Re: Hang-Time??? Ping: Rich Koerner


"Rich Koerner" <richk@xxxxxxxxxxxxx> wrote in message
news:462C68B2.AA81AA44@xxxxxxxxxxxxxxxx


Mike Rieves wrote:

"Rich Koerner" <richk@xxxxxxxxxxxxx> wrote in message
news:462B6494.53B9FF3D@xxxxxxxxxxxxxxxx


Guitarmakermark wrote:

Rich,

Thanks for the reply, I'm eager to read your upcoming post!

Thanks again,

Mark

Mark

I'm a little behind on things here in the shop, and I'll try to type
something up over the weekend.


Well, here is sit wishing I had the time to come up with something with
graphics as good as the ones
I found on another web site. So, for the sake of cutting down my time,
we
can refer to the
illustrations found there.

When you get to this web page, take your time going over the basic
theme
behind the different
magnetic devices shown, as you find your way down the page to the
section
on the loud speaker.
Note, that everything is based on the application of the SINE wave.

http://www.physclips.unsw.edu.au/jw/electricmotors.html





Special Note #1. Observe the continuous dynamic motion of the speaker
cone
in the dynamic speaker
graphic. This illustrates the resultant continuous motion that the
applied sine wave provides.

Special Note #2. Study the illustration in detail that shows the
internal
construction of the loud
speaker. When you return here, I'll go over things, explain to you
what
hang time is, and how hang
time kills speakers.

Which, over heats amplifiers too.



All the illustrations on the entire page show what happens when a sine
wave is applied to a coil of
wire held in different configurations of magnetic fields. If either
the
coil of wire or the
magnetic field is held stationary, the other will move in response to
the
alternating magnetic field
changes the applied sine wave produces in the coil of wire.

Looking at the dynamic speaker graphic and the internal construction
illustration, you can see how
the applied sine wave to that coil of wire, which sits in the
continuous
stationary magnetic field
of the voice coil's gap, moves the speaker cone in and out as the
alternating sine wave changes the
voice coil's magnetic field back and forth.

Back and forth, in and out, following the sine wave.

http://www.sprags.com/images/mainpower_sine_wave.jpg

Now, all the time this motion of the cone is going on, there is
particle
friction creating HEAT to
some level in the voice coil of the speaker. The more power that is
applied, the more the friction,
the higher the heat build up. At the extremes, heat with no place to
go,
will cause voice coil
assembly failure.

The internal construction illustration shows no real cooling system.
Yet,
there are some really
nice cooling methods that are used in today's speakers. Some use the
cone's movement to pump and
channel air flow through the voice coil gap to provide air cooling.
Some
also add a temperature
conducting fluid to the walls of the gap to transfer heat to the
speakers
structure away from the
coil. But, what ever the method of cooling, they all have their
practical
limits. Even under
ideal/normal operating conditions.

http://www.sprags.com/images/mainpower_sine_wave.jpg


Which, brings us to the impact that hang time has on a speaker.

Let's keep things simple, and forget about the many ways and places
that
the clipped sine wave can
unknowingly show up in the signal chain. For our example, we have a
sine/square wave generator on
the input of an amplifier.

We have that generator set to the sine wave mode, and are going to feed
the amp a 30 Hz signal.

http://www.sprags.com/images/mainpower_sine_wave.jpg

Then, we will turn up the volume to the rated output of the amplifier,
and
feed it to a speaker of
the same rating.

Now, if this is a perfect world, the manufacturer specs are not bogus,
and
we now have reached the
maximum performance level. Where, failure is NOT expected to show it's
ugly face.

Lets say, in 5 hours time, things are still running fine. Gee, no one
told lies for a change.

So, let's take a look at what is dynamically sitting in front of us.

We have an amp operating at max rating. Which should be considered
something that should not void
warranty. Likewise, the same with respect to the speaker under test
too.

Now let's say that this speaker under test uses cone pumped air to cool
it's voice coil, and there
is no over heating going on. Everything is just fine with BOTH the
amplifier and the speaker.

Therefore, we can conclude this setup lives up to what the specs as
stated, with a sine wave
applied.

