[[email protected]]

I tried a single Alpine Type R off my 1500W response monoblock but it
didn't hit hard enough so I've obtained a pair of Pioneer TSW1201DVC in
a custom sealed box. Now they are producing a decent amount of bass but
they are 4 ohm subs so presenting a 2 ohm amp load which means it's
only putting out 800W RMS @ 14.4V according to specs.

Seems a waste to run this amp at half what it's capable of and I have
the chance to pick up another pair of these subs cheaply so I was
wondering if running them isobaric would be a good option. I have never
run an isobaric setup but I don't want an enclosure that fills the boot
so it might be a good option. I'm thinking of mounting the 2nd pair of
subs on the outside of the box face to face with the current subs and
running them 180 degrees out of phase.

That way I'll have 4 drivers presenting a 1 ohm load they'll be sharing
1500W RMS in theory across the 4. HOw much louder is this likely to be?

THanks
LT.

[D.Kreft]

On Oct 18, 1:53 am, wrote:

Seems a waste to run this amp at half what it's capable of and I have
the chance to pick up another pair of these subs cheaply so I was
wondering if running them isobaric would be a good option. I have never
run an isobaric setup but I don't want an enclosure that fills the boot
so it might be a good option. I'm thinking of mounting the 2nd pair of
subs on the outside of the box face to face with the current subs and
running them 180 degrees out of phase.

First, note that you cannot simply slap one driver atop another--you
have to separate them with a ring of wood so that the surrounds don't
rub and wear a hole in your speakers. Depending upon the sub, a 3/4"
ring should suffice--you want enough for the surrounds to clear one
another, but no more--too much air trapped between the drivers is less
than optimal.

That way I'll have 4 drivers presenting a 1 ohm load they'll be sharing
1500W RMS in theory across the 4. HOw much louder is this likely to be?

Anytime you double power *or* piston area, you get a 3dB SPL gain. If
you double both power and piston area, you get a 6dB SPL gain. So here,
you're talking about maintaining the piston area as constant but
*almost* doubling the power (1500W / 4 speakers == 375 W/speaker)...so
you'll get close to a 6dB SPL gain. It'd be noticeable, but I can't
tell you if that's going to satisfy you.

Another thing to keep in mind is that typically, when you run an amp at
lower impedances, the amp is going to run much hotter and you're not
going to have as much dynamic headroom. Think about jumping on your bed
with vaulted ceilings--the amount of headroom (room to jump) you have
is going to be far less if you raise your bed by a meter--you'll be
higher in altitude when you jump, but if you jump just as hard, you'll
find yourself banging your head against the ceiling. This is exactly
what happens in an amp--when your baseline (low performance level, not
"bass line") is closer to the limits of the amplifier, it'll have less
"oomph" to give you when you have a really snappy bass-drum kick you
want to produce. You'll also have a reduced damping factor as well, so
your setup, while louder, might not sound as responsive or controlled
as it once used to. All this comes from experience.

If I were you, I'd visit a local, reputable dealer who knows a lot
about those subs and ask them what you should do in your particular
situation[1]--but going isobarik for the sake of achieving maximum SPL
is almost never the right reason.

You would do well to read the tutorials at the url below:

http://mobile.jlaudio.com/support_pages.php?page_id=146

-dan

[1] Actually, that's not entirely true. If I were you, and if I could
afford it, I'd switch to JL Audio subwoofers, paricularly the W7 line,
but that's just me. :-)

[Matt Ion]

D.Kreft wrote:

[1] Actually, that's not entirely true. If I were you, and if I could
afford it, I'd switch to JL Audio subwoofers, paricularly the W7 line,
but that's just me. :-)

I'll add a vote for JL - my 10" does a pretty decent job in my little hatchback

[Captain_Howdy]

Haven't you heard JL is for suckers.

In article zNsZg.155246$1T2.14117@pd7urf2no, Matt Ion
wrote:
D.Kreft wrote:

[1] Actually, that's not entirely true. If I were you, and if I could
afford it, I'd switch to JL Audio subwoofers, paricularly the W7 line,
but that's just me. :-)

I'll add a vote for JL - my 10" does a pretty decent job in my little hatchback

[almond]

you know i would build a better box first. try porting the box to the
manufacturers specs, that will help some. and running a 1 ohm stable
amp isn't really wasting it. it will run cooler longer, which keeps the
eqipment happy, it will draw less amps, which will keep you electrical
happy.
also wattage doesn't equal loud. i have an eclipse sw8200 running at
900 watts rms at 3 ohms, it has more output (in the same box, simple
drop in test) than my orion h2 at 2200 watts rms at 1 ohm.
sorry for the mispellings, i'm in a hurry.


--
almond

[GregS]

In article , almond wrote:

you know i would build a better box first. try porting the box to the
manufacturers specs, that will help some. and running a 1 ohm stable
amp isn't really wasting it. it will run cooler longer, which keeps the
eqipment happy, it will draw less amps, which will keep you electrical
happy.
also wattage doesn't equal loud. i have an eclipse sw8200 running at
900 watts rms at 3 ohms, it has more output (in the same box, simple
drop in test) than my orion h2 at 2200 watts rms at 1 ohm.
sorry for the mispellings, i'm in a hurry.

