Copper tape always comes in handy I this appilcation.
However I just remebered that copper also corrodes and oxidizes very easily and a while back WacharaC. had posted som pics of his copper tape charge ring corroding due to the high humidity of his area.
Just a liittle something when it comes to long term use. jer
P.S But don't be alarmed as there are ways to combat such issues.
However I just remebered that copper also corrodes and oxidizes very easily and a while back WacharaC. had posted som pics of his copper tape charge ring corroding due to the high humidity of his area.
Just a liittle something when it comes to long term use. jer
P.S But don't be alarmed as there are ways to combat such issues.
I'm so excited right now I almost started my build thread yeasterday. It would be jumping the gun though. I'm still sourcing the BOM.
I have a small question though. Bear with me if this is to stupid, I had a late night out last night and I'm feeling the aftermath...
To have an easy to drive ESL you want High reactance?
Large stators tend to lower the reactance?
If you do curved panels you'll have a bunch of horizontal spacers. What if you isolate them (the sections) from each other and then connect them in series? Is this a good idea or just plain stupid? It could be as easy as just not coating the membrane under the spacers?
I have a small question though. Bear with me if this is to stupid, I had a late night out last night and I'm feeling the aftermath...
To have an easy to drive ESL you want High reactance?
Large stators tend to lower the reactance?
If you do curved panels you'll have a bunch of horizontal spacers. What if you isolate them (the sections) from each other and then connect them in series? Is this a good idea or just plain stupid? It could be as easy as just not coating the membrane under the spacers?
On my first ESL's, I used 3M foam tape spacers and a copper foil tape charge ring made from 1/4" wide copper foil purchased from McMaster-Carr. Two years later I disassembled those panels to upgrade the diaphragm and stator coatings and I found no visible corrosion on the copper foil. Of course, my speakers are in a relatively non-humid, air-conditioned environment, the copper foil was encapsulated/sealed within the the foam tape spacers and the bias polarity was negative (I read someplace that a positive bias polarity can corrode the contact).
I'm wondering how screen stators would sound. I've never built or listened to any screen wire stator ESL's myself so perhaps my concerns about their relative low mass (in comparison to perf metal or larger-diameter wire stators) are unfounded. After all, a lot of people apparently build mesh stators and are happy with them.
Anyway, whenever I touch one of my .048" thick steel stators while the music is playing I can feel vibration in the metal. Based on that observation, I figure that 1) all stator materials vibrate to some extent, 2) any vibration has to generate distortion, 3) lower mass materials would be more prone to vibration, 4) and at some point of reduced stator mass, the resulting distortion would become objectionable.
Again, absent any actual experience building wire mesh ESL', my concerns are only conjecture. For all I know, their greater open area may even sound better than lesser open area perf stators. I would be interested to hear opinions from anyone who's listened to both wire mesh and perf metal stator ESL's. What do you think, Jer?
Last edited:
Makes sense that you would feel some vibration. Every action has a equal and opposite reaction.
Yes I agree ,Charlie,those are all very vaild points and I have read many discussions on this matter.
My theory is that metal panels are prone to ringing (as all materials by the way) at a particular frequency like a bell when it is struck and this can be difficult to dampen even after they have been coated.
Where as my plastic supported panels also have a particular ring aswell but they do seem to be dampened quite a bit naturaly and I have even thought of possibly a generous coat of polyurathene on the grid frame might just about stop it completely (almost) leavin just the diagphram resonance to contend with.
Yes, They do vibrate profusley as I have reached ther mechanical limits on several occasions.
But, I also currently do not have them mounted in a suitable frame yet.
Once they are properly mounted (as well as a metal panel) these issues would not be realivant.
And any energy created by the ringing would be minute' compared to the output of the panel.
However, I have read that this is very much a problem when using a metal panel that is too thin such as lincain.
Although I have never tried it yet,it would probably be okay for a smaller panel the size like mine.
