Need a big transformer...

So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

 

[Bob Eld]

<mrdarrett@gmail.com> wrote in message
news:37c1b825-37ff-4042-8c36-e5a294c4b4bb@s12g2000prg.googlegroups.com...

So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

There is no such thing as an RMS Watt. You mean 68Watts sign wave average.
Depending on the service, you can usually get by with about 1/4 of the
average sign wave power in the transformer which would be about 70 Watts.
100 Watt transformer would be a little better. Remember music reproduction
does not need continuous power. Any transformer can supply four times it's
rated power in short bursts. It's a matter of heating or how hot the
transformer is allowed to get. For your application a 100Watt transformer is
more than sufficient.

 

[Paul E. Schoen]

<mrdarrett@gmail.com> wrote in message
news:37c1b825-37ff-4042-8c36-e5a294c4b4bb@s12g2000prg.googlegroups.com...

So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

Why do you need a transformer? The amplifier has direct output drive with
+/- 28 VDC supplies. You can use two inexpensive 100 watt switchers, and
add a little series inductor and a really big capacitor on each, to take
care of HF switching noise and short burst peak power requirements. You
could even use 24 VDC supplies, which might be more common, and available
very cheap surplus or used.

Paul

 

[Phil Allison]

"Bob Eld"

mrdarrett@gmail.com
So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.


There is no such thing as an RMS Watt.

** Really ?

You mean 68Watts sign wave average.

** Which is precisely what the term " watts rms " means in audio amplifier
specs.

Depending on the service, you can usually get by with about 1/4 of the
average sign wave power in the transformer which would be about 70 Watts.
100 Watt transformer would be a little better. Remember music reproduction
does not need continuous power.

Any transformer can supply four times it's rated power in short bursts.

** The penalty is in the regulation factor, which is four times worse than
normal.

With a rectifier and capacitor input filter, the current peaks are higher
than the simple resistive load case - makes the DC regulation much poorer
than the AC figure quoted in the specs for the transformer.

Eg: a nominal 100VA transformer has a full ( resistive) load regulation of
10%.

So for a peak secondary voltage of 35 volts under load, the off load peak
voltage needs to be 39 volts for a resistive load and 43 volts for a
rectifier load - which is higher that the absolute max spec for the
LM3886.

With four times current overload, the AC peak output and DC supply voltages
falls drastically, to around 20 volts.

It's a matter of heating or how hot the
transformer is allowed to get. For your application a 100Watt transformer
is
more than sufficient.

** Not true at all.


BTW: The OP has made his usual dumb mistakes.

The 68 watt figure applies to a resistive amplifier load of 4 ohms and DC
supplies of ** +/- 28 volts **

A transformer with a 70 volt AC secondary ( plus CT) will produce DC rails
of +/- 54 volts off load.

Way to high - bye, bye LM3886.

The transformer he needs is one rated at 200VA with a 25-0-25 volt, 4 amp
secondary.

This will give him DC rails of +/- 37 volts off load, dropping to about
+/- 28 volts at full load.



........ Phil

 

On Mar 22, 3:34 pm, "Paul E. Schoen" <pst...@smart.net> wrote:

mrdarr...@gmail.com> wrote in message

news:37c1b825-37ff-4042-8c36-e5a294c4b4bb@s12g2000prg.googlegroups.com...



So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

Why do you need a transformer? The amplifier has direct output drive with
+/- 28 VDC supplies. You can use two inexpensive 100 watt switchers, and
add a little series inductor and a really big capacitor on each, to take
care of HF switching noise and short burst peak power requirements. You
could even use 24 VDC supplies, which might be more common, and available
very cheap surplus or used.

Paul

24VDC supplies are commonly used for...?

Or do you mean regular computer ATX supplies, -12-0-12?

Could I take two ATX supplies, and tie them like this to get -24-0-24?