Let's go to the speaker and take a close look at how it tracks that
sine
wave.


http://hem.passagen.se/communication/pic/pow313.gif

Looking at that sine wave on the right, as the cone follows the wave
from
it's point of rest to the
top peak, the cone moves in an outward direction reaching the top peak.
Then, instantly reverses
direction, and continues toward the inward peak. Where the cone, again
instantly reverses
direction, headed for that top peak again.

Thus, continuous air flow cooling the voice coil.


http://hem.passagen.se/communication/pic/pow313.gif

Now, let's follow what happens when the sine wave is switched to a
square
wave, and the speaker cone
tracks the square wave on the left.

The speaker cone first jumps out to its maximum distance at the top of
the
square wave, and is held
there at a complete stop for a moment in time. It makes no sound
sitting
there like that. Nor, is
there any air flow to cool the voice coil while the high current that
flows through the voice coil
holds it at a dead stop for that moment in time.

A moment in time where that voice coil is HEATING up from the high
current
holding it in place, with
no COOLING air flow going on, and making no sound.

After the end of that duration in time ends, the speaker jumps to its
inward maximum, and is held
there by the high current flow for a second moment in time. Where,
again
the voice coil is HEATING
up, there is again no sound, or air flow for cooling.


This is why I call this speaker operating condition Hang Time.

It hangs out and hangs in for that moment in time for the heat to build
up
in the voice coil
assembly.

There is one flaw in your "hang time" theory, air has inertia,

There is not enough air mass contained in the motor assembly *air cooling
design* to impact the
moving mass of the cone assembly.

We aren't talking about impacting the assembly, we're talking about the air
circulation cooling the assembly.

it doesn't
stop moving instantly just because the cone movement does. In point of
fact
the cone is still moving back and forth many times a second even at low
frequencies, often enough to keep air circulating around the voice coil
and
pole piece.

Ever take a JBL 130 or 140 and place it in a vise to hold it up free air.
Then place a candle flame
by the vent to check out the air flow there is at different frequencies,
and power levels.

I've actually done that with other speakers with vented pole pieces and
even when driven with square waves, you could see the air movement affecting
the flame.

At very low frequencies, that candle flame takes on a different appearance
from what the pure sine
wave produces compared to what a hard clip square wave produces.

Sure it does but it's still shows that the air is moving, in fact, because
of the increased cone velocity when reproducing a square wave, the air
actually moves more rapidly.

As you raise the test frequency, there comes a point where no change in
the flame's appearance can
be detected between the two wave shapes.

Of course, because at higher frequencies, the cone's velocity difference
between sine and square waves is less because the cone is too heavy to
matche the rise and fall times of the square wave.

The real problem is that when driven into heavy clipping, an amp
can put out much more than its rated power so the owner may not realize
that
his "fifty watt" amp is overloading his "fifty watt" speakers by a
considerable margin. If you put in more energy than a speaker's voice
coil
can dissipate, it will overheat, eventually resulting in failure,and if
the
inlsulation melts and the coils short, it can presents short circuit to
the
amp possibly resulting in its failure as well.

I mentioned that. But, you didn't get there yet.

I wasn't disagreeing with that part, just your "hangtime" theory.

If your hang time theory were correct, rock guitar players would be
replacing amps and speakers constantly, because an overdrinven guitar
sound
is mostly square-ish waveforms,

Guitars don't go down low enough to where hang time becomes heat building
to an *air cooled* speaker
with the high power amps common to bass players.

When you compare a 50 watt Marshall tube bass amp to a 50 watt Marshall
tube guitar amp, with the
same speakers used for both, you with find yourself in quite a different
playing field of
comparisons.

The Marshall bass amps aren't failure prone either, even though they're
overdriven as hard as the guyitar amps. I've seen quite a few rock bass
players who used marshall guitar amp cabinets, and they had no issues with
speaker failure.

and guitar amps are often driven at rated
power or above. For example, the reason a 50 Watt Marshall amp is so loud
is
that when heavily overdriven, they can put out around 200 watts.

Not so. The output tranny alone in a 50 watt marshall tube amp can't
support 200 watts. Not to
mention the power supply falls on it face at 50 watts. Most can't even
produce a 50 watt RMS sine
wave on a test load.

