Current vs voltage drive ESL? - Page 5 - diyAudio
Go Back   Home > Forums > Loudspeakers > Planars & Exotics

Planars & Exotics ESL's, planars, and alternative technologies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th October 2010, 09:39 PM   #41
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by bolserst View Post
The spacing, d , is not actually involved in defining the obtainable force per unit area.
Quote:
Originally Posted by Calvin View Post
That is just a part of the truth.
The formulas clearly show a anti-proportionality of attainable Forces to the distance d. It also shows that You need to increase the voltages to obtain the same Force-values when You increase d.
Since we must(!) stay below the maximum force which is defined by the flashover treshold, only the range below Fmax is of interest and here the 1/d proportionality applies. In other words. As long as the panel doesn´t flashover a decrease in spacing d will result in higher Force, hence higher SPL, if we keep Vpol and Vsig unaltered. On the other hand, if we increase d we need to increase Vpol and Vsig to rearrange for the same Force again. The choice of d should then only depend on the excursion needs. So, if we have chosen a value for d that meets the excursion demands, any further increase in d is not only obsolete, but actually counter productive, in that the required higher voltages (and associated with it increased Power) mean increased effort, inferior signal quality, less safety and speeded aging.
As far as I can tell reading your reply, your statements agree with my conclusion. The maximum obtainable force per unit area is defined by the flashover threshold for air, not the spacing d. As you say, if you increase d you need to increase Vsig and Vpol to get the same Force. But you can without risking flashover because the spacing is larger by the same ratio that the voltages need to be increased.

Choose the spacing required for the desired LF excursion and this defines the required Vpol and Vsig for maximum obtainable force. The maximum obtainable force would be the same no matter what spacing you chose. Only the Vsig & Vpol required to obtain it would be different.


Quote:
Originally Posted by bolserst View Post
2) If your transformer generates less than optimum stator voltages, turning up your bias beyond the optimum Vpol value will result in higher sensitivity and increased maximum SPL.
Quote:
Originally Posted by Calvin View Post
This is indeed correct, but who would design so in first place?
If I were to design a F1 car, I surely wouldn´t opt for a 125ccm Motorbike engine, but a 3.000ccm racing engine. Why when You can ?
A tranny that is incabable of supplying the optimal Vsig is simply not the right tranny for that certain application.
Two common reasons people use transformers resulting in less than optimum Vsig.
1) They choose to use cheap or readily available step-up transformers. Not everybody has the ability to design or $$$ to purchase an optimum transformer.
2) They are unable to achieve the bandwidth they want with a higher step-up ratio. Either limiting the HF due to parastics of leakage inductance and winding capacitance, or limiting the LF due to core saturation.
  Reply With Quote
Old 30th October 2010, 09:59 PM   #42
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
OK... but then with the best old step-up transformers you can get your hands on, you make biggish spacing, largish bias voltage, and then the old step-up transformer will be working at its most efficient, however limited that may be by your circumstances.

Is that wrong?
__________________
Dennesen ESL tweets, Dayton-Wright ESL (110-3200Hz), mixed-bass Klipschorn (35-110), and giant OB using 1960's Stephens woofer (18-35); Behringer DSP. HiFi aspirations since 1956
  Reply With Quote
Old 30th October 2010, 10:06 PM   #43
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by bentoronto View Post
I'd say hard to keep panels adequately stiff, especially as the surface dimensions grow. So there are hard limits a builder faces unless they can weld steel braces. But it is not too hard to increment voltages upwards with bigger spacing, if you have a solid mechanical base. With the high voltages around ESLs, I bet few people are rapping their knuckles on the panels or feeling them for vibration while playing - that's the way we always do it with speaker boxes. Ever wonder how badly the flimsy Quad panels rattle? Or sheets of perforated steel that everybody uses? You could trust your life to the strength of Dayton-Wright panels, by contrast. Why are we nutty about the solidity of speaker boxes but pay minimum attention to the rigidity of ESL panels?
Again you mention it is "not too hard to increment the voltages upward with bigger spacing".
Actually, in practice, it is.
Sure, there is not much difficulty in generating higher polarizing voltages. But, that is only have the equation. The other is the need for higher signal voltages. In my experience it is difficult to ramp up the signal voltage and retain the same upper and lower bandwidth. Now if you are only planning on using your ESL over a restricted bandwidth, such as a woofer and/or midrange, then generating higher signal voltages isn't as big a problem.

I totally agree about the necessity to build structurally sound ESL panels. I spend a great deal of effort ensuring they do not emit any undesirable vibrations. As has been mentioned elsewhere, it is generally quite embarrassing to apply a high level slow sine wave sweep to most ESLs, both DIY & professionally manufactured. It is a challenge as the stators need to be as acoustically transparent as possible, and some people prefer optically transparent as well.


Quote:
Yes, going big in any dimension means proportionally and absolutely less diaphragm displacement - but easier to achieve with spacing. Which, I am pretty sure, means better performance. The mathematics of ESL forces are not pretty, as best as I can remember Hunt. That's why big-voltage bias HAD to be introduced, eh. I think as soon as you start serious movement, you get into exponential instead of linear forces.
Hunt's research did show exponential forces for constant voltage ESLs.
But, must people build constant charge ESLs so that the forces are linear thru out the whole stroke right up until the diaphragm hits the stators. Refer to the Force equations Calvin and I have been posting. Force is a linear function of Vpol & Vsig. Position of the diaphragm in the gap is nowhere to be found in the dynamic portion of the equation. We didn't mention it, and probably should have, that these equations are applicable to constant charge ESLs only.
  Reply With Quote
Old 30th October 2010, 10:15 PM   #44
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by bentoronto View Post
OK... but then with the best old step-up transformers you can get your hands on, you make biggish spacing, largish bias voltage, and then the old step-up transformer will be working at its most efficient, however limited that may be by your circumstances.

Is that wrong?
No. Nothing "wrong" with that per se.
But your efficiency and maximum obtainable SPL will be more restricted than if you reduced the spacing and bias voltage to match the output capability of the step-up transformer.
  Reply With Quote
Old 30th October 2010, 11:36 PM   #45
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
Back to the armchair. There are very few things you have any clarity about before you build. You don't want to have more signal voltage than you can use. So you err on the side of spacing. You can readily get the bias voltage adjustable and potentially too high (before adjustment down). You build. Then you see how well your signal drives your panels.

With many unknowns (well, known precisely as models on paper, of course), the process will be iterative. Your bias is adjustable so that's helpful. So what don't you want to iterate? You sure don't want to iterate building panels. That leaves the signal drive to be changed upwards to meet the potential of your panels... assuming you left room (spacing) for upwards potentials.

Personal history: I've fooled with ESL over the years, but never with panel building. I have a hoard of Dayton-Wright panels (around 40 of them) as well as a complete speaker pair (8 identical panels per speaker). Lucky me, all I had to do with get bigger amps and bits of EQ to drive the collection of reversed tube output transformers I had collected. For about 20 years, I used my direct drive Sanders-like 2400 v B+ amp. Perhaps I am just promoting the kind of process I've gone through, starting with fine panels and adjusting the electronics to them.
__________________
Dennesen ESL tweets, Dayton-Wright ESL (110-3200Hz), mixed-bass Klipschorn (35-110), and giant OB using 1960's Stephens woofer (18-35); Behringer DSP. HiFi aspirations since 1956

Last edited by bentoronto; 30th October 2010 at 11:46 PM.
  Reply With Quote
Old 31st October 2010, 09:13 AM   #46
Calvin is offline Calvin  Germany
diyAudio Member
 
Calvin's Avatar
 
Join Date: Nov 2004
Location: close to Basel
Hi,

Quote:
1) They choose to use cheap or readily available step-up transformers. Not everybody has the ability to design or $$$ to purchase an optimum transformer.
The optimum transformer will be that which matches the panel´s demands best. That is not necessarily correlated to the attaced price tag and certainly no question of old or new stuff. It´s only a question of impedance matching and bandwidth.
If You look especially into older ESL Models -even of the big names- You´ll discover cheap transformers, that´s quality ranges don´t exceed or are even below that of nowadays standard E/I-core power trannies, let alone toroids.
Using panels with smallish d/s, hence low voltage demands and hence low U-value demands, generates the possibility to use less expensive but high quality Off-the-shelf trannies that outperform the old stuff quite easily. To the contrary it are the small production numbers, large d/s, high voltage, high-U fraction of trannies which need spezialized and complicated winding that costs loads of bucks.