But, I wouldn't suggest using it unless you take extra measures for support as the stuff is very thin and flexible. jer
My theory is that metal panels are prone to ringing (as all materials by the way) at a particular frequency like a bell when it is struck and this can be difficult to dampen even after they have been coated.
Where as my plastic supported panels also have a particular ring aswell but they do seem to be dampened quite a bit naturaly and I have even thought of possibly a generous coat of polyurathene on the grid frame might just about stop it completely (almost) leavin just the diagphram resonance to contend with.
Yes, They do vibrate profusley as I have reached ther mechanical limits on several occasions.
But, I also currently do not have them mounted in a suitable frame yet.
Once they are properly mounted (as well as a metal panel) these issues would not be realivant.
And any energy created by the ringing would be minute' compared to the output of the panel.
However, I have read that this is very much a problem when using a metal panel that is too thin such as lincain.
Although I have never tried it yet,it would probably be okay for a smaller panel the size like mine.
But, I wouldn't suggest using it unless you take extra measures for support as the stuff is very thin and flexible. jer
Last edited:
Questions are aways welcome as sometimes it is the only way to learn so always ask questions!
Yes, You do want a higher reactance is this makes for an easier load (higher impedence) on the amplifier.
Yes, normaly a larger panel warrants more capacitance therefore a lower reactance and a lower impedence.
However keep in mind that the spacing between the stators play a big factor as to how much capacitance the panel will have.
And with that said the transformation ratio with the panel capacitance plus the transformer capacitance determins the reflected imdedence that the amplifier will see.
I have never built a curved panel so I don't have any tips for you in that area.
I have series connected my panels and it does work but it reduces the overall efficientcy because the voltage across each panel is reduced.
Just like a resistor voltage divider only using capacitors instead.
I hope that helps. jer
Last edited:
Hi,
Reactance (X) is the imaginary part of the impedance (Z). It is the part that generates the phase shift.
Typical amplifiers work best into a resistive impedance (R) of a certain defined value -or value range, say 2-16Ohms.
Too far off of the resistive design centre value and the amp can´t supply its full power. And a large imaginary part may lead to oscillation or even fatal breakdown.
Any capacitance regardles of its value is a pure reactance featuring a 90° phase shift, hence no resistive component at all. What might be disturbing is the fact that the resistance (R) as well as the complex impedance (Z=R+jX) are measured in Ohms (the j indicating the imaginary nature of X).
The impedance of a capacitance drops linear with increasing frequency. Its value rising with lower capacitance values (1/(2pi*f*C). So a larger panel area means larger capacitance value, means lower impedance value. But the phase shift stays alway 90° -pure reactance.
In practise there are always resistive losses, so that the real capacitor only comes close to 90° phaseshift, but never actually reaches 90°.
And it is only the small resistive part which can be used for acoustic output.
The impedance of an ESL doesn´t match the required load impedance of the amplifier. So a impedance matching device is needed....the audio transformer. It transforms the high impedance values of the ESL to a much lower value the amp can work with. Beeing a complementary device to a capacitance and due to its lossy nature it shapes the impedance response from a high and linear-falling-with-rising-frequency-response to a lower value and a bell-shaped, kind of gaussian curve, which features also lower phaseshift values. In other words the transformer makes the impedance less reactive and more resistive (real) to the amplifier. So the more lossy the transformer the lower the phaseshift the amplifier sees, the easier it can drive the load.
Another method is to make the capacitance smaller with increasing frequency and in doing so to change the impedance response from linearly falling to a more constant value. This method is called electrical segmentation. Besides the positive influence on the impedance response it adds resistive components to the impedance and it ´reduces´ the active membrane area with rising frequency. This way Xou can shape the amplitude response and distribution character to a degree at the same and ease the amplifiers task.
jauu
Calvin
Sorry, but this is totally wrong.
Reactance (X) is the imaginary part of the impedance (Z). It is the part that generates the phase shift.
Typical amplifiers work best into a resistive impedance (R) of a certain defined value -or value range, say 2-16Ohms.
Too far off of the resistive design centre value and the amp can´t supply its full power. And a large imaginary part may lead to oscillation or even fatal breakdown.