That's strange, because I've measured close to 200 wats from an fifty watt
Marshall amp driving a resistive dummy load. Of course, that was back in the
early seventies when Marshall amps were practically bullet proof, and you
could still get good tubes for them. You may be right about the modern
Marshalls, I haven't run any tests on them.

Assumed 120 volt line is held constant.


You see
those amps run wide open all the time, and both speakers and amps rarely
fail.

That's because things are not as you say.

As I said, with modern Marshals you might be right, but back in the day,
things were exactly as I say.

Now, what you have to remember here is that the difference between a
sine
wave and a square wave of
the same size is that the square wave is twice the power.

The amp is over heating trying to deliver the power, and the speaker is
over heating trying to deal
with the amplifier's square wave heat generating madness.

Here's another flaw in your reasoning, transisters or tubes don't
have
to dissapate a great deal of heat when acting as switches,

You forget tubes are not direct coupled to their speakers as transistors
are.

Makes no difference, they're still in near-switching mode when passing
square waves, and you're also forgetting that tubes naturally current-limit
when driven hard.

But, that's not what is going on here. There is a 4 ohm load on the SS
output devices here.

So take that SS amp and switch it on with a 9 volt DC battery input for a
40 volt DC offset at the
speaker terminals, and drop a 4 ohm resistor across them. By what road
did the current take to heat
that 4 ohm resistor get there. The output devices.

But, since the transistor was in full conduction at that point, making its
resistance very low, a matter of a few milliohms at most, the resistor is
dissipating the vast majority of the power. Remember, you have a resistive
network, consisting of the resistor and the transister. If the resistor is 4
ohms and the transistor is 10 milliohms, the resistor is dissipating 99.75%
of the output power. With 40 volts, you have 400 watts of output power
(voltage squared over resistance), and the resistor is dissipating 399 watts
and the output transistor is only dissipating one watt. How is that going to
overheat the output transisters? Keep in mind that we don't use audio amps
for DC voltages. The output device is only dissipating a considerable amount
of power when it's changing state from conduction to non-conduction. If it's
changing state rapidly, as with a square wave, it's spending very little
time changing state. When it's reproducing a sine wave, it's always changing
state because voltage in a sine wave is constantly changing. That means that
a sine wave is much more demanding on output transistors than is a square
wave. Look at it this way, a transistor drives a speaker by changing its
resistance constantly. At cut-off, resistance is very high and no current is
flowing, so no energy is being dissipated, at full conduction, reisitance is
very low and maximum current is flowing, but since the output's resistance
is very low, the speaker is dissipating nearly all the energy. At half
conduction, the output's resistance is equal to the speaker's resistance,
and they are each dissipating half the energy in the circuit, so that is the
point of highest dissipation in the output device.

This, can not be done with a vacuum tube amplifier, as the output tranny
will not support a
continuous DC offset
at the speaker terminals. Yet, that 9 volts DC placed on the control
grids of the output stage will
send those tubes into full conduct, and they will croak trying to get that
output tranny to pass
that DC to the secondaries.

No, they won't because tube amps aren't direct coupled, they're ether
transformer coupled or they use coupling capacitors, and all the output
tubes will see is a brief spike as the battery is connected and another as
it's disconnected.

That's why vacuum tube amplifiers have a hard time producing hang time in
speakers at low
frequencies.

Nope, the fact that a transformer cannot pass DC is why they can't
reproduce a true square wave at low frequencies.

Speakers don't like DC of any kind. Even for the short moment in hang
time.

Since DC isn't an audio frequency there is absolutely no reason for
putting DC across a speaker. BTW DC for a short time isn't DC, it's either
pulsating DC or AC, depending on what you're referencing it to.

And, don't try to tell me, that ain't DC.

Sorry, I just did.

and with a square
wave, they are mostly in a switching function. With a square wave, the
energy is dissipated mostly by the load, in this case, the speaker. That
means that while a speaker is dissipating more energy when driven by a
square wave, the amp's outputs are not. A sine wave is actually almost
the
worst case condition for an amp because when switching as they do with a
square wave, the output devices are either cut off, not passing anything
or
full on, meaning all the power is going to the speaker. With a sine wave,
the output devices have to gradually turn on and off, presenting a
resistance to the current flowing through them and causing them to have
to
dissipate an appreciable part of the power because they are sharing the
load
with the speakers.
The real problem comes when the speaker impedance is low enough that
the
amplifer output impedance becomes an appreciable percentage of the total
impedance of the amp/speaker combination and it has to dissipate a large
percentage of the total load. Because speakers are dynamic devices this
often happens only at certain frequencies, and if the amp is called on to
supply high power at these frequencies, it has to dissipate a greater
share
of the load, sometimes enough to cause it to overheat.

