Playmaster amps - anyone own one?
Geoff
I hate Jaycar, it has so much cool stuff :)
I was talking to the guy about their Playmaster amp, and the catalogue gives
it a pretty good rap. The specs on paper look VERY impressive, of course
it's a power amp not an integrated amp so the pre section isn't included.
Does anyone own one and have comments to make about its sound compared to
other amps you have previously or currently own?
I mean it's pretty insane... 180W per channel. But it would be fun to make
it and I doubt many people ever get the clipping lights to come on!!
Geoff
Richard Goldsmith
again (250w) and IME it can only improve matters, better bass and imaging.
And you are less likely to fry your tweeters.
I too am curious about that kit, because like you, I think it looks good on
paper. But it would be nice to hear one.
Geoff <spamif...@cyberjunkie.com> wrote in message
news:4DV77.1540$t9....@news1.belrs1.nsw.optushome.com.au...
Trevor Wilson
"Geoff" <spamif...@cyberjunkie.com> wrote in message
news:4DV77.1540$t9....@news1.belrs1.nsw.optushome.com.au...
>
> I hate Jaycar, it has so much cool stuff :)\
**You'd hate it even more, if you saw the boss's collection of cars
(Ferraris are a popular choice).
>
> I was talking to the guy about their Playmaster amp, and the catalogue
gives
> it a pretty good rap. The specs on paper look VERY impressive, of course
> it's a power amp not an integrated amp so the pre section isn't included.
>
> Does anyone own one and have comments to make about its sound compared to
> other amps you have previously or currently own?
**Playmaster amps were designed by Electronic Australia staff. Using the
specified components (sometimes Jaycar and Dick Smith substitute
sub-standard components), they perform competently enough. The designs,
however, are not cutting edge quality. We're not talking ME or Krell
standards, here, but they should easily equal any of the Asian products
presently on offer, from the mass marketers.
>
> I mean it's pretty insane... 180W per channel. But it would be fun to
make
> it and I doubt many people ever get the clipping lights to come on!!
**"Fun" is a relative term. If you have some electronics experience, then
yes, it can be. Ensure the components are EXACTLY as specified by the
publishers. Counterfeit components are very common. Ask for a written
guarantee that the transistors supplied, are the correct ones. Incorrect
components may cause serious damage (to both speakers and amplifier). Just
place a message on aus.electronics about counterfeit transistors and watch
the fireworks. Many of us (myself included) have been caught by this
disgusting practice.
--
Trevor Wilson
http://www.rageaudio.com.au
Geoff
> again (250w) and IME it can only improve matters, better bass and imaging.
> And you are less likely to fry your tweeters.
Well the more powerful it is the more headroom you have. Also the bias
point will probably be a bit higher which means you stay in class A for
longer.
More wattage means I am more likely to fry my tweeters since I am using a
6db/oct x-o, sure there is a lot less danger from clipping destroying the
tweeters but I don't drive my 15W RMS amp to clipping, if I had a 5W RMS amp
I wouldn't make it clip either. Apart from destroying tweeters it sounds
horrible :)
Geoff
Geoff
> (Ferraris are a popular choice).
He deserves it. Jaycar sells all the right stuff, the price is good
considering it's a retail shop, and the staff at the shop I go to have been
there for ages and they virtually know every single product code and if you
ask for something they lay their hands right on it no matter how obscure
(except for the one new guy, he's painful). As far as quality goes Jaycar
sells some stuff that's OK, on the budget end, and some stuff that's not
bad, on their higher end (like their Vifa P17WJ/D25AG and the JV60 kit).
High end audio is still a bit of a niche market though, I can understand why
they don't go there.
> **Playmaster amps were designed by Electronic Australia staff. Using the
> specified components (sometimes Jaycar and Dick Smith substitute
> sub-standard components), they perform competently enough. The designs,
> however, are not cutting edge quality. We're not talking ME or Krell
> standards, here, but they should easily equal any of the Asian products
> presently on offer, from the mass marketers.
What better quality kits would you recommend? If I was building my own amp
I'd be happy with 50W at the low end, and 100W would be a good upper bound
(of course the more the merrier, just have to be careful not to blow up my
speakers).
> **"Fun" is a relative term. If you have some electronics experience, then
> yes, it can be. Ensure the components are EXACTLY as specified by the
> publishers. Counterfeit components are very common. Ask for a written
> guarantee that the transistors supplied, are the correct ones. Incorrect
> components may cause serious damage (to both speakers and amplifier). Just
> place a message on aus.electronics about counterfeit transistors and watch
> the fireworks. Many of us (myself included) have been caught by this
> disgusting practice.
Grey market transistors? Damn, they're what like 50c-$1 each? I guess in
quantities of 1000 or more...... that sucks though.
Geoff
Trevor Wilson
> > **You'd hate it even more, if you saw the boss's collection of cars
> > (Ferraris are a popular choice).
>
> He deserves it. Jaycar sells all the right stuff, the price is good
> considering it's a retail shop, and the staff at the shop I go to have
been
> there for ages and they virtually know every single product code and if
you
> ask for something they lay their hands right on it no matter how obscure
> (except for the one new guy, he's painful). As far as quality goes Jaycar
> sells some stuff that's OK, on the budget end, and some stuff that's not
> bad, on their higher end (like their Vifa P17WJ/D25AG and the JV60 kit).
> High end audio is still a bit of a niche market though, I can understand
why
> they don't go there.
**Oh absolutely. Gary is one smart operator.
>
> > **Playmaster amps were designed by Electronic Australia staff. Using the
> > specified components (sometimes Jaycar and Dick Smith substitute
> > sub-standard components), they perform competently enough. The designs,
> > however, are not cutting edge quality. We're not talking ME or Krell
> > standards, here, but they should easily equal any of the Asian products
> > presently on offer, from the mass marketers.
>
> What better quality kits would you recommend? If I was building my own amp
> I'd be happy with 50W at the low end, and 100W would be a good upper bound
> (of course the more the merrier, just have to be careful not to blow up my
> speakers).
**I havn't had to service a kit, for quite some time (ETI 5000 series crap,
was the last), so I cannot make any specific suggestions. Just make sure the
parts are Kosher.
>
> > **"Fun" is a relative term. If you have some electronics experience,
then
> > yes, it can be. Ensure the components are EXACTLY as specified by the
> > publishers. Counterfeit components are very common. Ask for a written
> > guarantee that the transistors supplied, are the correct ones. Incorrect
> > components may cause serious damage (to both speakers and amplifier).
Just
> > place a message on aus.electronics about counterfeit transistors and
watch
> > the fireworks. Many of us (myself included) have been caught by this
> > disgusting practice.
>
> Grey market transistors? Damn, they're what like 50c-$1 each? I guess in
> quantities of 1000 or more...... that sucks though.
**It sure does. Last time I bought some, they looked just like the real
thing, except the Base and Emitter leads were slightly different diameter,
to what I was accustomed to (the average punter would never know).
Susupicius, I stuck them on my breakdown tested and found that instead of a
180 Volt breakdown, the ones I was supplied demonstrated a breakdown of
around 130 Volts. YIKES! In a big amp, with the right sort of load
conditions, the damage would have been substantial.
These were output transistors and cost around $12.00 each. The counterfeits
were around $2.00 each. There is a significant temptation to rip off the
consumer accordingly. Both Dick Smith and Jaycar have been found to
distribute these counterfeits. I am not suggesting that they do so
deliberately. It's just that great pressure is placed on profitability and
they buy at the lowest price they can. Sometimes they do not check the
pedigree of the source. Common sense would dictate that their internal QC
would batch check smaples of transistors. Sadly, common sense doesn't seem
to be a valued commodity.
Paul R. Hunt
<spamif...@cyberjunkie.com> wrote:
>I was talking to the guy about their Playmaster amp, and the catalogue gives
>it a pretty good rap. The specs on paper look VERY impressive, of course
>it's a power amp not an integrated amp so the pre section isn't included.
Geoff, if you don't mind going completely DIY (i.e. PCB supplied
only), check out Anthony Holton's site
http://www.aussieamplifiers.com/
I haven't built any of his amps, but his site is well worth a look,
and he sells PCB for several of his designs: MOSFET power amps in the
200W/8ohm range.
Paul.
DeutschMick
>Grey market transistors? Damn, they're what like 50c-$1 each? I guess in
>quantities of 1000 or more...... that sucks though
Quality Trannies say Motorola MJL21193/4 can retail for $10,
depending on where you shop.
Fake ones may cost a few cents each to import, n-i-c-e margin :-(
Luckily, there's plenty of information, and buying through an
authorised channel helps.
Regards: Mick Deutsch, Project Audio, Canberra, AUSTRALIA.
/
/ - __0 m$ts...@pcug.org.zzz, mick.$ts...@hic.gov.au
[@\ _ \<,_ Searching for great sound! NO DAVE! I'm not
[@/ (_) / (_) Wind1900/XP compatible, this is the 21st century!
\ replace z's with au, currency symbol with deu
\
Geoff
"Paul R. Hunt" <prh...@dyson.brisnet.org.au> wrote in message
news:pdu1mtcr9mvpapbtk...@4ax.com...
Michael Seddon
on the AKSA kitset and the VERY soon to be released 100W AKSA which might
offer the cheap power with class that i need on my limited student budget.
Michael
Trevor Wilson <tre...@rageaudio.com.au> wrote in message
news:Zj587.65015$Xr6.2...@news-server.bigpond.net.au...
Patrick Turner
Trevor Wilson wrote:
> "Geoff" <spamif...@cyberjunkie.com> wrote in message
> news:4DV77.1540$t9....@news1.belrs1.nsw.optushome.com.au...
> > Hi,
> >
> > I hate Jaycar, it has so much cool stuff :)\
>
> **You'd hate it even more, if you saw the boss's collection of cars
> (Ferraris are a popular choice).
Trevor, you ARE a bloomin stirer.
> >
> > I was talking to the guy about their Playmaster amp, and the catalogue
> gives
> > it a pretty good rap. The specs on paper look VERY impressive, of course
> > it's a power amp not an integrated amp so the pre section isn't included.
> >
> > Does anyone own one and have comments to make about its sound compared to
> > other amps you have previously or currently own?
>
> **Playmaster amps were designed by Electronic Australia staff. Using the
> specified components (sometimes Jaycar and Dick Smith substitute
> sub-standard components), they perform competently enough. The designs,
> however, are not cutting edge quality. We're not talking ME or Krell
> standards, here, but they should easily equal any of the Asian products
> presently on offer, from the mass marketers.
>
> >
> > I mean it's pretty insane... 180W per channel. But it would be fun to
> make
> > it and I doubt many people ever get the clipping lights to come on!!
>
> **"Fun" is a relative term. If you have some electronics experience, then
> yes, it can be. Ensure the components are EXACTLY as specified by the
> publishers. Counterfeit components are very common. Ask for a written
> guarantee that the transistors supplied, are the correct ones. Incorrect
> components may cause serious damage (to both speakers and amplifier). Just
> place a message on aus.electronics about counterfeit transistors and watch
> the fireworks. Many of us (myself included) have been caught by this
> disgusting practice.
Yes, I have found TO92 fets that really were garden variety transistors,selling
for $5. Whoever it is in asia who does this swapsie skulduggery must
be laughing their heads off. Power devices have to be watched.
Patrick Turner.
Stephen Robertson
Amps. The amps I made that used the BF469/470 driver transistors
actualy all caught on fire after the 469/470 blew (did the designers
check the soar curves)
The amps that use mje340/350 are very reliable. Very hard to blow
the output mosfets (i've had them in Guitar/PA amps runing at 110%
power level for hours at a time) (I found the EAM6500 to be unstable however)
Jaycar/Dick Smith used to sell the amp module in the Jacar Playmaster180W
separately. I have one as a sub amp, its power is overkill for home use but
better too much than too little.(Ive also used it as a PA amp, very robust)
The basic design goes back over 15 years so dont expect hi-tech.
Geoff
> separately. I have one as a sub amp, its power is overkill for home use
but
> better too much than too little.(Ive also used it as a PA amp, very
robust)
They sell a 100W ultra low distortion module, I haven't checked the details
thoroughly though.
> The basic design goes back over 15 years so dont expect hi-tech.
My current amp is 26 years old, and if something 26 years old can sound good
then I don't care about hi-tech, only hi-fi :)
Geoff
Trevor Wilson
"Michael Seddon" <ref...@refused.com> wrote in message
news:bZ887.134$Ee....@news1.rdc1.nsw.optushome.com.au...
> Trevor, i would just like your opinion, or anyone elses opinion of course,
> on the AKSA kitset and the VERY soon to be released 100W AKSA which might
> offer the cheap power with class that i need on my limited student budget.
>
**I have yet to examine one of these products, but it is not an easy one,
for an amateur, because (AFAIK) no COMPLETE kit of parts exists. I would
imagine that only experienced kit builders should tackle it. Having said
that, all reports are that it is a very fine sounding product.
Grant Sellek
I agree with Trevor. AKSA is not a kit, in the sense of the Jaycar.
ASKA is a pcb with components for the pcb, and a heatsink. That leaves
an awful lot for the inexperienced builder to do (and parts to find)
-- chassis, switches, internal wiring, chassis layout design, earthing
layout, ...... the inexperienced can make some poor decisions with all
kinds of consequences for the finished item.
Grant Sellek
Adelaide
Andrew Reilly
> My current amp is 26 years old, and if something 26 years old can sound good
> then I don't care about hi-tech, only hi-fi :)
My second piece of audio gear was a Playmaster 60/60 kit. It's
not the most powerful amp in the world, and twelve years or so
later the carbon pots are all scratchy, but it still sounds
pretty nice. Nice enough that non-audio friends comment on it.
(Well, it's driving some very nice speakers now, and is being
helped along by a powered sub, but still...)
I probably wouldn't build one again---the saving just isn't
there---but as an electrical engineering student, it's just
something that you have to get out of your system... :-)
--
Andrew
Trevor
>
> "Geoff" <spamif...@cyberjunkie.com> wrote in message
> news:4DV77.1540$t9....@news1.belrs1.nsw.optushome.com.au...
> gives
> > it a pretty good rap. The specs on paper look VERY impressive, of
course
> > it's a power amp not an integrated amp so the pre section isn't
included.
> >
> > Does anyone own one and have comments to make about its sound compared
to
> > other amps you have previously or currently own?
Fine amplifier, you might also consider the newer 100W ULD amp by SC,
transistor rather than MOSFET.
> **Playmaster amps were designed by Electronic Australia staff. Using the
> specified components (sometimes Jaycar and Dick Smith substitute
> sub-standard components), they perform competently enough. The designs,
> however, are not cutting edge quality. We're not talking ME or Krell
> standards, here, but they should easily equal any of the Asian products
> presently on offer, from the mass marketers.
Of course ME and Krell don't sell anything at that price point, and I agree
it will probably beat any thing else at the same price. You do have to build
it though.
> > I mean it's pretty insane... 180W per channel. But it would be fun to
> make
> > it and I doubt many people ever get the clipping lights to come on!!
180 Watts is only a couple of dB more than the more "normal" 60-100 Watt
amps. Getting the clipping LED's to come on is easy, but it is sufficient
for most purposes.
> **"Fun" is a relative term. If you have some electronics experience, then
> yes, it can be. Ensure the components are EXACTLY as specified by the
> publishers. Counterfeit components are very common. Ask for a written
> guarantee that the transistors supplied, are the correct ones. Incorrect
> components may cause serious damage (to both speakers and amplifier). Just
> place a message on aus.electronics about counterfeit transistors and watch
> the fireworks. Many of us (myself included) have been caught by this
> disgusting practice.
Jaycar quote "Genuine Hitachi MosFET devices" in their catalog, so that can
be considered a Guarantee. I would be more worried about MJ15003/4's etc.
that appear to be rife with substitution.
Trevor.
Trevor
"Stephen Robertson" <ste...@nznet.gen.nz> wrote in message
news:9jr84r$gdr$1...@aklobs.kc.net.nz...
> Ive built a few of the ETI, EA,EAM, Silicon Chip, type Hitachi
> Amps. The amps I made that used the BF469/470 driver transistors
> actualy all caught on fire after the 469/470 blew (did the designers
> check the soar curves)
Never had problem's myself, because I always used MJE 340/350's. The price
is essentially the same.
> The amps that use mje340/350 are very reliable. Very hard to blow
> the output mosfets (i've had them in Guitar/PA amps runing at 110%
> power level for hours at a time) (I found the EAM6500 to be unstable
however)
Not when modified with appropriate gate resisors and capacitors. The PC
boards were wrong as well.
> The basic design goes back over 15 years so dont expect hi-tech.
The S/N ratio is the worst feature when used for HiFi IMO. 100 dB WRT 100W
is not so marvelous when playing at an average 1W.
The Silicon Chip ULD is spec'd at 117 dB WRT 100W. I haven't measured one of
those, but it does use a regulated input stage, so I would expect it to be
better.
Trevor.
Stephen Robertson
>They sell a 100W ultra low distortion module, I haven't checked the details
>thoroughly though.
>
The low distortion may be due to the amount negative feedback applied, rather
than being a low dist design??
Geoff
> transistor rather than MOSFET.
What are the different technical and sonic characteristics of BJTs and
MOSFETs? I don't mean the basic operation, I know how they work. But for
example the amplifier book I bought was obviously written when MOSFETs were
a novelty, and it raves about how they have a negative temperature
coefficient (helps avoid thermal runaway), and how this means the current
channel is spread evenly rather than concentrated in a hotspot as with BJTs.
They also said the MOSFET is much faster in its response because it's a
majority carrier rather than minority as with BJTs. They do admit some
limitations as far as circuit design does, but the disadvantages section
begins "Basically, there aren't any" and then spends 6 paragraphs discussing
the limitations... So I'll take all that with a grain of salt thanks (and
they admit that you can design around the BJT's "problem" of thermal runaway
anyway, you just need to operate your transistor well within its rating).
They also admit that BJTs can have 1/3 the distortion of a MOSFET when used
as a follower, in some circumstances... So what are the advantages and
disadvantages of each? And by which I mean the more important ones (ie
things that affect the sound!). If you can get around issues like thermal
runaway by using a bigger transistor, well it may be worth mentioning but
it's hardly a major issue because you can easily design around it and avoid
the problem.
