[spritzer]

Quote:

Originally Posted by g1981c /img/forum/go_quote.gif that square root thing i believe does not apply to infinite plane ... so where does the power go, it must go somewhere ... are you saying the power factor of electrostatic cans is close to zero ? so you're simply charging and discharging the capacitance without doing useful work ? ?

I think you aren't on the same page as me. The force goes to moving the diaphragm through electrostatic force with a bias of DC and AC driving voltages. The power is used to drive the film but since it's AC we a re dealing the impedance is variable and because the electrostatic force is so weak (drops by the square of the distance, i.e.1m x/1 and 2m x/4 and 3m x/9 etc.) that you need plenty of power to move it just a small distance. Add to that the drivers get more efficient the bigger they get (line source) so we have very small drivers with large spacing that requires heaps of power to get going.

A Stax Lambda is 120pF and if you drive them at 600v (half maximum voltage) at 20kHz you need a slew voltage of 75.4 v/us. The current needed is the SR*capacitance so 9.05mA.

 

[g1981c]

Quote:

Originally Posted by spritzer /img/forum/go_quote.gif I think you aren't on the same page as me. The force goes to moving the diaphragm through electrostatic force with a bias of DC and AC driving voltages. The power is used to drive the film but since it's AC we a re dealing the impedance is variable and because the electrostatic force is so weak (drops by the square of the distance, i.e.1m x/1 and 2m x/4 and 3m x/9 etc.) that you need plenty of power to move it just a small distance. Add to that the drivers get more efficient the bigger they get (line source) so we have very small drivers with large spacing that requires heaps of power to get going. A Stax Lambda is 120pF and if you drive them at 600v (half maximum voltage) at 20kHz you need a slew voltage of 75.4 v/us. The current needed is the SR*capacitance so 9.05mA.

so its not so much about driving the diaphragm as it is about driving the capacitance itself (which wouldnt appreciably change if the diaphragm wasnt mobile at all)

ok, i can buy that

another thing though is that energy content of music is centered at around 100 hz, and then it rolls of at something like 6db/octave towards the top so at 20 khz there is very little energy. in a 2 or 3-way speaker the tweeter receives no more than about 5% of energy.

then it doesnt really make sense to talk about 20 khz when you try to size up current demands. Stax claims 450V @ 1 khz ... so at 2 khz it may take twice the current for same voltage but also at 2 khz there will be ( not even ) half the energy in music as compared to bass frequencies so if Stax handles 450V @ 1 khz it is more likely to run out of juice below 1 khz than it is above it

this obviously depends on type of music, perhaps with some acoustic music you do need that extra current but my friend listens to music which is quite bass heavy so im wondering if slew rate should be of concern to him ...

also even if we assume that your calculation was not taken out of context so to speak then 9 ma @ 600 V is only 5 watts, so multiply that by 4 chanels is 20 watt, and Stax 727's 46 watt power dissipation still allows for this 20 W output with less than 50% efficiency ...

 

[spritzer]

Quote:

Originally Posted by g1981c /img/forum/go_quote.gif so its not so much about driving the diaphragm as it is about driving the capacitance itself (which wouldnt appreciably change if the diaphragm wasnt mobile at all) ok, i can buy that another thing though is that energy content of music is centered at around 100 hz, and then it rolls of at something like 6db/octave towards the top so at 20 khz there is very little energy. in a 2 or 3-way speaker the tweeter receives no more than about 5% of energy. then it doesnt really make sense to talk about 20 khz when you try to size up current demands. Stax claims 450V @ 1 khz ... so at 2 khz it may take twice the current for same voltage but also at 2 khz there will be ( not even ) half the energy in music as compared to bass frequencies so if Stax handles 450V @ 1 khz it is more likely to run out of juice below 1 khz than it is above it this obviously depends on type of music, perhaps with some acoustic music you do need that extra current but my friend listens to music which is quite bass heavy so im wondering if slew rate should be of concern to him ... also even if we assume that your calculation was not taken out of context so to speak then 9 ma @ 600 V is only 5 watts, so multiply that by 4 chanels is 20 watt, and Stax 727's 46 watt power dissipation still allows for this 20 W output with less than 50% efficiency ...

The driver is always pushing against the same field but the impedance changes with frequency so they need plenty of power at low and high F. The rules for dynamic speakers are worthless here as electrostatic behave in a very different way. Dynamics need very little power to reproduce the HF because they are a point source and a monopole. With electrostatics the current requirement is increased dramatically at the trouble spots and thats why most sound great in the midrange with almost anything.

Most of the power in electrostatic amps is wasted through heat as the Blue Hawaii draws some 130-150w and most of it is wasted. Their primary function is to create voltage and not current so even the tiny Stax basic amp that draws only 4w can drive the phones but the bass and treble are always rolled off. This could also be the function of high impedance so that has to be take into account.

