Electrostatic headphones - how does it work?
Peter
I was given a pair of MicroSeiki hi-fi electrostatic headphones (giver
does not know its origin). It has 4 pins terminated in a DIN-like
connector, and it comes without any HV energizer unit. A resistance
check shows about 2 mega-ohms for each earpiece. What I want to do is
to build a high voltage bias energizer to power this phone. Can anyone
guess what voltage I should use (don't worry, I KNOW how to build and
handle HV power supplies). And how do I superimpose the audio signal
onto the HV? I understand quite well how electrostatic SPEAKERS work
(like say the Quad 63) but have never handled e'headphones before. I
am really looking forward to bringing this baby up so I can compare
how they sound, compared to some of the regular moving coil headphones
(or my reference - a pair of Yamaha orthodynamic phones).
Thanks
Peter
donald
Peter,
The principle is the same as for the Quads, but from the tone of this
post, and the impression it gives, I would leave this project well
alone. Unless you feel like recreating the electric chair, find
yourself a proper energiser unit.
d
Guenter A. Scholz
Peter <audi...@hotmail.com> wrote:
>Hi,
>I was given a pair of MicroSeiki hi-fi electrostatic headphones (giver
Peter, make sure you first ascertain whether or not these are
indeed electrostatic and not electret. The latter has a permanent charge
on the diagphragm and does not need a polarizing supply. The DIN pin
should give you an indea which it is. If there are only 4 connections
chances are it is electret. The 'supply' then mearly steps up the audio
voltage.,..probably x10 or thereabouts. Of course some electrostatics
use the audio voltage as a source of the bias supply. Best is probably
to contact MicroSeiki.
-regards
Sam Goldwasser
Here's a description and typical circuit (from the audio+misc FAQ):
Electrostatic Loudspeakers and Headphones:
-----------------------------------------
While the vast majority of sound producing devices utilize a moving coil (or
at least are magnetically actuated), there are also some that are both very
expensive and which supposedly provide very accurate sound reproduction based
on electrostatic principles. Basically, a conductive membrane or diaphragm
is charged to a high DC voltage with respect to a fixed plate - several
hundred volts for a headphone driver but several thousand volts for a
wall-size loudspeaker. This sets up a bias field that draws the diaphragms
together and keeps them tight. The audio signal is boosted to a high AC
voltage and added to the DC bias. The resulting change in force results in
the movement of the diaphragm and thus the production of sound. The claimed
advantages of such an approach is that the sound emitting area can be very
large and the movement very small minimizing various types of distortions.
The DC bias voltage can be supplied by any number of means - directly from the
AC line via a transformer and/or voltage multiplier or using a high frequency
inverter. The current requirement is essentially 0.
(Note: There are also some 'planar magnetic' loudspeakers in existence which
may look similar to electrostatic types but have no external power source and
special circuitry inside. The planar voice coil is etched onto a thin
diaphragm suspended within the magnetic structure.)
The audio signal from the output of a normal amplifier may be boosted to a
higher voltage by another special amplifier which is part of the electrostatic
loudspeaker 'energizer' before being further boosted by a transformer to drive
the speakers themselves.
The energizer schematic below is for a pair of Radio Shack electrostatic
headphones. A room-size set of Thunderblasters would require a somewhat more
sophisticated set of electronics - probably including its own amplifier as well
as operating at much higher voltage and higher power - but the basic idea will
be the same.
o--o L Spkr Out 5.1K 1
Left Chan In o---o +-----/\/\---+-----o L Drive
\o---/\/\----+---+ ||( |
3.3 | )||( |
PTC 30 / )||( 100pF _|_
5W \ )|| +--+ 500V ---
/ )||( | |
| )||( | |
o------------------+---+ ||( | | 4
+--|---------+-----o L Ret
|
o--o R Spkr Out | 5.1K 3
Right Chan In o---o +--|--/\/\---+-----o R Drive
\o---/\/\----+---+ ||( | |
3.3 | )||( | |
PTC 30 / )||( | 100pF _|_
5W \ )|| +--+ 500V ---
/ )||( | |
| )||( | |
o------------------+---+ ||( | | 5
+--|---------+-----o R Ret
|
25K 1N4004 5M | 2
AC H o---/\/\---+-----+--|>|---+----/\/\---|---------------o Bias Com
| | _|_ 1/2W |
/ | 1uF --- |
100K \ | 160V | |
/ +--|--------+ |
\ | | 1uF _|_ |
25K | | | 160V --- |
AC N o---/\/\---+--+ | | 1M |
+--|<|---+----/\/\---+
1N4004 1/2W
The 3.3 ohm PTC (Positive Temperature Coefficient) thermistors are supposed
to provide some sort of protection for the transformer and its circuitry in
case the unit is fed with too much power. Now how could that happen? :)
However, if abused too much, they can fail as well. Problems may occur in the
voltage doubler and high value resistors sometimes just go bad on their own.