I know we're straying a bit from the usual discussion of "should I buy a
rotel or a marantz CD player" but I find technical discussions more
interesting ;)
Geoff
Geoff
of
> those, but it does use a regulated input stage, so I would expect it to be
> better.
I was interested to read that in JC's catalogue. Am I correct in my
understanding that the pre stage benefits from regulation because the supply
is much more stable (ie ripple is more easily avoided with an LM317 than
capacitors, assuming current draw is low?) but regulators can't keep up with
large transient changes in current draw, which is why capacitors are still
used to store energy for the output stage?
The catalogue quotes SNR as 117dB unweighted from 20-20kHz, and 123dB
A-weighted. I'm not sure which one really matters, but most mfgr's probably
state the A-weighted since it's higher :)
Geoff
Trevor Wilson
> > Fine amplifier, you might also consider the newer 100W ULD amp by SC,
> > transistor rather than MOSFET.
>
> What are the different technical and sonic characteristics of BJTs and
> MOSFETs? I don't mean the basic operation, I know how they work. But for
> example the amplifier book I bought was obviously written when MOSFETs
were
> a novelty, and it raves about how they have a negative temperature
> coefficient (helps avoid thermal runaway), and how this means the current
> channel is spread evenly rather than concentrated in a hotspot as with
BJTs.
**All quite true.
Fortunately (for BJTs) avoiding thermal runaway is easy enough. It only
requires one of two transistors placed in close proximity to the output
devices. Hotspots are not much of an issue, either.
Unfortunately (for MOSFETs) the negative tempco of gm is a (minor) strength
and a major weakness. In a standard Class A/B MOSFET amp, this
characteristic tends to 'choke off' transients. As a result most MOSFET
designs tend to sound as though they are lacking in dynamics.
> They also said the MOSFET is much faster in its response because it's a
> majority carrier rather than minority as with BJTs.
**Absolutely. A very useful aspect, if you're building amplifiers which are
required in the MHz region. Fortunately, BJTs work very well up to 200 or
300kHz or so, with no problem. BJTs with this capability have been easily
available for a couple of decades.
They do admit some
> limitations as far as circuit design does, but the disadvantages section
> begins "Basically, there aren't any" and then spends 6 paragraphs
discussing
> the limitations... So I'll take all that with a grain of salt thanks (and
> they admit that you can design around the BJT's "problem" of thermal
runaway
> anyway, you just need to operate your transistor well within its rating).
**Of course. Any competent engineer knows these limitations and works within
them.
>
> They also admit that BJTs can have 1/3 the distortion of a MOSFET when
used
> as a follower, in some circumstances...
**MOre like 1/10th, actually. For MOSFETs to demonstrate similar levels of
distortion, as BJTs operating at (say) 50mA, the MOSFETs would need to
operate at around 2 Amps, or so. IOW: Almost pure Class A.
So what are the advantages and
> disadvantages of each? And by which I mean the more important ones (ie
> things that affect the sound!). If you can get around issues like thermal
> runaway by using a bigger transistor, well it may be worth mentioning but
> it's hardly a major issue because you can easily design around it and
avoid
> the problem.
**MOSFETs are not available in perfectly symetrical pairs. BJTs are.
MOSFETs (in complementary pairs) cost several times that of equivalent BJTs.
If MOSFETs are run in almost pure Class A, they may offer some measureable
improvments over BJTs. At anything less than very high bias currents,
though, BJTs sound better and can deliver significantly more current, with a
more realistic level of dynamics.
>
> I know we're straying a bit from the usual discussion of "should I buy a
> rotel or a marantz CD player" but I find technical discussions more
> interesting ;)
**Me too.
Patrick Turner
mosfet amps and experienced no great difficulties,
produced very clear sound, and unless folks are told,
nobody can tell the difference between this and large
high powered tube amps operating in class A.
I have recently began work on 50 watt class A mosfet amps,
which seem also to do rather well with music.
You don't need as much NFB when you use class A.
Many bipolar designs I have heard do not have the neutrality
that I cherish. I prefer to not hear the amp, just the music.
I do like the character of mosfets, they are like genetically altered pentode
tubes
which have huge powerful muscles, and allowed themselves to be crammed into
a small package.
Their only disadvantage is that the required drive voltage changes with the
load,
but I prefer this to having the current drive changeing with the load.
The drive circuits for mosfet amps can be extremely simple.
As for sound differences, well, I hear so many claims upon claims,
and I refuse to comment.
The amps I build don't have much distortion, that is what I can say.
If you don't like mosfets because they are not so good as BJT's in follower
mode, then use 3 times the number, you won't blow them up,
as the heat dissipation is spread over more devices.
Remember that when follower designs are used, it is an example of an
application of negative feedback, and a lot of it, I might add.
Whether you use follower topology in output stages or not,
and how much global NFB is used, will affect the amp.
But steer clear of any design which produces crossover distortion,
and does not have adequately filtered rail supplies, especially
when the amp is working hard. When BJT designs get hot,
their thermal sensing circuits can work to under bias the
output devices, so as these will not have any gain in this condition
there is not an optimal benefit from NFB action.
What you should remember is that commercial amp makers
are more inclined to use BJTs because higher total amounts of NFB
can be used to clean up the mess that is the open loop response
with no NFB.
But with DIY, you could consider the more earnest approach of
building from first principles, making power supplies solid,
regulated, and get the thing to work without applying any NFB.
This is easier said than done, unless you build a tube circuit.
At the moment I like class A circuits, and why not, for DIY?
Why follow all the bloomin bean counters at Sony, Yamaha, Pioneer, etc,
who say that class A is a wank, and an expensive one at that?
If more power is needed, then class AB is a good option,
with a substantial part of operation in class A.
One must then stop wingeing, and buy decent heatsinks,
and plenty of devices.
It ain't what you got,
it's the way you use it
that counts.
Patrick Turner
Geoff
strength
> and a major weakness. In a standard Class A/B MOSFET amp, this
> characteristic tends to 'choke off' transients. As a result most MOSFET
> designs tend to sound as though they are lacking in dynamics.
I'm fuzzy on the reason behind this, the version of logic that exists in my
mind suggests the reason why this happens is because when a transient hits,
current tries to flow which heats up the junction which then increases in
resistance which reduces the peak. In a BJT current would try to flow and
concentrate in a particular pathway which heats it up more, but the -ve
coefficient means that current can flow more freely which helps "boost" the
transient slightly (which would compensate for the overall resistance of the
wiring in the amplifier trying to reduce the transients).
I'm pulling this all out the proverbial so let me know when I'm way off the
mark :)
> > They also said the MOSFET is much faster in its response because it's a
> > majority carrier rather than minority as with BJTs.
>
> **Absolutely. A very useful aspect, if you're building amplifiers which
are
> required in the MHz region. Fortunately, BJTs work very well up to 200 or
> 300kHz or so, with no problem. BJTs with this capability have been easily
> available for a couple of decades.
I assume by "fast" they're talking about a high slew rate. Why does a
"fast" MOSFET then have a poor transient response? Thinking about things on
the most basic level one would expect a slow device to be poor at
HF/transients and a fast device to be good at both. Unless of course there
is a major difference between a high frequency signal, and a transient.
> > They also admit that BJTs can have 1/3 the distortion of a MOSFET when
> used
> > as a follower, in some circumstances...
>
> **MOre like 1/10th, actually. For MOSFETs to demonstrate similar levels of
> distortion, as BJTs operating at (say) 50mA, the MOSFETs would need to
> operate at around 2 Amps, or so. IOW: Almost pure Class A.
I did notice that a number of the MOSFET designs in this book were class A
designs. Why do MOSFETs have a higher distortion? Is it due to a poorer
ability to switch on and off cleanly?
Geoff
Patrick Turner
Geoff wrote:
> > Unfortunately (for MOSFETs) the negative tempco of gm is a (minor)
> strength
> > and a major weakness. In a standard Class A/B MOSFET amp, this
> > characteristic tends to 'choke off' transients. As a result most MOSFET
> > designs tend to sound as though they are lacking in dynamics.
>
> I'm fuzzy on the reason behind this, the version of logic that exists in my
> mind suggests the reason why this happens is because when a transient hits,
> current tries to flow which heats up the junction which then increases in
> resistance which reduces the peak. In a BJT current would try to flow and
> concentrate in a particular pathway which heats it up more, but the -ve
> coefficient means that current can flow more freely which helps "boost" the
> transient slightly (which would compensate for the overall resistance of the
> wiring in the amplifier trying to reduce the transients).
>
> I'm pulling this all out the proverbial so let me know when I'm way off the
> mark :)
Er, about now. Try building some amplifiers, and all becomes clear.It really
won't unless you do.
> > > They also said the MOSFET is much faster in its response because it's a
> > > majority carrier rather than minority as with BJTs.
> >
> > **Absolutely. A very useful aspect, if you're building amplifiers which
> are
> > required in the MHz region. Fortunately, BJTs work very well up to 200 or
> > 300kHz or so, with no problem. BJTs with this capability have been easily
> > available for a couple of decades.
>
> I assume by "fast" they're talking about a high slew rate. Why does a
> "fast" MOSFET then have a poor transient response? Thinking about things on
> the most basic level one would expect a slow device to be poor at
> HF/transients and a fast device to be good at both. Unless of course there
> is a major difference between a high frequency signal, and a transient.
This business about transients is somewhat misleading.Square wave performance of
either tube, BJT or mosfets can be
as good as you want it to be.
> > > They also admit that BJTs can have 1/3 the distortion of a MOSFET when
> > used
> > > as a follower, in some circumstances...
> >
> > **MOre like 1/10th, actually. For MOSFETs to demonstrate similar levels of
> > distortion, as BJTs operating at (say) 50mA, the MOSFETs would need to
> > operate at around 2 Amps, or so. IOW: Almost pure Class A.
>
> I did notice that a number of the MOSFET designs in this book were class A
> designs. Why do MOSFETs have a higher distortion? Is it due to a poorer
> ability to switch on and off cleanly?
The mosfet has an initial logarithmic turn on characteristic.So what.
When you build a single ended circuit using just one power mosfet,
and another using a single transistor, and yet another using a triode,
and then a pentode,
the characteristics become meaningful.
Splendid amps can be made from what ever devices you choose to use.
BJTs have the highest current ability per device.
But say you want 100 watts into 8 ohms.
then you gotta have 28 V rms applied to the load,
this is only 3.375 amps rms.
Even old Hitachi 2SK 176 can easiliy provide
a peak current of 8 amps, and are suitable to use.
There are now much more current able mosfets about.
I have built 100 watt amps with two mosfets which were just as
"dynamic" as any other amp.
Patrick Turner.
> Geoff
Geoff
> and I refuse to comment.
:)
Probably the best way to avoid a flame war is to reply with "n/c" since
different people hear different things and have different biasses. Pity I'm
still a student or I'd build one of each and compare them.
> If you don't like mosfets because they are not so good as BJT's in
follower
> mode, then use 3 times the number, you won't blow them up,
> as the heat dissipation is spread over more devices.
I don't know whether or not I prefer BJTs or MOSFETs, I would have to hear
similarly designed examples of each before I could make a decision.
Theoretically speaking each one has something different to offer so I guess
it depends on how you use them as to how they will sound.
> Remember that when follower designs are used, it is an example of an
> application of negative feedback, and a lot of it, I might add.
I need to study amplifier design more to understand these things :)
Transistors have always confused me a little. Their basic operation is
obvious enough but designing a circuit around them is still a bit confusing,
it's been a while since I did any EE (and uni lectures always leave a lot to
be desired when it comes to explaining how stuff works)
> But with DIY, you could consider the more earnest approach of
> building from first principles, making power supplies solid,
> regulated, and get the thing to work without applying any NFB.
> This is easier said than done, unless you build a tube circuit.
I found an interesting (if above my head in a number of places) web page on
distortion in amplifiers, at http://www.dself.demon.co.uk/dipa.htm - would
agree with what this page says on making your amp as linear as possible
before applying NFB. But as for the rest of it... well I'll get back to you
when I understand it all :)
> At the moment I like class A circuits, and why not, for DIY?
> Why follow all the bloomin bean counters at Sony, Yamaha, Pioneer, etc,
> who say that class A is a wank, and an expensive one at that?
Class A does have a number of merits, the main "problem" being that it
consumes a large amount of power all the time and isn't very efficient. The
average consumer is more concerned about the power consumption and heat
issues, to their mind it's a waste of money if you mute the volume and the
amp is still using 150W of power. Plus, a cooler running amp is seemingly
better than one that feels hot like it's working really hard. Sound quality
doesn't really come into it for the average person :)
> If more power is needed, then class AB is a good option,
> with a substantial part of operation in class A.
> One must then stop wingeing, and buy decent heatsinks,
> and plenty of devices.
I like AB as a general idea. Under normal listening levels you can be
entirely in class A which means nice, low distortion figures. With a
flexible design you can probably pick your bias point according to
preference (turn it up in winter to keep the room warm hehe), so that you
only use class B when you really wind it up.
I strongly agree with the purchasing of multiple devices and large heatsinks
and sometimes cooling fans.
> It ain't what you got,
> it's the way you use it
> that counts.
So say lots of men with small p.......
haha. I agree though. Not much use in having something large and
impressive if you expect being large and impressive to speak for itself.
First impressions last, all of about 30 seconds!
I think I'm getting a bit OT though.
Geoff
Geoff
the
> > mark :)
>
> Er, about now. Try building some amplifiers, and all becomes clear.It
really
> won't unless you do.
I like to appreciate the theory behind things so I know why they should
work, and then the enjoyment of putting something together to prove that in
practice it does work. :)
Added benefit is theory is free, which is about all I can afford atm!
> Splendid amps can be made from what ever devices you choose to use.
Knowing how best to use devices is absolutely necessary though. If I tried
to build an amp from scratch knowing only what I do now I could probably do
an acceptable job but it would measure poorly compared with any decent
quality kit, even if the kit is just PCB + parts list.
Geoff
Trevor Wilson
> > Unfortunately (for MOSFETs) the negative tempco of gm is a (minor)
> strength
> > and a major weakness. In a standard Class A/B MOSFET amp, this
> > characteristic tends to 'choke off' transients. As a result most MOSFET
> > designs tend to sound as though they are lacking in dynamics.
>
> I'm fuzzy on the reason behind this, the version of logic that exists in
my
> mind suggests the reason why this happens is because when a transient
hits,
> current tries to flow which heats up the junction which then increases in
> resistance which reduces the peak. In a BJT current would try to flow and
> concentrate in a particular pathway which heats it up more, but the -ve
> coefficient means that current can flow more freely which helps "boost"
the
> transient slightly (which would compensate for the overall resistance of
the
> wiring in the amplifier trying to reduce the transients).
**BJTs possess a POSITIVE tempco, in relation to current flow. Heat 'em up
and they 'want' to pass more current, which heats 'em up, causing them to
'want' to pass more current. Etc. Transients tend to be slightly more
dynamic than real life as a result. MOSFETs tend to shut down, as they heat
up, causing a kind dynamic compression. The compression aspect of MOSFETs is
much more severe than the expansion aspect of BJTs.
>
> I'm pulling this all out the proverbial so let me know when I'm way off
the
> mark :)
**Except for the negative remark, pretty close.
>
> > > They also said the MOSFET is much faster in its response because it's
a
> > > majority carrier rather than minority as with BJTs.
> >
> > **Absolutely. A very useful aspect, if you're building amplifiers which
> are
> > required in the MHz region. Fortunately, BJTs work very well up to 200
or
> > 300kHz or so, with no problem. BJTs with this capability have been
easily
> > available for a couple of decades.
>
> I assume by "fast" they're talking about a high slew rate. Why does a
> "fast" MOSFET then have a poor transient response?
**Speed is one thing. The ability to handle asymetrical transients, is quite
another thing. In any case, the large amounts of Global NFB (absolutely
essential in Class A/B MOSFET designs) will take care of most measured,
repetitive signal situations. Pity music is not always the same as a bunch
of sine waves (or square waves).
Thinking about things on
> the most basic level one would expect a slow device to be poor at
> HF/transients and a fast device to be good at both. Unless of course
there
> is a major difference between a high frequency signal, and a transient.
**Not strictly speaking. Let's put aside speed, for the moment, since modern
BJTs are easily capable of handling ANY conceivable musical signal.
>
> > > They also admit that BJTs can have 1/3 the distortion of a MOSFET when
> > used
> > > as a follower, in some circumstances...
> >
> > **MOre like 1/10th, actually. For MOSFETs to demonstrate similar levels
of
> > distortion, as BJTs operating at (say) 50mA, the MOSFETs would need to
> > operate at around 2 Amps, or so. IOW: Almost pure Class A.
>
> I did notice that a number of the MOSFET designs in this book were class A
> designs. Why do MOSFETs have a higher distortion?
**Their 'knee', at low currents is much more pronounced than that of BJTs.
The 'knee' extends much farther up the current axis than does the 'knee' of
BJTs. It is the 'knee', or the kink in the amplification characteristic,
which causes the distortion.
Is it due to a poorer
> ability to switch on and off cleanly?
**No. The switching is fine. It is when they are called upon to amplify
linearly, which is where problems can occur.
Trevor Wilson
> > As for sound differences, well, I hear so many claims upon claims,
> > and I refuse to comment.
>
> :)
>
> Probably the best way to avoid a flame war is to reply with "n/c" since
> different people hear different things and have different biasses. Pity
I'm
> still a student or I'd build one of each and compare them.
**More the pitty, because, like all things in life, there are good MOSFET
designs and bad ones. Equally, there are good BJT designs and bad ones. A
bad BJT design will sound worse than a good MOSFET. In the ultimate sense,
however, for a Class A/B design, BJTs are well ahead. In fact, I can only
name one, highly rated Class A/B MOSFET design. All the rest are BJT ones.
**And this is where you can run into severe problems with MOSFETs. EACH
MOSFET pair must be heavily biased, in order to allow operation in the
linear region. One MOSFET pair = (say) 2 Amps bies, Two MOSFET pairs = 4
Amps and so on.
