Hi
I have been thinking about building ESLs on and off for many years now. The output of this has been zero so far, sorry to say.
One thing that bothers me is that I still don't really understand how they work... If you bear with me I will explain what the "problem" is. I will use the following ESL scenario, just the standard way to do it:
We have a cell with two stators (A and B) of perforated metal or whatever with a coated mylar membrane placed between them using plastic spacers. That is, a standard cell.
To drive the cell we use a transformer delivering a high voltage audio signal to the stators from its secondary winding.
The midpoint of the secondary winding is attached to the positive node of the bias supply. The negative node goes via a high value resistor to the membrane giving this a negative charge relative to the stators.
This is the way almost all ESL speakers are made.
If we look at this cell before we try to play music thru them (statically), the membrane is attracted to both stators by the fact that objects of different charge, i.e. positive and negative, are attracted to each other. To prevent the membrane to get stuck to one of the stators we stretch the membrane tight when we put the cell together.
Now we start to play music on the cell.
For the sake of argument, the mid point of the secondary winding is our reference point at zero volts.
Let’s say that the transformer delivers a balanced signal of 1kV p-p on the stators A and B.
Let’s say that the bias voltage at the stator is at 4kV negative.
At one point stator A will be at 1kV positive and stator B at 1kV negative.
The relative voltages, stator - membrane will be:
Stator A - membrane = 5kV
Stator B - membrane = 3kV
The membrane will thus still be attracted to both stators but as the voltage between stator A and the membrane is the highest the membrane will move towards stator A. After ½ a period the reverse will be true, and stator B will attract the membrane the most.
Questions:
1.
Is this really a push-pull speaker?
2.
We usually have a reference when talking about voltages, most often “ground” or GND. If we in the example above lift the midpoint of the secondary winding using a DC voltage of say 5kV positive (everything else being the same), then the membrane would be at 1kV positive relative to ground and the stators at 5kV positive. Will the field force, stator-membrane be one of repulsion instead of attraction?
A long mail. I’m missing out of something both crucial and basic. If you se my “problem” and can be of assistance please comment.
/Urban
I have been thinking about building ESLs on and off for many years now. The output of this has been zero so far, sorry to say.
One thing that bothers me is that I still don't really understand how they work... If you bear with me I will explain what the "problem" is. I will use the following ESL scenario, just the standard way to do it:
We have a cell with two stators (A and B) of perforated metal or whatever with a coated mylar membrane placed between them using plastic spacers. That is, a standard cell.
To drive the cell we use a transformer delivering a high voltage audio signal to the stators from its secondary winding.
The midpoint of the secondary winding is attached to the positive node of the bias supply. The negative node goes via a high value resistor to the membrane giving this a negative charge relative to the stators.
This is the way almost all ESL speakers are made.
If we look at this cell before we try to play music thru them (statically), the membrane is attracted to both stators by the fact that objects of different charge, i.e. positive and negative, are attracted to each other. To prevent the membrane to get stuck to one of the stators we stretch the membrane tight when we put the cell together.
Now we start to play music on the cell.
For the sake of argument, the mid point of the secondary winding is our reference point at zero volts.
Let’s say that the transformer delivers a balanced signal of 1kV p-p on the stators A and B.
Let’s say that the bias voltage at the stator is at 4kV negative.
At one point stator A will be at 1kV positive and stator B at 1kV negative.
The relative voltages, stator - membrane will be:
Stator A - membrane = 5kV
Stator B - membrane = 3kV
The membrane will thus still be attracted to both stators but as the voltage between stator A and the membrane is the highest the membrane will move towards stator A. After ½ a period the reverse will be true, and stator B will attract the membrane the most.
Questions:
1.
Is this really a push-pull speaker?
2.
We usually have a reference when talking about voltages, most often “ground” or GND. If we in the example above lift the midpoint of the secondary winding using a DC voltage of say 5kV positive (everything else being the same), then the membrane would be at 1kV positive relative to ground and the stators at 5kV positive. Will the field force, stator-membrane be one of repulsion instead of attraction?
A long mail. I’m missing out of something both crucial and basic. If you se my “problem” and can be of assistance please comment.
/Urban
1. Yes. The field that the membrane sees is the sum of the fields from the two stators.
2. Take an ESL and reverse the bias supply on the membrane. You will note that it operates exactly as it did before.
2. Take an ESL and reverse the bias supply on the membrane. You will note that it operates exactly as it did before.
1. Possibly this is not really an issue, but as I see it there is never any pushing action on the membrane, just pulling. This is a shame really , if repulsion and attraction could work together on the membrane, this would (I guess) lead to higher efficiency of the speaker. It's perhaps a pull-pull speaker ?
2. If you use a positive charge (instead of a negative) on the membrane, the speaker will play as before, but 180degee out of phase. This is not really my question though. Is a charge of an object relative to:
- itself ?