Trying to find replacements can be a treat but it appears that some PolySwitch
protectors fit the description and should be available from major electronics
distributors. For example, the 3.3 ohm device has a hold current spec of
0.17 A (.1 W) and a trip current of 0.34 A (.4 W). However, I don't know if
this is adequate for the headphones described above.
--- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Site Info: http://www.repairfaq.org/REPAIR/F_mirror.html
Peter
As to contacting Microseiki, I don't think they even exist any
more....
Peter
On 1 Nov 2000 13:41:01 GMT, sch...@watsci.uwaterloo.ca (Guenter A.
Peter
I just opened up the earpiece, and inside are 3 "prongs", and includes
2 pieces of 1-meg-ohm resistors per earpiece. My guess is these are
the current limiting resistors for the HT that feeds the membrane.
Wish me luck....I wish I could afford a Stax or a Senn electro
headphone, but I can't, so I guess I will tinker with this "gift".
Peter
On Wed, 01 Nov 2000 13:44:52 GMT, donald @pearce.uk.com (Don Pearce)
wrote:
Peter
start....again thanks.
Peter
On 01 Nov 2000 14:47:04 GMT, s...@saul.cis.upenn.edu (Sam Goldwasser)
wrote:
>Here's a description and typical circuit (from the audio+misc FAQ):
>
>Electrostatic Loudspeakers and Headphones:
>-----------------------------------------
>
>While the vast majority of sound producing devices utilize a moving coil (or
>
L.J. Timpert
I cracked my head over presumably the same headphone, but this was a
working example with the HV power supply that works with a 9 Volts
battery. As I remember well, it totally matches your description, and it
polarizes at around 200-250 Volts. I unscrewed the cover of the power unit
to examine it. It has some very special features from a design point of
view. I'll describe it below.
The first thing that struck me was the cheap type of construction. Not
quite the thing you would expect from an audiophile item such as this. It
had a very standard rotary switch for loudspeaker/headphone/OFF selection
and a pertinax PCB and normal wiring just like in your average Japanese
transistor radio. The transformers looked just like the normal power
transformer in such a radio, same size and no interleaved windings or
other audiophile tricks, just a secondary winding over a primary. The
winding configuration was a little different though. The primary was a
normal winding, the secondary was a center tapped winding whith much more
turns and thinner wire. It should be like this of course for a step up
transformer, but I think you could use just as well a standard power
transformer with two 110V primaries in this application, just to try it
out.
Now comes the tricky part: the wiring of the plug, it is a four pin
DIN type. If you look into the plug, you see four pins. Let's count them
clockwise. One would think that the left stators were connected to pins
one and two, but this is not the case. Instead, pins one and three are
connected to stator one and pins two and four to stator two (stators are
left or right, if one is left or right I don't remember, but this is easy
to find out once you have gotten some sound out of it).
Now for the biasing. Here comes the real trick to what they have done
to save a wire or two in the headphone cord. A normally biased headphone
would use at least five wires, otherwise six (two for each stator pair and
one for the HT bias, or an individual HV wire for each membrane) but this
one uses only four. The center tap of one transformer is put at 200 Volts
or something of that order, the center tap of the other one is connected
to ground. There is no bias wire in the headphone cord, instead the
resistors connected to the stator plates form a resistive divider such
that the center voltage is equal to the voltage applied to the center tap
of the transformer connected to this particular stator set. The center
point of this divider is connected to the membrane of the OTHER earpiece
!?!?!?! Yes, it's just like that. Let's consider the set that is connected
to the transfomer with it's center tap grounded. This is just like any
normal electrostat. The bias of around 200 Volts comes from the resistive
divider in the other earpiece which has its stators at high voltage. So
this unit operates just like one would expect with stators at zero volts
and HV on the membrane.