Geoff
Got my positives and negatives mixed up. I was thinking coefficient related
to resistance (resistance goes down -> negative) rather than related to
current (current goes up -> positive).
Geoff
> MOSFET pair must be heavily biased, in order to allow operation in the
> linear region. One MOSFET pair = (say) 2 Amps bies, Two MOSFET pairs = 4
> Amps and so on.
Maybe I should devlop the world's first actively cooled (ie, TEC/peltier)
class A mosfet amp..... it would be interesting, but given your TEC needs to
be rated at about double your load dissipation things could get pretty
toasty on the hot side.
Water cooled amp! excellent........ :)
Patrick Turner
Geoff wrote:
> > As for sound differences, well, I hear so many claims upon claims,
> > and I refuse to comment.
>
> :)
>
> Probably the best way to avoid a flame war is to reply with "n/c" since
> different people hear different things and have different biasses. Pity I'm
> still a student or I'd build one of each and compare them.
That is right, some folks enjoyment of hi-fi is different to others,some are
totally in the SE Triode and no NFB camp,
and some are all the way with BJT.
> > If you don't like mosfets because they are not so good as BJT's in
> follower
> > mode, then use 3 times the number, you won't blow them up,
> > as the heat dissipation is spread over more devices.
>
> I don't know whether or not I prefer BJTs or MOSFETs, I would have to hear
> similarly designed examples of each before I could make a decision.
The drive requirements for BJTs and Mosfets mean that designsare always going to
be different, and the measurements are
going to differ somewhat, and so the sound is going to change a bit.
On rec.audio.high-end news group these subjects are argued almost
continually to a deeper depth.
I think a recent conclusion by Howard Ferstler was that
if amps measure the same then there is no difference to the sound.
I dimly remember Peter Walker saying the same thing.
Such experts dismiss differences in cables, and offer large rewards
to people who can tell which cable is which in carefully
constructed double blind tests. Nobody has been paid yet.
I just enjoy music AND the technology, tubes, BJTs, mosfets,
the bloomin lot.
> Theoretically speaking each one has something different to offer so I guess
> it depends on how you use them as to how they will sound.
>
> > Remember that when follower designs are used, it is an example of an
> > application of negative feedback, and a lot of it, I might add.
>
> I need to study amplifier design more to understand these things :)
>
> Transistors have always confused me a little. Their basic operation is
> obvious enough but designing a circuit around them is still a bit confusing,
> it's been a while since I did any EE (and uni lectures always leave a lot to
> be desired when it comes to explaining how stuff works)
Well, you are not the only one to be a bit confused. Most peopleare, and trying
to learn about ohms law and all other simple three pronged equations
that you need to be familiar with to design anything that works can be daunting.
I am glad I learned a long time ago, BC, ( before I got computors ), so that
when I learned to share this knowledge I had some firm experience.
I have a lot of gear I have made myself, and I am still learning.
If you find that when you build a simple one transistor signal amplifier,
that you spend a week asking questions to yourself about why it is so,
then you are getting someplace. If you do not build stuff, and ask no questions,
you ought to go fishing.
> > But with DIY, you could consider the more earnest approach of
> > building from first principles, making power supplies solid,
> > regulated, and get the thing to work without applying any NFB.
> > This is easier said than done, unless you build a tube circuit.
>
> I found an interesting (if above my head in a number of places) web page on
> distortion in amplifiers, at http://www.dself.demon.co.uk/dipa.htm - would
> agree with what this page says on making your amp as linear as possible
> before applying NFB. But as for the rest of it... well I'll get back to you
> when I understand it all :)
>
> > At the moment I like class A circuits, and why not, for DIY?
> > Why follow all the bloomin bean counters at Sony, Yamaha, Pioneer, etc,
> > who say that class A is a wank, and an expensive one at that?
>
> Class A does have a number of merits, the main "problem" being that it
> consumes a large amount of power all the time and isn't very efficient. The
> average consumer is more concerned about the power consumption and heat
> issues, to their mind it's a waste of money if you mute the volume and the
> amp is still using 150W of power. Plus, a cooler running amp is seemingly
> better than one that feels hot like it's working really hard. Sound quality
> doesn't really come into it for the average person :)
Well, most folks think nothing of using power at the rate of 1 Kilowatt or more
perhour all day long, to sustain life in the suburbs. If you make a 50 watt
class A
amp, you will draw about 120 from the mains. Times 2 that is 240 watts.
But only for a couple of hours per day, and in liew of some other more
energy waste someplace else, So the cost of 100 watts of class A is not high.
Amp heat is due to bean counter mentality.
How many folks change their sound system every 7 years on average,
for some reason?
Divide the cost of that by the number of CDs they play,
and see how much it really costs to play a song or two.
Does any sailor in the Sydney to Hobart worry over costs
too much?
> > If more power is needed, then class AB is a good option,
> > with a substantial part of operation in class A.
> > One must then stop wingeing, and buy decent heatsinks,
> > and plenty of devices.
>
> I like AB as a general idea. Under normal listening levels you can be
> entirely in class A which means nice, low distortion figures. With a
> flexible design you can probably pick your bias point according to
> preference (turn it up in winter to keep the room warm hehe), so that you
> only use class B when you really wind it up.
Exactly. And there is a special significance here in Canberra ofusing amps to
keep warm on.
One has to worry over a smooth crossover between class A and class B.
> I strongly agree with the purchasing of multiple devices and large heatsinks
> and sometimes cooling fans.
Well designed amps at home should not require fans, even on a January day.And to
make 50 watts of class A you would need at least a heat sink
that would normally be found on a 300 watt class aB.
Most class aB solid amps are really not much over pure class B amps,
hence my unofficial use of class aB, rather than class AB, where
perhaps the first 50 out of a maximum of 100 watts is class A.
> > It ain't what you got,
> > it's the way you use it
> > that counts.
>
> So say lots of men with small p.......
>
> haha. I agree though. Not much use in having something large and
> impressive if you expect being large and impressive to speak for itself.
> First impressions last, all of about 30 seconds!
>
> I think I'm getting a bit OT though.
>
> Geoff
Regards, Patrick Turner.
Geoff
going to
> be different, and the measurements are
> going to differ somewhat, and so the sound is going to change a bit.
Sure, what I meant by similar design was that you've applied the same
general principles to the design, ie using a particular method for working
out how big your filter caps need to be, coupling things in a particular way
and so on. Generally designed by the one person with the one mindset trying
to achieve a similar result, so you minimise the number of differences
between the designs. Much like there's no point even comparing a Sony to a
Rotel and saying the sonic difference is due to one using a regulated power
supply and one uses large caps.
> I just enjoy music AND the technology, tubes, BJTs, mosfets,
> the bloomin lot.
As it should be.
Geoff
Patrick Turner
Trevor Wilson wrote:
> "Geoff" <spamif...@cyberjunkie.com> wrote in message
> news:jAR87.1774$t9....@news1.belrs1.nsw.optushome.com.au...
> > > As for sound differences, well, I hear so many claims upon claims,
> > > and I refuse to comment.
> >
> > :)
> >
> > Probably the best way to avoid a flame war is to reply with "n/c" since
> > different people hear different things and have different biasses. Pity
> I'm
> > still a student or I'd build one of each and compare them.
>
> > >
> > I like AB as a general idea. Under normal listening levels you can be
> > entirely in class A which means nice, low distortion figures. With a
> > flexible design you can probably pick your bias point according to
> > preference (turn it up in winter to keep the room warm hehe), so that you
> > only use class B when you really wind it up.
> >
> > I strongly agree with the purchasing of multiple devices and large
> heatsinks
> > and sometimes cooling fans.
>
> **And this is where you can run into severe problems with MOSFETs. EACH
> MOSFET pair must be heavily biased, in order to allow operation in the
> linear region. One MOSFET pair = (say) 2 Amps bies, Two MOSFET pairs = 4
> Amps and so on.
>
I think Trevor is grossly exagerating mosfet problems
to make BJTs look better and easier to use.
This business of having to use a minimum of 2 amp bias currents
is excessive for class AB operation.
I use about 80 Ma per device, and I get very low distortion.
In my 300 watt class AB amps I run this current, and I use six TOP3
flat pack mosfets per channel. So there is a reasonable amount of class A
before the amp reverts to class B.
There are no severe problems with mosfets, that an intelligent person cannot
get around.
The only time you would use 2 amps for a bias current is in a class A design,
with a low drain supply of 15 volts, to limit dissipation to about 30 watts.
I don't even use this high a current, an amp is plenty,
to swing a substantially linear transfer curve with a 35 ohm load
with a 35 volt supply.
Patrick Turner.
Patrick Turner
Geoff wrote:
If each result of ones efforts meant that there was little difference,then they
would either all be doing well with the signal,
or not, then all doing the same rotten thing, which hardly ever occurs.
All that is needed is that the power amp get across the sonic line by a wide
margin,
so that the other things present such as source and preamp, speakers,
and maybe cables will degrade the sound more.
There is a rule of diminishing returns going on here.
And then measurements don't explain all things.
Distortion measurements which are assessed simplistically are
not worth a cent.
1.0 % of thd caused by crossover distortion
will be a lot worse than 1.0 % of second harmonic distortion
at full power.
Patrick Turner.
Geoff
> Distortion measurements which are assessed simplistically are
> not worth a cent.
I haven't mentioned measurements yet, because I know that what you're about
to say next is true!
> 1.0 % of thd caused by crossover distortion
> will be a lot worse than 1.0 % of second harmonic distortion
> at full power.
...due to crossover distortion being present at significant levels for a
minority of the total measurement period, wheras low-order harmonics are
present for the whole cycle and can be inaudible or even desirable by some
(but either way being less offensive).
Patrick Turner
at
http://www.dself.demon.co.uk/dipa.htm
I really think Trevor should look at this page.
And so should anyone else with an eye to solid state
circuitry operation.
Douglas Self does not like Mosfets much.
But other designers who have contributed to
'Electronics, Wireless World' over the years
don't seem to mind them.
There are about 1,001 tips on circuits for transistors
to get distortion down to less than 0.01% at near full power
At least what Douglas Self is saying makes a lot of sense.
It won't be very palatable to the SET-without-NFB crowd.
Whether one of his designs using his circuit would be as good
sonically as he suggests it must be, is anyone's guess.
I would imagine there are printed circuit boards
available for one of Mr Self's Blameless Amplifiers.
You need to be ever so careful with solid state
circuit board layout if you are to come anywhere near
achieving low distortion, so in this case it would pay if you
were to email Douglas and ask him if he knows of a source
of boards which are ok. I am sure there would be one.
With this sort of amplifier, I doubt anyone could do better
for themselves unless they had lots of experience.
He states there are circuits for Trimodal, class B,
class AB and class A can be had by simply adjusting the bias.
He says, and I would agree, that class AB has less
distortion than class B whilst the amp stays in class A,
but more distortion when it goes above, which may not matter
if there is 50 watts of class A, and then a total of 150 of class AB.
Patrick Turner.
Trevor Wilson
"Patrick Turner" <in...@turneraudio.com.au> wrote in message
news:3B641596...@turneraudio.com.au...
**Not at all, Patrick. Just look at the curves. I've tried MOSFETs many
times, in many different configurations. Anything less than VERY HEAVY
biassing and they are a waste of time and effort. In ine case, I even
provided a client with two, otherwise identical power amps (300 Watt/ch).
One was an orignal GAS 500, using BJTs, in a rather uninspired topology. The
other was a high performance MOSFET circuit using 8 paralleled 2SK176/2SJ56
devices. The client returned the MOSFET amp, claiming that it had no
'grunt'. Yet it's output power was almost identical to that of the BJT
design. I even managed to obtain a frequency response out to 900kHz from the
thing. Very impressive. THD figures were commendably low and it could even
cope with 2 Ohm loads. Yet it still sounded like a typical Class A/B MOSFET
amp. Uninvolving and lacking in dynamics.
> This business of having to use a minimum of 2 amp bias currents
> is excessive for class AB operation.
> I use about 80 Ma per device, and I get very low distortion.
**By using Global NFB. Sure, that's easy enough to do.
> In my 300 watt class AB amps I run this current, and I use six TOP3
> flat pack mosfets per channel. So there is a reasonable amount of class A
> before the amp reverts to class B.
> There are no severe problems with mosfets, that an intelligent person
cannot
> get around.
**Yes, there are. Like it or not, the negative tempco of gm will suck the
life out of music, every time.
>
> The only time you would use 2 amps for a bias current is in a class A
design,
> with a low drain supply of 15 volts, to limit dissipation to about 30
watts.
> I don't even use this high a current, an amp is plenty,
> to swing a substantially linear transfer curve with a 35 ohm load
> with a 35 volt supply.
**Agreed.
Trevor Wilson
> I had a good look at Douglas Selfs web page on amplifiers
> at
> http://www.dself.demon.co.uk/dipa.htm
>
> I really think Trevor should look at this page.
> And so should anyone else with an eye to solid state
> circuitry operation.
**I agree, Patrick. I've read every word written by Self, ever written in
Eletronics World. What makes you think I haven't?
>
> Douglas Self does not like Mosfets much.
> But other designers who have contributed to
> 'Electronics, Wireless World' over the years
> don't seem to mind them.
**Sure. They don't much listen to music, thats all.
>
> There are about 1,001 tips on circuits for transistors
> to get distortion down to less than 0.01% at near full power
>
> At least what Douglas Self is saying makes a lot of sense.
**I agree with much that Self says. I alswo dissagree with a bunch of stuff,
too.
>
> It won't be very palatable to the SET-without-NFB crowd.
**And why should it be? Those sorts of amplifiers are highly coloured and
distorted. They are (generally) incapable of reproducing music with any sort
of fidelity. They sometimes sound 'nice' though. Of course, it ain't
reality, though.
>
> Whether one of his designs using his circuit would be as good
> sonically as he suggests it must be, is anyone's guess.
**I've built one. It was OK. Not an ME, of course, but competent enough.
>
> I would imagine there are printed circuit boards
> available for one of Mr Self's Blameless Amplifiers.
> You need to be ever so careful with solid state
> circuit board layout if you are to come anywhere near
> achieving low distortion, so in this case it would pay if you
> were to email Douglas and ask him if he knows of a source
> of boards which are ok. I am sure there would be one.
> With this sort of amplifier, I doubt anyone could do better
> for themselves unless they had lots of experience.
> He states there are circuits for Trimodal, class B,
> class AB and class A can be had by simply adjusting the bias.
>
> He says, and I would agree, that class AB has less
> distortion than class B whilst the amp stays in class A,
> but more distortion when it goes above, which may not matter
> if there is 50 watts of class A, and then a total of 150 of class AB.
>
**And if your output devices are carefully matched, you can keep distortion
nice and low, when operating in the Class B region too.
Trevor Wilson
>
> I think Trevor is grossly exagerating mosfet problems
> to make BJTs look better and easier to use.
> This business of having to use a minimum of 2 amp bias currents
> is excessive for class AB operation.
> I use about 80 Ma per device, and I get very low distortion.
> In my 300 watt class AB amps I run this current, and I use six TOP3
> flat pack mosfets per channel. So there is a reasonable amount of class A
> before the amp reverts to class B.
> There are no severe problems with mosfets, that an intelligent person
cannot
> get around.
**Then why bother with a Class A MOSFET amp, such as the one you are
developing?
If Class A/B MOSFET amps are perfect, wouldn't your project be a complete
waste of time?
Patrick Turner
Trevor Wilson wrote:
I don't think your circuit you used with the mosfets could have been
optimum.Well, you are entitled to your anecdotes and opinion, of course,but my
experience is different.
I know plenty of people who don't mind mosfet amps.
> > This business of having to use a minimum of 2 amp bias currents
> > is excessive for class AB operation.
> > I use about 80 Ma per device, and I get very low distortion.
>
> **By using Global NFB. Sure, that's easy enough to do.
So what is wrong with global NFB? Well appliedglobal NFB works very well,
providing the drive amp is
suitably configured.
> > In my 300 watt class AB amps I run this current, and I use six TOP3
> > flat pack mosfets per channel. So there is a reasonable amount of class A
> > before the amp reverts to class B.
> > There are no severe problems with mosfets, that an intelligent person
> cannot
> > get around.
>
> **Yes, there are. Like it or not, the negative tempco of gm will suck the
> life out of music, every time.
The mosfets' "suck" effect is not a problem.The negative temperature coefficient
of Gm is not an issue,as the thermal temp change even at 10 Hz at full power,
won't
have the Gm change during cycles. I suggest you read up a little
on the web page by Douglas Self, who I would say makes a little more sense
on solid state amps than both you or I do.
One just cannot express all the issues on any type of
amps in a few one line generalisations.
> >
> > The only time you would use 2 amps for a bias current is in a class A
> design,
> > with a low drain supply of 15 volts, to limit dissipation to about 30
> watts.
> > I don't even use this high a current, an amp is plenty,
> > to swing a substantially linear transfer curve with a 35 ohm load
> > with a 35 volt supply.
>
> **Agreed.
Patrick Turner.
Patrick Turner
Trevor Wilson wrote:
> "Patrick Turner" <in...@turneraudio.com.au> wrote in message
> news:3B643C8B...@turneraudio.com.au...
> > I had a good look at Douglas Selfs web page on amplifiers
> > at
> > http://www.dself.demon.co.uk/dipa.htm
> >
> > I really think Trevor should look at this page.
> > And so should anyone else with an eye to solid state
> > circuitry operation.
>
> **I agree, Patrick. I've read every word written by Self, ever written in
> Eletronics World. What makes you think I haven't?
Well your idea that devices either compress or expandduring cycles seems to be
one reason for you to draw to everybody's
attention the differences between Mosfets and BJT.
Douglas Self seems to think this effect you say happens, just does not.
But maybe your test results indicate otherwise.
> >
> > Douglas Self does not like Mosfets much.
> > But other designers who have contributed to
> > 'Electronics, Wireless World' over the years
> > don't seem to mind them.
>
> **Sure. They don't much listen to music, thats all.
Oh, common, you cannot say what the listening habitsare of such desingers are.
> >
> > There are about 1,001 tips on circuits for transistors
> > to get distortion down to less than 0.01% at near full power
> >
> > At least what Douglas Self is saying makes a lot of sense.
>
> **I agree with much that Self says. I alswo dissagree with a bunch of stuff,
> too.
I also think some of his methods seem to be extremein that he worries over even
the slightest distortion of any kind,
no matter how tiny and isignificant it may be.
But at least he is a perfectionist, and I don't mind.
At the end of the day, I doubt many folks would tell much
difference between an amp with 0.1% IMD, or 0.001% IMD,
but then it depends on the spectral analysis a bit, would it not?
> >
> > It won't be very palatable to the SET-without-NFB crowd.
>
> **And why should it be? Those sorts of amplifiers are highly coloured and
> distorted. They are (generally) incapable of reproducing music with any sort
> of fidelity. They sometimes sound 'nice' though. Of course, it ain't
> reality, though.
Well, it is reality, for many listeners.And to say otherwise is simply like
waving a pork chop around
in a synagogue, a pointless exercise.
> >
> > Whether one of his designs using his circuit would be as good
> > sonically as he suggests it must be, is anyone's guess.
>
> **I've built one. It was OK. Not an ME, of course, but competent enough.
>
> >
> > I would imagine there are printed circuit boards
> > available for one of Mr Self's Blameless Amplifiers.
> > You need to be ever so careful with solid state
> > circuit board layout if you are to come anywhere near
> > achieving low distortion, so in this case it would pay if you
> > were to email Douglas and ask him if he knows of a source
> > of boards which are ok. I am sure there would be one.
> > With this sort of amplifier, I doubt anyone could do better
> > for themselves unless they had lots of experience.
> > He states there are circuits for Trimodal, class B,
> > class AB and class A can be had by simply adjusting the bias.
> >
> > He says, and I would agree, that class AB has less
> > distortion than class B whilst the amp stays in class A,
> > but more distortion when it goes above, which may not matter
> > if there is 50 watts of class A, and then a total of 150 of class AB.
> >
>
> **And if your output devices are carefully matched, you can keep distortion
> nice and low, when operating in the Class B region too.
Well, I would hope so.
Patrick Turner.
Patrick Turner
Trevor Wilson wrote:
> "Patrick Turner" <in...@turneraudio.com.au> wrote in message
> news:3B641596...@turneraudio.com.au...
> >
> > I think Trevor is grossly exagerating mosfet problems
> > to make BJTs look better and easier to use.
> > This business of having to use a minimum of 2 amp bias currents
> > is excessive for class AB operation.
> > I use about 80 Ma per device, and I get very low distortion.
> > In my 300 watt class AB amps I run this current, and I use six TOP3
> > flat pack mosfets per channel. So there is a reasonable amount of class A
> > before the amp reverts to class B.
> > There are no severe problems with mosfets, that an intelligent person
> cannot
> > get around.
>
> **Then why bother with a Class A MOSFET amp, such as the one you are
> developing?
The difference between you and me is that I like to experiment and try new
things,and you don't, because you are so content with the designs of Peter
Stein.
I just like foolin around with things. I keep getting ideas, and I build on
them.
Some folks like playing around with cars and boats etc, with me it is
amplifiers,
and to a lesser extent, speakers.
I have had no need to ever buy more than one hi-fi reciever
in the last 30 years, I just go to the shed and make what I want.
It is a pity I cannot seem to find time to make a CD player.
> If Class A/B MOSFET amps are perfect, wouldn't your project be a complete
> waste of time?
Nothing is perfect, and I am serene about this fact of life.Of course the
project would NOT be a complete wase of time.Watching Survivor on TV, or
guzzling beer or pulling the pokies would
be a waste of time, for me, it would.
So I like to learn more, and confound the critics and doubters.
I now have the second 50 watt class A monobloc mosfet amp
ready to test. If music played through these devices is seriously musical,
engaging, intimate, smooth, dynamic, and detailed like the rest of my gear
then I will be happy until I want to try something else.
I have some novel ideas on current dumping which I would like to try out.
Patrick Turner.
Trevor Wilson
**Perhaps. I used them in THREE totally different topologies, however. All
sounded slightly different. ALL demonstrated similar dynamic problems.
Well, you are entitled to your anecdotes and opinion, of course,but my
> experience is different.
> I know plenty of people who don't mind mosfet amps.
**Sure. So do I. In many cases, they have not heard a properly designed BJT
amp. Significantly, EVERY MOSFET amp, I have ever heard, demonstrates
similar dynamic problems. Well, except one. The Pass designs don't seem to
possess the dynamic problems, inherent to every other MOSFET design. Maybe
it's due to the very high bias current used.
>
> > > This business of having to use a minimum of 2 amp bias currents
> > > is excessive for class AB operation.
> > > I use about 80 Ma per device, and I get very low distortion.
> >
> > **By using Global NFB. Sure, that's easy enough to do.
>
> So what is wrong with global NFB? Well appliedglobal NFB works very well,
> providing the drive amp is
> suitably configured.
**Having taekn part in some blind trials, of ME amplifiers, with and without
Global NFB, I can tell you that my preference is for Zero Global NFB. ZGNFB
amps sound cleaner, claered and more faithful to the original source.
Additionally, they provide significant improvements in EM radiation
rejection (very handy, for us city dwellers), which can enter an amp,
through the speaker connections.
>
> > > In my 300 watt class AB amps I run this current, and I use six TOP3
> > > flat pack mosfets per channel. So there is a reasonable amount of
class A
> > > before the amp reverts to class B.
> > > There are no severe problems with mosfets, that an intelligent person
> > cannot
> > > get around.
> >
> > **Yes, there are. Like it or not, the negative tempco of gm will suck
the
> > life out of music, every time.
>
> The mosfets' "suck" effect is not a problem.
**Well, Patrick, I say it is. Only one MOSFET amp, I've heard (and I've
listened to a LOT of them), doesn't exhibit compressive effects.
The negative temperature coefficient
> of Gm is not an issue,as the thermal temp change even at 10 Hz at full
power,
> won't
> have the Gm change during cycles. I suggest you read up a little
> on the web page by Douglas Self, who I would say makes a little more sense
> on solid state amps than both you or I do.
**Sure, but don't forget: I never said I agreed with everything Self says.
> One just cannot express all the issues on any type of
> amps in a few one line generalisations.
**Sure. What you need to do, is to listen to a PROPERLY designed, zero
Global NFB, BJT amp, beside your Class A/B MOSFET amp. Then you'll get the
idea.
Trevor Wilson
>
>
> Trevor Wilson wrote:
>
> > "Patrick Turner" <in...@turneraudio.com.au> wrote in message
> > news:3B641596...@turneraudio.com.au...
> > >
> > > I think Trevor is grossly exagerating mosfet problems
> > > to make BJTs look better and easier to use.
> > > This business of having to use a minimum of 2 amp bias currents
> > > is excessive for class AB operation.
> > > I use about 80 Ma per device, and I get very low distortion.
> > > In my 300 watt class AB amps I run this current, and I use six TOP3
> > > flat pack mosfets per channel. So there is a reasonable amount of
class A
> > > before the amp reverts to class B.
> > > There are no severe problems with mosfets, that an intelligent person
> > cannot
> > > get around.
> >
> > **Then why bother with a Class A MOSFET amp, such as the one you are
> > developing?
>
> The difference between you and me is that I like to experiment and try new
> things,and you don't, because you are so content with the designs of Peter
> Stein.
**I enjoy listening to other designs. And, as I've stated before, I've built
quite a few, over the years. I long ago realised that I could not build a
product, which was superior to Peter Stein's product, so I ceased. Nothing
wrong with that. I long ago realised that GMH could build a better
automobile, than I could, so I stopped trying to build an automobile. And
for the record: I have made many suggestions to Peter, over the last two
decades. Some of those suggestions have made it to market. Some have not.
> I just like foolin around with things. I keep getting ideas, and I build
on
> them.
> Some folks like playing around with cars and boats etc, with me it is
> amplifiers,
> and to a lesser extent, speakers.
**Good for you. Keep at it. Whatever you do, though, keep an open mind.
Consider a PROPERLY designed Zero Global NFB BJT amp. They really do sound
very accurate.
>
> I have had no need to ever buy more than one hi-fi reciever
> in the last 30 years, I just go to the shed and make what I want.
> It is a pity I cannot seem to find time to make a CD player.
**Awww. I'd like to see you do it with tubes. Even down to the servo loops,
DACs and 16kB of RAM. Now THAT would be interesting.
>
> > If Class A/B MOSFET amps are perfect, wouldn't your project be a
complete
> > waste of time?
>
> Nothing is perfect, and I am serene about this fact of life.Of course the
> project would NOT be a complete wase of time.Watching Survivor on TV, or
> guzzling beer or pulling the pokies would
> be a waste of time, for me, it would.
> So I like to learn more, and confound the critics and doubters.
**Your words:
----
There are no severe problems with mosfets, that an intelligent person
cannot get around.
----
This suggests you have nothing more to accoplish with MOSFETs. So, I ask
again: Why bother with a Class A MOSFET amp? Unless, of course, you agree
with me.......
>
> I now have the second 50 watt class A monobloc mosfet amp
> ready to test. If music played through these devices is seriously musical,
> engaging, intimate, smooth, dynamic, and detailed like the rest of my gear
> then I will be happy until I want to try something else.
> I have some novel ideas on current dumping which I would like to try out.
**As I have stated, I have no real objection to Class A MOSFET designs. They
are the only implentation, which obviates the more annoying shortcomings of
MOSFET designs.
Trevor Wilson
"Preamp" <pre...@spamblock.dingoblue.net.au> wrote in message
news:gecamt83sb90mhru3...@4ax.com...
> >> So what is wrong with global NFB? Well appliedglobal NFB works very
well,
> >> providing the drive amp is
> >> suitably configured.
> >
> >**Having taekn part in some blind trials, of ME amplifiers,
>
** <SNIP> repetitive part of post.
Tell you what: If you sit down and listen to one, in your system, for a
couple of days, I'll say nothing for a week, about ME.
Fair enough?
Patrick Turner
Trevor Wilson wrote:
>
>
> THD figures were commendably low and it could
> even
> > > cope with 2 Ohm loads. Yet it still sounded like a typical Class A/B
> MOSFET
> > > amp. Uninvolving and lacking in dynamics.
> >
> > I don't think your circuit you used with the mosfets could have been
> > optimum.
>
> **Perhaps. I used them in THREE totally different topologies, however. All
> sounded slightly different. ALL demonstrated similar dynamic problems.
>
> Well, you are entitled to your anecdotes and opinion, of course,but my
> > experience is different.
> > I know plenty of people who don't mind mosfet amps.
>
> **Sure. So do I. In many cases, they have not heard a properly designed BJT
> amp. Significantly, EVERY MOSFET amp, I have ever heard, demonstrates
> similar dynamic problems. Well, except one. The Pass designs don't seem to
> possess the dynamic problems, inherent to every other MOSFET design. Maybe
> it's due to the very high bias current used.
It depends on what you call high current.And on the design. I don't know what
Pass design you mean.
> >
> > > > This business of having to use a minimum of 2 amp bias currents
> > > > is excessive for class AB operation.
> > > > I use about 80 Ma per device, and I get very low distortion.
> > >
> > > **By using Global NFB. Sure, that's easy enough to do.
> >
> > So what is wrong with global NFB? Well appliedglobal NFB works very well,
> > providing the drive amp is
> > suitably configured.
>
> **Having taekn part in some blind trials, of ME amplifiers, with and without
> Global NFB, I can tell you that my preference is for Zero Global NFB. ZGNFB
> amps sound cleaner, claered and more faithful to the original source.
Yes, I see that this is possible.But I bet that the ME amp is configured with an
emitter follower output stage,
which can be made to produce low distortion any way, well, low measurements.
BJTs cannot ever have more than about 0.8 volts between base and emitter,
and the voltage gain inherent with BJT is vast compared to all other
devices, including mosfets. So even with an emitter follower output
stage there is a pile more NFB than in any other device used this way.
So all one needs to do is have a voltage drive amp which exhibits
low drive output impedance and good lineaity and your BJT amp should sound well
if you have a decent psu.
I use the same philosophy with my drive amp in my class AB mosfet
amp which has a source follower output stage.
I go to the extent of running them with a regulated
rail voltage well above that of the output stage, so that the voltage amp drive
is very low distortion for all of its work.
But because that amp has only differential pair, VAS, and output mosfets in
three simple wide bandwidth stages, there are no problems with global feedack
application, and excellent dynamics and stability.
I admit source follower mosfet stages have higher output
impedance than BJT follower stages, but so what,
this is easily compensated for.
> Additionally, they provide significant improvements in EM radiation
> rejection (very handy, for us city dwellers), which can enter an amp,
> through the speaker connections.
The mains here where I live are appalling with their RF content.When I
constructed mains filters for my C-tick acreditation folder
I found out about RF getting into things.
But none of my amps suffer from RF .
Speaker cables are good aerials for RF, and yes. any RF goes into the amp
via the NFB network, if you have one. So because global schemes of NFB
exist where there is a lot of gain, you would expect any RF to get amplified.
But no, it doesn't happen, because the open loop gain is only
high up to 5 Khz in my amps, afterwhich it rolls off at 6 Db per octave,
so by the time we get to 500 Khz, or the beginnining of the broadcast band,
the gain of the amp is very low.
In the transistor Linear Design reciever I bought 25 yrs ago,
I would ocassionally get a blast of demodulated RF coming from
some fellow with a cb radio driving past and transmitting.
So the circuit was acting as a crude de-facto radio.
But this has never happened in my own amps, since it does not,
I have not reasoned why not, as I am a busy person.
> > > > In my 300 watt class AB amps I run this current, and I use six TOP3
> > > > flat pack mosfets per channel. So there is a reasonable amount of
> class A
> > > > before the amp reverts to class B.
> > > > There are no severe problems with mosfets, that an intelligent person
> > > cannot
> > > > get around.
> > >
> > > **Yes, there are. Like it or not, the negative tempco of gm will suck
> the
> > > life out of music, every time.
> >
> > The mosfets' "suck" effect is not a problem.
>
> **Well, Patrick, I say it is. Only one MOSFET amp, I've heard (and I've
> listened to a LOT of them), doesn't exhibit compressive effects.
OK.
> The negative temperature coefficient
> > of Gm is not an issue,as the thermal temp change even at 10 Hz at full
> power,
> > won't
> > have the Gm change during cycles. I suggest you read up a little
> > on the web page by Douglas Self, who I would say makes a little more sense
> > on solid state amps than both you or I do.
>
> **Sure, but don't forget: I never said I agreed with everything Self says.
Self would say like someone else I know, that he only dealt with facts,and
nothing else but facts, and so on.....
> > One just cannot express all the issues on any type of
> > amps in a few one line generalisations.
>
> **Sure. What you need to do, is to listen to a PROPERLY designed, zero
> Global NFB, BJT amp, beside your Class A/B MOSFET amp. Then you'll get the
> idea.
Well, maybe one day. But we all might have own own ideas of what PROPERLYmeans.
What do you think about the Sziklai output stage configuation?
As opposed to just emitter follower operation.
I think it has some merits, and its use with mosfets would be interesting.
Patrick Turner. http://www.turneraudio.com.au
Patrick Turner
Preamp wrote:
> >> So what is wrong with global NFB? Well appliedglobal NFB works very well,
> >> providing the drive amp is
> >> suitably configured.
> >
> >**Having taekn part in some blind trials, of ME amplifiers,
>
> Fer chrissakes, can you possibly post for 24 hours and not mention ME ME ME ME
> ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME
> ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME
> ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME
> ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ME ...
Preamp, it is quite OK for a certain friendly correspondant to mention ME.There is
at least one alternative, TURNER AUDIO, http://www.turneraudio.com.au
But just because this may be the case, it does not mean huge increases in sales,
simply because we publicise our approach to audio design.
Patrick Turner.
Geoff
>
> ** <SNIP> repetitive part of post.
>
> Tell you what: If you sit down and listen to one, in your system, for a
> couple of days, I'll say nothing for a week, about ME.
>
> Fair enough?
Ladies and Gentlemen, I have great pleasure (and some amusement) in
presenting to you none other than Mr "OMFG can you shut up about ME" Preamp
vs Trevor "but ME is so good!" Wilson.
I find it amusing because I agree with both of you. TW does talk about ME
endlessly but ME does make a number of amazing products. Tough to call, I
would take TW up on his offer then at least you'll be able to disagree with
him based on experience, alternatively you can acknowledge that ME is good
but let TW know that the record has reached the end of its groove :)
I would talk endlessly about my homemade Vifa speakers with 10" ports with a
90 degree bend in them and my 26 year old amplifier complete with all
original components, but I don't think anybody would believe that my system
could sound remotely good, let alone be accurate.
(alter-ego's note: I doubt my system is really as good as I believe it is,
but that doesn't stop me from enjoying it!)
Geoff
Geoff
built
> quite a few, over the years. I long ago realised that I could not build a
> product, which was superior to Peter Stein's product, so I ceased. Nothing
> wrong with that.
Depends how you derive your enjoyment. If you enjoy your own creations
because they kick "amp X"'s arse, then you will easily get disappointed
because there's almost always someone just that bit higher than you. If
however you enjoy the fruits of your own labour on its own merits, and
attempting to beat your own PB, then you will enjoy building things forever.
Of course if you already have an amazingly good amp sitting in the middle of
your room it's tough to put up with something that has more flaws that you
made yourself.
> > I have had no need to ever buy more than one hi-fi reciever
> > in the last 30 years, I just go to the shed and make what I want.
> > It is a pity I cannot seem to find time to make a CD player.
>
> **Awww. I'd like to see you do it with tubes. Even down to the servo
loops,
> DACs and 16kB of RAM. Now THAT would be interesting.
Would be interesting, I'll give it that. Whether anyone is insane enough to
try is another question :)
Geoff
Patrick Turner
Trevor Wilson wrote:
Some time I might do something serious with BJTs.I will not tell you of my
ultimate approach, I don't know what it will be.