My opinion, reducing to half power means nothing as far as volume goes.
Power has little to do with SPL, but has very much to do with not clipping.
The real thing extra power gets you is headroom. I dare anybody
to reduce the volume to half power, from full power, by ear alone, without
using some kind of calibrated pad or attenuator. Can't be done.

greg

[almond]

i don't understand what that post was for? please elaborate.


--
almond

[D.Kreft]

On Oct 18, 11:18 am, (GregS) wrote:

Power has little to do with SPL, but has very much to do with not clipping.

Wow, so if I'm reading you correctly, a 500W amp will produce a
negligible difference in output compared to a 50W amp when applied to
the exact same speaker set-up?

The real thing extra power gets you is headroom.

How are you defining "headroom"?

I dare anybody
to reduce the volume to half power, from full power, by ear alone, without
using some kind of calibrated pad or attenuator. Can't be done.

That proves nothing more than "doubling or halving your power does not
result in a doubling of SPL." As a matter of fact, it takes a 10-fold
increase in power (or piston area) to just *double* the objective
loudness of a sound system. Just because one cannot precisely locate
the "half-power" (or 3dB down) point of a system does in no way imply
that the difference is not discernable. Because the Bel is nothing more
than a logarithmic ratio between two values (see
http://en.wikipedia.org/wiki/Decibel#Acoustics for how it applies to
sound), and most people don't understand that, it should come as no
surprise that the experiment you propose would produce the results you
suggest (i.e. "can't be done").

-dan

[GregS]

In article , almond wrote:

i don't understand what that post was for? please elaborate.

Which post? Please elaborate.

greg

[GregS]

In article . com, "D.Kreft" wrote:
On Oct 18, 11:18 am, (GregS) wrote:

Power has little to do with SPL, but has very much to do with not clipping.

Wow, so if I'm reading you correctly, a 500W amp will produce a
negligible difference in output compared to a 50W amp when applied to
the exact same speaker set-up?

The 50 w will be roughly half volume. What I'm trying to say, a 1000 watt
amp is going to sound very close to the same volume as a 500 watt amp.
One cannot set his volume at 1000 watts or 500 watts, so whats going
to really matter.

The real thing extra power gets you is headroom.

How are you defining "headroom"?

Not hearing clipping.

I dare anybody
to reduce the volume to half power, from full power, by ear alone, without
using some kind of calibrated pad or attenuator. Can't be done.

That proves nothing more than "doubling or halving your power does not
result in a doubling of SPL." As a matter of fact, it takes a 10-fold
increase in power (or piston area) to just *double* the objective
loudness of a sound system. Just because one cannot precisely locate
the "half-power" (or 3dB down) point of a system does in no way imply
that the difference is not discernable. Because the Bel is nothing more
than a logarithmic ratio between two values (see
http://en.wikipedia.org/wiki/Decibel#Acoustics for how it applies to
sound), and most people don't understand that, it should come as no
surprise that the experiment you propose would produce the results you
suggest (i.e. "can't be done").

The 3 dB was considered by many to be the minimin descernable
level change. It only has merit at certain levels. At real high levels and
real low levels, the 3 dB means little compared to 1 khz levels
at 90 dB SPL. The 10 dB or doubling or halfing, also is level and frequency dependant.
I however, might claim to feel the difference of 3dB more than I can hear the difference.
When making a speaker, sometimes a 1 or 2 dB change in the crossovers or
levels can make a noticible difference. It seems easier to compare levels in pink
noise spectrum, than using the 3 dB standard of minimum descernable level
change.

greg

[Brandonb]

Hold on a moment...

From what I'm readin gon the specs for the Pioneer TS W1201DVC... they
are only 4-ohm DVC subs. Which would mean if you did a parallel/parallel
wiring configuration for the pair of subs, going to a single mono-block
amp, that's a 1-ohm final load. Not 2-ohm. Adding the other 2 identical
subs would put it to 1/2-ohm, which would probably not make your amp
very happy.

Just kind of adding this as it seems to highly pertain to the discussion.

Brandonb


wrote:
I tried a single Alpine Type R off my 1500W response monoblock but it
didn't hit hard enough so I've obtained a pair of Pioneer TSW1201DVC in
a custom sealed box. Now they are producing a decent amount of bass but
they are 4 ohm subs so presenting a 2 ohm amp load which means it's
only putting out 800W RMS @ 14.4V according to specs.

Seems a waste to run this amp at half what it's capable of and I have
the chance to pick up another pair of these subs cheaply so I was
wondering if running them isobaric would be a good option. I have never
run an isobaric setup but I don't want an enclosure that fills the boot
so it might be a good option. I'm thinking of mounting the 2nd pair of
subs on the outside of the box face to face with the current subs and
running them 180 degrees out of phase.

