[Wink]

On Wed, 26 Mar 2008 21:59:32 +0000, Eeyore
wrote:



Wink wrote:

I'm going get a dozen of these Nichicons and sell my BGs on Ebay for
a grand.

That's far and away the most sensible thing you've said.

Graham

I was serious Graham, I ordered 12.

[Arny Krueger]

"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)

Did you mean All Electronics?

http://www.allelectronics.com/cgi-bi...APACITOR_.html

whoops, yes All Electronics.

[Arny Krueger]

"Wink" wrote in message

Ironically, note that the Jeff Rowland amp's power
supply caps are *not* Blackgates.
That's because BGs are no longer made. :]
They are Nichicon LQ-M series parts. I see some 3,300 uF
80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)

I'm going get a dozen of these Nichicons and sell my BGs
on Ebay for a grand.

Go for it!

[Wink]

On Wed, 26 Mar 2008 22:13:05 -0400, "Arny Krueger"
wrote:

"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)

Did you mean All Electronics?

http://www.allelectronics.com/cgi-bi...APACITOR_.html

whoops, yes All Electronics.

It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.

I've resolved the voltage on the 63 V BGs torment. I ordered toroids
with this sec: 50-43-0-43-50. The 43 V taps will run produce 56 V
under load. A compromise value, not excessively low.

Should I eject the BGs like a retching dog to Ebay hell, the 100 V
winding will allow a return to 63-ish volt operation, maybe with the
cool Nichicons.

212 Bucks for 2 custom wound toroids, 800 for the BGs, roughly 500 for
Hilligs cards and stuff. Exicon double-current mosfets, 140. So if
the amps turn ok I'll have some good monoblocks for under 2 Gs. Much
less if I can the BGs and go with Nichs, or the original Chemicons.

I think the Behringer Europower amp are much less. I think that's
you're using, I assume to drive your Maggies, which I've never heard,
by the way.
Dave

[Gareth Magennis]

"Arny Krueger" wrote in message
. ..
"Eeyore" wrote in
message
Arny Krueger wrote:

"Gareth Magennis" wrote in
"Eeyore" wrote

Do both power amps run off the same supply rails ?

Graham

Yes.

Makes sense, as an additional power supply would cost
$$$$.

I've seen it done otherwise with taps on the TX.

Yes, that saves you the cost of the second transformer, but it still costs
you diodes and a couple of caps.



Space is a bit tight on a powered monitor also.



Gareth.

[Eeyore]

Wink wrote:

"Arny Krueger" wrote:
"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)

Did you mean All Electronics?

http://www.allelectronics.com/cgi-bi...APACITOR_.html

whoops, yes All Electronics.

It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.

Well ..... you may not like how 20,000 uF sound ! It will change the clipping behaviour agin.

My advice to you is to get a larger amplifier so you're never close to clipping.


I've resolved the voltage on the 63 V BGs torment. I ordered toroids
with this sec: 50-43-0-43-50. The 43 V taps will run produce 56 V
under load. A compromise value, not excessively low.

Err .... if only life were that simple ! Capacitor input PSUs are not like resistive loads.


Should I eject the BGs like a retching dog to Ebay hell, the 100 V
winding will allow a return to 63-ish volt operation, maybe with the
cool Nichicons.

Or cool Panasonics or cool Samwhas or cool Rubycons or cool Nippon Chemi Cons !

The damn BRAND is IRRELEVANT !

Graham

[Eeyore]

Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine for rock and roll sound reinforcement though.

Graham

[jakdedert]

Eeyore wrote:

Wink wrote:

"Arny Krueger" wrote:
"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)
Did you mean All Electronics?

http://www.allelectronics.com/cgi-bi...APACITOR_.html
whoops, yes All Electronics.
It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.

Well ..... you may not like how 20,000 uF sound ! It will change the clipping behaviour agin.

Couldn't this strategy also overwhelm the the preceeding stages
(transformer/rectifier/fuse) by creating a huge inrush current upon turnon?

My advice to you is to get a larger amplifier so you're never close to clipping.