> > I have had no need to ever buy more than one hi-fi reciever
> > in the last 30 years, I just go to the shed and make what I want.
> > It is a pity I cannot seem to find time to make a CD player.
>
> **Awww. I'd like to see you do it with tubes. Even down to the servo loops,
> DACs and 16kB of RAM. Now THAT would be interesting.
No, no, no, that would be dead boring!It would completely bankrupt me, and the
device when finished would need
a private power station and could not fit on my block of land!
No, you are leading me astray, and you have been spotted!
I do prefer the idea of countless transistors switcthing on and off.
Actually though, ther is a bloke in the Melbourne Audio Club
who has built his own CD player with all the right chips,
and a decent transport mechanism, with gold plated pcbs,
and their review was very favourable, and the bloke is 70.
You just gotta sit down quietly for about a month and figure out
all the little steps it takes to get from A to B, and then go from there.
Electronics is about all these little things adding up to a whole,
daunting for 98% of the population, but everything can be understood.
> >
> > > If Class A/B MOSFET amps are perfect, wouldn't your project be a
> complete
> > > waste of time?
> >
> > Nothing is perfect, and I am serene about this fact of life.Of course the
> > project would NOT be a complete wase of time.Watching Survivor on TV, or
> > guzzling beer or pulling the pokies would
> > be a waste of time, for me, it would.
> > So I like to learn more, and confound the critics and doubters.
>
> **Your words:
> ----
> There are no severe problems with mosfets, that an intelligent person
> cannot get around.
> ----
>
> This suggests you have nothing more to accoplish with MOSFETs. So, I ask
> again: Why bother with a Class A MOSFET amp? Unless, of course, you agree
> with me.......
>
> >
> > I now have the second 50 watt class A monobloc mosfet amp
> > ready to test. If music played through these devices is seriously musical,
> > engaging, intimate, smooth, dynamic, and detailed like the rest of my gear
> > then I will be happy until I want to try something else.
> > I have some novel ideas on current dumping which I would like to try out.
>
> **As I have stated, I have no real objection to Class A MOSFET designs. They
> are the only implentation, which obviates the more annoying shortcomings of
> MOSFET designs.
I shall see, in the fullness of time,that I will have to keep looking out
to see more.
Patrick Turner.
Geoff
> >but let TW know that the record has reached the end of its groove :)
>
it is
> the only alternative. If it is not the only alternative, this makes
Trevor's
> continually utterances about it, close kin to continual advertisng, which
is
> against the intentions of this newsgroup as I understand them.
Just do what I do when you see those two capitalised letters together: nod,
smile and reach for the salt shaker. I know you never said ME wasn't good,
just that you were sick of hearing about it. If you have heard it, at least
you can agree that it's good at the same time as saying you're sick of
hearing about it. Softens the blow somewhat :)
When was the last time you heard a Christian raving about the merits of
being a Mormon or a Jehovas Witness? People will always concentrate on
their own beliefs, and tolerance is the ability to be diplomatic about the
situation even if you don't agree!
Geoff
Geoff
Maybe TW can instead say "my favourite amp is very accurate" ;)
> Smart advertising does not get the potential customer offside!
Agreed.
Patrick Turner
Preamp wrote:
> >
> >The mains here where I live are appalling with their RF content.When I
> >constructed mains filters for my C-tick acreditation folder
>
> As an aside, what's the deal with C-tick? Is it compulsory? Is it desirable, and
> if so why?
Before you can sell a manufactured amplifier you are required by law to have
aquired C-tick acreditation from the Australian Communications Authority.
This is gained by proving to them that your device does not add noise to the mains
supply
between 150 Khz and 30 Mhz, by means of direct injection.
It is as simple as that for amps with linear ( conventional ) power supplies,
with tranny and diodes charging a cap.
Amps with switchmode PSUs are harder to get approved.
I documented all my tests and presented my report and got an approval number.
It has special significance if you are a maker wanting to produce ten
thousand amps, you don't want 10,000 complaints, the fines are not nice,
if you have not complied.
Shops can get fined for stocking amps without the compliance C-tick Mark
on them.
There are approved test labs which charge about a grand for a minimal
approval test, but I did it myself, to the ACA's satisfaction.
A normal amp with a tranny PSU will often comply very easily.
And if you build one amp only and sell it as a work of art,
they don't come looking for you.
Safety is a seperate issue, and since no amps have ever killed anyone,
which is miraculous, there are no requirements for safety beyond the very basic
for electric equipment.
You do not have to have NATA approval, but if you sold 10,000,
you would be foolish not to gain the more expensive safety approval
which you are not allowed to do yourself.
This would help to insulate you from legal claims from goodness knows who,
very likely with an unapproved article, when there are 10,000
items out there, not just a handful.
Imagine if 10,000 SET amps were sold into average aussie homes tomorrow,
each one with an unenclosed 211 tube with 1200 volts in there!
The possibility for dead kids laying around is rather offputting for
the shareholders.
Australia is a well reasoned country.
Patrick Turner. http://www.turneraudio.com.au
Geoff
> >being a Mormon or a Jehovas Witness? People will always concentrate on
> >their own beliefs, and tolerance is the ability to be diplomatic about
the
> >situation even if you don't agree!
>
beliefs
> and commercial interests.
Spreading religion is a commercial interest though, it earns you brownie
points with your particular God and ensures you get good seats when you go
to the great white stadium in the sky ;)
Settle settle! I'm joking... Put away those lynching weapons.... heh.
It's a tough call to make, I think there's only one person who knows whether
it's raving about beliefs or whether it's a sales pitch (even if done so
subconsciously). One could even go for reverse psychology by mentioning
other brands and people start wondering why you're not pimping (whoops, that
just slipped out. I meant "promoting") your own brand. Then they have
natural curiosity and will be more receptive when you tell them how good it
is :)
Geoff
Patrick Turner
Preamp wrote:
> >
> >
> >Preamp wrote:
> >
> >> >
> >> >The mains here where I live are appalling with their RF content.When I
> >> >constructed mains filters for my C-tick acreditation folder
> >>
> >> As an aside, what's the deal with C-tick? Is it compulsory? Is it desirable, and
> >> if so why?
> >
> > Before you can sell a manufactured amplifier you are required by law to have
> >aquired C-tick acreditation from the Australian Communications Authority.
> >This is gained by proving to them that your device does not add noise to the mains
> >supply
> >between 150 Khz and 30 Mhz, by means of direct injection.
>
> Seems fair enough.
>
> >It is as simple as that for amps with linear ( conventional ) power supplies,
> >with tranny and diodes charging a cap.
>
> A should not be a problem...
>
> >Amps with switchmode PSUs are harder to get approved.
>
> Yeah, mayhap.
>
> >
> >I documented all my tests and presented my report and got an approval number.
> >It has special significance if you are a maker wanting to produce ten
> >thousand amps, you don't want 10,000 complaints, the fines are not nice,
> >if you have not complied.
>
> So if noone complains, then you get away with it? This would explain a store I
> know (and will not name) that doesn't have c-tick for some stuff being made and
> sold.
Well the shop and maker, if Australian, would be at risk of penalty.And they would be
prosecuted, if they did not produce a report
on their product. The ACA is not a merciless Big Brother.
but like a lot of regulatory authorities they have the power
to lay down the law if need be.
Only fools would pretend they don't matter.
> >
> >Shops can get fined for stocking amps without the compliance C-tick Mark
> >on them.
>
> Hmmm...
>
> > There are approved test labs which charge about a grand for a minimal
> >approval test, but I did it myself, to the ACA's satisfaction.
>
> Yes, I was told it costs in that order. I was also told that _ANY_ modification
> needs reapproval, and since a change in brand of any component might be
> considered a modification, that could get expensive quickly!
Well test labs tell you this line of crap to make it seem complexand above ordinary
mortals, which it is not.
But unless you knew about filters, electronics, and report writing,
you would have no chance of getting an acreditation.
If you did change something in a line of product, all you have to do
is show you still comply, and record it in your folder of compliance, to be held
at the business owners address, available for inspection at any time.
The test labs peddle the line that only they are able to give the
ok to gear, and while they perform tests they always find reasons
to add a cap here or there to get it to comply, which is a
modification, and has a minimum charge of $275 at least,
on top of a minimum of the grand of basic fees.
This is how they make their consultant fees.
They make no distinction between me and any well healed
manufacturer.
They sure are not struggling to make ends meet on a very low income
like me.
My approach was to avoid the consultant's fee, and learn all about
what the heck it was all about, and do the tests myself.
It all took several days of hard yakka.
>
>
> >A normal amp with a tranny PSU will often comply very easily.
> >And if you build one amp only and sell it as a work of art,
> >they don't come looking for you.
>
> Not worth their while...
>
> >
> >Safety is a seperate issue, and since no amps have ever killed anyone,
> >which is miraculous, there are no requirements for safety beyond the very basic
> >for electric equipment.
>
> This brings up another issue on the required trade certificates to build the
> thing. The best shot would seem to be an exemption for work done in a
> 'workplace'. The snag is getting your workshop recognised as a workplace. The
> catch 22 is that if 1 or 2 people work there, workplace health and safety do not
> required it to be inspected/(registered?), but then the electrical authorities
> don't seem to want to guarantee the exemption.
Well, workplace regulations would be a common sense thing.I could never afford to
employ anybody to handmake amps, including
the output trannies, unless they operated as a subcontractor,
for so much per item. I know of zero people prepared to spend
2 days making an output tranny for the dollars I get out of it.
It is not the sort of thing I could trust to many other folks,
as the risk of failures is down to me, not on any employee
of mine.
If I had a proper factory, and several people helping to make product,
then I would be silly not to comply to whatever regulations apply.
These costs of compliance here or there, and rent for premises
and so on should be of minimal concern, as the profits
from the efforts should be high, or why the heck do it?
As it is I have been operating my own sole trader business for
20 years, and never have any goons arrived from some
onerous department to investigate.
Care taken is the best insurance.
> >You do not have to have NATA approval, but if you sold 10,000,
> >you would be foolish not to gain the more expensive safety approval
> >which you are not allowed to do yourself.
> >This would help to insulate you from legal claims from goodness knows who,
> >very likely with an unapproved article, when there are 10,000
> >items out there, not just a handful.
>
> Which of course would affect your insurability.
>
> >
> >Imagine if 10,000 SET amps were sold into average aussie homes tomorrow,
> >each one with an unenclosed 211 tube with 1200 volts in there!
> >The possibility for dead kids laying around is rather offputting for
> >the shareholders.
>
> Haha... In the past I considered building and selling guitar amps (love of SET
> is much more recent). Then again, guitars make lovely connections between high
> voltages in amps and people holding the steel strings and fittings on the
> guitar, or sucking on the metal microphones. :-)
There should be isolation transformers for all pickups.
> >Australia is a well reasoned country.
>
> In some areas... but they have a huge blind spot when they require certification
> to wire a building in order to build a guitar amp.
But the wiring would need to comply no matter what you made, evenwater beds. It would
be simple to get such compliance.
If you had a fire, well, you know what smart lawyers can do....
Anyhow it would be very hard to compete with Fender and Marshall,
who get all the donkey work done on the Mexican side of the border,
where no rules apply, and people work for a dollar a day.
On the yank side, they put labels on and do the selling,
Hundreds at a time.
Jolida is made in asia someplace, and the cost of production is
of minor concern to the company.
Patrick Turner.
> >
> >Patrick Turner. http://www.turneraudio.com.au
> >
CHOKY
--
message sent by Choky,
Prodanovic Aleksandar
ch...@EUnet.yu
Yugoslavia
"Preamp" <pre...@spamblock.dingoblue.net.au> wrote in message
news:52oamtkatqdkmtmjj...@4ax.com...
> >>
> >> Haha... In the past I considered building and selling guitar amps (love
of SET
> >> is much more recent). Then again, guitars make lovely connections between
high
> >> voltages in amps and people holding the steel strings and fittings on the
> >> guitar, or sucking on the metal microphones. :-)
> >
> >There should be isolation transformers for all pickups.
>
> Never seen one yet.
Trevor Wilson
**His MOSFET designs. And yes, they operate at very high bias currents. Pass
knows the limitations of MOSFETs quite well.
>
> > >
> > > > > This business of having to use a minimum of 2 amp bias currents
> > > > > is excessive for class AB operation.
> > > > > I use about 80 Ma per device, and I get very low distortion.
> > > >
> > > > **By using Global NFB. Sure, that's easy enough to do.
> > >
> > > So what is wrong with global NFB? Well appliedglobal NFB works very
well,
> > > providing the drive amp is
> > > suitably configured.
> >
> > **Having taekn part in some blind trials, of ME amplifiers, with and
without
> > Global NFB, I can tell you that my preference is for Zero Global NFB.
ZGNFB
> > amps sound cleaner, claered and more faithful to the original source.
>
> Yes, I see that this is possible.But I bet that the ME amp is configured
with an
> emitter follower output stage,
**Well that would be one bet you would lose. Even more curious, considering
that you claim to have serviced an ME power amp. Did you not notice that ME
amplifiers (and preamplifiers) all operate in Common Emitter configuration?
**Let's try them across the road from Channel 2, shall we? One of my clients
tried 15 or 20 amplifiers, before he discovered that an ME would operate
without the annoying frame buzz, many Gore Hill residents are familiar with.
The only other amplifiers, which could cope with this siuation, were a
couple of tube amps. I believe that induced EM radiation, into speaker
cables, is one reason why Zero Global NFB amplifiers are preferred by many
listeners.
> Speaker cables are good aerials for RF, and yes. any RF goes into the amp
> via the NFB network, if you have one. So because global schemes of NFB
> exist where there is a lot of gain, you would expect any RF to get
amplified.
> But no, it doesn't happen, because the open loop gain is only
> high up to 5 Khz in my amps, afterwhich it rolls off at 6 Db per octave,
> so by the time we get to 500 Khz, or the beginnining of the broadcast
band,
> the gain of the amp is very low.
**Interesting things happen, when you're across the road from a Megawatt or
so, of TV signals.
>
> In the transistor Linear Design reciever I bought 25 yrs ago,
> I would ocassionally get a blast of demodulated RF coming from
> some fellow with a cb radio driving past and transmitting.
> So the circuit was acting as a crude de-facto radio.
> But this has never happened in my own amps, since it does not,
> I have not reasoned why not, as I am a busy person.
**A good thing, Patrick. Nonetheless, some areas of some cities can be
*difficult*.
**Sure. As a designer, I would have thought you'd be interested in other
products. I sure am. I know that Peter Stein tried many MOSFET circuits, way
back when they came to market. Their properties SEEMED to be irresistable.
The reality was something else, though. Over the years, I too, have tried
many MOSFET designs. They are very convenient and easy to get operational.
Generally speaking, it is easier to to reasonable sound from a MOSFET amp,
than from a BJT amp. However, at the very high end, BJTs rule. Care list all
the planet's high end amps, which use MOSFETs? I can only think of one.
>
> What do you think about the Sziklai output stage configuation?
**Actually, the very first SS amp, to attract my attention, used exactly
this type of output stage configuration. It was the mighty Marantz 1200B. A
derivative of the 240 and 250 power amps. It demonstrated a remarkable
clarity and quality, unlike any other amplifier, I had ever heard.
Collectors driving loads, is, IMO, far and away the best method to interface
with a loudspeaker. OF course, more effort is required, to accomplish this
task. The Marantz, for instance, required the use of carefully matched
output and driver transistors.
> As opposed to just emitter follower operation.
> I think it has some merits, and its use with mosfets would be interesting.
**It would certainly be interesting.
Trevor Wilson
> >> > Fer chrissakes, can you possibly post for 24 hours and not mention ME
> >>
> >> ** <SNIP> repetitive part of post.
> >>
> >> Tell you what: If you sit down and listen to one, in your system, for a
> >> couple of days, I'll say nothing for a week, about ME.
> >>
> >> Fair enough?
> >
> >Ladies and Gentlemen, I have great pleasure (and some amusement) in
> >presenting to you none other than Mr "OMFG can you shut up about ME"
Preamp
> >vs Trevor "but ME is so good!" Wilson.
> >
> >I find it amusing because I agree with both of you. TW does talk about
ME
> >endlessly but ME does make a number of amazing products. Tough to call,
I
> >would take TW up on his offer then at least you'll be able to disagree
with
> >him based on experience,
>
**I don't believe you. You are on record as one who has no respect for any
amplifier technology, other than the one you build yourself. It is extremely
unlikely that you will report honestly and fairly about what you hear with
an ME amplifier. Nonetheless, if the cost to me is not high, I will send
you a product, for evaluation.
>
>
> >alternatively you can acknowledge that ME is good
> >but let TW know that the record has reached the end of its groove :)
>
> I really never said it wasn't good.
**Not entirely true. Look through your old posts.
I do reject the notion, however, that it is
> the only alternative.
**ME is not the only alternative. And I have stated as much, many times. And
you know it.
If it is not the only alternative, this makes Trevor's
> continually utterances about it, close kin to continual advertisng, which
is
> against the intentions of this newsgroup as I understand them.
**I speak about ME, when appropiate.
remarkabel
Trevor Wilson wrote:
Does this not imply that the problem lies in the implemetation and not the
devices? I agree, however, that it also probably implies that implemtation
*may* be more onerous on the designer than BJT based amps.
Cheers,
Mark
remarkabel
Preamp wrote:
> >> > Fer chrissakes, can you possibly post for 24 hours and not mention ME
> >>
> >> ** <SNIP> repetitive part of post.
> >>
> >> Tell you what: If you sit down and listen to one, in your system, for a
> >> couple of days, I'll say nothing for a week, about ME.
> >>
> >> Fair enough?
> >
> >Ladies and Gentlemen, I have great pleasure (and some amusement) in
> >presenting to you none other than Mr "OMFG can you shut up about ME" Preamp
> >vs Trevor "but ME is so good!" Wilson.
> >
> >I find it amusing because I agree with both of you. TW does talk about ME
> >endlessly but ME does make a number of amazing products. Tough to call, I
> >would take TW up on his offer then at least you'll be able to disagree with
> >him based on experience,
>
> Done! and I promise to evaluate it honestly.
