Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

"No One Really"

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp?
They
sort of seem to be doing the same thing.


** Only if you think that connecting two identical batteries in series or
parallel is the same thing.




........... Phil

On Fri, 2 Sep 2005 16:59:58 +0200, No One Really wrote:

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

No. Bridging doubles the output voltage, while parallel doubles the output
current. You need to choose your speaker impedance to suit either
condition.

d

"Don Pearce" wrote in message

On Fri, 2 Sep 2005 16:59:58 +0200, No One Really wrote:

Looking at the datasheet for a National LM4780 audio
amplifier IC I was wondering: what's the difference
between a parallel and a bridged amp? They sort of seem
to be doing the same thing.

No. Bridging doubles the output voltage, while parallel
doubles the output current. You need to choose your
speaker impedance to suit either condition.

The pedantic version would be ;-)

Bridging doubles the maximum available output voltage, which
may lead to a doubling of maximum output current. The latter
eventuality needs to be carefully considered because it is a
common pitfall of bridged operation.

Parallel operation doubles the output current that is
available for delivery to the load.

You need to choose your speaker impedance to suit either
condition.

Or, you need to choose your mode of operation to suit your
speaker load and needs for dynamic range.

No One Really wrote:

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

You got your answer but I'll add that I wouldn't recommend that parallel
configuration. It involves a pair of current sharing resistors that increase
the output impedance.

Graham

"No One Really" wrote in message
news:1125673453.32b95b4cc31f94002d7a5ff15be71d6e@t eranews...
Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp?
They
sort of seem to be doing the same thing.



Take a close look to see how the speaker is hooked up in the bridge
configuration.
Tam

I read in sci.electronics.design that Pooh Bear
wrote (in
) about 'Difference between bridged and
parallel amp?', on Fri, 2 Sep 2005:

No One Really wrote:

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

You got your answer but I'll add that I wouldn't recommend that parallel
configuration. It involves a pair of current sharing resistors that increase
the output impedance.

One UK company sold/sells sound system amplifiers with two 60 W MOSFET
amp modules in parallel. It works, but it worried me a lot when I first
saw it. There are no current-sharing resistors and no cross-connections
of any sort. Not happy at full output at 20 kHz, though; smoke happens.
--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

On Fri, 2 Sep 2005 16:59:58 +0200, "No One Really"
wrote:

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

---
What no one has commented on so far is that for a given load
impedance, using paralleled outputs will do nothing as far as
increasing the power into the load goes, while using bridged outputs
will result in _quadrupling_ the power into the load! That is, of
course, assuming the power supply can deliver the increased
(doubled) current.

--
John Fields
Professional Circuit Designer

"No One Really" wrote in message
news:1125673453.32b95b4cc31f94002d7a5ff15be71d6e@t eranews...
Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp?
They
sort of seem to be doing the same thing.



As stated, output power is quadrupled because of the doubled voltage swing
and thus, doubled current in the output. However, one must pay attention to
the maximum current in the output devices. For example, an audio amplifier
IC may be rated to drive a 4 Ohm minimum load at rated supply voltage. When
bridged, it can no longer handle the 4 Ohm load. 8 Ohm would be the new
minimum. So, considering you are trying to get as much power from the amp by
using the lowest output impedance possible in any case, the maximum you can
gain by bridging is double the power.

John

"No One Really" wrote in message
news:1125673453.32b95b4cc31f94002d7a5ff15be71d6e@t eranews...
Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp?
They
sort of seem to be doing the same thing.

Use bridging when your amplifier is voltage limited, i.e. when it won't put
out enough voltage to drive your speaker to full power.

Don't use parallel amplifiers unless the feedback loop includes both
amplifiers, which it doesn't unless the amplifiers were specifically
designed for parallel operation.

Norm Strong

"No One Really" wrote in message
news:1125673453.32b95b4cc31f94002d7a5ff15be71d6e@t eranews...
Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp?
They
sort of seem to be doing the same thing.