---------- ----------
| ATX | | ATX |
| Supply 1 | | Supply 2 |
| | | |
---------- ----------
| | | |
-12V +12V -12V +12V
| |___________| |
| | |
| | |
-24V 0 +24V


(view in fixed width font)

Thanks,

Michael

 

[Bob Eld]

"Phil Allison" <philallison@tpg.com.au> wrote in message
news:64ls1qF2c2r0iU1@mid.individual.net...

"Bob Eld"

mrdarrett@gmail.com
So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.


There is no such thing as an RMS Watt.

** Really ?

You mean 68Watts sign wave average.


** Which is precisely what the term " watts rms " means in audio
amplifier
specs.

Yes I know, a bull shit term invented by uneducated technicians and
marketing types trying to impress the unwashed masses. The term has been
misused for many years but that doesn't make it right. RMS means root mean
square. Power isn't a root, power is already squared. Only current and
voltage can be the root. You, of all people, should know that. It's
electricity 101 stuff.

 

[Michael A. Terrell]

mrdarrett@gmail.com wrote:

24VDC supplies are commonly used for...?

Or do you mean regular computer ATX supplies, -12-0-12?

Could I take two ATX supplies, and tie them like this to get -24-0-24?

Not safely or reliably. The cases are tied to earth, and the
negative 12 volt supply is low current on an ATX supply.

---------- ----------
| ATX | | ATX |
| Supply 1 | | Supply 2 |
| | | |
---------- ----------
| | | |
-12V +12V -12V +12V
| |___________| |
| | |
| | |
-24V 0 +24V

(view in fixed width font)

--
aioe.org is home to cowards and terrorists

Add this line to your news proxy nfilter.dat file
* drop Path:*aioe.org!not-for-mail to drop all aioe.org traffic.

http://improve-usenet.org/index.html

 

[Eeyore]

mrdarrett@gmail.com wrote:

So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency,

Where did you get 50% from ? At full power, Class AB amps are near as
dammit 67% efficient.

I'd need at least, what, a 150W transformer, right?

NO.

2 things.

1: Your efficiency calculation was wrong. Not mention the stuff about VA
vs W !

2: Most of the time, an audio amplifier is NOT running at full power.
Choosing what % of full power you expect it to be used at is a design
choice. You could of course design for 'worst case' but that's fairly
silly in this context since it results in a huge TX.

This is *designer* territory laddy ! You want to know how I deal with it
?

Graham

 

[Eeyore]

"Paul E. Schoen" wrote:

Why do you need a transformer?

To run it from a mains supply of cousre !

Are you really that silly ?

Graham

 

[Eeyore]

"Michael A. Terrell" wrote:

the negative 12 volt supply is low current on an ATX supply.

Too damn right !

Graham

 

[Eeyore]

mrdarrett@gmail.com wrote:

My Kenwood is advertised as 50W per channel. Peak... what does that
mean?

It's marketing bollocks.

Graham

 

[Phil Allison]

"Bob Eld"

There is no such thing as an RMS Watt.

** Really ?

You mean 68Watts sign wave average.


** Which is precisely what the term " watts rms " means in audio
amplifier specs.

Yes I know, a bull shit term invented by uneducated technicians

** That is an entirely false comment.

The term " watts rms" = watts derived from continuous sine wave test.

It is a simply a verbal shorthand.

Do try to get over your rampant pedantry - fool.

( snip rest of you paranoid drivel on the matter)



........ Phil

 

[Phil Allison]

<mrdarrett@gmail.com
On Mar 22, 6:48 pm,

There is no such thing as an RMS Watt.


You're absolutely right!

** Like, how the FUCK would an ignorant ass like you know ?



....... Phil

 

On Mar 22, 6:48 pm, "Bob Eld" <nsmontas...@yahoo.com> wrote:

"Phil Allison" <philalli...@tpg.com.au> wrote in message

news:64ls1qF2c2r0iU1@mid.individual.net...





"Bob Eld"

mrdarr...@gmail.com
So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

There is no such thing as an RMS Watt.

** Really ?

You mean 68Watts sign wave average.

