CLS-II ... some crazy mod ideas

[GreggoEggo]

Getting ready to pick up some CLS-II speakers in good shape, going to make a long road trip to pick them up and then enjoy them "as is" for a few months... but then the insanity begins. I would greatly appreciate any feedback on some ideas I am kicking around:

1) Going to save up a little extra and purchase new panels from Martin Logan at some point. I could get CLS-II panels but wondering if they would build me CLS-IIz panels instead, and if there might be any benefit to that approach?

2) They seem to have the power supplies (does that include step up transformers?) for either CLS-II or CLS-IIz, was thinking about getting a new set of either model... and stripping out literally everything else in my electronics modules and soldering my cables direct. No crossover correction filters, no power on/off delay or music sensor, nothing... just pure straight wire from power amp to transformer and then to panel. I would use a power strip to physically turn on the modules when I play music, and turn them off just before I stop playing music so the charge dissipates. I would also use a miniDSP upstream (digital only system) to put a digital low pass filter at 18 kHz, a digital high pass filter at 80 Hz and then use room correction to eq the panels for everything in between, 80 Hz to 18 kHz

3) Build some custom dipole subwoofer towers to flank the outside of each panel with separate DAC/amp taking in a digital signal with a digital low pass set at 80 Hz and a rumble filter set at 30 Hz.

I would have to have a qualified electronics tech evaluate the module and ensure there are not any special components need to manage the impedance perhaps ??? but in terms of the other components it seems like it would be possible to deal with power manually and deal with eq and protection (frequency extremes) in the digital domain. Has anyone ever tried this or thought about it?

From some of the threads I have read through on here it seems like I would get more dynamics and volume bypassing the binding posts and other components and then also be able to adjust the panel eq to my taste without severely compromising the great music playback qualities of these panels?

If this is feasible, I would build a system around it like this:

Apple iPad
miniDSP Flex Digital
RME ADI-2 DAC FS into a pair of Wyred 4 Sound SX-1000R mono blocks into the CLS-II/z panels
Schiit BiFrost 2 into a N-core based power amp DIY kit into a DIY dipole subwoofer tower

 

[Brad225]

I believe I still have schematics for the power supplies of both CLSll and llx if you want them. If you do, send me you email address via the PM here and I will send them to you.

Brad

 

[GreggoEggo]

Thanks Brad! pm sent...

 

[Brad225]

I sent you a pm about interface rebuild.

 

[Don Camillo]

Nice!, I always like out of the box thinking... (but imo often simpler is better).

For my CLSII (I think I have the A- models) I was also thinking of doing something like this... (or a bit different).

- with Ripole subs. (The CLS fullrange from 80 or 100hz or so)

- or use only the mids from the CLS, and as tweeters the ribbons from Apogee Duettas (I have a pair that can easily be seperated from their bad bass panels). So the small ribbon as tall as the CLS right beside them. And with some OB 15" untill the mid begins from the CLS.

That means I would have to seperate the sound to the CLS (with a dsp).

Or disconnect the lows and highs from the CLS in the electronic box?

Can that be done without problems?

Or will it always be all 3 playing together?

My thoughts are that passive with the right high end components sounds better and less "digital/electronic" than a dsp. So if it can be done in the electronic box I am curious about how.

The more dynamic lower and fuller sound from a large cone woofer could be a plus.

But seperating the sound would also bring phase issues...

Does someone knows if this has been done with good results?

 

[Leporello]

Getting ready to pick up some CLS-II speakers in good shape, going to make a long road trip to pick them up and then enjoy them "as is" for a few months... but then the insanity begins. I would greatly appreciate any feedback on some ideas I am kicking around:

1) Going to save up a little extra and purchase new panels from Martin Logan at some point. I could get CLS-II panels but wondering if they would build me CLS-IIz panels instead, and if there might be any benefit to that approach?

2) They seem to have the power supplies (does that include step up transformers?) for either CLS-II or CLS-IIz, was thinking about getting a new set of either model... and stripping out literally everything else in my electronics modules and soldering my cables direct. No crossover correction filters, no power on/off delay or music sensor, nothing... just pure straight wire from power amp to transformer and then to panel. I would use a power strip to physically turn on the modules when I play music, and turn them off just before I stop playing music so the charge dissipates. I would also use a miniDSP upstream (digital only system) to put a digital low pass filter at 18 kHz, a digital high pass filter at 80 Hz and then use room correction to eq the panels for everything in between, 80 Hz to 18 kHz

3) Build some custom dipole subwoofer towers to flank the outside of each panel with separate DAC/amp taking in a digital signal with a digital low pass set at 80 Hz and a rumble filter set at 30 Hz.