 

[g1981c]

Quote:

Originally Posted by spritzer /img/forum/go_quote.gif The driver is always pushing against the same field but the impedance changes with frequency so they need plenty of power at low and high F. The rules for dynamic speakers are worthless here as electrostatic behave in a very different way. Dynamics need very little power to reproduce the HF because they are a point source and a monopole. With electrostatics the current requirement is increased dramatically at the trouble spots and thats why most sound great in the midrange with almost anything. Most of the power in electrostatic amps is wasted through heat as the Blue Hawaii draws some 130-150w and most of it is wasted. Their primary function is to create voltage and not current so even the tiny Stax basic amp that draws only 4w can drive the phones but the bass and treble are always rolled off. This could also be the function of high impedance so that has to be take into account.

i understand why high frequencies are difficult to drive

but i dont understand why the same phenomenon doesnt make it easy to drive low frequencies ?

lets say capacitance increases at low frequencies because of increasing excursion of the diaphragm ... but that effect should still be countered by the increase in the time the amplifier has to charge up this capacitance ...

?

 

[spritzer]

Quote:

Originally Posted by g1981c /img/forum/go_quote.gif i understand why high frequencies are difficult to drive but i dont understand why the same phenomenon doesnt make it easy to drive low frequencies ? lets say capacitance increases at low frequencies because of increasing excursion of the diaphragm ... but that effect should still be countered by the increase in the time the amplifier has to charge up this capacitance ... ?

You need more power to control the extra excursion of the mylar. The extra power isn't used to produce the bass but to regain and have full control of the drive unit. It's pretty much the same as with dynamics where it is easy to make huge amount of horrible bass but getting good bass is a bigger challenge then many manufacturers can handle.

 

[electrathecat]

Though you guys might theoretically be thread-jacking, I love this discussion (it's quite informative) and hope you guys keep it up. Since we're being technical, I was curious as to what makes the Omega IIs harder to drive (i.e., more power hungry) than the HE60s? Does this power-need difference lead to any impact on sound quality or character?

 

[spritzer]

Quote:

Originally Posted by electrathecat /img/forum/go_quote.gif Though you guys might theoretically be thread-jacking, I love this discussion (it's quite informative) and hope you guys keep it up. Since we're being technical, I was curious as to what makes the Omega IIs harder to drive (i.e., more power hungry) than the HE60s? Does this power-need difference lead to any impact on sound quality or character?

The SR-007 is much tougher to drive as the distances in the drivers are greater. The He60 drivers are just 4 pieces of molded plastic stuck together with more then a dozen of plastic screws. The conductive elements are all painted on and the diaphragm is glued to one of the stator structure, the back one to be precise. The conductive paint is nearly in contact with the film but it is so tightly stretched so that won't happen.

The SR-007 uses PCB material as stators with the side facing away from the diaphragm completely copper extruded and the electrodes then gold plated. The stator is thus fully insulated and the diaphragm can't arc. The diaphragm is fixed to a ring structure and all encased in a very sturdy enclosure that is designed to transfer all extra energy into the head. The He60 driver is simply clamped in place by two pieces of rubber, not a bad way of doing it but far from ideal.

The SR-007 driver is larger, the diaphragm isn't as tight and the housing is much better so that all leads to better bass handling but since the drivers are so far apart you need much more voltage to get things going. There are very different opinions surrounding the D/S spacing and it's correlation to bias voltage and how much voltage is needed on the stators and there is no right answer here.

One thing that owners and users should know is that while the bias voltage can be similar there can be other factors to consider, namely how big is the bleeder resistor. The bleeder controls the constant charge mode on the diaphragm which keeps it in line and lowers the distortion. Sennheiser used 10M while Stax uses 5.1M and that's why the old He90 phones need the extra resistor on the bias line in series with the one inside the amp.

To my ears the He60 is a very good headphone (listening ti it right now :he90

and I do like it more then the HE90 but it's far from being in the same class as the SR-007. There is an obvious midrange coloration and the highs, while clear and extended, are prone to reveal sibilance. No big issues but as a whole they are a Tier 3 phone in terms of performance but a nr.3 of my favorite phones of all times.

 

[luvdunhill]

Quote:

Originally Posted by spritzer /img/forum/go_quote.gif The bleeder controls the constant charge mode on the diaphragm which keeps it in line and lowers the distortion. Sennheiser used 10M while Stax uses 5.1M and that's why the old He90 phones need the extra resistor on the bias line in series with the one inside the amp.

have you experimented with using a neon bulb and a capacitor to control the constant charge mode? I know a lot of the ESL people do this on their bias supplies. I'm planning on playing with this some with my bias supply, but am waiting for the 60 VDC / 90 VAC neon to come in the mail...

Also, why not just increase the bias voltage on the SR-007, if it can handle it? I'm sure the 600v would be fine and might give some more control to the sound, besides the obvious volume increase.