** Which is precisely what the term " watts rms " means in audio
amplifier
specs.

Yes I know, a bull shit term invented by uneducated technicians and
marketing types trying to impress the unwashed masses. The term has been
misused for many years but that doesn't make it right. RMS means root mean
square. Power isn't a root, power is already squared. Only current and
voltage can be the root. You, of all people, should know that. It's
electricity 101 stuff.

You're absolutely right! National Semi uses the nomenclature
"continuous average output power". That's what I'd meant to say.

My Kenwood is advertised as 50W per channel. Peak... what does that
mean? For one second before the amplifier fries itself? For one
millisecond? For five seconds?

Then a web search (Kenwood itself didn't publish this) shows my
Kenwood can put out 22W RMS. Ah.
http://www.crutchfield.com/S-OzS0NnnnJC2/App/Product/Item/Main.aspx?I=113mp205&search=kenwood+mp205

Now if they'd said "continuous average output power" that would be
more correct..?

MD

 

On Mar 22, 6:52 pm, "Michael A. Terrell" <mike.terr...@earthlink.net>
wrote:

mrdarr...@gmail.com wrote:

24VDC supplies are commonly used for...?

Or do you mean regular computer ATX supplies, -12-0-12?

Could I take two ATX supplies, and tie them like this to get -24-0-24?

Not safely or reliably. The cases are tied to earth, and the
negative 12 volt supply is low current on an ATX supply.

---------- ----------
| ATX | | ATX |
| Supply 1 | | Supply 2 |
| | | |
---------- ----------
| | | |
-12V +12V -12V +12V
| |___________| |
| | |
| | |
-24V 0 +24V

(view in fixed width font)

--
aioe.org is home to cowards and terrorists

Add this line to your news proxy nfilter.dat file
* drop Path:*aioe.org!not-for-mail to drop all aioe.org traffic.

http://improve-usenet.org/index.html

Thanks!

Michael

 

[Paul E. Schoen]

<mrdarrett@gmail.com> wrote in message
news:0078d794-be10-40f4-b43c-252cdb98d58f@s19g2000prg.googlegroups.com...

On Mar 22, 3:34 pm, "Paul E. Schoen" <pst...@smart.net> wrote:
mrdarr...@gmail.com> wrote in message

news:37c1b825-37ff-4042-8c36-e5a294c4b4bb@s12g2000prg.googlegroups.com...



So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

Why do you need a transformer? The amplifier has direct output drive
with
+/- 28 VDC supplies. You can use two inexpensive 100 watt switchers, and
add a little series inductor and a really big capacitor on each, to take
care of HF switching noise and short burst peak power requirements. You
could even use 24 VDC supplies, which might be more common, and
available
very cheap surplus or used.

Paul


24VDC supplies are commonly used for...?

Or do you mean regular computer ATX supplies, -12-0-12?

Could I take two ATX supplies, and tie them like this to get -24-0-24?

---------- ----------
| ATX | | ATX |
| Supply 1 | | Supply 2 |
| | | |
---------- ----------
| | | |
-12V +12V -12V +12V
| |___________| |
| | |
| | |
-24V 0 +24V


(view in fixed width font)

Thanks,

Michael

I don't think the +/- 12 VDC in computer ATX supplies are the same current
rating. For instance, a 450W ATX has +12V at 15 amp, but -12V at 0.8A. and
usually you must also load the 5 VDC supply.

But you can probably get a truckload of old computer supplies for free, so
maybe you could series the +12V outputs of several of them to get what you
need. The older non-ATX supplies might be easier to use, as the ATX type
require some additional connections.

www.mpja.com has a 24.5 VDC at 5A supply for $14.95 (Item 12912-PS). They
also have standard 24V 4.5A (Item 16015-PS) for $32.80, and 28V 5.5A (Item
16033-PS) for $40.75. Also 24V 4.2A (Item 16853-PS) for $24.95.

24 VDC supplies are standard industrial control voltage supplies and can
probably be found for $20 or less on eBay or surplus. 28 VDC is more for
military equipment (I think), but may be found pretty cheap.

Also check www.jameco.com and www.herbach.com for surplus deals. And also
local hamfests, like the one coming up in Timonium, MD next weekend:
http://www.gbhc.org/

Good luck,

Paul

 

On Mar 22, 8:24 pm, Eeyore <rabbitsfriendsandrelati...@hotmail.com>
wrote:

mrdarr...@gmail.com wrote:
So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency,

Where did you get 50% from ? At full power, Class AB amps are near as
dammit 67% efficient.

Rough guess.

Ok, I glanced at the Power Dissipation vs. Output Power curve on page
15 of the datasheet.

I'd need at least, what, a 150W transformer, right?

NO.

2 things.

1: Your efficiency calculation was wrong. Not mention the stuff about VA
vs W !

Ok, thanks for the clarification.
http://www.etasys.com/Main/ABCs/CalcVAWATTS.asp

2: Most of the time, an audio amplifier is NOT running at full power.
Choosing what % of full power you expect it to be used at is a design
choice. You could of course design for 'worst case' but that's fairly
silly in this context since it results in a huge TX.

You're right. I'm designing it for worst-case because I know my
brother-in-law is good at blowing up power transformers. If I'm going
to risk that kind of behavior, I might as well overdesign the power
supply.

This is *designer* territory laddy ! You want to know how I deal with it
?

Graham

Are you offering to explain?

Michael

 

On Mar 22, 10:12 pm, "Paul E. Schoen" <pst...@smart.net> wrote:

mrdarr...@gmail.com> wrote in message

news:0078d794-be10-40f4-b43c-252cdb98d58f@s19g2000prg.googlegroups.com...



On Mar 22, 3:34 pm, "Paul E. Schoen" <pst...@smart.net> wrote:
mrdarr...@gmail.com> wrote in message

news:37c1b825-37ff-4042-8c36-e5a294c4b4bb@s12g2000prg.googlegroups.com...

So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

Why do you need a transformer? The amplifier has direct output drive
with
+/- 28 VDC supplies. You can use two inexpensive 100 watt switchers, and
add a little series inductor and a really big capacitor on each, to take
care of HF switching noise and short burst peak power requirements. You
could even use 24 VDC supplies, which might be more common, and
available
very cheap surplus or used.

Paul

24VDC supplies are commonly used for...?

Or do you mean regular computer ATX supplies, -12-0-12?

Could I take two ATX supplies, and tie them like this to get -24-0-24?

---------- ----------
| ATX | | ATX |
| Supply 1 | | Supply 2 |
| | | |
---------- ----------
| | | |
-12V +12V -12V +12V
| |___________| |
| | |
| | |
-24V 0 +24V

(view in fixed width font)

Thanks,

Michael

I don't think the +/- 12 VDC in computer ATX supplies are the same current
rating. For instance, a 450W ATX has +12V at 15 amp, but -12V at 0.8A. and
usually you must also load the 5 VDC supply.

But you can probably get a truckload of old computer supplies for free, so
maybe you could series the +12V outputs of several of them to get what you
need. The older non-ATX supplies might be easier to use, as the ATX type
require some additional connections.

www.mpja.comhas a 24.5 VDC at 5A supply for $14.95 (Item 12912-PS). They
also have standard 24V 4.5A (Item 16015-PS) for $32.80, and 28V 5.5A (Item
16033-PS) for $40.75. Also 24V 4.2A (Item 16853-PS) for $24.95.

24 VDC supplies are standard industrial control voltage supplies and can
probably be found for $20 or less on eBay or surplus. 28 VDC is more for
military equipment (I think), but may be found pretty cheap.

Also checkwww.jameco.comandwww.herbach.comfor surplus deals. And also
local hamfests, like the one coming up in Timonium, MD next weekend:http://www.gbhc.org/

Good luck,

Paul

Thank you for the tips.

Michael

 

On Mar 22, 10:12 pm, "Paul E. Schoen" <pst...@smart.net> wrote:

mrdarr...@gmail.com> wrote in message

news:0078d794-be10-40f4-b43c-252cdb98d58f@s19g2000prg.googlegroups.com...