>
> >alternatively you can acknowledge that ME is good
> >but let TW know that the record has reached the end of its groove :)
>
> I really never said it wasn't good. I do reject the notion, however, that it is
> the only alternative. If it is not the only alternative, this makes Trevor's
> continually utterances about it, close kin to continual advertisng, which is
> against the intentions of this newsgroup as I understand them.
But TW really does believe in the merits of ME amplification. Sure he is in a
difficult situation as a retailer whose major product line is ME. So how do you
know that TW is not stating his opinuion as Trevor Watson and not the proprietor of
Rage Audio? Patrick Turner has regularly praised the performance of his own amps
which he sells commercially (albeit on a different scale) and no-one seems to have
a go at his 'bias'. (by the way Patrick I am not having a go just using you as an
example, I apologise for any offence I may have caused)
Cheers,
Mark
>
>
> >
> >I would talk endlessly about my homemade Vifa speakers with 10" ports with a
> >90 degree bend in them and my 26 year old amplifier complete with all
> >original components, but I don't think anybody would believe that my system
> >could sound remotely good, let alone be accurate.
> >
> >(alter-ego's note: I doubt my system is really as good as I believe it is,
> >but that doesn't stop me from enjoying it!)
>
> same here :-)
>
> >
> >Geoff
> >
> >
> >
> >
remarkabel
Preamp wrote:
> >
> >When was the last time you heard a Christian raving about the merits of
> >being a Mormon or a Jehovas Witness? People will always concentrate on
> >their own beliefs, and tolerance is the ability to be diplomatic about the
> >situation even if you don't agree!
>
> No problems. My annoyance is about the need to draw a distiction between beliefs
> and commercial interests.
Fiar enough. But who is to draw the line, so to speak?
Cheers,
Mark
Trevor Wilson
"remarkabel" <mark...@bit.net.au> wrote in message
news:3B65E62E...@bit.net.au...
**Sure. And I have said as much. In theory, MOSFETs are lower in distortion
than BJTs. That lower distortion, however, requires higher bias currents.
Ironically, at higher bias currents, the negative tempco of gm problem,
disappears, as well. IOW: Class A MOSFET amps, may offer the best of all
possible solutions, for audio amplifiers.
Trevor
> I was interested to read that in JC's catalogue. Am I correct in my
> understanding that the pre stage benefits from regulation because the
supply
> is much more stable (ie ripple is more easily avoided with an LM317 than
> capacitors, assuming current draw is low?) but regulators can't keep up
with
> large transient changes in current draw, which is why capacitors are still
> used to store energy for the output stage?
Essentially, yes. A regulated supply will nearly always have less ripple,
however a regulated supply for the output stage will increase cost
substantially for the same output, so a regulated input stage is a good cost
effective compromise.
> The catalogue quotes SNR as 117dB unweighted from 20-20kHz, and 123dB
> A-weighted. I'm not sure which one really matters, but most mfgr's
probably
> state the A-weighted since it's higher :)
Yes, but 6 dB is not really much difference at that level. Either is
substantially better than the 100dB quoted for the Series 3.
Trevor.
CHOKY
lets see what is copyed from rec.audio.tech :
"CP" <c...@1pnickc.freeserve.co.uk> wrote in message
news:i042mt4ad3silt6e5...@4ax.com...
> Basically, what is the best (solid state) amp set up: decent AC mains
> PSU + Class A amp or Batteries plus Class AB amp. Money and space mean
> Batteries + Class A is impractical for me.
**You're basic defintions are too broad to make a sensible suggestion. Class
A power amp operation is unnecessarily wasteful of energy and no better
sounding than a properly designed and biased Class A/B amplifier. Likewise,
a properly designed power supply will provide performance which is at least
the equal of batteries. Given the inflexibility of battery supplies, an AC
supply makes more sense.
>
> Been experimenting running budget audiophile gear off batteries. So
> far AMC Dac 8 and Cambridge Audio A1. Amazed at the increase in
> quality.
>
> I don't have the time / money to do the same experimenting with a
> Class A amp, though I found on the net a Jean Hiraga designed Class A
> DIY amp (The Monster) that uses a battery smoothed mains PSU and I
> could also run this off a high quality online UPS. I was also thinking
> of adding linear regulators to the supply rails.
**Sounds like a dumb idea, to me. Why not just add a bigger power
transformer and (lots) more filter capacitance?
>
> Put it another way, what gives the most benefit - Class A or Battery
> power?
**Neither.
>
> Still on the subject of The Monster, are there inexpensive more modern
> transistor substitutes that could be used to improve performance?
**I am unfamiliar with the design.
--
Trevor Wilson
http://www.rageaudio.com.au
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
,,,,end of copy
--
message sent by Choky,
Prodanovic Aleksandar
ch...@EUnet.yu
Yugoslavia
"Trevor Wilson" <tre...@rageaudio.com.au> wrote in message
news:S5m97.75261$Xr6.3...@news-server.bigpond.net.au...
Trevor
>
> "Patrick Turner" <in...@turneraudio.com.au> wrote in message
> news:3B651917...@turneraudio.com.au...
> > of Gm is not an issue,as the thermal temp change even at 10 Hz at full
> > power, won't
> > have the Gm change during cycles. I suggest you read up a little
> > on the web page by Douglas Self, who I would say makes a little more
sense
> > on solid state amps than both you or I do.
I said the same thing in reply to TW months ago, amazing how these things
keep repeating themselves.
> **Sure, but don't forget: I never said I agreed with everything Self says.
But can you tell us just how fast you believe the die temperature changes?
What frequency would it be expected to track the input signal. I have seen
no evidence that it can track a 20 Hz signal or above. It seems you hear a
difference (which may be true or not for your samples) but fail to prove
that it is caused by NTC as you repeatedly claim.
> > One just cannot express all the issues on any type of
> > amps in a few one line generalisations.
>
> **Sure. What you need to do, is to listen to a PROPERLY designed, zero
> Global NFB, BJT amp, beside your Class A/B MOSFET amp. Then you'll get the
> idea.
Exactly, you soon get the idea that any topology can be made to work or not
depending on the ability of the designer, and quality of the final product.
Trevor.
remarkabel
remarkabel wrote:
<snip>
> So how do you know that TW is not stating his opinuion as Trevor Watson and not the
> proprietor of
> Rage Audio?
<snip>
Wilson, oops, sorry.....
Trevor Wilson
"Trevor" <trevor@home> wrote in message
news:3b6660fb$0$20909$7f31...@news01.syd.optusnet.com.au...
>
> "Trevor Wilson" <tre...@rageaudio.com.au> wrote in message
> news:T5a97.73836$Xr6.3...@news-server.bigpond.net.au...
> >
> > "Patrick Turner" <in...@turneraudio.com.au> wrote in message
> > news:3B651917...@turneraudio.com.au...
> >> The negative temperature coefficient
> > > of Gm is not an issue,as the thermal temp change even at 10 Hz at full
> > > power, won't
> > > have the Gm change during cycles. I suggest you read up a little
> > > on the web page by Douglas Self, who I would say makes a little more
> sense
> > > on solid state amps than both you or I do.
>
> I said the same thing in reply to TW months ago, amazing how these things
> keep repeating themselves.
**Sure enough. Nevertheless, the negative tempco of gm, is measurable. It
does explain the compressive effects which some listeners complain of. If
you have an alternative explanation, then I welcome it.
>
> > **Sure, but don't forget: I never said I agreed with everything Self
says.
>
> But can you tell us just how fast you believe the die temperature changes?
**The silicon can warm up, very quickly. I don't have the data, but I would
think that due to it's inherent protection mechanism, the temperature rise
would be in the order of milliseconds. I'll need to do some homework on this
one.
> What frequency would it be expected to track the input signal. I have seen
> no evidence that it can track a 20 Hz signal or above. It seems you hear a
> difference (which may be true or not for your samples) but fail to prove
> that it is caused by NTC as you repeatedly claim.
**I accept that I have not provided adequate proof. More homework.
>
> > > One just cannot express all the issues on any type of
> > > amps in a few one line generalisations.
> >
> > **Sure. What you need to do, is to listen to a PROPERLY designed, zero
> > Global NFB, BJT amp, beside your Class A/B MOSFET amp. Then you'll get
the
> > idea.
>
> Exactly, you soon get the idea that any topology can be made to work or
not
> depending on the ability of the designer, and quality of the final
product.
**Oh, absolutely! On hearing the early Perreaux and ETI 5000 amplifiers, one
could easily disregard MOSFET amplifiers, as a complete joke. More
sophisticated designs, which I am familiar with (EHT, later ETI designs,
Redgum, Pass, Spectral and others) demonstrate significant sonic
improvements, to those early designs.
Trevor Wilson
"CHOKY" <ch...@eunet.yu> wrote in message news:9k5lth$a94$1...@news.eunet.yu...
> Somebody talk different things on different places;
> lets see what is copyed from rec.audio.tech :
>
**Not at all. My words are entirely consistent, with my beliefs and
measurements. If you feel that is not the case, then please feel free to
highlight any logical inconsistencies, you may find.
Patrick Turner
Trevor Wilson wrote:
> "
> > > amp. Significantly, EVERY MOSFET amp, I have ever heard, demonstrates
> > > similar dynamic problems. Well, except one. The Pass designs don't seem
> to
> > > possess the dynamic problems, inherent to every other MOSFET design.
> Maybe
> > > it's due to the very high bias current used.
> >
> > It depends on what you call high current.And on the design. I don't know
> what
> > Pass design you mean.
>
> **His MOSFET designs. And yes, they operate at very high bias currents. Pass
> knows the limitations of MOSFETs quite well.
I assume you mean the Nelson Pass Zen amps, with onemosfet for the active device
and another for a CCS, and that's all,
and it has 12 Db of shunt NFB, to get the Ro down to about
theb same as an SET amp, with no NFB.
A friend is trying to build a pair, he has had 15 seconds of music,
followed by a sullen smoky silence.
But anyone who tries to get a TOP3 style mosfet
to dissipate 51 watts at idle for class A, deserves this.
I think the design is stupid.
And then the CCS dissipates 51 watts too.
But I use four mosfets which dissipate 25 watts each
in my class A amp, amp they are happy with that.
When you put a 2 ohm load on the amp and
drive it up to clip, the idle current backs off, with the temperature
rise, real nice it is.
But when you use the correct load of 5 ohms, and stay in class A,
there's no temperature rise. There is little compression,
just the usual inceasing 3H distortion which all amps give.
> .But I bet that the ME amp is configured
> with an
> > emitter follower output stage,
>
> **Well that would be one bet you would lose. Even more curious, considering
> that you claim to have serviced an ME power amp. Did you not notice that ME
> amplifiers (and preamplifiers) all operate in Common Emitter configuration?
I just pulled out my circuit of the "75" amp, and yes, you are correct,The load
is connected to the collector circuit. It has been 18 mths
since I fixed this amp.
Gee, it has a lot of consecutive stages though, does it not?
No wonder you don't/can't use global NFB.
My 300 watter has half the stages.
Also, my input amp is similar to ME, with a pair
of symetrical paralleled differential amps, one with N and one with P devices,
and then you have complementary pair gain block, same as
I have used. I got the ideas from Electronics(Wireless World), and developed
them
for my own use.
But after the VAS, the similarity ends, my amp is just so simple.
>
>
> >
> > > Additionally, they provide significant improvements in EM radiation
> > > rejection (very handy, for us city dwellers), which can enter an amp,
> > > through the speaker connections.
> >
> > The mains here where I live are appalling with their RF content.When I
> > constructed mains filters for my C-tick acreditation folder
> > I found out about RF getting into things.
> > But none of my amps suffer from RF .
>
> **Let's try them across the road from Channel 2, shall we? One of my clients
> tried 15 or 20 amplifiers, before he discovered that an ME would operate
> without the annoying frame buzz, many Gore Hill residents are familiar with.
> The only other amplifiers, which could cope with this siuation, were a
> couple of tube amps. I believe that induced EM radiation, into speaker
> cables, is one reason why Zero Global NFB amplifiers are preferred by many
> listeners.
Well, apart from youself, nobody else reckons RF ingress is a majorproblem with
global NFB. The majority of low grade Consumer
electronics rely on Global NFB, no probs.
If you are getting RF bothers in problem areas, fit RF filters, simple.
>
>
> > Speaker cables are good aerials for RF, and yes. any RF goes into the amp
> > via the NFB network, if you have one. So because global schemes of NFB
> > exist where there is a lot of gain, you would expect any RF to get
> amplified.
> So the circuit was acting as a crude de-facto radio.
>
> > > **Sure. What you need to do, is to listen to a PROPERLY designed, zero
> > > Global NFB, BJT amp, beside your Class A/B MOSFET amp. Then you'll get
> the
> > > idea.
> >
> > Well, maybe one day. But we all might have own own ideas of what
> PROPERLYmeans.
>
> **Sure. As a designer, I would have thought you'd be interested in other
> products. I sure am. I know that Peter Stein tried many MOSFET circuits, way
> back when they came to market. Their properties SEEMED to be irresistable.
> The reality was something else, though. Over the years, I too, have tried
> many MOSFET designs. They are very convenient and easy to get operational.
> Generally speaking, it is easier to to reasonable sound from a MOSFET amp,
> than from a BJT amp. However, at the very high end, BJTs rule. Care list all
> the planet's high end amps, which use MOSFETs? I can only think of one.
Well, I know I havn't the glitsy profile, but my own Mosfet amps wouldcomply
with High End; they do similar things with music when compared to
so many other designs.
> > What do you think about the Sziklai output stage configuation?
>
> **Actually, the very first SS amp, to attract my attention, used exactly
> this type of output stage configuration. It was the mighty Marantz 1200B. A
> derivative of the 240 and 250 power amps. It demonstrated a remarkable
> clarity and quality, unlike any other amplifier, I had ever heard.
> Collectors driving loads, is, IMO, far and away the best method to interface
> with a loudspeaker. OF course, more effort is required, to accomplish this
> task. The Marantz, for instance, required the use of carefully matched
> output and driver transistors.
>
> > As opposed to just emitter follower operation.
> > I think it has some merits, and its use with mosfets would be interesting.
>
> **It would certainly be interesting.
The Sziklai pair has some attractive advantages.Say you have a gain of 10 in the
small drive transistor,
with its load being about 100 ohms.
Then you have a common source mosfet with and it has a gain of
about 7 driving an 8 ohm load, in the class B part of the cycle.
The total gain is 7 X 10, and this is reduced to just less
than unity by the Sziklai arragement, so distorion and noise
will be reuced by about 70 times, and Ro will be very low.
But because we use Mosfets, the need for a load of 100 ohms
is not so necesary, and it could be more, say 470 ohms,
and then the gain of the output stage would be
50 X 7 and the reduction of distortion, and Ro
would be similar to a BJT stage, IMHO.
And the small drive transistor would have a higher input impedance,
not affected by base currents from the output transistors,
and not affected by capacitances from the mosfet gates.
Of course to get to near zero distortion levels, class A operation
would be nice. And once you have class A, the dependance on ultra close matching
disappears, as there is no crossover distortion at all,
the sweet spot is a mile wide, and only when you use extremely high levels
does the distortion begin to raise its ugly head.
In my class A mosfet amps amps there are only two consecutive stages,
the diff pair drive amp, and the mosfets, that is all.
So I cannot see any benefit in reducing the complexity further,
and the NFB is applied in two loops already.
I am using a total of only 23 Db of NFB.
Ro is 0.2 ohms, and distortion is 0.12% at full power,
and it contains no crossover distortion.
So when I hear folks say I must not use global
NFB, I do hear them, but I think they have not
percieved what I am doing, just handing out their solutions
which may have benefits in their circuits,
but it cannot be simplistic universal fix all.
Patrick Turner.
Patrick Turner
the designer of ME products, and not only from our dear Trevor W.
Perhaps, like so many business people, he's just too flat out to
keep his show on the road to be able to spare several hours each
evening answering queries from hi-fi enthusiasts.
The world of the Net is an entirely free and uncensored one, and one which
has no resemblance to any other medium of the past.
With this in mind, there is no need for any one to be upset
my marketting of one particular product, be it ME, or VAF speakers,
or even myself.
I would like to see all other makers come forward to the discussion
and explain their design ideas.
It would be nice to hear from some overseas designers, such as Tim
Paravicini, Allen Wright, Bill Johnson, and so on.
I can understand their reluctance if they thought they might get howled down
by folks who make nothing and market nothing.
Since I am a maker of equipment, and a repairist of equipment,
it is impossible for me to keep the experience of trading
entirely out of the discussion.
It would be nice if suddenly there were a thousand customers
knocking at the door after talking about audio design for awhile.
But let me assure you all that this has not happened.
The discussions are part of the fun one can have with audio,
there is no need for things to be taken absolutely seriously.
Patrick Turner.
Geoff
> knocking at the door after talking about audio design for awhile.
> But let me assure you all that this has not happened.
In my case it's because interest and money are relatively independent of
each other. In fact I probably have more interest when I have no money
because if I'm not buying new stuff then I'm thinking about new stuff ;)
When I do have money I'll be posting a "which amp should I buy" question
rather than trying to start debates on BJTs vs FETs out of curiosity!
Geoff
CHOKY
TW:
**Sure. And I have said as much. In theory, MOSFETs are lower in distortion
than BJTs. That lower distortion, however, requires higher bias currents.
Ironically, at higher bias currents, the negative tempco of gm problem,
disappears, as well. IOW: Class A MOSFET amps, may offer the best of all
possible solutions, for audio amplifiers.
--
Trevor Wilson
http://www.rageaudio.com.au
TW on rec.audio.tech:
> I don't have the time / money to do the same experimenting with a
> Class A amp, though I found on the net a Jean Hiraga designed Class A
> DIY amp (The Monster) that uses a battery smoothed mains PSU and I
> could also run this off a high quality online UPS. I was also thinking
> of adding linear regulators to the supply rails.
**Sounds like a dumb idea, to me. Why not just add a bigger power
transformer and (lots) more filter capacitance?
>
> Put it another way, what gives the most benefit - Class A or Battery
> power?