I've resolved the voltage on the 63 V BGs torment. I ordered toroids
with this sec: 50-43-0-43-50. The 43 V taps will run produce 56 V
under load. A compromise value, not excessively low.

Why? With proper voltage filter caps, the original would appear to be
adequate. The 63 volt rails were only--marginally--problematic to the
63 volt rated magic caps.

Err .... if only life were that simple ! Capacitor input PSUs are not like resistive loads.

Should I eject the BGs like a retching dog to Ebay hell, the 100 V
winding will allow a return to 63-ish volt operation, maybe with the
cool Nichicons.

Or cool Panasonics or cool Samwhas or cool Rubycons or cool Nippon Chemi Cons !

The damn BRAND is IRRELEVANT !

Graham

[GregS[_3_]]

In article , jakdedert wrote:
Eeyore wrote:

Wink wrote:

"Arny Krueger" wrote:
"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)
Did you mean All Electronics?


http://www.allelectronics.com/cgi-bi...F_80_V_ELECTRO
LYTIC_SNAP-IN_CAPACITOR_.html
whoops, yes All Electronics.
It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.

Well ..... you may not like how 20,000 uF sound ! It will change the clipping
behaviour agin.

Couldn't this strategy also overwhelm the the preceeding stages
(transformer/rectifier/fuse) by creating a huge inrush current upon turnon?

Thats why some amplifiers have a slow start circuit.

Also, while monitoring the rails at the circuit, any fluctuations
are caused by the wiring and power supply. Any changes with various
capicitors can be monitored here. If there is no change, I don't
think there is any way the output could change.

greg


My advice to you is to get a larger amplifier so you're never close to
clipping.


I've resolved the voltage on the 63 V BGs torment. I ordered toroids
with this sec: 50-43-0-43-50. The 43 V taps will run produce 56 V
under load. A compromise value, not excessively low.

Why? With proper voltage filter caps, the original would appear to be
adequate. The 63 volt rails were only--marginally--problematic to the
63 volt rated magic caps.

Err .... if only life were that simple ! Capacitor input PSUs are not like
resistive loads.

Should I eject the BGs like a retching dog to Ebay hell, the 100 V
winding will allow a return to 63-ish volt operation, maybe with the
cool Nichicons.

Or cool Panasonics or cool Samwhas or cool Rubycons or cool Nippon Chemi Cons
!

The damn BRAND is IRRELEVANT !

Graham

[Arny Krueger]

"Wink" wrote in message

I think the Behringer Europower amp are much less.

Lots. The A500 is $200, the EP1500 is $280, and the EP2500 is $350.

But I personally own none of the above. A close friend owns 4 EP2500 for his
subwoofers.

I think that's you're using, I assume to drive your
Maggies, which I've never heard, by the way.

The Maggies aren't mine, they belong to a (different) close friend. He
still uses the Hafler amp with them, AFAIK.

[Arny Krueger]

"Eeyore" wrote in
message
Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine
for rock and roll sound reinforcement though.

As we both know, they are QSC clones. Very precise clones, down to an
amazing level of detail.

They pass stright wire bypass tests, though.

[Eeyore]

Arny Krueger wrote:

"Eeyore" wrote
Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine
for rock and roll sound reinforcement though.

As we both know, they are QSC clones. Very precise clones, down to an
amazing level of detail.

They pass stright wire bypass tests, though.

I'd like to hear that for myself. Some earlier QSCs sounded distinctly
less than pure to me.

Graham

[Arny Krueger]

"Eeyore" wrote in
message
Arny Krueger wrote:

"Eeyore" wrote
Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine
for rock and roll sound reinforcement though.

As we both know, they are QSC clones. Very precise
clones, down to an amazing level of detail.

They pass stright wire bypass tests, though.

I'd like to hear that for myself. Some earlier QSCs
sounded distinctly less than pure to me.

Given Graham that you are in the UK, and I am in the Midwest US, you're
pretty much on your own! ;-)

You could warm up with the USA 850 - generated files from the PCABX web
site.

[Wink]

It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.

Well ..... you may not like how 20,000 uF sound ! It will change the clipping behaviour agin.