That way I'll have 4 drivers presenting a 1 ohm load they'll be sharing
1500W RMS in theory across the 4. HOw much louder is this likely to be?

THanks
LT.

[Eric Desrochers]

D.Kreft wrote:

On Oct 18, 11:18 am, (GregS) wrote:

Power has little to do with SPL, but has very much to do with not clipping.

Wow, so if I'm reading you correctly, a 500W amp will produce a
negligible difference in output compared to a 50W amp when applied to
the exact same speaker set-up?

50 to 500 watts is only 10 dB. Subjectively, 10 dB is like doubling the
volume, to most people ear. It was determined experimentally long ago.

The real thing extra power gets you is headroom.

How are you defining "headroom"?

There are not many definition. That's the difference between the actual
volume and the maximum volume. The reserve of power if you will.

If you feed some speakers by the 25 + 25 watts head unit, you may obtain
a volume that is adequate most of the time when near maximum power. But
when there is a sudden peak in the music, it will be clipped, it won't
go higher. Go to a 100 + 100 watts amp and you now have 6 dB of
headroom. IE, you most likely won't set the volume knob higher but the
peaks will now pass unnaffected. Hence better sound, more dynamics.

In live-sound applications, we routinely use amplifiers twice as much
powerful as the nominal power of the speakers. We seldom damage
anything. That's because clean, short peaks are always better than the
harsh signal from an underpowered amplifier. Also, live sound is more
dynamic, less compressed that CD playback so the average power level is
lower even if the peaks are higher.

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[D.Kreft]

On Oct 18, 7:06 pm, (Eric Desrochers) wrote:

50 to 500 watts is only 10 dB. Subjectively, 10 dB is like doubling the
volume, to most people ear. It was determined experimentally long ago.

You're not telling me anything I don't already know...I pointed this
out later in the same post to which you replied.

If you feed some speakers by the 25 + 25 watts head unit, you may obtain
a volume that is adequate most of the time when near maximum power. But
when there is a sudden peak in the music, it will be clipped, it won't
go higher. Go to a 100 + 100 watts amp and you now have 6 dB of
headroom. IE, you most likely won't set the volume knob higher but the
peaks will now pass unnaffected. Hence better sound, more dynamics.

Well you see, I think this is a somewhat flawed analysis because you
are assuming that after you upgrade the amp, your base listening volume
will be unchanged. My experience shows that more often than not, people
who upgrade from say a head unit to a good 50x2 will have a tendency to
listen to their systems at a greater base volume. The difference is
less pronounced when you start dealing with more power, obviously,
because there comes a point beyond which it becomes physically
uncomfortable to listen to the system.

If you assume that the base listening volume will not change, then yes,
I fully agree that a larger amp will buy you more dynamic headroom (not
to mention less distortion, cooler operation and greater damping).
However, if you bump up your base listening volume with an amp upgrade,
your headroom does not change. Follow what I'm saying?

In live-sound applications, we routinely use amplifiers twice as much
powerful as the nominal power of the speakers. We seldom damage
anything. That's because clean, short peaks are always better than the
harsh signal from an underpowered amplifier.

Right. Square waves (resulting from clipped signals) contain a *lot* of
energy--prime fodder for a burnt voice coil.

I've said a million times to customers and fellow enthusiasts that the
quickest way to destroy a speaker is to put an amp on it that is too
weak to suit your needs. The second fastest way to destroy a speaker is
to dramatically overpower it to the point where you run into serious
thermal issues.

Also, live sound is more
dynamic, less compressed that CD playback so the average power level is
lower even if the peaks are higher.

Yep, which is the reason you never, ever, ever want to plug your Strat
into your B&O home speakers. :-)

-dan

[MOSFET]

I've said a million times to customers and fellow enthusiasts that the
quickest way to destroy a speaker is to put an amp on it that is too
weak

Oh boy, Dan, now you've gone and done it!!! What a can of worms this one
ALWAYS is.

Yet again, another OFTEN disputed car audio axiom: "Underpowering a speaker
CAN be just as damaging to the speaker as overpowering".

This question seems to come up (and then is subsequently HEAVILY debated)
about every six months or so ever SINCE THE INCEPTION OF RAC.

I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.

In other words, as long as an amp does not exceed a speakers capability, it
doesn't matter what the amp is playing (even HEAVILY clipped music or
distortion).

Dan, is this not the case? Can a 10 watt per channel amp somehow damage a
subwoofer capable of hundreds of watts?

Again, I'm no EE, so if you choose to tackle this subject please explain for
us non-engineers.

Thank you, neighbor!!!!

Nick

[Eric Desrochers]

MOSFET wrote:

I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.

It's very simple and is another thing generally known.

Simply, the rms power of a square wave is twice that of a sine wave for
equal amplitude *peak*. Your 100 watts rms amp is really puting 200
watts if totally clipped. The dreaded "maximum power" quoted by
manufacturer (which is always twice the rms power) is indeed correct,
alas for 100% distorsion!

