Linkwitz LX521.4 - new build and impressions

[jhenderson0107]

Nice. Any thoughts on cabling yet?....

Thanks. Yes, I CNC'd a hole in the top of the woofer housing for an eight-way Speakon prior to assembly.

Also, I CNC'd a plate for the rear base of the woofer housing to accept 4x banana jacks. I previously ran 8 conductor cable in wall to each location plus short Speakon-4x banana plug pigtails for each speaker (a carryover from my previous Linkwitz Orions). The cabling is visible on the floor behind each black speaker carcass within the image of the previous post.

The cabling and drivers within the woofer box will be hidden by the front/rear grills.

 

[jhenderson0107]

Image of LX-521 final assembly and in use:

 

[Nate29]

I just went through and read all 10 pages of comment. Love every min of it

I've been a fan of Linkwitz' work over the years and started out with building the LxMini many years ago.

I have always wanted to build a set of these Lx521 but never got the wife's approval to place them in the living room and my office/listening rooms over the years has been rather small.

However, I came back to oplug suppport recently and read up on Dave's work about a hybrid (you can re-use some of the Lxmini components etcs). TLDR, ordered a pair of the seas woofer (they're quite expensive now, a bit more than I remember back then, but I suppose Covid inflation happened ).

What I currently have is not quite Lx521, and Dave calls them the "hybrid lx521". You can read about the work in detail if you have lx521 owner/user access on the oplug support forum. Basically, the top 2 tweeters are removed , mid-range is crossover, and using only 1 (or 2) seas woofer.

Some advantages being :
1) Need only 4 amp channels vs 8-10 of the original Lx521 design
2) Can use smaller MiniDSP 2x4HD vs Flex 8/4x10HD (discontinued)
3) 2 Seas woofer vs 4 ($400 each on Madisound)
4) Minor, but smaller footprint since you can place the front baffle on a stand (like a bookshelf)

Now, this is not exactly Lx521, Neither I or anyone on oplug claims that. But those that have experiment on both suggest that it's 90% of what a full lx521 system could be.

I messed up the cut on the original baffle and ended up with this shape . The wife said she prefers it and it's fine since the top tweeters aren't being used in this hybrid setup. Though, Linkwitz really harped on staying true to the front baffle shape now matter how ugly it looks. I might go back and make a new front baffle at some point after the holiday season.

So far, this hybrid setup sounds amazing and I can only guess that the full lx521 system can be better if correctly setup in a decent sized room. I do hope to build the full 521's one day to complete this journey. Looking forward to it.

 

[jhenderson0107]

I used Audiolense XO v6.21 with multi-seat measurements of the LX-521's in my living room to generate 65K-tap FIR filters providing room-correction. Below are a set of representative smoothed measurements from the MLP.

Three target curves were used to synthesize correction filters, named Tilt, Trained and Olive - the latter two loosely modeled from the Toole preference results chart.

resulting in the corresponding correction simulations below

The resultant convolution filters were loaded in Roon presets to allow near-real-time comparison. All three filters provide easily-discernable audible changes when enabled. Subjectively, I like the Trained or Tilt responses best for most music, the latter recessing the midrange slightly compared to the former. The bass bump of the Olive configuration is excessive at normal listening volumes, but great for lower-volume listening or bass-deficient recordings.

The subjective changes in tonal balance were significant. We listened critically to a variety of favorite albums and tracks for two hours last night and were fully engaged in the music.

It was a worthwhile effort/experiment.

 

[jhenderson0107]

Following up on my Audiolense tests above: Simply taming the room modes at 45, 75 and 105 Hz in my room with PEQ filters sounds (at least) as good to me as the 65k convolved FIR Trained or Tilt responses above.

In summary, the unmolested bass performance of the LX-521s as-designed is wonderful and little/no frequency-dependent correction (above 300 Hz) is really required/desirable.

Or perhaps I'm just indiscriminate/easily pleased.

 

[bumpy48]

Can users please give a review of the tweeters

I need to know specifically if they are smooth and free of any aggresiveness. Tweeters that are not will trigger my tinnitus.