My advice to you is to get a larger amplifier so you're never close to clipping.
Not going through that again.


I've resolved the voltage on the 63 V BGs torment. I ordered toroids
with this sec: 50-43-0-43-50. The 43 V taps will run produce 56 V
under load. A compromise value, not excessively low.

Err .... if only life were that simple ! Capacitor input PSUs are not like resistive loads.
Huh? 43 x 1.414 = 60.8 V
No load condition of course.

With this particular amp, PSU components, and biased to 350 mils, That
60.8 drops to 56 volts. The correction factor is .921

I approached this multiple ways actually. I empirically comfirmed my
calculations by setting the Variac to produce 86 AC no load volts at
the xfmr secondary. I then measured that voltage under load -- minimal
change, as you would expect, still ~ 86 Vac. This produces +/- 56
volts DC rail voltage under load. (Of course the AC mains voltage is
varying slightly the entire time. Insignificant)

The Grand Unification coefficient for predicting DC rail voltage under
load for a given xfmr secondary output is 112/86 = 1.302
That's with 10,000 ufd BGs.

The stock Hafler xfmr's sec output is 97 volts @ 120Vac mains input.
With the 27 kufd Chemicons I just pulled, this produces +/- 63 Vdc at
350 mils bias -- yes, idling, not dynamic conditions. 126/97 = 1.298

1.302
1.298

Looks like you're right, it isn't a resistive load, but about as
predictable. Yes I know I've included too many significant digits, or
digits that are not significant.

The idea is to moderately under-rate the caps without going to
excessively low voltages. Not terribly critical as Arny et al have
delicately expressed.

Should I eject the BGs like a retching dog to Ebay hell, the 100 V
winding will allow a return to 63-ish volt operation, maybe with the
cool Nichicons.

Or cool Panasonics or cool Samwhas or cool Rubycons or cool Nippon Chemi Cons !

The damn BRAND is IRRELEVANT !

Yes it is.

[Wink]

Well ..... you may not like how 20,000 uF sound ! It will change the clipping behaviour agin.

Couldn't this strategy also overwhelm the the preceeding stages
(transformer/rectifier/fuse) by creating a huge inrush current upon turnon?
Well, the caps I pulled to make room for the now infamous 10,000 ufd
BGs were 27,000ufd. Never a problem.

I've resolved the voltage on the 63 V BGs torment. I ordered toroids
with this sec: 50-43-0-43-50. The 43 V taps will run produce 56 V
under load. A compromise value, not excessively low.

Why? With proper voltage filter caps, the original would appear to be
adequate. The 63 volt rails were only--marginally--problematic to the
63 volt rated magic caps.

I was getting new xfmrs anyway. So I selected voltages that allow a
return to the original config with proper voltage caps, like Arny's
Nichicons. (Brand matters Graham), and avoid the marginal
problematics should I decide to stick with the BGs.

Why not? I had the option of getting any voltages I want at a price
comparable to Parts Express pricing on Avel Lindbergs.
Dave
-----------------------
The damn BRAND is IRRELEVANT !
Yes it is Graham, so is Can colour. Are you in Liverpool?
Graham

[Eeyore]

Wink wrote:

On Thu, 27 Mar 2008 09:54:31 -0400, "Arny Krueger"
wrote:

"Eeyore" wrote
Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine
for rock and roll sound reinforcement though.

As we both know, they are QSC clones. Very precise clones, down to an
amazing level of detail.

They pass stright wire bypass tests, though.

I heard they're very good. I am curious - amazing detail compared to
what?

He means that Behringer copied the QSC design very accurately. The
electronics is as close to a perfect clone as you'll get.

Even the general mechanical design is very similar.
http://www.live-audio.com/messages/archive7/91505.html

Graham

[Wink]

On Thu, 27 Mar 2008 11:18:35 +0000, Eeyore
wrote:



Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine for rock and roll sound reinforcement though.

Graham
They sound different from more refined amps?

[Wink]

On Thu, 27 Mar 2008 09:54:31 -0400, "Arny Krueger"
wrote:

"Eeyore" wrote in
message
Wink wrote:

I think the Behringer Europower amp are much less.