So a 50 watts amp clipped to square wave into a 100 watts rms speaker
should be safe but a 100 watts rms clipped into the same 100 watts
speaker would probably burn it.

Guitar amps reproduce heavily clipped sound and don't blow the speaker.
The great guitar solo in the metallica tune is not blowing anything even
if it is technically nearly a square wave! That's not the square wave
per se that destructs things, but the increase of actual power
generated.

The other thing to considere is that music have an average-to-peak
ratio. While clipping, you increase the average level. Tweeters in
particular are delicate and are designed to receive normal music, since
the average level is lower. Most tweeters rated in the 100 watts range
really won't tolerate more than about 5-10 watts continous sine wave for
any amount of time. A clipped signal increase the average level,
possibly above what the tweeter can take.

That's some of the thing we learn in electronic college!

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[D.Kreft]

On Oct 18, 8:53 pm, "MOSFET" wrote:

I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.

You're right, of course. I think you'd be pretty hard-pressed to blow a
500W RMS subwoofer with a 25W head unit...unless, of course, your head
unit were basically dumping DC to your driver over an extended period
of time--that could get he voice coil hot enough to do some damage. The
problem comes about when you have lots of power being applied, but not
enough cone movement to keep the coil cool, such as the case when you
apply DC power or you try to play lots of square waves.

But yeah, what you've got here is a case of me not qualifying my
statements heavily enough (gosh, saying anything on USENET always takes
so bloody long because no matter what is said, someone is going to poke
a hole in it).

Dan, is this not the case? Can a 10 watt per channel amp somehow damage a
subwoofer capable of hundreds of watts?

Highly unlikely, again, assuming we're not passing straight DC to the
drivers.

Let's put it this way...you're likely to do more damage to a 100W
speaker with a 30W amp played with obscene amounts of clipping than you
are using a 170W amp played cleanly.

Ack. The more I think of it, the more I wish I could redact that post
or just pull it altogether. I sincerely apologize in advance if this
drags-up another stupid debate. I I know--if someone tries to debate,
we'll all just put our fingers in our ears and say very loudly
"LALALALALALALALALALALALALALALA I CAN'T HEAR YOU!"

-dan

[almond]

thank you for twisitng my words. no 500 compared to 50 will not sound
the same. but in my car with my experience and on wini sd models the
sw8200 at 900 watts rms at 3 ohms and the h2 at 2200 watts st 1 ohm the
sw8200 is louder. the amp is still putting out 46.9 volts, the amperes
go down with the resistance. and of course spl is directly related to
wattage but there are many other parameters to look at.


--
almond

[Matt Ion]

Eric Desrochers wrote:
MOSFET wrote:


I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.


It's very simple and is another thing generally known.

Simply, the rms power of a square wave is twice that of a sine wave for
equal amplitude *peak*. Your 100 watts rms amp is really puting 200
watts if totally clipped. The dreaded "maximum power" quoted by
manufacturer (which is always twice the rms power) is indeed correct,
alas for 100% distorsion!

I don't know that it would actually be DOUBLED, but DC power is definitely
greater than RMS power - I believe the voltage difference is approximately a .7
factor (RMS voltage of a sine wave is .7 times the peak voltage).

Digging back in my brain about 20 years to 12th Grade Physics... if your top
rail is running, for the sake of calculation, at 40VDC, a max-power sine wave at
40V peak-to-peak would have an RMS voltage of 28V. Put that across a four ohm
load, and (assuming a purely resistive load, for ease of illustration) with
P=V^2/R, you get P=28*28/4 or 196W RMS.

Now put the straight output of the rail into the sub - what would happen if your
signal goes straight DC - and you get P=40*40/4, or... yup, Eric is right, it is
actually MORE than doubled - it's 400W.

Assuming, of course, my memory of Physics 12 is working properly

Now if you've got a 300W sub (speaker wattage ratings actually being an
indication of how much heat they can dissipate), and this amp that's probably
rated 200W RMS, the odd clipping won't be a problem, but extreme clipping for
extended periods will definitely produce output beyond the rating of the sub.

So a 50 watts amp clipped to square wave into a 100 watts rms speaker
should be safe but a 100 watts rms clipped into the same 100 watts
speaker would probably burn it.

Add to that the fact that when as long as the signal is clipped and effective
DC, the cone isn't moving, isn't getting any airflow across the coil, and thus,
no cooling, meaning your coil is going to get very hot, very fast.

Another consideration is that the output stages of the amp are only designed to
handle so much current; in cheaper amps, they're probably running close to max
with a clean RMS signal; with a badly-clipped signal and the corresponding
increase in current, you stand a good chance of blowing one of more of those as
well. Worse, the increase in heat generated by all that current can lead to
THERMAL breakdown of the transistors, which will also kill them quite readily.
Often these two conditions will "feed" each other, accelerating the death of the
semiconductor - heat starts the thermal failure, which leads to an increase of
current, which causes the generation of more heat, and so on.