I would have to have a qualified electronics tech evaluate the module and ensure there are not any special components need to manage the impedance perhaps ??? but in terms of the other components it seems like it would be possible to deal with power manually and deal with eq and protection (frequency extremes) in the digital domain. Has anyone ever tried this or thought about it?

From some of the threads I have read through on here it seems like I would get more dynamics and volume bypassing the binding posts and other components and then also be able to adjust the panel eq to my taste without severely compromising the great music playback qualities of these panels?

If this is feasible, I would build a system around it like this:

Apple iPad
miniDSP Flex Digital
RME ADI-2 DAC FS into a pair of Wyred 4 Sound SX-1000R mono blocks into the CLS-II/z panels
Schiit BiFrost 2 into a N-core based power amp DIY kit into a DIY dipole subwoofer tower

AFIK II and IIz panels are the same, just the interfaces are different. Some report the II sounds better, but the IIZ is more amp friendly. Worth considering the IIZ if you're thinking about tubes (I'm unfamiliar with the monoblocks you mentioned). Soldering in speaker leads is way too tweaky for me. I use locking banana plugs and don't lose sleep over it. Eliminating 0.1 ohm certainly isn't going to improve dynamics, or anything audible, IMO.

The turn-on delay is rather annoying (has nothing to do with the purity once they're fired up, though). It was never an issue with my Ascents, but I think it's a matter of the larger panels and higher capacitance. I've thought about making it 12 volt triggered. If you just turn off the power to them, you lose the standby voltage. I guess it's a matter of debate whether turning them full off versus turning them on from standby makes an audible difference until the charge spreads uniformly (but if you're tweaky enough to want to solder your speaker leads...). ML recommends leaving them on in standby, versus on all the time, to reduce dust collection. When I had Acoustats, they recommended leaving them on all the time, but they had a fabric cover over them. I find that with the ML's in standby, they don't collect a lot of dust. But when I do vacuum them, I let them discharge first.

I don't have a schematic for the II's, but found one somewhere for the CLS (and I imagine there are different versions of each). My interface differs in component count, but the basic idea looks the same: there's a bias transformer that makes about 300V ac give or take. There's a voltage multiplier diode stack, 10 or 12 sections, don't remember the exact number. They're fed from the transformer via a voltage divider, one arm of which is shunted by a triac. When the triac is turned off, you get full voltage. The triac is triggered by an op amp comparator sensing circuit.

I wouldn't try to out-design Martin Logan on the eq neetwork. They know what they're doing. Personally, I'd go crazy trying to get them to sound right using DSP to replicate their function. That's what we paid ML the big bucks for. The different sized segments in the panels create a staggered set of poles in the electroacoustic transfer function, which the eq network compliments.

They're beautiful speakers just as is. I'm probably not going to part with them until they're prized from my cold dead hands.

 

[Leporello]

Nice!, I always like out of the box thinking... (but often simpler is better).

For my CLSII (I think I have the A- models) I was also thinking of doing something like this... (or a bit different).

- with Ripole subs. (The CLS fullrange from 80 or 100hz or so)

- or use only the mids from the CLS, and as tweeters the ribbons from Apogee Duettas. With OB 15" untill the mid begins from the CLS.

That means I would have to seperate the sound to the CLS (with a dsp) or disconnect the lows and highs...

My thoughts are that passive with the right high end components sounds better and less "digital/electronic" than a dsp.

The more dynamic sound from a cone woofer could be a plus.

But seperating the sound would also bring phase issues...

Does someone knows if this has been done with good results?

Closest thing I've heard about is the legendary stacked Quads/Deca tweeter/Hartley subwoofer system. My approach is to take it on faith the people who design speakers for a living--particularly ESL's--know way more about it than I do. Yes, taking a beautiful phase coherent panel and mixing it with separate tweeters, and crossovers, will create phase issues, to say the least.

 

[GreggoEggo]

Thanks all for the comments so far, very insightful and continue to give me more to think about. Of course, once I get these in my house (picking them up in July) I will just live with them "as is" for a while and see what I think. I am planning on buying new panels from ML by the end of the year and have already been in touch with them regarding cost and lead times.