They're not exactly a refined design by a LONG way. Fine
for rock and roll sound reinforcement though.

As we both know, they are QSC clones. Very precise clones, down to an
amazing level of detail.

They pass stright wire bypass tests, though.

I heard they're very good. I am curious - amazing detail compared to
what?

[jakdedert]

GregS wrote:
In article , jakdedert wrote:
Eeyore wrote:
Wink wrote:

"Arny Krueger" wrote:
"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)
Did you mean All Electronics?


http://www.allelectronics.com/cgi-bi...F_80_V_ELECTRO
LYTIC_SNAP-IN_CAPACITOR_.html
whoops, yes All Electronics.
It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.
Well ..... you may not like how 20,000 uF sound ! It will change the clipping
behaviour agin.
Couldn't this strategy also overwhelm the the preceeding stages
(transformer/rectifier/fuse) by creating a huge inrush current upon turnon?

Thats why some amplifiers have a slow start circuit.

Also, while monitoring the rails at the circuit, any fluctuations
are caused by the wiring and power supply. Any changes with various
capicitors can be monitored here. If there is no change, I don't
think there is any way the output could change.

Hopefully he's reading this. It's a 'good' amp that the OP has
frankensteined into 'something' else. Based on what he's said here, it
may or may not have startup current limiting. I don't believe the
original Hafler on which it is based did. It wasn't originally intended
to have 20,000uf hanging off the bridge....

jak

greg

[Wink]

On Thu, 27 Mar 2008 21:50:03 -0500, jakdedert
wrote:

GregS wrote:
In article , jakdedert wrote:
Eeyore wrote:
Wink wrote:

"Arny Krueger" wrote:
"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)
Did you mean All Electronics?


http://www.allelectronics.com/cgi-bi...F_80_V_ELECTRO
LYTIC_SNAP-IN_CAPACITOR_.html
whoops, yes All Electronics.
It turns out they're a very good size. I probably can pack 20-ish kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.
Well ..... you may not like how 20,000 uF sound ! It will change the clipping
behaviour agin.
Couldn't this strategy also overwhelm the the preceeding stages
(transformer/rectifier/fuse) by creating a huge inrush current upon turnon?

Thats why some amplifiers have a slow start circuit.

Also, while monitoring the rails at the circuit, any fluctuations
are caused by the wiring and power supply. Any changes with various
capicitors can be monitored here. If there is no change, I don't
think there is any way the output could change.

Hopefully he's reading this. It's a 'good' amp that the OP has
frankensteined into 'something' else. Based on what he's said here, it
may or may not have startup current limiting. I don't believe the
original Hafler on which it is based did. It wasn't originally intended
to have 20,000uf hanging off the bridge....
Nope, no current limiting at all. I'm using Ixys FRED 600V, 12 amp
diodes in the bridge. I had 27,000 hanging off that bridge for
several years. No deaths.

However -- this is not a desirable situation. I'm now using only
10,000uf Black Gates, but am going to install a soft start circuit of
some type.

I've had Haflers for 25+ years. They are good amps, but can be
improved upon. A stock Parasound HCA1000a creams a stock 200 or 220
Dave

jak

greg

[Serge Auckland[_2_]]

"Wink" wrote in message
...
On Thu, 27 Mar 2008 21:50:03 -0500, jakdedert
wrote:

GregS wrote:
In article , jakdedert
wrote:
Eeyore wrote:
Wink wrote:

"Arny Krueger" wrote:
"Wink" wrote in message

I see some 3,300 uF 80 volt
Nichicon LQ-M caps for sale on the Active Electronics
web site for less than $1.60 each... ;-)
Did you mean All Electronics?


http://www.allelectronics.com/cgi-bi...F_80_V_ELECTRO
LYTIC_SNAP-IN_CAPACITOR_.html
whoops, yes All Electronics.
It turns out they're a very good size. I probably can pack 20-ish
kufd
a side in there, but would start at 3300 and experiment. I ordered a
dozen.
Well ..... you may not like how 20,000 uF sound ! It will change the
clipping
behaviour agin.
Couldn't this strategy also overwhelm the the preceeding stages
(transformer/rectifier/fuse) by creating a huge inrush current upon
turnon?

