive crossover tweaking in the chamber that he went through for each loudspeaker produced and that drove the costs of the product through the roof.

 

Chris

[ClaudeJ1]

Quote from John Dunlavy: "And we find that in order to reproduce those sounds with a level of accuracy such that you can not literally hear any difference between the live and the recorded sound, you have to have a speaker that exhibits almost perfect impulse and step responses. The only way to do that is to time-align the drivers very, very accurately, usually within a matter of a few microseconds, then use a minimum-phase, first-order crossover network and get everything right. And you have to have an on-axis response of better—well better—than ±2dB."

 

Looks to me that you have followed this advice in tweaking YOUR system, Chris!

[Chris A]

Dunlavy was right, in my experience.  That's the article that started me down the road to zero-phase-growth crossovers and much flatter SPL/phase (i.e., transfer function) response: