In article , ~misfit~
wrote:
Somewhere on teh intarwebs Arny Krueger wrote: [snip]
As time marched on, fairly inexpensive and modest-sized output devices
with excellent SOA and reasonble high frequency resposnse made it
relatively inexpensive to build very high powered audio power
amplifiers.

Arny, would you care to give a rough time-frame on this? I have a few
old SS amps (and might grab a few more if I see them at bargain prices)
but would love to know roughly when, in your opinion, SS amps became
stable and fairly clean across the frequency scale. Thanks.

Arny will no doubt have his own views in that, but I'll add mine FWIW.

I don't think the progression was as clear as it might be. Devices evolved
as the demands and expectations increased. So back in the 1960s people
would have been quite happy with a 20W power amp, but by the 1980s expected
rather more, and for it to deliver more current into awkward loads as
speakers got less efficient and speaker designers threw their problems over
the fence for amp designers to pick up. :-)

Also some designers and designs dealt with issues like stabilty or
delivering high currents better than others.

So circa 1970 you might expect to see amps that could deliver 40 watts per
channel in the UK. But the details would have varied.

Buy a Quad 303 and you got decent heatsinks and fairly well limited design
that would survive most loads - but at the cost of current limiting at a
few amps.

Buy an Armstrong 621 and you'd have got much the same 'headline' power of
around 40 Watts. But the heatsinks were small, so not suited to large
continued currents. OTOH no specific current limiters (other than fuses) so
would give you higher peak currents into demanding loads than a 303.

Since real music tends to be occasional peaks, say, over a factor of 10
higher than the average, you can argue that for many it was a better
tradeoff. But you could certainly find others who be better off with the
303.

Both designs were stable OK.

By c1980 powers in the Uk might now have drifted up to more like, say, 70W
and the devices around were much better. FWIW using paired output devices
in parallel I managed over 200W per channel continuous (and 30A rms
continuous) by then in a design that was pretty robust. Doing that would
have been a nightmare a decade or more earlier as the transistors weren't
available. Also unconditionally stable.

Yet you could also buy an early Naim amp. The heatsinking was 'adequate'
but it wasn't unconditionally stable and was known (by the trade) to
oscillate into some loadings. And some blew up in use, almost surely for
that reason. However many people liked it provided they used it with the
speakers and cables that Naim 'approved' - i.e. sounded OK and avoided
instability.

So some designers had fixed matters like stability many years before
others. And - like things like heatsinks and current demand/limiting - some
judged the issues differently to others.

Yer pays yer money... :-)


Most of the money ended up in the heat sinks, power supply
transformers and filter capacitors. The former are continuing to
shrink as improved output devices improve their reliabiity at higher
operating temperatures and power supplies are making greater use of
switchmode technology.

I've always wondered why amplifiers never (at least AFAIK) embraced the
use of heat-pipe technology to move the heat to side or back of case
passive radiators.

TBH I never saw the point of these in a power amp. I simply went for big
heatsinks. With high powers in the past you tended to need paralleled
output devices anyway, which spread out the dissipation. And you needed a
lot of sinks to get rid of what pipes could have conveyed to them, anyway!


I've actually used a heatpipe cooler from a laptop for just this purpose
(although it needs a fan on the radiator, as that's how it was designed,
the fin area isn't big enough to passive) when I re-housed a basic
subwoofer amp from a nasty plastic housing (previously mounted on the
front of the driver's box, with connections on the back) into a re-used
bookshelf speaker box.

The Shuttles I use also employ heat pipes to get the CPU waste from the CPU
to a big sink and large fan. They let you get a lot of heat out of a small
volume. But a CPU is somewhat different to a power transistor.

BTW I disconnected the fan on the Shuttle I use in the audio system, and
use a SSD to get a setup that makes no mechanical noises. This is fine
provided I only have doing basic tasks like playing audio files.

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

"~misfit~" wrote in message
...


Thanks for elucidating Arny. Interesting what you say about Switch-Mode
Power Supplies (SMPSs). I've been dabbling in PC building on-and-off for a
decade and, when I last looked you could get SMPSs that were outputting
1.2 KW.

People like Crown and QSC have been building 5+ KW audio power amps with
SMPS for years.

I asked a friend, who owns a bunch of Perreaux gear and also builds his
own amps using toroids why he couldn't use a SMPS and he said "too noisy".
(A man of few words. LOL, I remember once we were in a discussion, a few
of us, and someone who didn't know him well asked him how he felt - He
said "With my hands". g )

Pretty much the typical kind of know-nothing, hate real progress, knee-jerk
response one expects from high end audiophiles.

Considering the price he was paying for big toroids (and that he was using
a few large caps anyway) I wondered why not just filter better and maybe
make a split-system, with the SMPS in a seperate box, perhaps on the
floor....

The people who build real power amps for real people have been using SMPS
for years. They were scooped up by people who did crazy things like move
their equipment around from place to place and worried about what it
weighed.

I suspect that the traditional power supplies were cheaper than SMPS up
until the past year or two.

Now adays you can't even buy a router or a portable MP3 player without a
SMPS wall wart. Solid evidence that they are just plain cheaper.

"~misfit~" wrote in message
...
Somewhere on teh intarwebs Arny Krueger wrote:
[snip]
As time marched on, fairly inexpensive and modest-sized output
devices with excellent SOA and reasonble high frequency resposnse
made it relatively inexpensive to build very high powered audio power
amplifiers.

Arny, would you care to give a rough time-frame on this? I have a few old
SS amps (and might grab a few more if I see them at bargain prices) but
would love to know roughly when, in your opinion, SS amps became stable
and fairly clean across the frequency scale. Thanks.

Late 60s-early 70s.

Most of the money ended up in the heat sinks, power
supply transformers and filter capacitors. The former are continuing
to shrink as improved output devices improve their reliabiity at
higher operating temperatures and power supplies are making greater
use of switchmode technology.

I've always wondered why amplifiers never (at least AFAIK) embraced the
use of heat-pipe technology to move the heat to side or back of case
passive radiators.

Easy enough to put the devices and the heat sinks where you want them. Heat
sinks often end up on the backs or sides of amp cases, so no need to move
the heat around.

I've actually used a heatpipe cooler from a laptop for just this purpose
(although it needs a fan on the radiator, as that's how it was designed,
the fin area isn't big enough to passive) when I re-housed a basic
subwoofer amp from a nasty plastic housing (previously mounted on the
front of the driver's box, with connections on the back) into a re-used
bookshelf speaker box.

Heatpipes are amazing things. I've a small collection of coolers from
laptops that I've put together to illustrate the impact heatpipes have had
on cooling and removing heat. They all use small fans and go from a
(relatively) massively-finned aluminium plate (when CPUs only had about
20W of heat to shed) to a slightly smaller ali plate incorporating a ~4mm
heatpipe in a groove. Then, as CPUs got very hot (at least Intel ones) I
have a heat-remover (I don't know a more accurate name for this one
really) that's got an ali 'girder' and *three* 4mm heatpipes ending with
two fans and radiators (I'm guessing for redundancy) and a third, separate
fan blowing out the side, from the time when Pentium 4s were wasting near
100W of heat. The next cooler, from merely 18 months later, still moving
the same amount of heat is a single 8mm heat-pipe 5cm long, no ali casting
at all between the heat source and the single variable-speed fan blown
copper radiator.

Laptops and PCs are specialized enough, low-powered and costly enough that
when there is a technical problem to solve, the money is there to use a
relatively expensive solution. Not so with power amps. Lots of heat to move
and less costly solutions work well enough.

Somewhere on teh intarwebs Arny Krueger wrote:
"~misfit~" wrote in message
...


Thanks for elucidating Arny. Interesting what you say about
Switch-Mode Power Supplies (SMPSs). I've been dabbling in PC
building on-and-off for a decade and, when I last looked you could
get SMPSs that were outputting 1.2 KW.

People like Crown and QSC have been building 5+ KW audio power amps
with SMPS for years.

Interesting to know.

I asked a friend, who owns a bunch of Perreaux gear and also builds
his own amps using toroids why he couldn't use a SMPS and he said
"too noisy". (A man of few words. LOL, I remember once we were in a
discussion, a few of us, and someone who didn't know him well asked
him how he felt - He said "With my hands". g )

Pretty much the typical kind of know-nothing, hate real progress,
knee-jerk response one expects from high end audiophiles.

Yep.

Considering the price he was paying for big toroids (and that he was
using a few large caps anyway) I wondered why not just filter better
and maybe make a split-system, with the SMPS in a seperate box,
perhaps on the floor....

The people who build real power amps for real people have been using
SMPS for years. They were scooped up by people who did crazy things
like move their equipment around from place to place and worried
about what it weighed.

I suspect that the traditional power supplies were cheaper than SMPS
up until the past year or two.

That was a big part of the reason I raised the question with my friend (who,
apart from being an audiophile, started working part-time, after school
designing and building burgalar alarms for a security system when he was 12
y/o, 38 years ago). That's why I ask him things, he's my 'go-to guy' for all
things electronic.

However, when he told me how much a toroidial transformer he'd just bought
cost I raised the question of using a SMPS instead as even high-end computer
ones (which was my field of experience at the time) were a fraction of the
cost.

Now adays you can't even buy a router or a portable MP3 player
without a SMPS wall wart. Solid evidence that they are just plain
cheaper.

Yup, I've noticed that too. They (the wall warts) also seem much more
efficient than the older ones.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

Somewhere on teh intarwebs Jim Lesurf wrote:
In article , ~misfit~
wrote:
Somewhere on teh intarwebs Arny Krueger wrote:


As time marched on, fairly inexpensive and modest-sized output
devices with excellent SOA and reasonble high frequency resposnse
made it relatively inexpensive to build very high powered audio
power amplifiers. Most of the money ended up in the heat sinks,
power supply transformers and filter capacitors. The former are
continuing to shrink as improved output devices improve their
reliabiity at higher operating temperatures and power supplies are
making greater use of switchmode technology. As power device speed
continues to improve, switchmode amplifier circuits operate at
higher and higher frequencies, which in turn makes filtering the
switching noise out of their outputs a more economic and effective
operation.

Thanks for elucidating Arny. Interesting what you say about
Switch-Mode Power Supplies (SMPSs). I've been dabbling in PC
building on-and-off for a decade and, when I last looked you could
get SMPSs that were outputting 1.2 KW. I asked a friend, who owns a
bunch of Perreaux gear and also builds his own amps using toroids
why he couldn't use a SMPS and he said "too noisy".

Considering the price he was paying for big toroids (and that he was
using a few large caps anyway) I wondered why not just filter better
and maybe make a split-system, with the SMPS in a seperate box,
perhaps on the floor....

I think this starts off with a situation where a 'new' approach simply
isn't up to the task and then slowly glides into one where people use
whichever approach they are more comfortable with.

So I tend to still prefer to design/think in terms of 'traditional'
transformer/rectifier/caps PSUs as I know how to get what I'd want
that way, and what the snags are.

However in terms of RF hash I do tend to feel there is a 'prevention
is better than cure' argument. Why put up with having to filter and
suppress it if an alternative that works well doesn't generate it.
But this may just be me being old fashioned. :-) I can see the
advantages of SM PSUs in terms of being efficient, small, etc.

Slainte,

Jim

Heh! Yeah, I see your point Jim. However, as Arny just said (essentially),
it seems to be the way of the future.

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

Somewhere on teh intarwebs Arny Krueger wrote:
"~misfit~" wrote in message
...
Somewhere on teh intarwebs Arny Krueger wrote:
[snip]
As time marched on, fairly inexpensive and modest-sized output
devices with excellent SOA and reasonble high frequency resposnse
made it relatively inexpensive to build very high powered audio
power amplifiers.

Arny, would you care to give a rough time-frame on this? I have a
few old SS amps (and might grab a few more if I see them at bargain
prices) but would love to know roughly when, in your opinion, SS
amps became stable and fairly clean across the frequency scale.
Thanks.

Late 60s-early 70s.


Cheers.

Most of the money ended up in the heat sinks, power
supply transformers and filter capacitors. The former are
continuing to shrink as improved output devices improve their
reliabiity at higher operating temperatures and power supplies are
making greater use of switchmode technology.

I've always wondered why amplifiers never (at least AFAIK) embraced
the use of heat-pipe technology to move the heat to side or back of
case passive radiators.

Easy enough to put the devices and the heat sinks where you want
them. Heat sinks often end up on the backs or sides of amp cases, so
no need to move the heat around.

Yet, that said, most of my amplifiers have bloody great heatsinks in the
middle of the chassis, with perforated top and botom covers. I know that
higher-end stuff quite often has heatsink fins on the back, and rack-mount
comercial [and Perreaux] amps have them down the sides. However most of my
stuff is 80s - 90s Japanese and all the heat is in the middle, of a
component that is often in a stack. I see that the newer stuff's not much
different. Heat-pipe tech is well-known and inexpensive now, it just seems
to make sense to me to use it to put the heat to the outside. shrug

I've actually used a heatpipe cooler from a laptop for just this
purpose (although it needs a fan on the radiator, as that's how it
was designed, the fin area isn't big enough to passive) when I
re-housed a basic subwoofer amp from a nasty plastic housing
(previously mounted on the front of the driver's box, with
connections on the back) into a re-used bookshelf speaker box.

Heatpipes are amazing things. I've a small collection of coolers from
laptops that I've put together to illustrate the impact heatpipes
have had on cooling and removing heat. They all use small fans and
go from a (relatively) massively-finned aluminium plate (when CPUs
only had about 20W of heat to shed) to a slightly smaller ali plate
incorporating a ~4mm heatpipe in a groove. Then, as CPUs got very
hot (at least Intel ones) I have a heat-remover (I don't know a more
accurate name for this one really) that's got an ali 'girder' and
*three* 4mm heatpipes ending with two fans and radiators (I'm
guessing for redundancy) and a third, separate fan blowing out the
side, from the time when Pentium 4s were wasting near 100W of heat.
The next cooler, from merely 18 months later, still moving the same
amount of heat is a single 8mm heat-pipe 5cm long, no ali casting at
all between the heat source and the single variable-speed fan blown
copper radiator.

Laptops and PCs are specialized enough, low-powered and costly enough
that when there is a technical problem to solve, the money is there
to use a relatively expensive solution. Not so with power amps. Lots
of heat to move and less costly solutions work well enough.

Yep. See above. IMO it would really benefit a component that's often stacked
to *not* rely on convection through the middle to remove heat. Oh well, I'm
not an amplifier designer in a company with the wherewithall to utilise the
idea. ;-)

Cheers,
--
Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)

"~misfit~" wrote in message
...

However, when he told me how much a toroidial transformer he'd just bought
cost I raised the question of using a SMPS instead as even high-end
computer ones (which was my field of experience at the time) were a
fraction of the cost.

The toroids he was talking about are no doubt boutique items.

People like Behringer are putting toroids into $400 2500 watt (2 ohm) power
amps!

"Arny Krueger" = Compewter Geek Know All


People like Crown and QSC have been building 5+ KW audio power amps with
SMPS for years.

** But they are not regulated PSUs - but simple square wave inverter
supplies that have no voltage regulation and output 50/60 Hz ripple.

The availability of rugged IGBTs has made the difference in terms of
reliability.

The only exceptions to this in audio amplifiers are the tiny few examples
with " active PFC corrected " supplies that regulate and deliver pure DC.


I suspect that the traditional power supplies were cheaper than SMPS up
until the past year or two.


** Still cheaper in fact, real watt for real watt - long as the
manufacture is in China.


Now adays you can't even buy a router or a portable MP3 player without a
SMPS wall wart. Solid evidence that they are just plain cheaper.


** Arguing from one example to another is a classic blunder in thinking.

SMPS adaptors are only cheaper when a REGULATED DC supply of 10 watts or
more is needed.

When an unregulated DC supply or an AC ( low voltage) supply is needed -
iron transformer based types are still cheaper.



..... Phil

"Phil Allison" wrote in message
...

"Arny Krueger" = Compewter Geek Know All


People like Crown and QSC have been building 5+ KW audio power amps with
SMPS for years.

** But they are not regulated PSUs - but simple square wave inverter
supplies that have no voltage regulation and output 50/60 Hz ripple.

Not necessarily true. I have before me the QSC service manual for their
Powerlight series of power amps with SMPS. Referring to the power supply
schematic on page 80:

(1) The IRGPC50U IGBT power amplifiers in the amp's power supply that
convert line voltage to 100 KHz square waves derive their power from
filtered DC from a 40 amp full wave bridge rectifier.

(2) The IRGPC50U IGBT power amplifiers in the amp's power supply are
controlled by a SG2535 PWM that provides voltage regulation, etc.

"~misfit~" wrote in message
...

Yep. See above. IMO it would really benefit a component that's often
stacked to *not* rely on convection through the middle to remove heat. Oh
well, I'm not an amplifier designer in a company with the wherewithall to
utilise the idea. ;-)

This requirement is usually met by simply pulling air from the front, and
exhausting it out the back. The sides and bottom of the cases provide no
vents.