The main difference is the input connections. In the parallel amplifier
echematic, the inputs of the two channels ae the same, but in the bridged
amplifier schematic, the input of channel B is inverted wrt to the input of
channel A.

And so the outputs of the bridged amplifier are also inverted wrt to each
other; and by connecting the speaker across the outputs for both channels,
it will receive twice the voltage (one terminal at +V, one at -V, rather
than one terminal at +V and the other at 0V).

Tim

On Fri, 02 Sep 2005 17:15:39 +0100, John Woodgate wrote:

I read in sci.electronics.design that Pooh Bear
wrote (in
) about 'Difference between bridged and
parallel amp?', on Fri, 2 Sep 2005:

No One Really wrote:

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

You got your answer but I'll add that I wouldn't recommend that parallel
configuration. It involves a pair of current sharing resistors that increase
the output impedance.

One UK company sold/sells sound system amplifiers with two 60 W MOSFET
amp modules in parallel. It works, but it worried me a lot when I first
saw it. There are no current-sharing resistors and no cross-connections
of any sort. Not happy at full output at 20 kHz, though; smoke happens.

Isn't this the point where someone usually jumps in and says that MOSFET's
don't need current sharing resistors because resistance goes up with
temperature? This person might even cite an excellent and well known
electronics text.

And then someone else, possibly even one of the authors of that text jumps
in and says "No, when used in the linear region, virtually all power
MOSFET's DO need current sharing resistors?"

Just wondering.

--Mac

Mac wrote:

On Fri, 02 Sep 2005 17:15:39 +0100, John Woodgate wrote:

I read in sci.electronics.design that Pooh Bear
wrote (in
) about 'Difference between bridged and
parallel amp?', on Fri, 2 Sep 2005:

No One Really wrote:

Looking at the datasheet for a National LM4780 audio amplifier IC I was
wondering: what's the difference between a parallel and a bridged amp? They
sort of seem to be doing the same thing.

You got your answer but I'll add that I wouldn't recommend that parallel
configuration. It involves a pair of current sharing resistors that increase
the output impedance.

One UK company sold/sells sound system amplifiers with two 60 W MOSFET
amp modules in parallel. It works, but it worried me a lot when I first
saw it. There are no current-sharing resistors and no cross-connections
of any sort. Not happy at full output at 20 kHz, though; smoke happens.

Isn't this the point where someone usually jumps in and says that MOSFET's
don't need current sharing resistors because resistance goes up with
temperature? This person might even cite an excellent and well known
electronics text.

And then someone else, possibly even one of the authors of that text jumps
in and says "No, when used in the linear region, virtually all power
MOSFET's DO need current sharing resistors?"

Just wondering.

To answer your question, I've found that the audio specific lateral mosfets made by
Hitachi and also similar ones from Semelab and Exicon share current very well
without any ballast Rs..

Other mosfets types have very variable treshold voltages and can't be used reliably
in this way.

But that's only in the application where those devices are in a single amplifier.

Paralleling entire amplifiers is wholly unwise with current sharing Rs. Mainly
since they won't have exactly the same voltage gain, given normal component
tolerances. The feedback loop will win over everything.

Graham

I read in sci.electronics.design that Mac wrote (in
) about 'Difference between
bridged and parallel amp?', on Sun, 4 Sep 2005:

I wrote:

One UK company sold/sells sound system amplifiers with two 60 W MOSFET
amp modules in parallel. It works, but it worried me a lot when I first
saw it. There are no current-sharing resistors and no cross-connections
of any sort. Not happy at full output at 20 kHz, though; smoke happens.

Isn't this the point where someone usually jumps in and says that
MOSFET's don't need current sharing resistors because resistance goes
up with temperature? This person might even cite an excellent and well
known electronics text.

And then someone else, possibly even one of the authors of that text
jumps in and says "No, when used in the linear region, virtually all
power MOSFET's DO need current sharing resistors?"

Just wondering.

These are not paralleled devices but complete 60 W amplifier cards with
the outputs paralleled. The output source impedance of each is quite
low, due to negative feedback, but not audiophool low. Nevertheless,
each output must look like a near short-circuit to the other amplifier.
--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

"François Yves Le Gal" wrote:

On Sun, 4 Sep 2005 23:10:59 +0100, John Woodgate
wrote:

Nevertheless,
each output must look like a near short-circuit to the other amplifier.

Then what about amps using multiple output devices in parallel ? They all
must look like near short-circuits to each other, don't them ?

John was pointing out correctly the difference between paralleling devices
inside and outside a closed feedback loop.

Graham

"François Yves Le Gal" wrote in
message
On Sun, 4 Sep 2005 23:10:59 +0100, John Woodgate
wrote:

Nevertheless,
each output must look like a near short-circuit to the
other amplifier.

Then what about amps using multiple output devices in
parallel ?

Almost always done inside the same feedback loop.

Almost always done with current-balancing resistors.

They all must look like near short-circuits to each other,
don't them ?

One reason why amps in parallel look more like short
circuits to each other are the individual feedback loops.

"François Yves Le Gal"
Some pommy dickhead

John was pointing out correctly the difference between paralleling devices
inside and outside a closed feedback loop.

I know. But you can parallel two amp modules, use proper current balancing
resistors and derive proper feedback (if needed).



** There is at least one proper, engineering type way to do it.

The two or more power stages are made to have unity gain - ie 100% NFB.

Then, a single voltage amp drivis them all.






........... Phil

"François Yves Le Gal" wrote:

On Mon, 05 Sep 2005 00:11:32 +0100, Pooh Bear
wrote:

John was pointing out correctly the difference between paralleling devices
inside and outside a closed feedback loop.

I know. But you can parallel two amp modules, use proper current balancing
resistors and derive proper feedback (if needed).

That's true but the result is inferior to a single amp designed for the
purpose.

Graham

"François Yves Le Gal"

Probably. But who cares nowadays with very high quality 1 Kw/4 Ohm class D
amp modules - with built-in SMPS, of course! - costing less than USD 200
in
OEM quantities ?



** Care to reveal the merchant ?

Tres interessant.




............ Phil

"François Yves Le Gal" wrote:

On Mon, 05 Sep 2005 12:46:38 +0100, Pooh Bear
wrote:

That's true but the result is inferior to a single amp designed for the
purpose.

Probably. But who cares nowadays with very high quality 1 Kw/4 Ohm class D
amp modules - with built-in SMPS, of course! - costing less than USD 200 in
OEM quantities ?

I guess that depends on the THD figures and so on that you don't mind.


Graham

"François Yves Le Gal" wrote in
message

On Mon, 5 Sep 2005 22:26:51 +1000, "Phil Allison"
wrote:

Care to reveal the merchant ?

The manufacturer... Scandinavian-based. Ice something.
:-)

Not exactly a KW for $1.98 but still impressive:

http://www.hypex.nl/

http://www.adireaudio.com/Home/UcDAmps.htm

On Mon, 05 Sep 2005 00:55:32 +0200, François Yves Le Gal
wrote:

On Sun, 4 Sep 2005 23:10:59 +0100, John Woodgate
wrote:

Nevertheless,
each output must look like a near short-circuit to the other amplifier.

Then what about amps using multiple output devices in parallel ? They all
must look like near short-circuits to each other, don't them ?

---
No, they must look like infinite impedances to each other. That is,
they should sink and source current into only the load an never into
or out of one another.

--
John Fields
Professional Circuit Designer

"François Yves Le Gal" wrote in
message news
On Mon, 5 Sep 2005 09:07:04 -0400, "Arny Krueger"
wrote:

Not exactly a KW for $1.98 but still impressive:

http://www.hypex.nl/

The Hypex UCD modules are excellent, but don't include
the (SM)PS.