You know, I really wanted to put the focus on the speaker and not the amp,
to keep things simple.

Keeping it simple and explaining it wrong are two different things, a fact
that I'm exceedingly aware of, thanks to Jim Carr and others. You're
explanation might sound good to someone who doesn't know basic electronics,
but it's wrong, and the theory supports me on this.

Look, hang time heating of the voice coil is a given. It's been around
too long cooking speakers
with all manner of SS amp. From the *air cooled* subs in PA and SR
systems, to the speakers in the
first SS bass amps to hit the music stores.

You're the first person I've ever seen talking about "hangtime" in
speakers, and the facts simply don't support you. As I said previously, when
producing a square wave, the speakers are dissipating nearly all the energy,
that and the fact that a square wave has twice the power of a sine wave of
the same voltage is why speakers get fried, it has nothing to do with
"hangtime" causing reduced air circuation In fact, when reproducing a square
wave, the speaker cone moves faster, causing the air to move faster as well,
this increase in velocity offsets the time when the speaker cone is
"standing still". You might want to note that because of inertia, the
speaker doesn't actually stand still, a speaker cannot reproduce a true
square wave at audio frequencies.

Not to mention finding croaked speakers out of normal center position due
to continuous DC offset
from the SS amps too.

When you overdrive a speaker to the point that the voice coil jumps out of
the gap it's rarely because of DC, except in the cases when an output has
shorted and put the full DC supply voltage across the speaker. All decent
guitar and bass amps have circuits that sense DC and either prevent it from
reaching the outputs or they shut down the amp when they sense DC. There is
no continuous DC offset in any normally functioning guitar, bass, or PA amp,
they all have circuits that prevent DC offset from occurring. The only time
you see DC at an amp's output is when a component fails, and even then, the
protective circuitry nearly always shuts down the amp.

I really have a problem with the cheap SS bass amp designs that allow the
user to unknowingly create
hang time conditions that toast their speakers.

Could you name a couple of those amps?

In the face of today's technology, there is no freaking reason to make a
SS amp that will produce
hang time any longer.

Again, name a couple of the amps that you think will pass DC to the speakers

With a little thought, and few more parts, and that SS amp is now bullet
proof, and the speakers are
safe.

Nearly all modern amps have those parts and they protact the speakers from
DC. What they won't protect the speakers from is the user turning the amp
wide open and running more power into the speaker than it can handle.

The original poster would not have asked, and we would not be having this
dialog.


The OP's problem had nothing to do with hang time and the amp he has won't
pass DC to the outputs unless it's malfunctioning.


.

Relevant Pages

  • Re: Hang-Time??? Ping: Rich Koerner
    ... that everything is based on the application of the SINE wave. ... some level in the voice coil of the speaker. ... We have an amp operating at max rating. ...
    (alt.guitar.bass)
  • Re: The Limits of the LP
    ... what my ears tell me, ... This square wave bullshit has been debunked AGAIN ... less your ear actually hears a square and it actually hears a sine wave. ...
    (rec.audio.opinion)
  • Re: The Limits of the LP
    ... This square wave bullshit has been debunked AGAIN ... less your ear actually hears a square and it actually hears a sine wave. ... design of the filter, 44.1 or 48k are perfectly adequate to COMPLETELY ...
    (rec.audio.opinion)
  • Re: Hang-Time??? Ping: Rich Koerner
    ... that everything is based on the application of the SINE wave. ... some level in the voice coil of the speaker. ... We have an amp operating at max rating. ...
    (alt.guitar.bass)
  • Re: fourier anlysis of square wave...
    ... I convert a sine wave to a square wave using a zero crossing detector ... his zero crossings that align with sample points. ...
    (comp.dsp)