I am a big believer in DSP, and I think there are many reasons that have nothing to do with actual performance that has limited their use in the higher end of this industry. However, I will be the first to admit that I am in way over my head with most of this stuff and also lack common sense more often than I would like. For example, here is my very first DIY speaker project:



Way way way too ambitious of a project but man what an experience... and my first time using DSP and I was blown away with what is possible and how it performed. Not world class in my case, but easily better than just about any 10k or under pair of commercial speakers I had heard at the time. That DIY project set me back around 6k however, and then eventually got scrapped and sold off after a few moves (those things are super heavy, with sand filled chambers and epoxy coated baltic birch plywood).

But the Javelin project was fun as I love doing my own industrial design, and it was really instructive around the benefits of DSP and individual speaker drivers connected directly to power amps.

I do agree that the ML design and engineering team deserves some respect and I probably shouldn't mess with their creation, but once I get new panels my remaining concern is the lifetime and serviceability of the electronics modules and if I can have them simplified and also make things a little more dynamic and even a little more friendly to the poor power amp driving these things... it might be worth the experimentation.

I just need to find a talented EE that can evaluate the schematics for me and possibly do the work on the modules when I am ready. Someone in Iowa, Kansas, Minnesota or Illinois so I can drive stuff out to them for contract work and then pick them up when it is done. These things are too much of a hassle to ship commercial.

 

[Brandon Hartwick]

You might want to run some of these ideas by the guys over on DIYaudio.com

Nothing against the members on here but there's a lot of guys on there that actually build speakers from scratch, including ESLs.

 

[Don Camillo]

Closest thing I've heard about is the legendary stacked Quads/Deca tweeter/Hartley subwoofer system. My approach is to take it on faith the people who design speakers for a living--particularly ESL's--know way more about it than I do. Yes, taking a beautiful phase coherent panel and mixing it with separate tweeters, and crossovers, will create phase issues, to say the least.

Yes thats true, the ML guys really know what they are doing. And I know nothing about electrostats... (But a bit about speakers and crossovers).

Also one thing is that I havent heard my CLS working correctly so far in my room (as they are defective at the moment).
So the real beauty of them I have still to discover.

But I am a fan of them for years and actually back then I thought I would never want or need a sub (as I am not a bass-head)....

But when mine play they lack too much in the lows to mids and tbh when you have heard a good clean ultra low bass, woww that is serious fun... (but for me the least important of the frequence spectrum).

I should get mine restored first and then see what they can do in my room...
thanks.

 

[GreggoEggo]

You might want to run some of these ideas by the guys over on DIYaudio.com

Nothing against the members on here but there's a lot of guys on there that actually build speakers from scratch, including ESLs.

Yes, definitely... I am going to hit up DIYaudio and TechTalk as I used to post often on both forums and had pages and pages of feedback and ideas as I documented my Javelin speaker project back in the day. Also hoping those communities can help me find a talented EE that might be interested in working with me on this if I decide to go through with it.

 

[Don Camillo]

Thanks all for the comments so far, very insightful and continue to give me more to think about. Of course, once I get these in my house (picking them up in July) I will just live with them "as is" for a while and see what I think. I am planning on buying new panels from ML by the end of the year and have already been in touch with them regarding cost and lead times.

I am a big believer in DSP, and I think there are many reasons that have nothing to do with actual performance that has limited their use in the higher end of this industry. However, I will be the first to admit that I am in way over my head with most of this stuff and also lack common sense more often than I would like. For example, here is my very first DIY speaker project:



Way way way too ambitious of a project but man what an experience... and my first time using DSP and I was blown away with what is possible and how it performed. Not world class in my case, but easily better than just about any 10k or under pair of commercial speakers I had heard at the time. That DIY project set me back around 6k however, and then eventually got scrapped and sold off after a few moves (those things are super heavy, with sand filled chambers and epoxy coated baltic birch plywood).

But the Javelin project was fun as I love doing my own industrial design, and it was really instructive around the benefits of DSP and individual speaker drivers connected directly to power amps.

I do agree that the ML design and engineering team deserves some respect and I probably shouldn't mess with their creation, but once I get new panels my remaining concern is the lifetime and serviceability of the electronics modules and if I can have them simplified and also make things a little more dynamic and even a little more friendly to the poor power amp driving these things... it might be worth the experimentation.

I just need to find a talented EE that can evaluate the schematics for me and possibly do the work on the modules when I am ready. Someone in Iowa, Kansas, Minnesota or Illinois so I can drive stuff out to them for contract work and then pick them up when it is done. These things are too much of a hassle to ship commercial.