Any capacitance regardles of its value is a pure reactance featuring a 90° phase shift, hence no resistive component at all. What might be disturbing is the fact that the resistance (R) as well as the complex impedance (Z=R+jX) are measured in Ohms (the j indicating the imaginary nature of X).
The impedance of a capacitance drops linear with increasing frequency. Its value rising with lower capacitance values (1/(2pi*f*C). So a larger panel area means larger capacitance value, means lower impedance value. But the phase shift stays alway 90° -pure reactance.
In practise there are always resistive losses, so that the real capacitor only comes close to 90° phaseshift, but never actually reaches 90°.
And it is only the small resistive part which can be used for acoustic output.
The impedance of an ESL doesn´t match the required load impedance of the amplifier. So a impedance matching device is needed....the audio transformer. It transforms the high impedance values of the ESL to a much lower value the amp can work with. Beeing a complementary device to a capacitance and due to its lossy nature it shapes the impedance response from a high and linear-falling-with-rising-frequency-response to a lower value and a bell-shaped, kind of gaussian curve, which features also lower phaseshift values. In other words the transformer makes the impedance less reactive and more resistive (real) to the amplifier. So the more lossy the transformer the lower the phaseshift the amplifier sees, the easier it can drive the load.
Another method is to make the capacitance smaller with increasing frequency and in doing so to change the impedance response from linearly falling to a more constant value. This method is called electrical segmentation. Besides the positive influence on the impedance response it adds resistive components to the impedance and it ´reduces´ the active membrane area with rising frequency. This way Xou can shape the amplitude response and distribution character to a degree at the same and ease the amplifiers task.
jauu
Calvin
Last edited:
Even though I've been in a zombie state the last couple of days from lack of sleep due to attending a large convention I really appreciate this.
It's good stuff!
My mind might not comprehend it fully at the moment but it does ring a bell.
Complex loads!
(I have the -j vs j relationship pictured in my mind...)
Before I embarrass myself too much I think there are two questions that need an answer.
1. What makes an easy load for an amplifier? (I was thinking the lower the capacitance the better but it would appear nothing is as easy as it looks?)
- What kind of load do we want?
- How can we make the ESL easier on the amp?
- What kind of amp do we need to drive the ESL?
2. If a transformer acts as a "counterweight" in a positive sense and balancing out the capacitive part of the load is a good thing, how come everyone is considering direct drive the holy grail of amplification?
It's good stuff!
My mind might not comprehend it fully at the moment but it does ring a bell.
Complex loads!
(I have the -j vs j relationship pictured in my mind...)
Before I embarrass myself too much I think there are two questions that need an answer.
1. What makes an easy load for an amplifier? (I was thinking the lower the capacitance the better but it would appear nothing is as easy as it looks?)
- What kind of load do we want?
- How can we make the ESL easier on the amp?
- What kind of amp do we need to drive the ESL?
2. If a transformer acts as a "counterweight" in a positive sense and balancing out the capacitive part of the load is a good thing, how come everyone is considering direct drive the holy grail of amplification?
This might help you to have a little insight of the types of problems of driving esl's.
http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design.html#post2088330
And this is a summary of my findings,
http://www.diyaudio.com/forums/planars-exotics/181041-my-2-nd-esl-attempt-2.html#post2520508
Although it may seem discoraging, don't be discoraged.
The reasoning behind my study was to find out exactly how things worked since there was no solid information as to what the exact limits are, and, to take the mystery out of what it takes to drive esl's ,sort of speaking.
Yes, a direct drive amplifier is the ultimate goal but I must remind, all, that working at those high of voltages and current levels to properly drive even a small panel is quite dangerous and can be very deadly and catastropic should a mishap occur!
I have driven my little panel to its mechanical limits on several occasions and the voltages required to do this was on the order of more than 26kv peak to peak.
It is hard enough to contain and work with voltages in the 5kv to 10kv range let alone trying to control them at high current levels.
As the panel size goes up the capacitance goes up and therefore the current requirements go up drasticaly aswell.
To the point that it is at very very dangerous and lethal levels.
However it does not say that it can't be done as others have done it but it is just not practical, Although it is one of my goals to acomplish aswell.
If you are worried about any distortions and losses being introduced because of the transformer, These can be delt with by a proper transformer build up.
And the better the core material the lower the THD.
And the bigger the core there is less chance of core saturation at the lower frequency's.
It is this saturation that is most notable as distortion that also occur in tube amplifiers.
But even with using what is commonly available the results have been amazing for the some of us that have tried it.
It is in my plans to show such THD figures sometime in the near future.
In the preliminary test that I have done, I assure you that they are much less then .5% to .1% as a round about maximum figure and with real results at least 1/10 of that at a normal listening level, using a common power toriod transformer as a stepup transformer.
Those test were from a panel being mic'ed and included all of the electronics in the chain.
Most ( But not all) typical cone speaker system types can't even come close to these figures on an average.
And if you have a system with even as much as 2% THD or less through out the audio bandwidth on an average, then you have a really super nice speaker system.
As with all speakers, There are some compromises to be delt with.
But IMHO it is all worth it in the end. jer
http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design.html#post2088330
And this is a summary of my findings,
http://www.diyaudio.com/forums/planars-exotics/181041-my-2-nd-esl-attempt-2.html#post2520508
Although it may seem discoraging, don't be discoraged.
The reasoning behind my study was to find out exactly how things worked since there was no solid information as to what the exact limits are, and, to take the mystery out of what it takes to drive esl's ,sort of speaking.
Yes, a direct drive amplifier is the ultimate goal but I must remind, all, that working at those high of voltages and current levels to properly drive even a small panel is quite dangerous and can be very deadly and catastropic should a mishap occur!
I have driven my little panel to its mechanical limits on several occasions and the voltages required to do this was on the order of more than 26kv peak to peak.
It is hard enough to contain and work with voltages in the 5kv to 10kv range let alone trying to control them at high current levels.
As the panel size goes up the capacitance goes up and therefore the current requirements go up drasticaly aswell.
To the point that it is at very very dangerous and lethal levels.
However it does not say that it can't be done as others have done it but it is just not practical, Although it is one of my goals to acomplish aswell.
If you are worried about any distortions and losses being introduced because of the transformer, These can be delt with by a proper transformer build up.
And the better the core material the lower the THD.
And the bigger the core there is less chance of core saturation at the lower frequency's.
It is this saturation that is most notable as distortion that also occur in tube amplifiers.
But even with using what is commonly available the results have been amazing for the some of us that have tried it.
It is in my plans to show such THD figures sometime in the near future.
In the preliminary test that I have done, I assure you that they are much less then .5% to .1% as a round about maximum figure and with real results at least 1/10 of that at a normal listening level, using a common power toriod transformer as a stepup transformer.
Those test were from a panel being mic'ed and included all of the electronics in the chain.
Most ( But not all) typical cone speaker system types can't even come close to these figures on an average.
And if you have a system with even as much as 2% THD or less through out the audio bandwidth on an average, then you have a really super nice speaker system.
As with all speakers, There are some compromises to be delt with.
But IMHO it is all worth it in the end. jer
To avoid any misconceptions...
I'm building a standard ESL per previous recommendations.
My questions are for greater understanding, I like to know what I'm actually doing.
I've tried to fight it but I have to admit.
I am an engineer at heart.
Step one is always to set up the goal-specifications and here knowledge is key.
I'll take a serious look at the links and re-read your posts a few times.
I'm building a standard ESL per previous recommendations.
My questions are for greater understanding, I like to know what I'm actually doing.
I've tried to fight it but I have to admit.
I am an engineer at heart.
Step one is always to set up the goal-specifications and here knowledge is key.
I'll take a serious look at the links and re-read your posts a few times.
Very good, I know exactly where you are coming from.
I contemplated my build for many many years and it wasn't until I actualy built one that it all became clear and how simple it really is.
Only back when I did it in 2003 trying to find a suitable transformer to drive them with was quite a task.
And it was only recently that this has been solved and I started up my project again.
Good luck and keep us posted on your results. jer
I contemplated my build for many many years and it wasn't until I actualy built one that it all became clear and how simple it really is.
Only back when I did it in 2003 trying to find a suitable transformer to drive them with was quite a task.
And it was only recently that this has been solved and I started up my project again.
Good luck and keep us posted on your results. jer
Hi,
famous last words: " Ohh, ....thats easy!"
I can hardly think of anything other appearing so simple on one hand and beeing so tricky on the other than a good ESL. Nailing together a board that puts out sounds is indeed easy ..... getting it to perform at the limits of this technology and on a constant consistant niveau makes the difference.
Q:
- What makes an easy load for an amplifier?
A: -Whatever load the amplifier is designed to work into!
The typical audio amplifier works as fed-back constant voltage source best into a purely resistive load. The range of allowed phase shift the load introduces, depends on the design and construction of the amplifier. If the phase shift aproximates or even exceeds the safety range the amp becomes unstable. Its impulse response is spoiled and eventually it may oscillate.
- How can we make the ESL easier on the amp?
A: - For the above described typical audio amplifier the only means to take are to ensure, that under no circumstances the load demands exceed the amps capabilities. In most cases making the load appear more resistive (i.e. reducing phase shift) helps. This could mean either a dedicated series power resistor ahead of the audio transformer, or a more lossy audio transformer (increased value of stray inductance) or means within the panel like electrical segmentation, etc. The somehow weird appearing thing is, that a better transformer makes the amp´s life harder.
- What kind of amp do we need to drive the ESL?
A: -Since the amp needs to supply for the usable real power into the load as well as the reactive power that circles within the amp, the powersupply, the power stage and cooling devices need to be designed with sufficient headroom. A phase shift of +-45° already ask for 1A of reactive current for every 1A of real current supplied to the load. The relationship quickly rises with increasing phase shift (60°->1.7, 75°->3.7, 80°->5.6). Amplifiers with simple protection circuits/limiters that measure the voltage over a Emitter- or Source-resistor often limit much too early and prevent the amp from driving the load properly (think of the power cube diagrams printed in some magazines).
Then there are methods to increase the phase shift safety margin of an amplifier, for example by modificating the feedback path, reducing the amps bandwidth, breaking up the global feedback path, adding resitors between amps output and load etc.
Especially the last named measures can more often be found in Tube amps than in Solid State. Low bandwidth, no global feedback, resistive component of the audio output transformer, etc. are more typical in the Tube world. Tubes are a good starting point when it comes to properly driving ESLs. As such the amplifier I found the most stable and best sounding are the single-ended class-A non-nfb Triodes of KR Audio.
My ESLs reach phaseshift values of up to 85° which unfortunately means that the number of amps capable to drive the ESLs properly and also sounding well is rather small. The KR Audio Krionzilla with just ~50W of specced power not only remain stable up to highest volume levels but sounds exceptionally well and really brings the utter best out of the ESLs to shine.
If one omits with the audio transformer alltogether, moving into the world of direct drive, one can and one needs to design the amplifier differently to the standards applying to low voltage audio.
If You want to drive a capacitive load from a constant voltage amplifier it has its issues as explained before. In the end it needs a current to charge and discharge the ESLs capacitance. So an amplifier working in the ´current domain´ may show better behaviour. You could for example design an amplifier, that doesn´t work as constant voltage source but as a constant current source.
Q:
- ....how come everyone is considering direct drive the holy grail of amplification?
A:- It might be, because a good HV-DD-amp is equally seldom to find.
There are many myths circling the audio world and quite alot of those myths are told in the ESL-world. A DD-amp beeing the best to drive an ESL properly is such a myth. The truth is, that it ranges from ´absolutely true´ to ´absolute BS´.