Most often a transistor will go "open" when it blows, breaking the circuit, and
you end up with just a dead amp (personally, I've far more often seen the power
supply transistors fry, which either blows the fuse or just shuts down the amp,
but we're talking theoreticals here), but they occaisionally also go "short", no
longer controlling output power based on input, but simply dumping current
straight through, and producing pure DC at the output, which as we've already
noted, is far worse than simple clipping.

Better amps will have beefier output stages and better cooling, so it's more
likely to handle such abuses, but all electronics have their limits...

Guitar amps reproduce heavily clipped sound and don't blow the speaker.
The great guitar solo in the metallica tune is not blowing anything even
if it is technically nearly a square wave! That's not the square wave
per se that destructs things, but the increase of actual power
generated.

Well, guitar amps also typically aren't clipping the output itself - the
waveform is clipped BEFORE it's fed to the output stages, and in most tube
designs, it's also going through a transformer before the speaker jack, which
helps smooth the clipping a bit.

All that being said, MOSFET is also correct that you're probably not going to
blow a 100W sub with a 10W amp no matter how much you clip it... the danger is
in amps that are only SLIGHTLY "under-powered" in RMS ratings, but are still
capable of producing excessive power under clipping conditions.

*Phew*... that WAS going to be short and simple and I found myself continually
going back as I thought of more stuff to add I think I need another drink!

[MOSFET]

"LALALALALALALALALALALALALALALA I CAN'T HEAR YOU!"

-dan


What? Huh? YOU'LL HAVE TO SPEAK UP, I'VE GOT MY FINGERS IN MY EARS.

Nick

[MOSFET]

THANKS GUYS!!!

That did clear some things up for me. I appreciate it.

Nick

"MOSFET" wrote in message
m...
I've said a million times to customers and fellow enthusiasts that the
quickest way to destroy a speaker is to put an amp on it that is too
weak

Oh boy, Dan, now you've gone and done it!!! What a can of worms this one
ALWAYS is.

Yet again, another OFTEN disputed car audio axiom: "Underpowering a
speaker
CAN be just as damaging to the speaker as overpowering".

This question seems to come up (and then is subsequently HEAVILY debated)
about every six months or so ever SINCE THE INCEPTION OF RAC.

I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous
to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will
faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.

In other words, as long as an amp does not exceed a speakers capability,
it
doesn't matter what the amp is playing (even HEAVILY clipped music or
distortion).

Dan, is this not the case? Can a 10 watt per channel amp somehow damage a
subwoofer capable of hundreds of watts?

Again, I'm no EE, so if you choose to tackle this subject please explain
for
us non-engineers.

Thank you, neighbor!!!!

Nick

[MOSFET]

Add to that the fact that when as long as the signal is clipped and
effective
DC, the cone isn't moving, isn't getting any airflow across the coil, and
thus,
no cooling, meaning your coil is going to get very hot, very fast.



Guys, I'm sorry I'm so dense when it comes to physics matters involving
electricity.

Matt, you're saying there is a type of "sound" (that has to do with DC
current) that will cause the voice coils to heat up, AND YET there will be
little movement of the cone (and this lack of movement would certainly
exacerbate possible over-heating issues with the voice coil).

I am a bit confused on this point. If you don't mind, could you please
explain that just a bit more for a "non-EE" type of person because I find
that very interesting and it sounds like it MAY lend credence to Dan's
original statement that underpowering can be dangerous to speakers.

Thanks,

Nick

[Matt Ion]

MOSFET wrote:
Add to that the fact that when as long as the signal is clipped and

effective

DC, the cone isn't moving, isn't getting any airflow across the coil, and

thus,

no cooling, meaning your coil is going to get very hot, very fast.




Guys, I'm sorry I'm so dense when it comes to physics matters involving
electricity.

Matt, you're saying there is a type of "sound" (that has to do with DC
current) that will cause the voice coils to heat up, AND YET there will be
little movement of the cone (and this lack of movement would certainly
exacerbate possible over-heating issues with the voice coil).

ANY current through a resistive path will generate heat - think an electric
stove element, a baseboard heater, a lightbulb filament, etc. HEAT is the
primary killer of speaker coils - at best it melts the ultra-thin insulation on
the coil wires and causes a short; at worst it melts the wire and creates an
open circuit.

If the audio is clipping, it's moving the speaker to one extreme, and keeping it
there. The movement of the coil normally creates airflow across the coil that
helps cool it; when the movement stops (because of a clipped signal), so does
the cooling (it also partially dissipates heat to the magnet pole plate, but
without movement only the narrow region of the coil that's in close proximity to
the plate will be cooled).

If the clipping is minor, the "stops" are briefer and less heat builds; the more
severe the clipping, the longer the "stop" and the more damage is done.

I am a bit confused on this point. If you don't mind, could you please
explain that just a bit more for a "non-EE" type of person because I find
that very interesting and it sounds like it MAY lend credence to Dan's
original statement that underpowering can be dangerous to speakers.