Thats why some amplifiers have a slow start circuit.

Also, while monitoring the rails at the circuit, any fluctuations
are caused by the wiring and power supply. Any changes with various
capicitors can be monitored here. If there is no change, I don't
think there is any way the output could change.

Hopefully he's reading this. It's a 'good' amp that the OP has
frankensteined into 'something' else. Based on what he's said here, it
may or may not have startup current limiting. I don't believe the
original Hafler on which it is based did. It wasn't originally intended
to have 20,000uf hanging off the bridge....
Nope, no current limiting at all. I'm using Ixys FRED 600V, 12 amp
diodes in the bridge. I had 27,000 hanging off that bridge for
several years. No deaths.

However -- this is not a desirable situation. I'm now using only
10,000uf Black Gates, but am going to install a soft start circuit of
some type.

I've had Haflers for 25+ years. They are good amps, but can be
improved upon. A stock Parasound HCA1000a creams a stock 200 or 220
Dave

jak

greg



How does it do that? Does it have flatter frequency response, lower
distortion, lower noise that aren't already below the threshold of hearing?
Unless you can quantify the differences, statements like "creams a stock 200
or 220" are not only meaningless, but come across as uninformed and
audiophoolery.

S.

--
http://audiopages.googlepages.com

[Arny Krueger]

"Serge Auckland" wrote in
message
"Wink" wrote in message
...

I've had Haflers for 25+ years. They are good amps, but
can be improved upon. A stock Parasound HCA1000a creams
a stock 200 or 220 Dave

How does it do that?

Perception and sighted listening tests.

I've heard them both. They are both very good.

Does it have flatter frequency
response, lower distortion, lower noise that aren't
already below the threshold of hearing?

Nope.

Unless you can
quantify the differences, statements like "creams a stock
200 or 220" are not only meaningless, but come across as
uninformed and audiophoolery.

If Wink ever did a proper listening test and believed the results, he'd have
a lot to think about. But he already laid out the logic-proof box he's
worked himself into, right near the start of the thread. He doesn't believe
in AB tests.

[Andre Majorel]

On 2008-03-26, Don Pearce wrote:

Although they are small, tweeters need just as much power as bass
drivers on average music material. But they are going to get a great
deal more upset than a bass unit when asked to mop up a lot of DC.

Why ? Because they cannot dissipate heat as effectively ? Or
because they somehow have more power to dissipate ?

--
André Majorel URL:http://www.teaser.fr/~amajorel/
(Counterfeit: )
"I drink, I smoke, I gamble, I chase girls--but postal chess is
one vice I don't have." -- Mikhail Tal

[Arny Krueger]

"Andre Majorel" wrote in message


On 2008-03-26, Don Pearce wrote:

Although they are small, tweeters need just as much
power as bass drivers on average music material. But
they are going to get a great deal more upset than a
bass unit when asked to mop up a lot of DC.

Why ? Because they cannot dissipate heat as effectively ?

Yes.

Tweeter voice coils are usually designed to have as little weight as
possible.

Or because they somehow have more power to dissipate ?

No.

[[email protected]]

On Apr 1, 2:36 pm, "Arny Krueger" wrote:
"Andre Majorel" wrote in message

On 2008-03-26, Don Pearce wrote:
Although they are small, tweeters need just as much
power as bass drivers on average music material. But
they are going to get a great deal more upset than a
bass unit when asked to mop up a lot of DC.
Why ? Because they cannot dissipate heat as effectively ?

Yes.

Tweeter voice coils are usually designed to have as
little weight as possible.

The answer, while technically correcvt, skips a lot
of important detail.

The important consideration here is the weight of
the voice coil itself which, in modern direct radiator
tweeters, often comprises a substantial portion of
the total moving mass, so minimizing its weight
is important in being able to get sufficient efficiency
out of such a small radiating area.