Well Greg, that is something to be proud of, really!

For a first project you nailed it!

I habe been building diy speakers for over 3 decades now and wish I had gotten into dsp earlier. I have done all the crossovers (parallel and series) passive, with calculations and tweaking by ear.
Later I just used my ears... faster everytime... You learn a lot by doing that, but it takes long time!...

A Dsp has a LOT of benefits. Absolutely.
They use Dsp in SQ cars (sound quality: main goal) for the last 15 years and have improved much since then.

I untill now have it all passive and although with good crossover parts (I also have Satori in my car) the sound is really nice analog and natural, but it cant beat a good Dsp set up right in terms of time alignment, linearity and so.

So I can follow your thoughts.

But there is a pretty big difference in the quality of Dsp's.

I bought a Helix recently with good Dacs inside. Although its for my car to play with this summer, I thought about using it at home some time... and see what it can bring.

 

[GreggoEggo]

Well Greg, that is something to be proud of, really!

...

Thank you, appreciate the kind words. I agree about all your points on DSP and general state of the industry... I like miniDSP mostly because I think there price points, support, software interfaces, etc... are just so extremely good. I think the internal DAC implementations have gotten better, and I am generally in the camp that feels DAC differences have gotten smaller as literally everything on the lower end has gotten much better while the high end is now IMHO just a matter of taste much more so than absolute performance.

But yes, I think the case work, power supplies and DAC/analog circuits in the miniDSP product line are a little compromised from a truly high end audio perspective even though the performance is remarkable and for most people good enough if not even better than needed. I am probably on the border line in that regard. To play it safe and scratch my neurotic hifi itch, I am going to just go with the miniDSP Flex Digital (even if I leave the speakers alone, will use that device for subwoofer integration and bass management), then probably move my RME ADI-2 DAC FS into subwoofer duties (and full range headphone duties with different settings) and some good class D mono blocks for the subs, and for the CLS-II my latest thinking is the Flex Digital into Schiit Gungnir into Freya+ into mono blocks (maybe the Wyred 4 Sound SX-1000R) but those plans change often as I go through different cycles of day dreaming

 

[Don Camillo]

Yeah I know what you mean...

I've got some wishes too but you can spend so much that you want or have... to spend,

30 yrs ago if I would have had a million, then I would spend 80% on audio... it was that important (but I didnt have much to spend) today its max 20%.... or less...

And now I buy almost everything used...

Yes the MiniDsp is a nice one. And the recent ones are indeed probably a lot better improved over the early ones.

Maybe I'll go the same route with the minidsp.... who knows. If so Ill let you know.

 

[john65b]

Hello Greggo - I am on DIYAudio along with a few others...there are a few members there that really know what they are doing when it come to electrostatic speakers...taught me tons... a member there - bolserst - was extremely knowledgeable and helpful...

I also have played quite a bit with the ML Panels...replacing Bias rings and also replacing the diaphragms CLS Panels with 6 micron mylar. I also tried Aluminized 6 micron mylar (what that does is changes from a Constant Charge to Constant Voltage)...

I have two sets of CLS and CLSII panels that have mylar replaced...I used Sodium Lauryl Sulfate (found in hand soap) as a coating and worked great. I am planning to build CLS panels with special Velcro so I can easily remove, separate and recoat whenever I needed to...

I am in West Sub of Chicago, and while not a EE, a mechanical Engineer...who is a bit nuts about ESLs...

I have tons of various parts too (pair of custom made Cherry and Oak CLS Frames) ...

Anyway, below are my first panels and full range 3000v interfaces...standard components that are easy to find...

 

[Brandon Hartwick]

@john65b do you have a build thread anywhere detailing this project? It looks very interesting and I'd love to learn more about what you've done here.

 

[john65b]

Hello Brandon...I will have to dig up the DIYAudio thread...this was about 10 years ago...let me post back.....

 

[john65b]

Hello Brandon,

Here is a thread that shows the restoration of a CLS type panel. Tricky but doable...

DIYAudio link

Also of note - the mylar on the CLS stator is tensioned vertically - not horizontally. And exactly what the tension is is still a mystery to me...there is a % vertical elongation that I finally found, but never tried..,yet...

 

[john65b]

And another thread that explains the Physics around ESL speakers... kinda technical but worth reading if you want to fully understand operation...

DIYAudio link