 

[luvdunhill]

Quote:

Originally Posted by electrathecat /img/forum/go_quote.gif Though you guys might theoretically be thread-jacking, I love this discussion (it's quite informative) and hope you guys keep it up. Since we're being technical, I was curious as to what makes the Omega IIs harder to drive (i.e., more power hungry) than the HE60s? Does this power-need difference lead to any impact on sound quality or character?

while we're off topic, have you compared your KGSS DLX to a Blue Hawaii by chance?

 

[spritzer]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif have you experimented with using a neon bulb and a capacitor to control the constant charge mode? I know a lot of the ESL people do this on their bias supplies. I'm planning on playing with this some with my bias supply, but am waiting for the 60 VDC / 90 VAC neon to come in the mail... Also, why not just increase the bias voltage on the SR-007, if it can handle it? I'm sure the 600v would be fine and might give some more control to the sound, besides the obvious volume increase.

They do this mostly to monitor the charge on the panels and to burn of the excess charge after the power is turned off. It will work with the headphones but I discharge them into my hands many times every day and there is nothing wrong with me...

I'm sure they can handle the extra voltage but there are less and less benefits when you climb above 600v and it looks like a log10 graph after 1000v. You have to throw a lot of voltage at the driver to get a small small increase in dB's which is why Sound Lab uses 12kV...

 

[g1981c]

Quote:

Originally Posted by spritzer /img/forum/go_quote.gif You need more power to control the extra excursion of the mylar. The extra power isn't used to produce the bass but to regain and have full control of the drive unit. It's pretty much the same as with dynamics where it is easy to make huge amount of horrible bass but getting good bass is a bigger challenge then many manufacturers can handle.

when you say control, are you referring to amplifier's output impedance ? i was unable to find any measure of output impedance in the specs for either KGSS or 727, which is disturbing

 

[luvdunhill]

Quote:

Originally Posted by spritzer /img/forum/go_quote.gif and there is nothing wrong with me...

say that again three times in a row

The neon bulb should also regulate when the panel goes into constant charge mode. Kinda like a regulator. Also, it can diagnose problems with the panels if it's always on. It should blink, not stay on until the panel is discharged. I'll report back in a week or so

 

[electrathecat]

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif while we're off topic, have you compared your KGSS DLX to a Blue Hawaii by chance?

Unfortunately, I've never had the pleasure of listening to a Blue Hawaii, though I'm certainly looking forward to it.

 

[spritzer]

Quote:

Originally Posted by g1981c /img/forum/go_quote.gif when you say control, are you referring to amplifier's output impedance ? i was unable to find any measure of output impedance in the specs for either KGSS or 727, which is disturbing

It's a combination of better a power supply design, lower impedance, higher output swing, larger current sources etc. People have been making their own Stax amps since the 60's and Stax has always encouraged it posting designs and selling the connectors.

Quote:

Originally Posted by luvdunhill /img/forum/go_quote.gif say that again three times in a row The neon bulb should also regulate when the panel goes into constant charge mode. Kinda like a regulator. Also, it can diagnose problems with the panels if it's always on. It should blink, not stay on until the panel is discharged. I'll report back in a week or so

There is nothing wrong with me,
there is nothing wrong with me,
there is something wrong with me... damn...

It does just that and is used for larger panels due to the huge resistors needed. Keep us posted.

 

[g1981c]

Quote:

Originally Posted by spritzer /img/forum/go_quote.gif People have been making their own Stax amps since the 60's and Stax has always encouraged it posting designs and selling the connectors.

when comparing KGSS with 727 the headamp obviously is the more heavy duty device. it has larger transformer and heatsinks, i also like the stepped attenuators.

however, i wonder if the circuit design is as good. i know nothing of the KGSS circuit but i have some thoughts on the 727 circuit based on their description.

what i like (assuming im understanding it right) is that it can amplify balanced input directly to push-pull output realizing greater circuit simplicity while still getting the benefits of common mode rejection.

another thing i like is its claimed ability to handle 30 volts of input, because some DACs can have a pretty strong output and it would be nice to be able to use ALL of the swing the DAC can put out to get optimum SNR. for example my Benchmark DAC is internally attenuated (using jumpers) down by about 20 decibel, because it will clip most electronics without this attenuation, and it would be nice to have a unit where such attenuation is not necessary. on the other hand if the volume control is purely passive then i guess it doesnt matter.

yet another thing i like about 727 is that it is direct coupled.

what i do NOT like about 727 is their claimed NON-NFB output stage. even if the emitter follower output stage may not really need any feedback it should still benefit from global NFB around it ... would you agree ? to me this seems like a gimmick that favors original design over absolute performance.

what do you think about the design of 727 vs KGSS ? ( im rather paranoid about tubes, im looking at the most performance to be had from solid state )