But now for the other unit, any electrostat needs to be biased, right?
Well, it is. The membrane of this one is connected to the resistive
divider inside the earpiece we just talked about, so this one is biased at
0 Volts, but it's stator plates are at HV here. The electric fields in
this earpiece are thus reversed with respect to the other earpiece and the
AC voltage should be applied in reverse phase in order to get the total
phase of the headphone right. This is because the forces are reversed on
this membrane due to the reversed bias field. All by all this is a clever
trick to spare some wires and thus make the connecting cord a little
thinner.
So now comes the connection. The plug goes like this: [
--1---2---3---4--] in this text where the numbers represent the pins
counted clockwise. Connect it like this:[--neg. stator one---pos. stator
two---pos stator one---neg. stator two--] so the outermost pins are
negative. By this I just indicate the phase: when used with a mono signal,
the negs are in phase whith each other and so are the pos'es. This is the
proper connection at which both membranes are acoustically in phase.
To test if the headphone works properly, do the following. You'll heed
at least one transformer with two 110 Volt primaries and a 12 Volt
secondary, an audio setup with a source and an amp and some HV source that
can deliver 150 to 200 Volts. Oh yeah, for your safety, a 2.2 Megs
resistor is a nice thing too.
Connect the primaries in series as in a 220 Volts configuration and
connect the center tap to gound. Now hook up the secondary to the output
of your poweramp with a 4.7 Ohm/2Watts resistor in series. Open the volume
a little, play some music and check if there is AC present between the
ends of the primary, which is actually your secondary since you use it the
other way round. If you don't measure any AC, reverse phase of one of the
110 Volts halves and try again. You should measure some now. Now is the
time to connect the stator plates on one earpiece to the ends of the (now)
secondary winding (the high voltage winding) so you connect for example
one end to pin one and the other end to pin three. Now your stator is
wired, and you connect the wires of the other stator together and, through
the 2.2 Meg resistor, to the HV source. Set it at 150 Volts and play some
music. Turn up the volume a bit and you should be able to hear something
now. If you have an identical transformer, hook it up to the other stator
set just like the first earpiece, but now attach the center tap of the HV
to +150 with the 2.2 Meg resistor and you can enjoy stereo! I saw in
another post that this type of thing resembles an electric chair, but this
is utter nonsense when done properly. The key to safety is the 2.2 Meg
resistor. This limits the current when you short the supply to a safe low
value that wil only tingle a bit, even when applied to your head. Just
don't exceed the voltage rating of ther resistor (which is 200 Volts for
the smallest ones and 300-500 Volts for the types of 1 Watt and up) and
you are perfectly safe. Of course, I can take no responsibility if you die
when using this device. But the cause is not likely to be the headphone
supply when this happens ;-) so you're still quite safe.
If this all works, you can consider building a dedicated little (tube)
amp for it. Keep us informed! Regards,
Jurgen Timpert
Sean McCartan
across the ears. Sounds like a tight SLAP to me.
咨菌egu�
http://abone.superonline.com/~soysal2/index2.htm
--
Joe
The 80's were a blur. I'm just lucky I didn't brain my damage.
"Peter" <audi...@hotmail.com> wrote in message
news:39fde293...@news.pacific.net.sg...
Dave Tweed
> If they are electret, it will work exactly like an electret condenser
> microphone in reverse. Find out how an electret condenser mike works and
> engineer it in reverse. I have in front of me an electret condenser mike
> element. It has exactly two wires, one red and the other black, for the
> obvious polarity of the system. The most obvious wiring is a battery in series
> with a resistor (in the case of this element 2.2K) and the connection between
> the element and resistor is the microphone ground and a cap on the other end
> of the resistor has to be the signal output.
>
> If you had an electret condenser speaker, a similar system will be in place,
> but a signal is fed into the cap instead of a signal taken from the cap. But
> you'd have a system diagramme like this:
>
> cap |
> -----||--+----------||------------------
> Audio / |
> I/O 2.2K \ battery to electret condenser element
> / (or power supply)
> ---------+------------------------------
> |
> __|__
> ---
> ground
Not quite. The battery isn't there to bias the electret condensor, it's
actually the power supply for a FET preamplifier (really an impedance
converter) that's built into the microphone element. This lowers the
output impedance of the microphone from the megohm range to whatever
value you put in as the load resistor (2.2K in this case). This makes
it a better match for the cable to the amplifier.