Quote:
OK... but then with the best old step-up transformers you can get your hands on, you make biggish spacing, largish bias voltage, and then the old step-up transformer will be working at its most efficient, however limited that may be by your circumstances.
Don´t be surprised, but I don´t share this opinion.
There´s no reason why an old transformer should be capable of better performance than a new one or easier to get Your hands on. There´s also no reason why the best old transformers should only be of high-U value. It still applies that the lower the transformation factor could be the better the results will be. The best transformer is the one that matches the impedance values best and allows for sufficient bandwidth.
There are to my knowledge only few companies that used very high U-values of >>100 in their StepUps over the complete working range.
Dayton was one, Quad another. Acoustat and Audiostatic used high U values just over a restricted range in bass and lower mids.
The Daytons needed high-U-trannies because of the large d/s spacing which led to a low capacitance, hence highish impedance level of the panels. The Quads 57 needed a lot of U to drive the resistors that should lead to a current drive of the panels. Because of their d/s being reasonably small the Quads needed much less U when voltage driven.
The same basically applies to the Quad 63 and its heirs. Its smallish d/s needed much lower values of U if it were voltage driven and if they omitted with the crude delay line and used resistor segmentation instead (the difference would be that the impedance value and as such the need for high U-trannies would decrease considerably).
To me it seems that the problem You Ben and many others experience is, that You rather only think in voltage terms instead of thinking in impedance terms first and then look at the voltage levels.

If You want to build a panel all dimensions can be calcualted for the specced dynamic range and bandwidth. The only real variable and unknown factor is how much ´safety-margin´ in d/s You add to cater for dimensional aberrations of the stators, spacers and the x-offset from the polarizing voltage. This is indeed a factor of experience, but be sure its much less than 100% of the d/s for a decent panel ;-)
So even with a FR-Panel of reasonbaly proportioned size You should be able to get away with 3mm of d/s or less. If its more You should closely check the design again. ;-)

jauu
Calvin

Last edited by Calvin; 31st October 2010 at 09:23 AM.
  Reply With Quote
Old 31st October 2010, 10:26 AM   #47
markusA is offline markusA  Sweden
diyAudio Member
 
markusA's Avatar
 
Join Date: Dec 2009
Location: Gothenburg
Blog Entries: 1
I'll try this gain, please point me in the right direction.
Calculating current, impedance, driving voltages and the driving force on the diaphragm I understad somewhat but I can't make the leap to SPL?

Naturally I understand that more Force will lead to higher sound levels but I still can't make the mathematical connection to SPL?

Oh, I almost forgot... there's one thing in the force equation that puzzles me.
The X factor, i.e. the diaphragm offset. How do you attain the diaphragm offset? (I assume this is the bowing of the diaphragm because of the polarizing voltage?)

Last edited by markusA; 31st October 2010 at 10:30 AM.
  Reply With Quote
Old 31st October 2010, 11:26 AM   #48
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
Back here on Planet Earth, most of us are pretty clueless about what to expect from a step-up transformer, even in the highly unlikely event we had a bunch of parameters for that transformer. As Markus suggests.

The widespread advice at DIYaudio about finding some "old" (meaning here, "off the shelf") toroid power transformer... just how does that connect to sophisticated talk (some would say, gobbeldygook) about cores and U (I don't know what "U" is)? New or old, many of us work with what we find at the local electronic surplus store or can scrounge from an old tube amp, etc. Only manufacturers can (or need to) calculate the fine-points of transformers they order by the carton... after a lot of practical experimentation to see if engineering theory bears much relation to performance.

And to assess our progress, we deal with voltage because we can measure voltage.

With wider spacing, as Calvin points out, the panels become more invisible as a load as their capacitance decreases.

Footnote: if the folks at Acoustat chose the dinky-looking transformer that Calvin disparages (as well as many other vendors), I guess that was their best judgment that it was a reasonable match to their system and not because they really wanted to introduce an especially weak link. Dayton-Wright's 18kg transformer aside, these often do look shoddy, like most hidden-away transformers do.
__________________
Dennesen ESL tweets, Dayton-Wright ESL (110-3200Hz), mixed-bass Klipschorn (35-110), and giant OB using 1960's Stephens woofer (18-35); Behringer DSP. HiFi aspirations since 1956
  Reply With Quote
Old 3rd November 2010, 05:08 PM   #49
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by markusG View Post
I'll try this gain, please point me in the right direction.
Calculating current, impedance, driving voltages and the driving force on the diaphragm I understad somewhat but I can't make the leap to SPL?
Naturally I understand that more Force will lead to higher sound levels but I still can't make the mathematical connection to SPL?
From the lack of replies on this subject, you probably have gathered that the answer is:
“It’s complicated”
I will try to give a brief summary of what is involved as I understand it from reading the book Acoustics by Beranek.
The standard method used to calculate SPL from known driving Forces is to create an electrical circuit that is analogous to the mechanical and acoustic properties of the transducer you want to analyze. Force is replaced by voltage, mass by inductance, compliance by capacitance, damping by resistance. The actual values of the components can be calculated based on physical properties. Acoustic output is then proportional to current thru the real part of the radiation impedance.
For dynamic drivers some simplifications can be made because the moving mass of the driver dominates the other components in the circuit. For ESLs this is not the case. The dominate circuit components are those representing the radiation impedance which, as Calvin has pointed out, is a complex function of diaphragm size and shape which has no closed form solution. Also, this type of modeling does not take into account directivity as you go up in frequency and the wavelength gets smaller than the size of the ESL diaphragm.
For reference, I had posted some plots of radiation impedance for some diaphragm shapes in another thread you had started.
No0b ESL size question