**Neither.
**********************end of copy***********
Choky: It's logical?
On rec.audio.tech adwice was-"don't mess with A class",and on aus.hi-fi
"maybe is A class best of all".
Notice that I say nothing about classes and what class is best,just seems
that you confront with your self.But ,maybe I am wrong...Maybe you talk thinking
of diff. between BJT and Mosfets..
cheers to all mates!
--
message sent by Choky,
Prodanovic Aleksandar
ch...@EUnet.yu
Yugoslavia
"Trevor Wilson" <tre...@rageaudio.com.au> wrote in message
news:S5m97.75261$Xr6.3...@news-server.bigpond.net.au...
>
"CHOKY" <ch...@eunet.yu> wrote in message news:9k5lth$a94$1...@news.eunet.yu...
>> Somebody talk different things on different places;
>> lets see what is copyed from rec.audio.tech :
>>
>**Not at all. My words are entirely consistent, with my beliefs and
>measurements. If you feel that is not the case, then please feel free >to
>highlight any logical inconsistencies, you may find.
Trevor Wilson
"CHOKY" <ch...@eunet.yu> wrote in message news:9k624v$emk$1...@news.eunet.yu...
> ***********copy***********
>
> TW:
>
> **Sure. And I have said as much. In theory, MOSFETs are lower in
distortion
> than BJTs. That lower distortion, however, requires higher bias currents.
> Ironically, at higher bias currents, the negative tempco of gm problem,
> disappears, as well. IOW: Class A MOSFET amps, may offer the best of all
**I stand by this comment.
>
>
> --
> Trevor Wilson
> http://www.rageaudio.com.au
>
> TW on rec.audio.tech:
>
>
> > I don't have the time / money to do the same experimenting with a
> > Class A amp, though I found on the net a Jean Hiraga designed Class A
> > DIY amp (The Monster) that uses a battery smoothed mains PSU and I
> > could also run this off a high quality online UPS. I was also thinking
> > of adding linear regulators to the supply rails.
>
> **Sounds like a dumb idea, to me. Why not just add a bigger power
> transformer and (lots) more filter capacitance?
**And this one.
>
> >
> > Put it another way, what gives the most benefit - Class A or Battery
> > power?
>
> **Neither.
**And this one.
> **********************end of copy***********
>
> Choky: It's logical?
**Sure.
> On rec.audio.tech adwice was-"don't mess with A class",and on aus.hi-fi
> "maybe is A class best of all".
**Your truncation of a communication, in another forum, does not cover the
entirety of the discussion. IOW: Taken out of context, it is possible to
misquote anyone.
As you well know, the originator of that thread, asked a bunch of incoherent
and illogical questions. My answers were backed up by other, highly
qualified posters. Class A power amps and battery operation are
incompatible.
I have said it many times. Class A amplification is unnnecessary and
wasteful, for audio products (MOSFET, or BJT). In a purely technical sense,
MOSFETs running pure Class A, *MAY* provide the highest level of technical
performance, of any SS device. However, as has been proven many times, the
sound of a PROPERLY designed Class A/B amplifier is indistiguishable from a
similar Class A product.
> Notice that I say nothing about classes and what class is best,just
seems
> that you confront with your self.But ,maybe I am wrong...Maybe you talk
thinking
> of diff. between BJT and Mosfets..
**Your feeble trolls are not welcome. If you want to engage in a meaningful
discussion, then, feel free to contribute. If all you want to do is to
attack my every word, then you will rapidly lose my interest.
Trevor Wilson
> It would be nice to hear something from time to time from Peter Stein,
> the designer of ME products, and not only from our dear Trevor W.
> Perhaps, like so many business people, he's just too flat out to
> keep his show on the road to be able to spare several hours each
> evening answering queries from hi-fi enthusiasts.
**Peter is unable to access this news group. He is also pretty busy. He is
also lousy at promoting his own products. He is WAY too modest.
>
> The world of the Net is an entirely free and uncensored one, and one which
> has no resemblance to any other medium of the past.
> With this in mind, there is no need for any one to be upset
> my marketting of one particular product, be it ME, or VAF speakers,
> or even myself.
**I agree. Provided the promotion is on-topic, of course.
> I would like to see all other makers come forward to the discussion
> and explain their design ideas.
> It would be nice to hear from some overseas designers, such as Tim
> Paravicini, Allen Wright, Bill Johnson, and so on.
> I can understand their reluctance if they thought they might get howled
down
> by folks who make nothing and market nothing.
**Promoting in an Australian NG is not worth their while.
Trevor Wilson
> >**Your words:
> >----
> >There are no severe problems with mosfets, that an intelligent person
> >cannot get around.
> >
> >This suggests you have nothing more to accoplish with MOSFETs. So, I ask
> >again: Why bother with a Class A MOSFET amp? Unless, of course, you agree
> >with me.......
>
mean
> there is not room for improvement in all sorts of ways, in any technology
being
> discussed in the present forum....
**Oh, I agree with that. Nothing is perfect. Patrick has indicated, on
numerous occasions, that his 300 Watt MOSFET amp, IS perfect, though. If it
is perfect then pursuing another technology, is a logical inconsistency. If
it is not perfect, then he should admit that there are other ways of doing
things, which may provide better performance.
Trevor Wilson
> >> Done! and I promise to evaluate it honestly.
> >
> >**I don't believe you.
>
**I can only go by your constant attacks. You clearly don't like what I say,
nor what I stand for. I don't trust anyone who deflects logical arguments,
with personal attacks.
>
> >You are on record as one who has no respect for any
> >amplifier technology, other than the one you build yourself.
>
> Utter rubbish!
**I'll expand that to include any person who supports your belief system,
then. I disagree with your concept of high fidelity. You, therefore, seem to
have no problems with attacking EVERYTHING I say. Although others may
provide a product, which is diametrically opposed to your ideals, you do not
attack them, because they also support some of your ideals.
>
>
> >It is extremely
> >unlikely that you will report honestly and fairly about what you hear
with
> >an ME amplifier. Nonetheless, if the cost to me is not high, I will send
> >you a product, for evaluation.
>
>
> Trevor, I just sent you a nice email explaining that, after some thought,
I
> decided it was unfair to ask you to send your expensive ME amp off to me,
> especially when I can hardly be considered to be a prospective customer...
**As others on this NG can testify, sales are NOT my prime aim. I regularly
loan ME products, to interested parties, fully in the knowledge that a sale
will not be the result. Do not assume I was trying to sell you an amplifier.
>
> but after this little outburst from you...
>
>
> ... the thought comes to me to tell you to get stuffed!
>
**As is your right to do so. I am sorry that you are uncomfortable with my
opinion of you. Your constant insults are bound to cause me to view you is
such a light.
Trevor Wilson
**Me too. It is very simple, however. It is also highly coloured sounding.
As you would expect from such a simple design. I was referring to his Aleph
series of products, which allegedly run Class A (but actually don't). They
use MOSFETs biassed on very heavily. They also sound pretty decent.
>
> And then the CCS dissipates 51 watts too.
> But I use four mosfets which dissipate 25 watts each
> in my class A amp, amp they are happy with that.
> When you put a 2 ohm load on the amp and
> drive it up to clip, the idle current backs off, with the temperature
> rise, real nice it is.
> But when you use the correct load of 5 ohms, and stay in class A,
> there's no temperature rise. There is little compression,
> just the usual inceasing 3H distortion which all amps give.
>
> > .But I bet that the ME amp is configured
> > with an
> > > emitter follower output stage,
> >
> > **Well that would be one bet you would lose. Even more curious,
considering
> > that you claim to have serviced an ME power amp. Did you not notice that
ME
> > amplifiers (and preamplifiers) all operate in Common Emitter
configuration?
>
> I just pulled out my circuit of the "75" amp, and yes, you are correct,The
load
> is connected to the collector circuit. It has been 18 mths
> since I fixed this amp.
> Gee, it has a lot of consecutive stages though, does it not?
**Sure. Each stage is configured to be as linear as it is possible to be.
When a manufacturer aims for low distortion, wide bandwidth, low output
impedance, etc, etc, WITHOUT the use of Global NFB, lots of very linear
stages are generally required. Don't assume that more stages, automatically
translate to poor sound quality, though. An amplifier needs to be as simple
as possible. And no simpler.
> No wonder you don't/can't use global NFB.
> My 300 watter has half the stages.
**Yep. Global NFB makes life much easier. ME has made the commitment to
build a low distortion, wide bandwidth, low output impedance product,
without resorting to Global NFB.
>
> Also, my input amp is similar to ME, with a pair
> of symetrical paralleled differential amps, one with N and one with P
devices,
> and then you have complementary pair gain block, same as
> I have used. I got the ideas from Electronics(Wireless World), and
developed
> them
> for my own use.
**Peter Stein developed that little design, more than 26 years ago.
> But after the VAS, the similarity ends, my amp is just so simple.
**So is the ME. Blindingly simple, in fact. Sadly, it needs to be (slightly)
more complex, due to MEs use of BJTs, rather than the easier and cheaper,
MOSFETs. Nevertheless, ME has found MOSFETs to be unsatisfactory, for a
number of reasons. Reasons, incidentally, that I and thousands of ME
customers agree with.
>
> >
> >
> > >
> > > > Additionally, they provide significant improvements in EM radiation
> > > > rejection (very handy, for us city dwellers), which can enter an
amp,
> > > > through the speaker connections.
> > >
> > > The mains here where I live are appalling with their RF content.When I
> > > constructed mains filters for my C-tick acreditation folder
> > > I found out about RF getting into things.
> > > But none of my amps suffer from RF .
> >
> > **Let's try them across the road from Channel 2, shall we? One of my
clients
> > tried 15 or 20 amplifiers, before he discovered that an ME would operate
> > without the annoying frame buzz, many Gore Hill residents are familiar
with.
> > The only other amplifiers, which could cope with this siuation, were a
> > couple of tube amps. I believe that induced EM radiation, into speaker
> > cables, is one reason why Zero Global NFB amplifiers are preferred by
many
> > listeners.
>
> Well, apart from youself, nobody else reckons RF ingress is a majorproblem
with
> global NFB.
**Untrue. Many manufacturers recognise it as a problem, hence their drive to
eliminate Global NFB where possible.
The majority of low grade Consumer
> electronics rely on Global NFB, no probs.
**Nope. There are serious problems with many consumer grade products. When I
lived with my parents, in an outer Sydney suburb, I noted a curious beep, in
my hi fi, at regular intervals (that is, AFTER I stopped using tubed
equipment). I couldn't figure out what it was, until I heard the same 'beep;
when I drove past Kingsford-Smith Airport. It was one of the long range
radars. The particular radar employed an ERP of around 1MWatt and used a
modulation frequency of about 600Hz. At my previous home, I was in a direct,
uninterupted line of sight of the radar. I tried many different amplifiers.
All exhibited the same fault (except the tubed ones). I finally managed to
eliminate the problem, by using shielded speaker cables. Substituting an ME
amplifier achieved the same result (albeit at a higher cost).
> If you are getting RF bothers in problem areas, fit RF filters, simple.
**Sure. If you know that it the problem, of course. The risk is that
sometimes RF filters can interfere with sound quality.
**Then I would be interested in hearing one.
**Correct. And that is why ME use Common Emitter configuration, for their
output stages. To enable complete isolation, between load and drive. It
allows the Voltage amplifier to 'see' a purely resistive load. That way, the
Voltage amp can be made highly linear, with ease.
>
> Of course to get to near zero distortion levels, class A operation
> would be nice. And once you have class A, the dependance on ultra close
matching
>
> disappears, as there is no crossover distortion at all,
> the sweet spot is a mile wide, and only when you use extremely high levels
> does the distortion begin to raise its ugly head.
**Sure, but it is MUCH better for the environment and the end user's pocket,
to match components, thus obviating the need for Class A operation.
>
> In my class A mosfet amps amps there are only two consecutive stages,
> the diff pair drive amp, and the mosfets, that is all.
> So I cannot see any benefit in reducing the complexity further,
> and the NFB is applied in two loops already.
> I am using a total of only 23 Db of NFB.
> Ro is 0.2 ohms, and distortion is 0.12% at full power,
> and it contains no crossover distortion.
> So when I hear folks say I must not use global
> NFB, I do hear them, but I think they have not
> percieved what I am doing, just handing out their solutions
> which may have benefits in their circuits,
> but it cannot be simplistic universal fix all.
**You need to approach amplifier design, in the way you see appropriate for
you.
CHOKY
interpret of true)
on some poor DIY's question,that DIY are not seller or buyer primary.
That is person really interested in own hobby.
That poor DIY fellow asked simple q.-"A class without or AB class with battery
supply-what is better?"
You just tell him that both are not good.
Qualified posters? He is not trying to find father on net-just asked
simple question,and what he got?
" Nah.your questions are dumb!"
You say in that way that Jean Hiraga is dumb,also.
You don't respect DIY,because you see that some other people build
better equipment than you and stopped building stuff,but you still patronize
people with your facts.Knowledge is not enough ,you lost passion; and DIY is
passion.
When you rejuvenate your passion,then you will not be so serious about
hifi stuff.
And-when I have nothing to add in some thread,I don't try.
And you finally lost my interest.
--
message sent by Choky,
Prodanovic Aleksandar
ch...@EUnet.yu
Yugoslavia
"Trevor Wilson" <tre...@rageaudio.com.au> wrote in message
Trevor Wilson
"CHOKY" <ch...@eunet.yu> wrote in message news:9k78sb$sbv$1...@news.eunet.yu...
> As you wish-but remember ,when you give answer (not only
your
> interpret of true)
> on some poor DIY's question,that DIY are not seller or buyer primary.
> That is person really interested in own hobby.
> That poor DIY fellow asked simple q.-"A class without or AB class with
battery
> supply-what is better?"
> You just tell him that both are not good.
**My answers were consistent with logic and reason. If you disagree, or have
an alternative suggestion, then present it. Neither idea was any good and I
provided the reasons why. Battery operation is wasteful and inefficient.
Modern regulators are a much better approach (if necessary). Class A
operation is wasteful and unecessary.
> Qualified posters? He is not trying to find father on net-just
asked
> simple question,and what he got?
> " Nah.your questions are dumb!"
**Liar! Please provide an EXACT quote from me, where I stated such a thing.
His questions were based on ill-informed knowledge. With the exception your
idiotic trolls, few questions are dumb.
> You say in that way that Jean Hiraga is dumb,also.
**Another lie. Please provide an exact quote, where I stated such a thing. I
do not know Jean Hiraga, nor am I aware of that person's ideas.
> You don't respect DIY,because you see that some other people
build
> better equipment than you and stopped building stuff,but you still
patronize
> people with your facts.Knowledge is not enough ,you lost passion; and DIY
is
> passion.
**Not at all. I began my quest in hi fi as a DIY enthusiast. I still perform
DIY stuff. The original poster asked some questions (presumably, because
they lacked the requisite knowledge) and I asnwered accordingly. My answers
were backed up by two other, highly qualified individuals. What problems do
you have with their answers?
Are you questioning them, as well?
Or do you imagine that I am a 'softer' target?
Why do you attack me and not Stewart Pinkerton and Richard Pierce?
**Your continual trolling is becoming tiresome. If you have a point to make,
then do so.
> When you rejuvenate your passion,then you will not be so serious
about
> hifi stuff.
> And-when I have nothing to add in some thread,I don't try.
> And you finally lost my interest.
**Don't you mean: "I'll stop trolling, when others see me for what I really
am."?
I am always ready and willing to discuss the facts. If you want to continue
attacking me, personally, then I will respond in kind, or kill-file you,
when I become bored with your ridiculous attacks.
Geoff
regularly
> loan ME products, to interested parties, fully in the knowledge that a
sale
> will not be the result. Do not assume I was trying to sell you an
amplifier.
Yep. Trev's very generous with his equipment.
Trevor
>
> "Trevor" <trevor@home> wrote in message
> news:3b6660fb$0$20909$7f31...@news01.syd.optusnet.com.au...
changes?
>
> **The silicon can warm up, very quickly. I don't have the data, but I
would
> think that due to it's inherent protection mechanism, the temperature rise
> would be in the order of milliseconds. I'll need to do some homework on
this
> one.
>
> > What frequency would it be expected to track the input signal. I have
seen
> > no evidence that it can track a 20 Hz signal or above. It seems you hear
a
> > difference (which may be true or not for your samples) but fail to prove
> > that it is caused by NTC as you repeatedly claim.
>
> **I accept that I have not provided adequate proof. More homework.
I am very impressed by the honesty of your reply. I will be interested in
the outcome of your investigations.
> **Oh, absolutely! On hearing the early Perreaux and ETI 5000 amplifiers,
one
> could easily disregard MOSFET amplifiers, as a complete joke. More
> sophisticated designs, which I am familiar with (EHT, later ETI designs,
> Redgum, Pass, Spectral and others) demonstrate significant sonic
> improvements, to those early designs.
Very true. When cost is a consideration (as it usually is) personal
preference as to what compromises can be made, will vary. Fortunately the
differences for any competent amplifier design will be very much smaller
than with speaker design. That doesn't stop us from arguing relative merit
though I guess, and our knowledge increases.
Trevor.
DeutschMick
>**Peter is unable to access this news group. He is also pretty busy. He is
>also lousy at promoting his own products. He is WAY too modest.
Having met the man once or twice in a store I worked in,
and at a HiFi show, I can support this statement.
If Peter's hyperbole was proportional to the product quality
his firm seems to deliver, versus what the marketing machines
of many other companies continually thrust at us, he would
probably never shut up.
People would queue up at his door at midnight on the release night
of a new product, without even knowing what it was, which allegedly
happened when a particular software firm released another new, and
fraudulent version of their software in 1995 ** see below:-(
And thus we have Trevor, who would appear to hold a high opinion
of the man and his products, and is more than happy to tell us
about it.
Don't we then all stand a better chance of making an informed decision?
Better 'marketing' from someone who gives a stuff, and actually seems
to understand its design, than all the cynical crap in the media, from
utter ignoramuses, IMHO.
This applies to you too, Patrick.
You're both a lot more interesting than an amplifier ad in a
mag, or on TV, and a little more likely to tell us something
we can USE :-)
<soapbox_mode>
Marketing at its best . . .