In the case of an electret speaker element, you wouldn't need a bias
supply at all, just a way to transform the low output impedance of the
amplifier (low voltage, high current) to the high impedance of the
element (high voltage, low current). A transformer is the obvious
choice, but does have some drawbacks. It would at least tell you whether
you're on the right track, and if that works, you could build an active
driver that puts the transformer in the feedback loop to compensate for
some of those drawbacks.
If the transformer alone doesn't work, then they're probably ordinary
condensor elements that require a high-voltage DC bias supply in addition
to the transformer.
-- Dave Tweed
Bloody Viking
Peter (audi...@hotmail.com) wrote:
: I just opened up the earpiece, and inside are 3 "prongs", and includes
: 2 pieces of 1-meg-ohm resistors per earpiece. My guess is these are
: the current limiting resistors for the HT that feeds the membrane.
: Wish me luck....I wish I could afford a Stax or a Senn electro
: headphone, but I can't, so I guess I will tinker with this "gift".
Have fun!. I remember once being given an old tube-type PA system amplifier
with a bad tube in the final. Got the old thing working, and had lots of fun
with it. Its transformer eventually died but I had some great times with it,
even using it once for a school play where I played the part of the sound
technician.
As far as modulating the high voltage DC with your music, you can get ideas
from the ARRL handbook with how hams modulate the AM transmitter's Class C
final. The audio is fed to a step-up transformer in series with the power
supply feeding the plates of the final's tubes. Becuse of this, hams also know
how to build BIG audio finals to use to modulate the transmitter! ("final" ==
final stage of an amplifying machine, normally used in RF work) BTW, the ARRL
Handbook is also good for designing FM antennas, and you get to do the math.
(HINT: FM is horizontally polarised and ham is vertically polarised)
There is another way to modulate DC, and that's with a resistor-capacitor
setup, which will sound better than the transformer like a ham uses. But
grounds become an issue with this.
--
FOOD FOR THOUGHT: 100 calories are used up in the course of a mile run.
The USDA guidelines for dietary fibre is equal to one ounce of sawdust.
The liver makes the vast majority of the cholesterol in your bloodstream.
Bloody Viking
Guenter A. Scholz (sch...@watsci.uwaterloo.ca) wrote:
: Peter, make sure you first ascertain whether or not these are
: indeed electrostatic and not electret. The latter has a permanent charge
: on the diagphragm and does not need a polarizing supply. The DIN pin
: should give you an indea which it is. If there are only 4 connections
: chances are it is electret. The 'supply' then mearly steps up the audio
: voltage.,..probably x10 or thereabouts. Of course some electrostatics
: use the audio voltage as a source of the bias supply. Best is probably
: to contact MicroSeiki.
If they are electret, it will work exactly like an electret condenser
microphone in reverse. Find out how an electret condenser mike works and
engineer it in reverse. I have in front of me an electret condenser mike
element. It has exactly two wires, one red and the other black, for the
obvious polarity of the system. The most obvious wiring is a battery in series
with a resistor (in the case of this element 2.2K) and the connection between
the element and resistor is the microphone ground and a cap on the other end
of the resistor has to be the signal output.
If you had an electret condenser speaker, a similar system will be in place,
but a signal is fed into the cap instead of a signal taken from the cap. But
you'd have a system diagramme like this:
cap |
-----||--+----------||------------------
Audio / |
I/O 2.2K \ battery to electret condenser element
/ (or power supply)
---------+------------------------------
|
__|__
---
ground
The part that sucks is that in the case of the microphone the battery case
must be shielded or add RF chokes or otherwise get creative. But a system like
above is the only imaginable way to extract a signal from a two-wire electret
condenser microphone. If I'm wrong, please correct me!. I know that the above
_will_ modulate an LED with sound for a fibre-optic linkage.