On Mar 22, 3:34 pm, "Paul E. Schoen" <pst...@smart.net> wrote:
mrdarr...@gmail.com> wrote in message

news:37c1b825-37ff-4042-8c36-e5a294c4b4bb@s12g2000prg.googlegroups.com...

So let's say I want to build an LM3886 amplifier.
http://www.national.com/mpf/LM/LM3886.html

At 68W RMS, assuming 50% efficiency, I'd need at least, what, a 150W
transformer, right?

And if I build two for stereo, that would be 300W required. 70V
secondary, center-tapped, to get +/- 35V.

Those things are expensive on Mouser ($60+). Especially considering
the LM3886 only costs $7 or so.

Any suggestions where I can get such a large transformer for cheap?

Thanks,

Michael

Why do you need a transformer? The amplifier has direct output drive
with
+/- 28 VDC supplies. You can use two inexpensive 100 watt switchers, and
add a little series inductor and a really big capacitor on each, to take
care of HF switching noise and short burst peak power requirements. You
could even use 24 VDC supplies, which might be more common, and
available
very cheap surplus or used.

Paul

24VDC supplies are commonly used for...?

Or do you mean regular computer ATX supplies, -12-0-12?

Could I take two ATX supplies, and tie them like this to get -24-0-24?

---------- ----------
| ATX | | ATX |
| Supply 1 | | Supply 2 |
| | | |
---------- ----------
| | | |
-12V +12V -12V +12V
| |___________| |
| | |
| | |
-24V 0 +24V

(view in fixed width font)

Thanks,

Michael

I don't think the +/- 12 VDC in computer ATX supplies are the same current
rating. For instance, a 450W ATX has +12V at 15 amp, but -12V at 0.8A. and
usually you must also load the 5 VDC supply.

But you can probably get a truckload of old computer supplies for free, so
maybe you could series the +12V outputs of several of them to get what you
need. The older non-ATX supplies might be easier to use, as the ATX type
require some additional connections.

www.mpja.comhas a 24.5 VDC at 5A supply for $14.95 (Item 12912-PS). They
also have standard 24V 4.5A (Item 16015-PS) for $32.80, and 28V 5.5A (Item
16033-PS) for $40.75. Also 24V 4.2A (Item 16853-PS) for $24.95.

24 VDC supplies are standard industrial control voltage supplies and can
probably be found for $20 or less on eBay or surplus. 28 VDC is more for
military equipment (I think), but may be found pretty cheap.

Also checkwww.jameco.comandwww.herbach.comfor surplus deals. And also
local hamfests, like the one coming up in Timonium, MD next weekend:http://www.gbhc.org/

Good luck,

Paul

If I tie the secondaries to get +/-24V, two of these might actually do
the trick.
http://www.mpja.com/prodinfo.asp?number=7846+TR

If I only wanted to power one LM3886, two of the smaller ones could
work:
http://www.mpja.com/prodinfo.asp?number=7845+TR

Pretty cheap. Cheaper than Mouser, at least...

Michael

 

[Bob Eld]

"Phil Allison" <philallison@tpg.com.au> wrote in message
news:64m13rF2bvvaqU1@mid.individual.net...

"Bob Eld"

There is no such thing as an RMS Watt.

** Really ?

You mean 68Watts sign wave average.


** Which is precisely what the term " watts rms " means in audio
amplifier specs.

Yes I know, a bull shit term invented by uneducated technicians


** That is an entirely false comment.

The term " watts rms" = watts derived from continuous sine wave test.

It is a simply a verbal shorthand.

Idiotic shorthand for saying "sine wave"? That's absurd. If they wanted to
say sine wave why wouldn't they say "sine wave" and not some goofy reference
to something that is a nonsequitur.

BTW notice than nobody in the commercial or industrial world uses the term.
It is only used in the consumer arena where ignorant sales types try to
impress the natives with insider technical talk.

When was the last time you saw an electric motor rated in RMS Watts? Motor
manufacurers arn't that stupid.