** xxxx is a complete rewrite from the ground up.
- Caught Intel out big time with the PentiumPro, didn't it?
and
** "xxxx is the first real 32 bit operating system for the PC.",
THIS one's the best cynical marketing statement I have ever seen!
(taken from their in-house magazine, which I have kept for posterity.)
How often do you see 3 lies in one sentence? Proof?
* It wasn't a "REAL" OS, but sat on top of something else, and
was only thinly disguised.
* It wasn't 32 bit if you looked under the bonnet, which
is why 32-bit optimised Pentium Pros didn't perform at
all well with it.
* and .. It wasn't first either, OS/2, Linux and other *nix
variants were 32 bit earlier, in some cases 5 YEARS earlier.
And still they come . . .
</soapbox_mode>
Regards: Mick Deutsch, Project Audio, Canberra, AUSTRALIA.
/
/ - __0 m$ts...@pcug.org.zzz, mick.$ts...@hic.gov.au
[@\ _ \<,_ Searching for great sound! NO DAVE! I'm not
[@/ (_) / (_) Wind1900/XP compatible, this is the 21st century!
\ replace z's with au, currency symbol with deu
\
Patrick Turner
Trevor Wilson wrote:
> "Preamp" <pre...@spamblock.dingoblue.net.au> wrote in message
> news:1pdamtspohckv9glk...@4ax.com...
> > >**Your words:
> > >----
> > >There are no severe problems with mosfets, that an intelligent person
> > >cannot get around.
> > >----
> > >
> > >This suggests you have nothing more to accoplish with MOSFETs. So, I ask
> > >again: Why bother with a Class A MOSFET amp? Unless, of course, you agree
> > >with me.......
> >
> > Quite obviously because, although they can be easily got around, that does
> mean
> > there is not room for improvement in all sorts of ways, in any technology
> being
> > discussed in the present forum....
>
> **Oh, I agree with that. Nothing is perfect. Patrick has indicated, on
> numerous occasions, that his 300 Watt MOSFET amp, IS perfect, though.
Whoa, there boy, Steady on, old chap, When did I ever sayone of my creations was
perfect? I have often said that nothing is
perfect.
> If it
> is perfect then pursuing another technology, is a logical inconsistency. If
> it is not perfect, then he should admit that there are other ways of doing
> things, which may provide better performance.
I doubt my 300 watt class AB mosfet amp is perfect.It sure is not as imperfect
as a lot of other things it may be compared with.
And even if I were to achieve the bloomin impossible, and create
a peice of perfection, it does not mean that following alteranive endeavours
are inconsistent, or illogical, or a waste of time.
Perhaps there is some warping of your logic.
You may recal that you posted a few weeks ago that you would be disgusted if ME
brought out a line of tube amps.
But McIntosh and Quad are still making tube gear, as they have a flexible
attitude towards the desires of their fans and customers.
Why be so locked up with only one way to achieve something?
And I have always been intersested in other designs and ways of doing things.
After absorbing the circuit of the ME75, at this point in time,
I doubt I would ever try the circuit ideas used.
Patrick Turner.
Trevor
> Trevor Wilson wrote:
> > If it
> > is perfect then pursuing another technology, is a logical inconsistency.
If
> > it is not perfect, then he should admit that there are other ways of
doing
> > things, which may provide better performance.
Assuming for a moment it was perfect (which I don't believe he claimed) then
would it not be logical to try and achieve the same thing at a lower cost?
No logical inconsistency that I can see.
> And even if I were to achieve the bloomin impossible, and create
> a peice of perfection, it does not mean that following alteranive
endeavours
> are inconsistent, or illogical, or a waste of time.
> Perhaps there is some warping of your logic.
Agreed, there may be more than one way to achieve "perfection". More likely
none at all, but that's another argument :-)
Trevor.
Trevor Wilson
"Patrick Turner" <in...@turneraudio.com.au> wrote in message
news:3B67ACCD...@turneraudio.com.au...
>
>
**That was a little disingenuous of me. Let me put it another way: What
possible improvements do you imagine will accrue from a transformer coupled,
Class A, MOSFET amplifier? Presumably, you are working towards some kind of
goal, with a theory to back it up?
>
> > If it
> > is perfect then pursuing another technology, is a logical inconsistency.
If
> > it is not perfect, then he should admit that there are other ways of
doing
> > things, which may provide better performance.
>
> I doubt my 300 watt class AB mosfet amp is perfect.It sure is not as
imperfect
> as a lot of other things it may be compared with.
**Very probably.
>
> And even if I were to achieve the bloomin impossible, and create
> a peice of perfection, it does not mean that following alteranive
endeavours
> are inconsistent, or illogical, or a waste of time.
> Perhaps there is some warping of your logic.
**Perhaps.
>
> You may recal that you posted a few weeks ago that you would be disgusted
if ME
> brought out a line of tube amps.
> But McIntosh and Quad are still making tube gear, as they have a flexible
> attitude towards the desires of their fans and customers.
**It could certainly be argued that Quad are only in it for the money. They
dropped out of tube amp manufacture, some 40 years ago, only to ressurrect
the design rather recently. That smacks of commerical oportunism, rather
than any kind of honesty. Audio Research do much the same thing, but, to
their credit, they never stopped building tube products.
> Why be so locked up with only one way to achieve something?
**If a designer has the conviction that he has got it right, then why not?
>
> And I have always been intersested in other designs and ways of doing
things.
>
> After absorbing the circuit of the ME75, at this point in time,
> I doubt I would ever try the circuit ideas used.
**I don't blame you. To do it right, requires and enormous investment. To
match enough output devices, to build one amplfier, for instance, requires
the purchase of several thousand transistors. It ain't cheap, until you are
building lots of amps. Fortunately, ME does build lots of amps.
Trevor Wilson
"Trevor" <trevor@home> wrote in message
>
> "Patrick Turner" <in...@turneraudio.com.au> wrote in message
> news:3B67ACCD...@turneraudio.com.au...
> > Trevor Wilson wrote:
> > > If it
> > > is perfect then pursuing another technology, is a logical
inconsistency.
> If
> > > it is not perfect, then he should admit that there are other ways of
> doing
> > > things, which may provide better performance.
>
> Assuming for a moment it was perfect (which I don't believe he claimed)
then
> would it not be logical to try and achieve the same thing at a lower cost?
> No logical inconsistency that I can see.
**Agreed. I have admitted my misdeed to Patrick. However, I very much doubt
that Patrick can build a 50 Watt, transformer coupled, Class A, MOSFET
amplifier, for less than a 300 Watt conventional design. I'm sure Patrick
can confirm or deny this comment.
Geoff
> difference?
Hey let me try....
You're an arsehole!
yep, sounds like a personal attack to me. And I meant it too, ya dickhead
;)
hehe.
Geoff
Geoff
> of the man and his products, and is more than happy to tell us
> about it.
Yep. unfortunately it sometimes comes across as if he is trying to push it
as the best option, ie "why would you consider anything else", rather than
pushing it as equipment that sounds amazing at an affordable (well, almost,
since i'm still a student) price...
> Don't we then all stand a better chance of making an informed decision?
Yep. I wouldn't have known about ME without this group, so it's another
product I know about when the time comes to buy a new amp.
> Better 'marketing' from someone who gives a stuff, and actually seems
> to understand its design, than all the cynical crap in the media, from
> utter ignoramuses, IMHO.
And someone who will instantly and truthfully answer questions about the
product even down to the circuit level, which is a hell of a lot more than
you'll find from 99% of other companies.
> This applies to you too, Patrick.
Uh oh, you're bring singled out. Run! Hide! :)
> You're both a lot more interesting than an amplifier ad in a
> mag, or on TV, and a little more likely to tell us something
> we can USE :-)
Even if we know what your answer is before you speak it ;)
Geoff
Geoff
> preference as to what compromises can be made, will vary. Fortunately the
> differences for any competent amplifier design will be very much smaller
> than with speaker design. That doesn't stop us from arguing relative merit
> though I guess, and our knowledge increases.
Every bit helps. And it may well help explain those elusive, subtle
differences we think we can hear but somehow can't quantify (like when I
first realised what a good speaker sounded like, and I could pick a speaker
that sounded great but if somebody asked me to listen to their crap one I
couldn't pin anything specific on it other than it simply sounded
ordinary...)
Geoff
>
> **If a designer has the conviction that he has got it right, then why not?
Narrow-mindedness is always a bad idea. Sure, you need to be able to focus
on a concept and throw all your weight behind it in an attempt to make it
"perfect" (where perfect means optimal). But if your mind is closed to
other options then you're living in your own version of reality.
And if you're on the wrong track, you won't realise it for a LONG time,
compared with an open minded person who looks at something else, looks at
their own... hang on a minute... looks at something else, looks at their
own... shit, that was silly of me!
Also, the more options you consider in your quest, the more certain you can
be that your way is the right way. But don't expect other people to take
your word for it, they won't. But at least you know in your own mind what
is right, and that is all that matters. Except imparting knowledge on
others, if and only if they are receptive and want to hear it.
That was a tad philosopical for a wednesday, I'd like to balance this post
by saying you all suck! :)
Geoff
Geoff
screw you, hippy!
Geoff
> >
> >screw you, hippy!
>
aw, nuts :P
Rick Stadelmaier
Thats probably the most accurate thing anyone has said to him :)
Rick.
--
Rick Stadelmaier
Equinox Audio
http://www.equinoxaudio.com.au
equ...@pip.com.au
Sydney, Australia
Patrick Turner
Trevor Wilson wrote:
Maybe the sound will be excellent, approaching perfection.Taking all things into
account, and locking all the bean counters out
of my shed, I can achieve very good fidelity with OPT, mosfets,
and Class A, and not much total NFB, and a whole lot of
serenity from making something unique which nobody else has attempted.
There are some common attributes between tube PP amps and the
mosfet amp. Tube amps, as you know, have OPT, cap coupling,
and all these percieved problems, yet there are so many good ones around.
So, if you build an SS amp with mosfets, similar to a tube amp,
it should sound well too. And mosfets are just like
pentodes which have spent their life at the gym, with huge muscles,
ie, about 100 times the Gm, and yet can be operated by applying a gate voltage,
which is as easy as driving a triode or pentode grid,
except that only 1 vrms is required for drive.
But methinks one can say, as a manufacturer, that for tubes, we have this
design,and for SS, we have that design, and so on, this shows a maker is
versatile,
and enjoys diversity. It would not be easy to offer many products
to a niche market, as diversity costs more, a company has to run two
companies within one, if they decide to do tube gear, and not only SS.
There can be "rightness" for whatever devices are chosen, and hopefully
the prospect of making dollars drives such commercial endeavours, not just
fascination, which drives DIY.
I am always on the lookout for what is a commercial oportunity,
just like yourself, and such oportunism is a healthy attitude.
With Quad, they have been around for so, so long,
it would seem such a pity to see them break completely from the past.
I somehow also believe I could have made their new 40 watt
mono blocs slightly better, but I was not ever going to be chosen to do that
design.
That really would have been an honour.
It would be like being asked to go out and design a new E-Type Jaguar,
not that I lust for one, but making anything that is crafted is
fascinating, for the craftsman, and hopefully for the shareholders.
> > And I have always been intersested in other designs and ways of doing
> things.
> >
> > After absorbing the circuit of the ME75, at this point in time,
> > I doubt I would ever try the circuit ideas used.
>
> **I don't blame you. To do it right, requires and enormous investment. To
> match enough output devices, to build one amplfier, for instance, requires
> the purchase of several thousand transistors. It ain't cheap, until you are
> building lots of amps. Fortunately, ME does build lots of amps.
I am unfamiliar with accurate matching of transistors,and there is more to that
than you say, and I doubt ME pays as
much for matched devices compared to what I pay at DSE,
or Jaycar, where I may be purchasing what is left over after
the close matched ones have been sold elsewhere.
It is not a big deal. The use of matched devices is commendable,
but alone it is not all that is required for fidelity with SS designs,
IMHO.
Patrick Turner.
Patrick Turner
Trevor Wilson wrote:
Well, There are several ways to achieve 50 watts of class A, with mosfetsor
BJTs.
All will mean having a heatsink about the same size as a 300 watt amp.
I purchased my 300 watt Al cast heatsinks from Clerke and Serverne at Hornsby.
The size is 300mm long, 150mm high, and there are 30 fins about 40mm long
To make 50 watts class A you will have to have 110 watts of draw from the
PSU, and all this heat will be converted to heat, with no signal.
But for a two channel system, it is still only 220 watts, equivalent
to a few light bulbs, and not a big deal.
The amp a 50 watt class A ability will need a good PSU,
and with todays knowledge, there is no reason why a 220 watt
lightweight switchmode PSU could not be employed,
which would be cheaper to make than a conventional arrangement,
and it has been done with other makers now as a matter of routine.
But for one off designs, the total effort to make a monoblock
50 watt class A design would be about the same as a 300 watt
class AB, providing a complementary output stage is used.
OPT inclusion is not for everyone, but it is the only extra major thing,
along with the stiffer bigger box.
I thought of innitially converting my stereo class AB mosfet amp to pure class
A,
and figured it could be done with exactly the same circuit, but with
plus and minus 25 volt rails instead of 70 volt rails, and just increasing the
bias
through the 6 mosfets to about 0.7 amps each.
I would then get about 16 vrms into a load of 5.3 ohms,
which is 48 watts, of pure class A.
The power transformer I wound for this amp is able to allow some
reconfiguration.
My experience with complementry class A pairs of n and p type
mosfets meant that I was able to easily make an amp using source
follower mode, and only get 1% of distortion at 25 watts from just two devices,
with no other type of NFB, besides follower action.
The nature of the resulting amp is about equivalent to a good triode PP amp,
in technical terms of Ro and N&D.
However due to conditions for good linearity and a decent voltage swing,
and load tolerance, more than 2 mosfets (or BJTs ) are required,
and 6 would be a nice number.
With more devices, the distortion can be reduced to 0.3%,
without any global NFB.
However, some global may be needed to help reduce noise,
which will be the reason why some global may have to be used.
The bandwidth with a simple drive amp can be much wider than is required,
and by careful critical damping within the amp it can be stabilised
easily, yet deliver a blameless HF response.
The fact that n and p devices are not matched in class A means that
the THD contains a bit of 2H distortion, and not just the dominant 3H.
At low levels of operation the THD declines towards zero, and crossover
distortion just does not occur, so you do not need so much extra global NFB,
just to reduce distortion.
Since mosfets are cheap and available, I suspect the high current
types of Exicon mosfets would be good, and because flat pack
types are easy to bolt onto a large heatsink, multiple devices are no trouble,
and may as well be used.
There are limitations with complementary pairs for load matching,
and as the load value declines from the ideal class A load, the amp
works in more Class AB, which increases distortion a lot.
With my OPT, I can re jig the OPT secondaries so that 1.25 ohms can be driven,
with the same winding losses of about 5% as when matched to the present 5 ohms,
and the power is still class A.
With an OPT, things become very difficult for the average home constructor,
as there are about 10 elements in the OPT design which have to be weighed up
before a worthwhile design can be wound.
If you want that beautiful creamy class A sound, then a complementary
pair output will suffice.
I see no reason why BJT or mosfets would not be able to deliver,
depending on the approach taken with implementation.
I hope the explanation of some of these salient points of amplifier
design answer some of your questions.
Trevor Wilson
>
> I am unfamiliar with accurate matching of transistors,and there is more to
that
> than you say, and I doubt ME pays as
> much for matched devices compared to what I pay at DSE,
> or Jaycar, where I may be purchasing what is left over after
> the close matched ones have been sold elsewhere.
**ME either purchase direct from manufacturers, or from their agents. Buying
retail, is a last resort. In any case, ALL component matching is performed
in-house. It is done at realistic currents and at normal amplifier operating
temperatures. No semiconductor manufacturer would be prepared to do this.
> It is not a big deal. The use of matched devices is commendable,
> but alone it is not all that is required for fidelity with SS designs,
> IMHO.
**I agree. A good topology is vital. However, for correct functioning of an
ME amplifier, matched transistors MUST be used. That way, crossover
distortion problems, commonly associated with Class A/B amplifiers, is
eliminated. Further, no muting system is required, with an ME amplifier, as
there are no turn on, turn off thumps, due to the almost perfect DC
conditions, present in the amplifier.
remarkabel
Patrick Turner wrote:
It was my understanding that there are still major limitations on switchmode PSU's
in audio applications. Am I off beam here? Also, are there really significant
savings to be had by using them?
remarkabel
a broader view on things you are less focussed on what your core
values/business/whatever is. This is turn means it is less likely that you will
achieve the level of 'perfection' you strive for. In this case Peter Stein has
obviously made a (very) educated decision on what is the road to 'perfection'
for his company and products. I think it is clear from the success ME have
enjoyed over a quarter of a century that Peter Stein has not taken "the wrong
track". I would also be surprised if he does not regular assess competitors
products and technology to ensure he is still where he aims to be.
Cheers,
Mark
Geoff
what he has about how to do things right. If he is narrow-minded now
because it helps him concentrate then it seems to be working but it's quite
possible that one day he'll start going in the wrong direction. However I
doubt this is the case, since as you say he evaluates the other products on
the market and is obviously outward-looking to make sure his product is
always as good as it can be relative to alternatives.
However, I'll make you read a paragraph again to clarify my post:
> > Narrow-mindedness is always a bad idea. Sure, you need to be able to
focus
> > on a concept and throw all your weight behind it in an attempt to make
it
> > "perfect" (where perfect means optimal). But if your mind is closed to
> > other options then you're living in your own version of reality.
There is a distinction between being able to focus, and being narrow-minded.
People who are narrow-minded don't accept that there is another solution to
the problem, they believe they have the answer and to hell with the rest.
Pretty foolish, I think you'll agree. You need to be able to give 100% on
your concept of the right solution, without ignoring what's happening around
you. Sound tough? Well life was never meant to be easy, especially if you
want to be successful :)
Geoff
"remarkabel" <mark...@bit.net.au> wrote in message
news:3B688165...@bit.net.au...