The radiating area has to be small in order to
achieve a reasonable dispersion at the short
wavelengths required, thus the diameter of the
tweeter and hence the voice coil has to be
appropriately small.

Further, at the frequencies over which the tweeter
works, the mechanical excursion is small and thus
the voice coil is also short.

All of that results in several things:

1. The voice coil, being small in diameter, short
in length and light in weight, must be wound
fairly small gauge wire to achieve the necessary
electrical and electro-mechanical properties
needed. It's not at all uncommon for the voice
coil to use 36 gauge wire (a two-layer 1" coil
with a length of .17" wound with 36 gauge wire
will result in a voice coil with a DC resistance
of just about 7 ohms and a Bl product in the
realm of about 5 with typical magnet structures
and weight about 0.6 grams, just about what
you want for a mid-efficiency tweeter).

2. The small, low mass structures end up with low
heat capaciites compared to woofer assemblies,
the result being large temperature rises for the
same disiipated power compared to woofers.

3. The thermal resistance of these assemblies is
usually substantially higher than larger drivers,
further exacerbating the temperature rise for
a given input power (which, actually, is the
definition of thermal resistance).

4. With no other heat transportt mechanisms,
the maximum available current carrying capacity
of the smaller gauge wire used in tweeter voice coils
is substantially less than tthe wire found in woofers.
By some standards, for example, the maximum
current capacity for 36 guage wire is on the order
of a few teths of an amp, while that of a 27 gauge
wire of the type found in a typical 6-8" woofer is
8 times as much by the same criteria.

[Mr.T]

wrote in message
...
On Apr 1, 2:36 pm, "Arny Krueger" wrote:
"Andre Majorel" wrote in message

On 2008-03-26, Don Pearce wrote:
Although they are small, tweeters need just as much
power as bass drivers on average music material. But
they are going to get a great deal more upset than a
bass unit when asked to mop up a lot of DC.
Why ? Because they cannot dissipate heat as effectively ?

Yes.

Tweeter voice coils are usually designed to have as
little weight as possible.

The answer, while technically correcvt, skips a lot
of important detail.

The important consideration here is the weight of
the voice coil itself which, in modern direct radiator
tweeters, often comprises a substantial portion of
the total moving mass, so minimizing its weight
is important in being able to get sufficient efficiency
out of such a small radiating area.

The radiating area has to be small in order to
achieve a reasonable dispersion at the short
wavelengths required, thus the diameter of the
tweeter and hence the voice coil has to be
appropriately small.

Further, at the frequencies over which the tweeter
works, the mechanical excursion is small and thus
the voice coil is also short.

All of that results in several things:

1. The voice coil, being small in diameter, short
in length and light in weight, must be wound
fairly small gauge wire to achieve the necessary
electrical and electro-mechanical properties
needed. It's not at all uncommon for the voice
coil to use 36 gauge wire (a two-layer 1" coil
with a length of .17" wound with 36 gauge wire
will result in a voice coil with a DC resistance
of just about 7 ohms and a Bl product in the
realm of about 5 with typical magnet structures
and weight about 0.6 grams, just about what
you want for a mid-efficiency tweeter).

2. The small, low mass structures end up with low
heat capaciites compared to woofer assemblies,
the result being large temperature rises for the
same disiipated power compared to woofers.

3. The thermal resistance of these assemblies is
usually substantially higher than larger drivers,
further exacerbating the temperature rise for
a given input power (which, actually, is the
definition of thermal resistance).

4. With no other heat transportt mechanisms,
the maximum available current carrying capacity
of the smaller gauge wire used in tweeter voice coils
is substantially less than tthe wire found in woofers.
By some standards, for example, the maximum
current capacity for 36 guage wire is on the order
of a few teths of an amp, while that of a 27 gauge
wire of the type found in a typical 6-8" woofer is
8 times as much by the same criteria.


All perfectly correct, however something you didn't mention is that a
tweeter is never "asked to mop up a lot of DC" unless something is seriously
broken, and I don't just mean the amplifier. (but that has to be broken too
of course for DC to appear at the output)
With any passive crossover, capacitors will block DC from the tweeter, and
in any decent active/bi-amped system, some form of DC prevention to the
tweeter is considered a good idea also.