[Captain_Howdy]

Now Misfit is going to be afraid to crank up his swap shop gear.

[GregS]

In article , (Eric Desrochers) wrote:
MOSFET wrote:

I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.

It's very simple and is another thing generally known.

Simply, the rms power of a square wave is twice that of a sine wave for
equal amplitude *peak*. Your 100 watts rms amp is really puting 200
watts if totally clipped. The dreaded "maximum power" quoted by
manufacturer (which is always twice the rms power) is indeed correct,
alas for 100% distorsion!

Who is going to play at square wave volume. Actually I found when
i used a bandpass box, depending on the frequency, you can get
heavy clipping with really hearing it, since its a bandpass.
Free extra power.

greg

[GregS]

In article jVDZg.157777$5R2.37673@pd7urf3no, Matt Ion wrote:
Eric Desrochers wrote:
MOSFET wrote:


I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.


It's very simple and is another thing generally known.

Simply, the rms power of a square wave is twice that of a sine wave for
equal amplitude *peak*. Your 100 watts rms amp is really puting 200
watts if totally clipped. The dreaded "maximum power" quoted by
manufacturer (which is always twice the rms power) is indeed correct,
alas for 100% distorsion!

I don't know that it would actually be DOUBLED, but DC power is definitely
greater than RMS power - I believe the voltage difference is approximately a .7

factor (RMS voltage of a sine wave is .7 times the peak voltage).

Digging back in my brain about 20 years to 12th Grade Physics... if your top
rail is running, for the sake of calculation, at 40VDC, a max-power sine wave
at
40V peak-to-peak would have an RMS voltage of 28V. Put that across a four ohm
load, and (assuming a purely resistive load, for ease of illustration) with
P=V^2/R, you get P=28*28/4 or 196W RMS.

What really happens is, most power amps will not be able to provide the extra
power trying to make a square wave at the same peak voltage, so maximum
power is more likely to be less than double of a sine wave.

greg

[Matt Ion]

GregS wrote:
In article jVDZg.157777$5R2.37673@pd7urf3no, Matt Ion wrote:

Eric Desrochers wrote:

MOSFET wrote:



I am no EE. But I have ALWAYS been under the assumption (based on what I
hear from the folks here) that underpowering a speaker is NEVER dangerous to
that speaker. Even if an amp is SEVERELY clipping, even if the signal is
100% clipped, the speaker does not know what clipping is and will faithfully
reproduce this noise as long as the power does not exceed the speaker's
power specifications.


It's very simple and is another thing generally known.

Simply, the rms power of a square wave is twice that of a sine wave for
equal amplitude *peak*. Your 100 watts rms amp is really puting 200
watts if totally clipped. The dreaded "maximum power" quoted by
manufacturer (which is always twice the rms power) is indeed correct,
alas for 100% distorsion!

I don't know that it would actually be DOUBLED, but DC power is definitely
greater than RMS power - I believe the voltage difference is approximately a .7

factor (RMS voltage of a sine wave is .7 times the peak voltage).

Digging back in my brain about 20 years to 12th Grade Physics... if your top
rail is running, for the sake of calculation, at 40VDC, a max-power sine wave
at
40V peak-to-peak would have an RMS voltage of 28V. Put that across a four ohm
load, and (assuming a purely resistive load, for ease of illustration) with
P=V^2/R, you get P=28*28/4 or 196W RMS.


What really happens is, most power amps will not be able to provide the extra
power trying to make a square wave at the same peak voltage, so maximum
power is more likely to be less than double of a sine wave.

True, it will depend on the performance of the amp's power supply... still, it
will be APPROXIMATELY doubled...

[MOSFET]

If the audio is clipping, it's moving the speaker to one extreme, and
keeping it
there. The movement of the coil normally creates airflow across the coil
that
helps cool it; when the movement stops (because of a clipped signal), so
does
the cooling (it also partially dissipates heat to the magnet pole plate,
but
without movement only the narrow region of the coil that's in close
proximity to
the plate will be cooled).


Interesting! That's exactly what I wanted to know. Thanks Matt.

Nick

[D.Kreft]

On Oct 19, 5:43 am, (GregS) wrote:

i used a bandpass box, depending on the frequency, you can get
heavy clipping with really hearing it, since its a bandpass.
Free extra power.

If you don't care how you define "free", yeah.

This is the very reason why I used to shy away from giving "bass
monkeys"
designs for bandpass enclosures unless I had some sense that they had
a clue what they were doing. The fact that you don't hear the
distortion from
a bandpass enclosure often doesn't mean "free bass", it means "more
trips
to the stereo store to buy replacement subs".

-dan

[Eric Desrochers]

Matt Ion wrote:

I don't know that it would actually be DOUBLED, but DC power is definitely
greater than RMS power

Hi Matt! Great to have this nice conversation without name calling or
blurbs of no-sense!