The concept of direct drive bears certain technical advantages. It allows for tighter control of the load, it omits with the complex coupling device, the audio transformer, and it allows for specialized amplifier topologies which are better suited to drive a capacitive load. In theory DD is clearly the better way to go. But we live in a real world and in practical DD has its own and serious problems. Besides safety issues already mid sized panels typically require very high drive voltages. 10kV+ are not unreal. You can´t use easy to source, off-the shelf standard power devices. There are some rather exotic tubes (at least exotic to the audio world) that may be useable. Things even get worse in the Solid State world. All power devices capable of several kV supply voltage are intended for switching useage not for linear amplification. Most of theses devices are intended to switch dozens of amperes of current (so they feature high internal capacitance values and they are rather slow) while we only need less than 1-2A. Cascading the power devices breaks the voltage barrier, but this trick also doesn´t come for free. Further Development steps and the upcoming SIC-technology might provide us with a solution in near future, but at the time suitable Solid State power devices are not at hand.
The amplifier topologies generally accepted as well sounding would result in terribly inefficient DD-amplifiers generating enormous amounts of heat. If on the other hand a low voltage audio amp would feature a topology as used in some better known DD-amps, probabely none of the HighEnders yelling "DD is best" would dare to hook up such a amp in their system.
It´d be hard to top the results of a well matched ESL-Transformer combination with a DD amplifier. And it would be of course much costier.
jauu
Calvin
famous last words: " Ohh, ....thats easy!"
I can hardly think of anything other appearing so simple on one hand and beeing so tricky on the other than a good ESL. Nailing together a board that puts out sounds is indeed easy ..... getting it to perform at the limits of this technology and on a constant consistant niveau makes the difference.
Q:
- What makes an easy load for an amplifier?
A: -Whatever load the amplifier is designed to work into!
The typical audio amplifier works as fed-back constant voltage source best into a purely resistive load. The range of allowed phase shift the load introduces, depends on the design and construction of the amplifier. If the phase shift aproximates or even exceeds the safety range the amp becomes unstable. Its impulse response is spoiled and eventually it may oscillate.
- How can we make the ESL easier on the amp?
A: - For the above described typical audio amplifier the only means to take are to ensure, that under no circumstances the load demands exceed the amps capabilities. In most cases making the load appear more resistive (i.e. reducing phase shift) helps. This could mean either a dedicated series power resistor ahead of the audio transformer, or a more lossy audio transformer (increased value of stray inductance) or means within the panel like electrical segmentation, etc. The somehow weird appearing thing is, that a better transformer makes the amp´s life harder.
- What kind of amp do we need to drive the ESL?
A: -Since the amp needs to supply for the usable real power into the load as well as the reactive power that circles within the amp, the powersupply, the power stage and cooling devices need to be designed with sufficient headroom. A phase shift of +-45° already ask for 1A of reactive current for every 1A of real current supplied to the load. The relationship quickly rises with increasing phase shift (60°->1.7, 75°->3.7, 80°->5.6). Amplifiers with simple protection circuits/limiters that measure the voltage over a Emitter- or Source-resistor often limit much too early and prevent the amp from driving the load properly (think of the power cube diagrams printed in some magazines).
Then there are methods to increase the phase shift safety margin of an amplifier, for example by modificating the feedback path, reducing the amps bandwidth, breaking up the global feedback path, adding resitors between amps output and load etc.
Especially the last named measures can more often be found in Tube amps than in Solid State. Low bandwidth, no global feedback, resistive component of the audio output transformer, etc. are more typical in the Tube world. Tubes are a good starting point when it comes to properly driving ESLs. As such the amplifier I found the most stable and best sounding are the single-ended class-A non-nfb Triodes of KR Audio.
My ESLs reach phaseshift values of up to 85° which unfortunately means that the number of amps capable to drive the ESLs properly and also sounding well is rather small. The KR Audio Krionzilla with just ~50W of specced power not only remain stable up to highest volume levels but sounds exceptionally well and really brings the utter best out of the ESLs to shine.
If one omits with the audio transformer alltogether, moving into the world of direct drive, one can and one needs to design the amplifier differently to the standards applying to low voltage audio.