- ground ?
- another charged object ?
Think about the membrane in my example. If we would lift the whole "system" to 8kV positive via the midpoint of the secondary winding, the membrane would be at 4kV positive relative to the surounding air and the stators at 8kV positive. Are these objects (membrane and stators) both possitivly charged now, repelling each other ?
/Urban
2. If you use a positive charge (instead of a negative) on the membrane, the speaker will play as before, but 180degee out of phase. This is not really my question though. Is a charge of an object relative to:
- itself ?
- ground ?
- another charged object ?
Think about the membrane in my example. If we would lift the whole "system" to 8kV positive via the midpoint of the secondary winding, the membrane would be at 4kV positive relative to the surounding air and the stators at 8kV positive. Are these objects (membrane and stators) both possitivly charged now, repelling each other ?
/Urban
Look at it this way:
The diaphragm has a more or less constant charge that is held in place by the very high resistance of the "sort-of-conductive" coating that allows the charge to be put there in the first place.
In a parallel plate capacitor, the electric field between the plates is constant at all positions between the plates (ignoring fringing effects at the edges). The field "E" is a vector field, usually represented by arrows pointing from + to -. The force "F" on a charge "q" in an electric field "E" is simply q*E. Since E is a vector field, F is a vector pointing in the same direction as E.
When the voltage reverses polarity on the stators, the E field reverses direction, so the force on the diaphragm does likewise.
I put some links to some physics sites that illustrate the E field in a parallel plate capacitor in another thread on "bare metal vs. insulated stator ESLs".
The low distortion of an ESL is due to the constant magnitude of the E field between the plates, and because the charge on the diaphragm is spread over the entire surface and held in place by the high resistance of the diaphragm coating (forcing q to remain constant). You can make an ESL with a metalized diaphragm and a single plate (as was done in the very early days of ESLs- 1890's if I recall correctly) that will produce sound but it will be distorted because the E field from a single charged plate is not uniform (and because the charge on the diaphragm is mobile due to the low resistance of the diaphrgam).
I_F
The diaphragm has a more or less constant charge that is held in place by the very high resistance of the "sort-of-conductive" coating that allows the charge to be put there in the first place.
In a parallel plate capacitor, the electric field between the plates is constant at all positions between the plates (ignoring fringing effects at the edges). The field "E" is a vector field, usually represented by arrows pointing from + to -. The force "F" on a charge "q" in an electric field "E" is simply q*E. Since E is a vector field, F is a vector pointing in the same direction as E.
When the voltage reverses polarity on the stators, the E field reverses direction, so the force on the diaphragm does likewise.
I put some links to some physics sites that illustrate the E field in a parallel plate capacitor in another thread on "bare metal vs. insulated stator ESLs".
The low distortion of an ESL is due to the constant magnitude of the E field between the plates, and because the charge on the diaphragm is spread over the entire surface and held in place by the high resistance of the diaphragm coating (forcing q to remain constant). You can make an ESL with a metalized diaphragm and a single plate (as was done in the very early days of ESLs- 1890's if I recall correctly) that will produce sound but it will be distorted because the E field from a single charged plate is not uniform (and because the charge on the diaphragm is mobile due to the low resistance of the diaphrgam).
I_F
You have to think about charges and not so much about voltages. If you charge the diaphragm ... I walked away from my computer and I_forgot posted
to add to that - say you add some electrons to the diaphragm (negative bias supply - but it is done through a large resistor and the diaphragm has a high surface resistance so don't think voltage, it is charged, the voltage will change if you move it!), now you make a plate negative - that is you add some electrons to it there is now a repulsion (a push) between the two and of course if the other plate has been made positive (by removing electrons) it will attract the diaphragm - so it is push pull.
but I think it is easier to think about the plates creating an electric field and the charged diaphragm has a force on it depending the strength and direction of the field.
There are many ways to think about things and quite often one way makes things appear simple and another will just confuse! And of course sometimes we just apply incorrect ideas and get really confused. The force comes from a charge in an electric field so analyse it that way.
Hope this hand waving with no math helps
Bill
to add to that - say you add some electrons to the diaphragm (negative bias supply - but it is done through a large resistor and the diaphragm has a high surface resistance so don't think voltage, it is charged, the voltage will change if you move it!), now you make a plate negative - that is you add some electrons to it there is now a repulsion (a push) between the two and of course if the other plate has been made positive (by removing electrons) it will attract the diaphragm - so it is push pull.
but I think it is easier to think about the plates creating an electric field and the charged diaphragm has a force on it depending the strength and direction of the field.
There are many ways to think about things and quite often one way makes things appear simple and another will just confuse! And of course sometimes we just apply incorrect ideas and get really confused. The force comes from a charge in an electric field so analyse it that way.
Hope this hand waving with no math helps
Bill
- Status
- Not open for further replies.