Actually, DC and RMS ARE equal power. RMS power is derived from PEAK
power as to get an equivalent way to rate AC systems compared to DC.

1 watt DC = 1 watt AC RMS = 2 watts AC PEAK, as long as it's a sine
wave.

- I believe the voltage difference is approximately a .7 factor (RMS
voltage of a sine wave is .7 times the peak voltage).

RMS mean Root Mean Squared. Hence, the value you are looking at is
0.7071067... !

Add to that the fact that when as long as the signal is clipped and
effective DC, the cone isn't moving, isn't getting any airflow across the
coil, and thus, no cooling, meaning your coil is going to get very hot,
very fast.

I must would oppose to this therory! Yes, the top of the clipped wave
form is flat like DC, but it's duration is so short. The VC is still
moving a lot. I would argue that the increased effective power is what
are killing them.

Another consideration is that the output stages of the amp are only
designed to handle so much current; in cheaper amps, they're probably
running close to max with a clean RMS signal; with a badly-clipped signal
and the corresponding increase in current, you stand a good chance of
blowing one of more of those as well. Worse, the increase in heat
generated by all that current can lead to THERMAL breakdown of the
transistors, which will also kill them quite readily. Often these two
conditions will "feed" each other, accelerating the death of the
semiconductor - heat starts the thermal failure, which leads to an
increase of current, which causes the generation of more heat, and so on.

This is also a false assumption. The plot thickens!! We may lost some
non-EE here, but here it goes.

Transistors are like valves in a water faucet. In a class AB, they goes
between fully open to fully closed, following the wave form of the input
signal. When totally closed the curent flowing in them is near 0 so the
power in them is 0. And when full conducting, the voltage across them
is near 0 so the power is also near 0. When partially conducting, they
must dissipate the heat created by the curent passing into them time the
voltage applied (rail voltage).

So, while reproducing a full power square wave, at any time, the
transistors are all either fully closed or fully open. They are in
partial conduction for a very small duration while switching. That
means, you guessed it, nearly no heat dissipation and doubled efficiency
(from 45 to 90%, for example)!! The power supply won't even notice and
the output transistors are all too happy to send a nice, full power
square wave!

This is basically how class-D amps works, and how they get their
impressive efficiency numbers.

Before Captain gets totally bent out of shape, it must be said that all
this talking, while technically correct, is not really usefull in real
life. Anyone with a ear and some sense won't allow such kind of
distorsion in his/her system.

All that being said, MOSFET is also correct that you're probably not going
to blow a 100W sub with a 10W amp no matter how much you clip it... the
danger is in amps that are only SLIGHTLY "under-powered" in RMS ratings,
but are still capable of producing excessive power under clipping
conditions.

That's correct. I would personnally prefere over-amping. As long as
YOU have the control over the volume knob, are capable of discipline and
have some ear to detect stress in the system.

--
Eric (Dero) Desrochers http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[Captain_Howdy]

Nice thread and I don't say this too often in here.

Transistors are like valves in a water faucet. In a class AB, they goes
between fully open to fully closed, following the wave form of the input
signal. When totally closed the curent flowing in them is near 0 so the
power in them is 0. And when full conducting, the voltage across them
is near 0 so the power is also near 0. When partially conducting, they
must dissipate the heat created by the curent passing into them time the
voltage applied (rail voltage).

So, while reproducing a full power square wave, at any time, the
transistors are all either fully closed or fully open. They are in
partial conduction for a very small duration while switching. That
means, you guessed it, nearly no heat dissipation and doubled efficiency
(from 45 to 90%, for example)!! The power supply won't even notice and
the output transistors are all too happy to send a nice, full power
square wave!

This is basically how class-D amps works, and how they get their
impressive efficiency numbers.


Bent out of shape, never.

Before Captain gets totally bent out of shape, it must be said that all
this talking, while technically correct, is not really usefull in real
life. Anyone with a ear and some sense won't allow such kind of
distorsion in his/her system.

I would say yes and no. If the amplifier outputs burn up from being over
driven for a lone period of time. In most amplifiers when the outputs burn up
they start feeding DC voltage right into the speaker. Most often the power
supply will a pop it's fuse and it's all good as far as the speaker goes, but
remove or over rate the fuse and it's speaker vs the power supply and from one
10 watt amplifier to another it's a hard call how things would turn out.


All that being said, MOSFET is also correct that you're probably not going
to blow a 100W sub with a 10W amp no matter how much you clip it... the
danger is in amps that are only SLIGHTLY "under-powered" in RMS ratings,
but are still capable of producing excessive power under clipping
conditions.

I also prefer over amping, but around here the trend seems to be power
matching or under amping for a false sense of security when it comes to
speaker protection. In fact not long ago there was some crazy guy in here that
popped a tweeter and started blaming it on his amplifier because it was a few
watts over the tweeter's power rating.

That's correct. I would personnally prefere over-amping. As long as
YOU have the control over the volume knob, are capable of discipline and
have some ear to detect stress in the system.