If You want to drive a capacitive load from a constant voltage amplifier it has its issues as explained before. In the end it needs a current to charge and discharge the ESLs capacitance. So an amplifier working in the ´current domain´ may show better behaviour. You could for example design an amplifier, that doesn´t work as constant voltage source but as a constant current source.
Q:
- ....how come everyone is considering direct drive the holy grail of amplification?
A:- It might be, because a good HV-DD-amp is equally seldom to find.
There are many myths circling the audio world and quite alot of those myths are told in the ESL-world. A DD-amp beeing the best to drive an ESL properly is such a myth. The truth is, that it ranges from ´absolutely true´ to ´absolute BS´.
The concept of direct drive bears certain technical advantages. It allows for tighter control of the load, it omits with the complex coupling device, the audio transformer, and it allows for specialized amplifier topologies which are better suited to drive a capacitive load. In theory DD is clearly the better way to go. But we live in a real world and in practical DD has its own and serious problems. Besides safety issues already mid sized panels typically require very high drive voltages. 10kV+ are not unreal. You can´t use easy to source, off-the shelf standard power devices. There are some rather exotic tubes (at least exotic to the audio world) that may be useable. Things even get worse in the Solid State world. All power devices capable of several kV supply voltage are intended for switching useage not for linear amplification. Most of theses devices are intended to switch dozens of amperes of current (so they feature high internal capacitance values and they are rather slow) while we only need less than 1-2A. Cascading the power devices breaks the voltage barrier, but this trick also doesn´t come for free. Further Development steps and the upcoming SIC-technology might provide us with a solution in near future, but at the time suitable Solid State power devices are not at hand.
The amplifier topologies generally accepted as well sounding would result in terribly inefficient DD-amplifiers generating enormous amounts of heat. If on the other hand a low voltage audio amp would feature a topology as used in some better known DD-amps, probabely none of the HighEnders yelling "DD is best" would dare to hook up such a amp in their system.
It´d be hard to top the results of a well matched ESL-Transformer combination with a DD amplifier. And it would be of course much costier.
jauu
Calvin
Last edited:
Again very well put, Calvin!
Just to make myself clearer When I said simple I ment the actual construction of the panel is fairly simple.
However driving them as I have found out is not easy as it seems and there are many factors and guidelines that must be considered in order to have a sucessful build and to achieve great results.
I understand completely were you are coming from when you talk about dd amplifier output devices and have been following what is coming up in the future.
As I was just scoping out some hvfets the other day and had found their gate capacitances were on the order of 10 to 100 times more than capacitance of an average size panel of 1' X 4'.
So it will be very interresting to see what the future lies in solid state devices.
Thank you for pointing how the current requirements go up due to the phase shift as I have not sat down to go through those calculations yet.
It is alot more than I had anticipated and explains why some amplifiers have such a hard time keeping up.
I knew it was high but I didn't relize how much higher it is, and, this gives me a very good insight of my next amplifier design for driving them even with low voltages through a transformer.
I have been contemplating a very high current class a or ab design.
But I haven't decided yet as I have reading about how the THD can be much higher at the transistion from class a to class ab than the THD of a plain class b push pull design.
I have been thinking about something like susan parkers design as it is very simple and since there needs to be an transformer in the output anyway.
It looks even more Intriguing when I had finaly saw the THD figures of much less than .05% for the amplifier at normal levels under say 10 or 20 watts which is plenty for comfortable listening.
Or some thing along those lines to lessen the need for a complicated amplifier design that requires special parts.
Just a few thoughts although there are many very good designs to consider. jer
Just to make myself clearer When I said simple I ment the actual construction of the panel is fairly simple.
However driving them as I have found out is not easy as it seems and there are many factors and guidelines that must be considered in order to have a sucessful build and to achieve great results.
I understand completely were you are coming from when you talk about dd amplifier output devices and have been following what is coming up in the future.
As I was just scoping out some hvfets the other day and had found their gate capacitances were on the order of 10 to 100 times more than capacitance of an average size panel of 1' X 4'.