MrT.

[[email protected]]

On Apr 2, 12:41 am, "Mr.T" MrT@home wrote:
All perfectly correct, however something you didn't
mention is that a tweeter is never "asked to mop up
a lot of DC" unless something is seriously broken,
and I don't just mean the amplifier. (but that has to
be broken too of course for DC to appear at the output)
With any passive crossover, capacitors will block
DC from the tweeter, and in any decent active/bi-amped
system, some form of DC prevention to the tweeter is
considered a good idea also.

NEVER say never!

DC current through tweeters is certainly a pathological
case, but that does not eliminate it from the realm of
possibilities. One of the consequences is that the same
DC current can often be more destructive of a tweeter
than the same AC RMS current. The reason being is
thatv the mechanical position of the voice coil gets
prefferentially biased on one direction or another,
rather than averaging around the 0 mechnical point.
The result is that the thermal resistance can be
increased either because it no longer sees the
metal in the gap as a sink, or, in the case of
magnetic fluid cooling, it actuall can move out of
the fluid. There is also the issue of mechanical
strain.

I once was asked by a client, a US distribution arm of
a French driver company, to evaluat about 1500 tweeters
that were returned by a dealer as "defective," and the
dealer was refusing to pay the now 120-day overdue bill.
Every single one of them showed substantial thermal
damage inthe form of melting in the plastic dome
material where the voice coil former (aluminum)
was joined to the dome. There was also some of
the ferrofluid splattered on the back of the dome and
around the front surface of the magnet structure.
All 1500 had the same failure.

When I tested unused samples of the same model,
I was able to drive them to more than 4 times their
rated power with in-band signals of every kind with
no failures.

However, if I simply hooked the tweeter to a car
battery and left them there for 1 minute, I was
able to EXACTLY duplicate the same appearance.
The voice coil got biased enough that the majority
of it was outside the fluid, withn only the rearmost
portion of the former in the fluid. At that point, the
voice coil was able to get hot enough to glow
visibly under room light, melting the plastic on the
dome and boiling the residue fluid off the voice
coil spattering it around ther area.

Yes, something was SERIOUSLY broken. The dealer,
a car installer in the southwest, didn't have the money
to pay for the order, so decided to methodically
connect every one of them to a car battery, burn them
up, and return them unpaid as "defective."

[Arny Krueger]

wrote in message


I once was asked by a client, a US distribution arm of
a French driver company, to evaluat about 1500 tweeters
that were returned by a dealer as "defective," and the
dealer was refusing to pay the now 120-day overdue bill.
Every single one of them showed substantial thermal
damage inthe form of melting in the plastic dome
material where the voice coil former (aluminum)
was joined to the dome. There was also some of
the ferrofluid splattered on the back of the dome and
around the front surface of the magnet structure.
All 1500 had the same failure.

When I tested unused samples of the same model,
I was able to drive them to more than 4 times their
rated power with in-band signals of every kind with
no failures.

Hmm, sounds like some Audax tweeters I've used in the past.

However, if I simply hooked the tweeter to a car
battery and left them there for 1 minute, I was
able to EXACTLY duplicate the same appearance.
The voice coil got biased enough that the majority
of it was outside the fluid, withn only the rearmost
portion of the former in the fluid. At that point, the
voice coil was able to get hot enough to glow
visibly under room light, melting the plastic on the
dome and boiling the residue fluid off the voice
coil spattering it around ther area.

Yes, something was SERIOUSLY broken. The dealer,
a car installer in the southwest, didn't have the money
to pay for the order, so decided to methodically
connect every one of them to a car battery, burn them
up, and return them unpaid as "defective."

This sort of thing happens not infrequently, and sometimes fraud is not the
reason.

Most hard drive vendors demand that devices returned in warranty will have
failed a diagnostic test using a program they provide. It's no secret that
hard drives can easily fail in actual use, but pass the diagnostic, even
when run in the failing computer. There are procedures for ensuring that
the hard drive will always fail the diagnostic, none of which need be
discussed right now.