ESL, Ribbon THD

[a.wayne]

a.wayne: I share Lukas's confusion about why a sealed enclosure should yield a slower rise time for the woofer. The rise time will be determined by the upper frequency limit of the woofer. That is determined by the frequency and slope of the low pass filter used for the woofer, rather than by the nature of the enclosure, unless the woofer is used to such a high frequency that it's limited by its own high frequency extension.

Are you perhaps thinking of the time it takes for the woofer's output to settle back to zero after it's stimulated by a step response, for example? That settling time would be affected by the enclosure type because the enclosure affects the Q and tuning frequency of the woofer-enclosure system.

Of course once you add the low frequency room resonances to the picture the details of the woofer's behavior may get washed out. I'm inclined to think Earl Geddes' approach to the lowest frequencies may be the most sensible---multiple subwoofers arranged in the room to work with the room's resonances and smooth the overall response.

Few

Hello Few ,
By Using Earl's approach are you suggesting adding 1 extra woofer to the setup ? would this be randomly placed ?

I have added an additional subwoofer (s) in the past .
If using one , i have found the best placement to be in the center of the sound stage ...

I still stand by my original assessment , that a sealed enclosure will never match or setup well with a dipole speaker system , IMO..

 

[Calvin]

Hi,

@Martin
You´re right in so far, that the listening room spoils the theoretical performance abit. So the 6dB/3B won´t apply fully here, especially in smaller rooms. In a 30m² room the only possible way to notice this effect is to listen in very close distance to the speaker. Remind that a 3dB difference in loudness is just recognizable. So to hear the effect over distance You´d need longer distances, say >10m. Additionally You normally hear at a certain place and distance without much variation. So the variation is smaller than in theory, but walking through in a long big room the different character becomes obvious.

@Wayne

W- H ratio, very important with a flat panel , less so with a curve panel ..

Why could....should that be?

2.83V-issue. sorry, my text seems to have been a bit long. The third sentence read: "I normally use 2.83V......"

2.83 v with Impedance , magnitude and phase could get no clearer...

I disagree as long as there is no measuring distance stated! Measuring line sources at 1m is not very practical and the results are hardly comparable to the measurements of global distributing speakers. In fact most line sources don´t measure too well at all at this distance. One should understand, that the ´industry standard´ is good for industry standard level, like typical car tests are ok to evaluate a towncar. But like a formula1 car electrostats are a completely different league, which needs different standards.

regarding bass: no one was talking about big basses up till now! Talk was about ´heavy´ paper versus light film. And the weight of a ESL diaphragm is very light under all circumstances compared to normal speaker membranes.

Sealed enclosures have too slow a rise time and with the mass of it's radiated energy being from the front will have an acoustic center that is so far off the radiating pattern of a dipole, it will never be coherent ..

The rise time depends on the upper bandwidth limit of the driver and not on the cabinet style. I don´t understand the second part. What has mass to do with energy, radiation pattern, acoustic centre and coherence?
Like Lukas and Few I strongly disagree with what You say about vented systems. The impression of a precise bass is correlated to the group delay. The shorter the value the better. Dipoles beeing the first (~5-6ms), closely followed by CB(~6-10ms) and with due distance vented systems (>>10ms). I don´t regard dipoles as vented systems since they don´t introduce additional mass-springs (resonators) into the system. The lower You go in frequency the more the cylindrical dipole pattern transforms into a 2-lobes dipole pattern and the more similar to the global pattern of a normal bass it becomes. So the lower You go in crossover frequency (<100Hz) the easier it will be to use ´classical´ bass technologies with good success, with a tower of CBs giving the best results. With crossover frequencies >100Hz the dipole bass is the only technology that allows for seamless integration of panel and bass.

We will present the new systems at the HighEnd09 in Munic at the end of may and hope to find a american distributor or dealer. I don´t know when and where we will show in the US.
I visited the CES in January and don´t think we will attend there again. Maybe the Denver show, maybe somewhere else..but nothing fixed yet.
I think its not fair to ask me about a sonic comparison to ML. I wouldn´t be a good developer, if I wouldn´t be convinced of what I´m designing. I´m in the lucky situation that I can design basically without budget limitations. So our systems feature design elements and construction details, which others won´t have. They could rather be compared to the former ML Statement II, because its very much the same design concept. A CLX or Summit are for example conceptually different matters.
And in the end its always the personal taste of the listener that counts. Some will prefer our systems some the MLs. We had a good chat with the Boys of ML at last year´s HighEnd and they obviously liked very much what they heard.

@Lukas

It will likely have lower efficiency , but what is the problem of using higher turns ratio step-up ?

The problem is that efficiency should be the prime goal!
And just incase You don´t believe me.....RS also claimed for it
Ask a transformer designer about what he thinks is better and he´ll answer You that lower is always better. Losses, capacitances, coupling factor, ease of winding, insulation, drive requirements.....basically every parameter gets worse with increasing T-factor. Sonically the higher You step up the less dynamic and open it sounds. You´re giving up on the dynamic range and won´t be able to recover the loss by pushing more watts into the speaker. It simply won´t work.

Still , I think that a properly designed sealed box with active equalization should outperfom vented , bandpass, and similar boxes in transient response. Is this audible ? I don't know

Agreed! Vented is for HiFi, Dipole and CB for music

jauu
Calvin

 

[Bazukaz]

The problem is that efficiency should be the prime goal!
And just incase You don´t believe me.....RS also claimed for it
Ask a transformer designer about what he thinks is better and he´ll answer You that lower is always better. Losses, capacitances, coupling factor, ease of winding, insulation, drive requirements.....basically every parameter gets worse with increasing T-factor. Sonically the higher You step up the less dynamic and open it sounds. You´re giving up on the dynamic range and won´t be able to recover the loss by pushing more watts into the speaker. It simply won´t work.

jauu
Calvin [/B]

Hi Calvin,

Eefficiency is not the prime goal for everyone . If I'll get somewhere around 85 db/2.83V efficiency or more , I will be statisfied .

I agree with you that a single transformer with high turns ratio is difficult and probably not very practical to design. But in that case more transformers can be used(up to a certan degree , for sure). Mains toroids are not very expensive , and some of them have really good perforcmance in audio band (One example I tested had <0.05% THD in desired frequency range , even when driven not too far from core saturation).

Segmented wire panel will have much lower capacitance , because it is not wasted by spacers and also because of electrical segmentation. So capacitive loading on amplifier should not increase much , despite of higher turns ratio.

Regards,
Lukas.

 

[Calvin]

Hi,

thats why I wrote ´should´......If You want to build the best panel (regardless of flat, curved or segmented style) You have to aim for highest efficiency, point. If You´re content with something mediocre...go ahead
With efficiency I rather meant the panel itself and the combination of panel+tranny in second place. A high efficiency panel is always one with comparatively high capacitance, simply because of more membrane area and/or smaller d/s. The efficiency of the panel reflects by a lower transformation factor to the amplifier side for the same SPL. You can of course fudge a bit with this factor, if impedance issues ask for it, but if You can choose between a tranny doing 1:50 or one doing 1:100 opt for the lower value, i.e. the more efficient panel.

jauu
Calvin

 

[bear]

I think we're talking about sensitivity mostly.

Efficiency is a slightly different animal. May or may not matter for us.

One of the key problems in an ESL is the relationship between the gap, the bias voltage and the maximum excursion vs. freq the diaphragm can manage (assuming unlimited drive for the moment), which sets the max SPL possible.

If ur at "85dB" SPL/1m (whatever that really means) one starts to be concerned that you can reach high enough SPL levels to avoid clipping on peaks (never mind if you like playing things way way too loud).

Which is why I asked Calvin about his apparently extremely high SPL quote... unusually high for an ESL.

Thinking out loud, is that for the ESL section alone?

_-_-bear

 

[Calvin]

Hi,

the 110dB@4m@50W is for one panel. We all didn´t trust the measurement at first. We assumed that there might have been a failure in setup or a wrong value in one of the menues. So everything was checked again and prooved to be allright.
With Dr. A.Goertz performing the measurement I don´t have the slightest doubts about the professionalism and correctness of the values, but am just happy that my panels perform so well
And everybody who heard these panels would shurely agree that they perform with outstanding dynamics.

jauu
Calvin

 

[Bazukaz]

So , as promised , I did THD measurements of my panels. Measurements were carried out in my living room, signal level was around 2-4 Vrms. THD is slightly higher than that measured from transformer's secondary winding, but room noise (computer , etc) as well as microphone adds some.

Regards,
Lukas.