Have a look at say,
http://www.icepower.bang-olufsen.com/sw2049.asp

The issue of price was raised:

http://www.audiokabel.dk/da-amps.htm

3900 Kroener exchanges out at about $650. About 6 DKK per
USD

"François Yves Le Gal" wrote in
message
On Mon, 5 Sep 2005 09:55:07 -0400, "Arny Krueger"
wrote:

The issue of price was raised:

http://www.audiokabel.dk/da-amps.htm

3900 Kroener exchanges out at about $650. About 6 DKK per
USD

That's the retail price for a *complete* amp, including
case, connectors, packaging et al.

Agreed.

The OEM module - here a 500 w/4 ohm version - costs much
less, say less than USD 150 FOB in OEM quantities.

Let's hope so, or the guys at
http://www.audiokabel.dk/da-amps.htm are running a
money-losing operation.

Now for a little bit of fun.

Let's have a look at the web site of a manufacturer using
ASP modules:

Say, Red Dragon Audio.

http://www.reddragonaudio.com/main.htm

Their SA 1000 is sold direct for USD 1,799 (plus tax and
shipping)

http://www.reddragonaudio.com/sa1000.htm


Pricey for a 600 wpc/8 ohm amp.

It uses two 1000ASP modules housed in a very nice
aluminum box fitted with sturdy sockets, jacks, binding
posts and feet,, which looks identical to the Entry 130
manufactured in China and sold in North America by
Digital Analogue DIY for USD 130 (Qty one, end user
price).

http://www.digitalanaloguediy.com/entry130.html

Isn't that the same gif on both web sites?

Hmmm. Some modules - minimum order is half a pallet, 80
modules, so you have to front something like 30 G's, some
nice boxes, a couple of feet of internal wiring, a mains
cable, a sturdy UPS-proof box and a laser-printed user
manual

Presto, a new "manufacturer" is born...

i.e., red dragon audio! ;-)

(All this in order to show that the wide availability of
OEM modules will have the same effect on the audio market
than it had on PC's: we'll see more and more state of the
art products at affordable prices. We'll also see
manufacturers trying to differentiate their offerings
with tweaked or custom solutions).

Could be.

On Mon, 05 Sep 2005 14:44:08 +0200, François Yves Le Gal
wrote:

On Mon, 5 Sep 2005 22:26:51 +1000, "Phil Allison"
wrote:

Care to reveal the merchant ?

The manufacturer... Scandinavian-based. Ice something.

Aah, are you referring to the B&O ICE modules? Proving very popular
now that they've proved their chops in the Lab 5 speakers.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

"François Yves Le Gal" wrote in
message news
On Mon, 5 Sep 2005 13:59:55 -0400, "Arny Krueger"
wrote:

Isn't that the same gif on both web sites?

It awfully looks like the same .gif.

I can't be affirmative as I've not run an ABX double blid
triple Dutch test...
:-)

I think it was clever how they avoided the need for a new
gif by keeping the front panel clean. I do like at least a
power indicator on the front panel, though. ;-)

On Mon, 05 Sep 2005 13:59:55 -0400, Arny Krueger wrote:

"François Yves Le Gal" wrote in
http://www.reddragonaudio.com/main.htm

http://www.digitalanaloguediy.com/entry130.html

Isn't that the same gif on both web sites?

No, they're the same, but they're jpg's. ;-)

$ ls en*
entry130_front.jpg-1 entry130_front.jpg-2
richgrise@thunderbird:~
$ diff entry130_front.jpg-1 entry130_front.jpg-2
richgrise@thunderbird:~
$

Cheers!
Rich

"John Woodgate" wrote in message
news
One UK company sold/sells sound system amplifiers with two 60 W MOSFET amp
modules in parallel. It works, but it worried me a lot when I first saw
it. There are no current-sharing resistors and no cross-connections of any
sort. Not happy at full output at 20 kHz, though; smoke happens.

Amplifiers run on smoke. When the smoke comes out they stop running. Very
simple!