 

[suttondesign]

Can users please give a review of the tweeters

I need to know specifically if they are smooth and free of any aggresiveness. Tweeters that are not will trigger my tinnitus.

SEAS 27TFFNC/G (H1396-04) | HiFiCompass

This tweeter is used in the top level dipole project LX521 designed by world famous Siegfried Linkwitz: https://www.linkwitzlab.com/LX521/Description.htm https://www.madisoundspeakerstore.com/4-way-speaker-kits/lx521-linkwitz-lab-open-baffle-4-way-kit/

hificompass.com

 

[bumpy48]

Yes thanks very much
I've read most of the advertising and technical stuff, but much is focussed on the LX521 complete speaker. I would like to hear users reviews of the tweeters alone. I am looking to install back to back tweeters in my own DIY OBs.

 

[jhenderson0107]

I would like to hear users reviews of the tweeters alone. I am looking to install back to back tweeters in my own DIY OBs.

The LX-521s cross over to the tweeters at about 7.5kHz. There's little meaningful musical content above that so the tweeters merely contribute a sense of "air" or sparkle. Davey built an LX-521 baffle sans the tweeter, modified the crossover and it provided ~95% of the full experience. So, it's difficult to review this tweeter based solely on an LX-521 owner's experience.

That said, the frequency response, directivity and sensitivity of this tweeter is documented in its datasheet. That should provide enough to allow you to decide whether to prototype it in your DIY speakers.

 

[bumpy48]

Thanks.
I fully appreciate that no meaningful musical content exists over the tweeter and it is just the 'air' that I am seeking, and yes it is that missing 5% that I want.

I am having pretty good success with back to back Celestion HF1300 tweeters with 4uF capacitors, and cant really afford to prototype 4 x LX521 tweeters without just a few words of encouragement about the absence of agression.

Being in the UK I have no chance of auditioning them but hold a high regard for the designer of the LX521 and have read nothing about the LX521 itself being aggressive.

One additional question. Am I right in saying the LX521 tweeters are in series giving a combined impedance of 8ohms?

 

[jhenderson0107]

Am I right in saying the LX521 tweeters are in series giving a combined impedance of 8ohms?

They should be wired 180 degrees out-of-phase to compensate for their mechanical orientations. On an OB system, all drivers should move in the same direction when voltages are applied. With one tweeter facing forwards and another backwards, they must be cross-wired in parallel.

 

[bumpy48]

They should be wired 180 degrees out-of-phase to compensate for their mechanical orientations. On an OB system, all drivers should move in the same direction when voltages are applied. With one tweeter facing forwards and another backwards, they must be cross-wired in parallel.

I,m not that skilled at this elctronic lark, but doesn't this diagram show the tweeters in series?

 

[JP]

Indeed.

 

[jhenderson0107]

My apologies, I stand corrected. That's what I get for relying on memory. But do notice that the polarity of the tweeter wiring is indeed reversed in compensation for their drivers facing in opposite directions.

 

[bumpy48]

Just checking
As the drivers are 4 ohms each, in series do they present a nice 8 ohm impedence to the amplifier and for calculations for the high pass filter are they treated as a single 8 ohm diver?

 

[fluid]

If they actually presented a 4 ohm load then yes twin series would give you 8, Vituix agrees.

But in reality they behave nothing like that. So basic calculations are likely to steer you wrong. Measurements are your friend. You could even trace the response and impedance and see the effect of any filter in Vituix. Not as good as a real measurement but better than nothing.

 

[Jokerbre]

 

[Jokerbre]

Don't worry about the power of the amplifier. Any amplifier will do for that purpose. There won't be more than 1-3W. What should be taken into account when choosing an amplifier is that distortions must not increase with frequency.

 

[Jokerbre]

But it is more realistic that it will be mW. I've never seen the point of measuring at 5W. a more realistic picture is obtained when measuring at 100mW, 500mW, 1W. at 5W almost all of them work well.

 

[bumpy48]

Great info, but could I refer back to my original question please.

What are peoples impression of the tweeters themselves.

I would like to hear that they are airy, smooth and relaxing, and not that they are strident, piercing, harsh, peaky and tinnitus inducing.
Thanks