Bloody Viking
Sam Goldwasser (s...@saul.cis.upenn.edu) wrote:
: While the vast majority of sound producing devices utilize a moving coil (or
: at least are magnetically actuated), there are also some that are both very
: expensive and which supposedly provide very accurate sound reproduction based
: on electrostatic principles. Basically, a conductive membrane or diaphragm
: is charged to a high DC voltage with respect to a fixed plate - several
: hundred volts for a headphone driver but several thousand volts for a
: wall-size loudspeaker. This sets up a bias field that draws the diaphragms
: together and keeps them tight. The audio signal is boosted to a high AC
: voltage and added to the DC bias. The resulting change in force results in
: the movement of the diaphragm and thus the production of sound. The claimed
: advantages of such an approach is that the sound emitting area can be very
: large and the movement very small minimizing various types of distortions.
The actual advantage is that across the diaphragm the force is distributed
over a wide area evenly. In the case of the trusty dynamic speaker, the force
comes concentrated at the voice coil so the radiating surface must be rigid,
as the cones are in fact designed. If you dismantle headphone speaker elements
that are dynamic, you find an odd surface that looks like a round surface with
the centre pushed in to be attached to the spider and voice coil. Since this
is cheap to manufacture, the result is decent sound, certainly compared to
taking, say, pocket radio speakers and ramming into earmuffs. The dynamic
headphone speaker element design attempts to replicate the electrostat design,
and isn't too bad at getting a small speaker to attain good bass when near an
ear.
In the dynamic headphone speaker element, the entire surface acts both as the
radiating surface AND that edge designed to allow movement. The ratio of the
two behaviours depends on distance from the voice coil.
Meanwhile, an electrostat speaker has the conducting diaphragm and two
gratings, hooked to a bipolar supply. Each grating is hooked to a leg of the
supply and the diaphragm hooked to the centre tap, which should be ground. The
modulation is in series with the centre tap and diaphragm, causing it to move
back and forth as polarity changes.
With that scheme, a pair of speakers would need a 5 pin fitting, one for
ground, one for V++, one for V--, one for Left In, and one for Right in. Since
the original poster has a 4-pin DIN fitting, the casing of the fitting MIGHT
be the system ground. I would have to inspect the headset to determine if they
are electrostats.
Your scheme causes the problem of pre-existing tension on the diaphragm,
causing a higher resonant frequency and a resultant bass loss. No good for
speakers, let alone a much smaller headset.
I have never heard of electret condenser speakers until now, but any
connection will have to be compliant with the left-ground-right paradigm, even
if it meant using transformers, a no-no for a stereo headset. Transformers are
a no-no simply becuse of their fidelity losses. Good audio transformers are
hard to design and are expensive as a result. The impedance of either electret
condenser or electrostat is real high, causing a dependancy on good audio
transformers, hence the high cost of such methods. Tubes could help alleviate
the problem for a headset to make a transformer-free coupling, or use "fetron"
devices equivalent to the 12AX7 dual tube. A normal > electrostat interface
should use a fetron type device to avoid them transformers.
Bloody Viking
Sean McCartan (n...@chance.anon) wrote:
: mmmmmmmm, sounds like a potential candidate for the Darwin Awards - KV
: across the ears. Sounds like a tight SLAP to me.
Those are always fun... unless you are the award winner that is. There are
other creative ways to earn a Darwin Award. (: Get stupid about DIY Medicine
and you can win one. Who's going to get the Darwin Award of the Millennium?
In late 1979, a wacko came CLOSE to winning a Darwin Award, but was a little
too good at his homebrew Medicine. I'm referring to the 1979 Self-Surgeon.
This bugger did an 8-hour transabdominal surgery to denervate his adrenals.
The adrenals weren't accessible due to pain, so he patched himself up and
called the 911 of the time for "a rupture". The real surgeons were amazed at
how clean and professional the job was!
But fucking with high voltage is a good way to win one too, and doesn't take
hittin' the med libraries. However, high-power low voltage with its high amps
can also cause a Darwin Award due to fire. You don't fuck with electricity
unless you know what you are doing. Darwin's hand can come out of that wire
and choke you!
Guenter A. Scholz
Bloody Viking <nos...@ripco.com> wrote:
>
>must be shielded or add RF chokes or otherwise get creative. But a system like
>above is the only imaginable way to extract a signal from a two-wire electret
>condenser microphone. If I'm wrong, please correct me!. I know that the above
speakers sitting on your head :-)).......ergo 4 wires?? necest pas
-gueter