So it will be very interresting to see what the future lies in solid state devices.
Thank you for pointing how the current requirements go up due to the phase shift as I have not sat down to go through those calculations yet.
It is alot more than I had anticipated and explains why some amplifiers have such a hard time keeping up.
I knew it was high but I didn't relize how much higher it is, and, this gives me a very good insight of my next amplifier design for driving them even with low voltages through a transformer.
I have been contemplating a very high current class a or ab design.
But I haven't decided yet as I have reading about how the THD can be much higher at the transistion from class a to class ab than the THD of a plain class b push pull design.
I have been thinking about something like susan parkers design as it is very simple and since there needs to be an transformer in the output anyway.
It looks even more Intriguing when I had finaly saw the THD figures of much less than .05% for the amplifier at normal levels under say 10 or 20 watts which is plenty for comfortable listening.
Or some thing along those lines to lessen the need for a complicated amplifier design that requires special parts.
Just a few thoughts although there are many very good designs to consider. jer
Calvin, I've said it before and I'll say it again... You are my hero.
Still I'll readily admit that this is indeed scr*wing with my head.
I've lived with the conception that a lower capacitance is, for the lack of better words, easier to drive?
A low capacitance will however result in a greater reactance which will create a greater phase shift, actually making it a harder load to drive?
I've looked at Impedance as a "black box" up until now. A greater impedance used to be a good thing in my book?
Following this new line of reasoning you are scr*wed either way? If it's only the real part of the impedance that count I'm lost at the moment?
A greater capacitance -> Lower impedance and smaller phase shift.
A lower capacitance -> Higher impedance but higher phase shift.
It's like you're damned if you do and damned if you don't?
Having said that it only means I'll be hitting the books much harder. I hate feeling stupid.
Funny thing about the amp designing issue. Being somewhat electrically challenged and always trying to stick to the kiss principle, pretty much all my initial thoughts start with tube amps and zero feedback. (If possible)
Possibly some sort of push-pull design or a double SE in a PP configuration?
In my world it appears there is only one constant. The more I learn the more I realize how little I really know.
I hope this line on questions aren't considered OT? I don't want to be seen as hijacking the thread...
Still I'll readily admit that this is indeed scr*wing with my head.
I've lived with the conception that a lower capacitance is, for the lack of better words, easier to drive?
A low capacitance will however result in a greater reactance which will create a greater phase shift, actually making it a harder load to drive?
I've looked at Impedance as a "black box" up until now. A greater impedance used to be a good thing in my book?
Following this new line of reasoning you are scr*wed either way? If it's only the real part of the impedance that count I'm lost at the moment?
A greater capacitance -> Lower impedance and smaller phase shift.
A lower capacitance -> Higher impedance but higher phase shift.
It's like you're damned if you do and damned if you don't?
Having said that it only means I'll be hitting the books much harder. I hate feeling stupid.
Funny thing about the amp designing issue. Being somewhat electrically challenged and always trying to stick to the kiss principle, pretty much all my initial thoughts start with tube amps and zero feedback. (If possible)
Possibly some sort of push-pull design or a double SE in a PP configuration?
Up until this moment I've never encountered anyone advocating a current drive for ESL's. Building a constant current amp with a 3kV capability sounds tricker than the usual run of the mill voltage amplifier though.
In my world it appears there is only one constant. The more I learn the more I realize how little I really know.
I hope this line on questions aren't considered OT? I don't want to be seen as hijacking the thread...
I too get confused with the term "reactance", when it is used with phaze in the eqution so I need to hit the books some more.
But I can tell you this,
The higher the capacitance is the lower the impedence is, as the frequence gets higher the impedence gets lower aswell, and, therefore more current must flow inorder to charge and discharge the capacitor within the given time. jer
But I can tell you this,
The higher the capacitance is the lower the impedence is, as the frequence gets higher the impedence gets lower aswell, and, therefore more current must flow inorder to charge and discharge the capacitor within the given time. jer
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.