“The easy way”
In post#8 of this thread kavermei already posted a link to an SPL calculator for ESLs. Some people questioned its validity, but I have found that when used with proper inputs and understanding it’s limitations it gives accurate results. It is based on “The Walker Equation” that Peter Walker developed using something called the reciprocity principle which completely side-steps all the complications mentioned above. If interested, the complete derivation can be found in Section 3.3.2 of the Baxandall paper.
As a sanity check of the online calculator, I compared the results for a 5cm wide line source with those calculated using the line source spreadsheet I had posted here:
experiences with ESL directivity?

See Attachment 1) & 2) for comparison. The agreement is quite good, both in SPL magnitude and in the expected 3dB/octave slope for an ESL line source. I know several posters mentioned never having seen measurements like this before. I have seen measurements like this many times. I am away from home on business right now, but happen to have a print out of a measurement I took last year of a 2" wide ESL line source. I have scanned it in as Attachment 3). The 3dB/octave slope is evident. The SPL magnitude is not shown in the measurement, but it did agreed within 2dB of the calculated values. I consider this pretty good agreement considering that the voltage loss in the stator insulation is not accounted for in the Walker Equation.


If you are interested in what the standard sensitivity would be for 2.83Vrms(1Watt/8ohm) change the Drive Voltage to 2.83 x (step-up ratio). See Attachment 4) for the SPL of the ESL line source from Attachment 1) driven with 300Vrms. Obviously the response still has the 3dB/octave slope which you will need to equalize out. So, the SPL “rating” for 2.83Vrms input will depend on how low you want to crossover your ESL. For example, if you wanted to cross over ~600Hz, your dB/2.83V = 70dB. This low value gives you an idea why it is generally advisable to make your ESL as large as possible. Each time you double the width of the line source, you increase this value by 6dB. Of course making your ESL wide reduces the dispersion of high frequencies which is generally not a desirable thing, although some people like it.

I hope this quick summary helped rather than adding confusion.
Attached Images
File Type: jpg ESL_sim_5000.jpg (50.2 KB, 152 views)
File Type: jpg ESL_line_8000p.jpg (122.4 KB, 153 views)
File Type: jpg ESL_line_measured.jpg (93.8 KB, 158 views)
File Type: jpg ESL_sim_300.jpg (49.8 KB, 151 views)
  Reply With Quote
Old 3rd November 2010, 06:52 PM   #50
markusA is offline markusA  Sweden
diyAudio Member
 
markusA's Avatar
 
Join Date: Dec 2009
Location: Gothenburg
Blog Entries: 1
I have been to the library and got myself some "light" reading.
Loudspeaker and Headphone handbook + an AES article.
It'll probably be a little while before I've managed to process all the information but I'm working on it.

Your summation did indeed clear somethig up for me. I had asked myself that exact question which you answer. Bases on the 2.83Vrms reference, depending on the step-up, what spl can one expect at what XO-frequency.

Reading the sim/curve that way helps a lot.
Looking at it like a regular frequency plot it's just confusing since I've never seen a speaker behaving that way.

It does make me wonder exactly what I'm seing in this grapgh and what makes it different from the spl plots I'm used to seeing. (The "flat" ones...)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
ESL Headphone direct drive Bazukaz Planars & Exotics 22 28th April 2012 10:07 PM
CG to drive DIY ESL SM7UYJ Chip Amps 6 19th May 2009 07:23 AM
ESL Direct Drive Atom666 Planars & Exotics 6 5th October 2008 08:09 AM
mix current drive with voltage drive at LF? capslock Multi-Way 0 27th April 2006 10:17 AM
2 x LM4780 in parallel to drive ESL panel Padel Chip Amps 6 3rd December 2004 04:44 AM


New To Site? Need Help?

All times are GMT. The time now is 07:56 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2