[Matt Ion]

Eric Desrochers wrote:

Actually, DC and RMS ARE equal power. RMS power is derived from PEAK
power as to get an equivalent way to rate AC systems compared to DC.

1 watt DC = 1 watt AC RMS = 2 watts AC PEAK, as long as it's a sine
wave.

In electrical theory, yes. However, the amp's rails run at a particular DC
voltage; the peak-to-peak output of a wave form can be UP TO that voltage, but
the RMS voltage will be whatever lesser level (.707 of the peak for a sine, more
or less for more complex waveforms).

Remember we're talking about a "floating" output with a split-rail configuration
- one channel runs from the "+" rail to common, the other from common to the "-"
rail, so the /peak/ can only be up to the the DC rail.

- I believe the voltage difference is approximately a .7 factor (RMS
voltage of a sine wave is .7 times the peak voltage).


RMS mean Root Mean Squared. Hence, the value you are looking at is
0.7071067... !

Right... .7 is close enough for our calculations here.

Add to that the fact that when as long as the signal is clipped and
effective DC, the cone isn't moving, isn't getting any airflow across the
coil, and thus, no cooling, meaning your coil is going to get very hot,
very fast.


I must would oppose to this therory! Yes, the top of the clipped wave
form is flat like DC, but it's duration is so short. The VC is still
moving a lot. I would argue that the increased effective power is what
are killing them.

Yes, but the point was, the MORE the signal is clipped, the LONGER the cone is
stopped at each extreme, allowing it more time to heat up. Worst-case, you end
up with a square wave. It's not the heat alone that kills the speaker, but it
CONTRIBUTES to the failure.

All that being said, MOSFET is also correct that you're probably not going
to blow a 100W sub with a 10W amp no matter how much you clip it... the
danger is in amps that are only SLIGHTLY "under-powered" in RMS ratings,
but are still capable of producing excessive power under clipping
conditions.


That's correct. I would personnally prefere over-amping. As long as
YOU have the control over the volume knob, are capable of discipline and
have some ear to detect stress in the system.

[GregS]

In article , (Eric Desrochers) wrote:
Matt Ion wrote:

I don't know that it would actually be DOUBLED, but DC power is definitely
greater than RMS power

Hi Matt! Great to have this nice conversation without name calling or
blurbs of no-sense!

Actually, DC and RMS ARE equal power. RMS power is derived from PEAK
power as to get an equivalent way to rate AC systems compared to DC.

1 watt DC = 1 watt AC RMS = 2 watts AC PEAK, as long as it's a sine
wave.

I was just reading about AC surge supressors. The term RMS cannot
be derived from DC or a spike, or so I read. You need a repetative
signal to compute it, technically. But your explanation is otherwise
common.


greg

[Eric Desrochers]

GregS wrote:

I was just reading about AC surge supressors. The term RMS cannot
be derived from DC or a spike, or so I read. You need a repetative
signal to compute it, technically. But your explanation is otherwise
common.


greg

Hello! "Derive" is probably not the best word to describe what I was
trying to say. I meant we need a way to express how much power a sine
wave contain versus DC.

Considere a 10 volts DC source applied to a resistor. The resistor will
dissipate some power, hence produce some heat.

Now take a sine wave with a peak of 10 volts. You will find that the
power and heat is less.

While all this was researched, decades ago, they tried to find a way to
express the sine wave energy that is equivalent to DC. It turned out,
if you increased the peak by the root mean square of 2 (ie 1.41) you
ended with a average power equall to DC.

So 10 DC contain as much energy as 14.1 volts AC peak, which is
expresses as 10 volts RMS.

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[D.Kreft]

On Oct 24, 5:47 pm, (Eric Desrochers) wrote:

So 10 DC contain as much energy as 14.1 volts AC peak, which is
expresses as 10 volts RMS.

All very correct, but in the interests of maintaining a long-standing
r.a.c. traditon of being hopelessly anal-retentive w.r.t. nit-picky
details, it should be highlighted that this *only* works for sine
waves. Music (well, unless you consider sine waves "music") carries
with it far less energy because of its peaky, jagged waveforms....so a
+14.1V peak voltage is going to have a far lower RMS equivalent.

Okay, that's my unsolicited, gratuitous, superflous $0.02. Carry on.

-dan

[Eric Desrochers]

D.Kreft wrote:

All very correct, but in the interests of maintaining a long-standing
r.a.c. traditon of being hopelessly anal-retentive w.r.t. nit-picky
details, it should be highlighted that this *only* works for sine
waves. Music (well, unless you consider sine waves "music") carries
with it far less energy because of its peaky, jagged waveforms....so a
+14.1V peak voltage is going to have a far lower RMS equivalent.

Okay, that's my unsolicited, gratuitous, superflous $0.02. Carry on.

-dan

Seeing how modern music is multi-band compressed, then hard limited, if
not outright clipped, in the name of maximum loudness, it's begining to
look more like white noise!

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95