Tube with Power IC Output Stage - JLTi

[Joe Rasmussen]

Hi Dimitri

Tried the link you gave but it no longer seems
online?

I hear what you say, but to include the tube
stage is not practicable if you look closely at
the circuit. The buffer (unity gain) has no way
of easily accepting feedback.

I am loath to use a triode gain stage as there
is already too much feedback - I want to go LESS
not more. Also I've avoided 'warming up' the
sound - and adding a stage that increases 2nd
harmonic may well sweeten things up (and other
tricks). Actually lessening feedback did the
job for me and got the BALANCED result I was
seeking.

The circuit I've developed involved a lot of
pros and cons - and I'm happy where it stands.
It cannot NOW be easily changed. As for using
feedback to correct any anomaly I have to say
I'm not in favour of it. I suspect very much that
the basic goodness starts from the open loop
characteristics of the 'clone chip, and that
feedback basically - being an opamp - is a way
of setting gain.

As for further developments... for sure, but I
don't want to start from scratch but rather build
on the foundation I've got.

----------------------------------

Now for a different tack:

To both of you, Dimitri and Dave (and others).

The inverted 'clone circuit that Thorsten
supplied has served the DIY fraternity well. If
I may, and hopefully Thorsten won't mind too much,
I am going to post a very reconsidered version
- not tube - of the inverted clone. I will
prepare this thoughtfully over the next week or
two. It will include one interesting area that
can be tweaked, a kind of voicing mechanism.
They key is the way feedback is applied and also,
my main concern with Thorten's, the stablising of
the feedback loop as the level is changed.

It will incorporate some of the benefits of my now
commercial (not DIY) JLTi.

There will be full diagrams and component values
and suggestions on which can be tweaked. I am also
going to do some "Lynn Olson first 1W" type
harmonic distortion measurements (these are actual
dist/freq graphs).

I would like to see others try this new version
(can somebody think of a name to give it?) and
get feedback - we might be able to give this
whole topic a bit of a new boost.

It should be fun!

Stay posted.

Joe

PS: James Taylor (of Fire and Rain fame) is here
in Sydney at the moment. I heard him yesrerday
say, on a local radio station: "War is failure,
not a strategy."

[planet10]

Originally posted by Joe Rasmussen
Tried the link you gave but it no longer seems online?

Dimitri had mistyped the url... it is fixed now.

dave

[planet10]

Originally posted by Joe Rasmussen

I am going to post a very reconsidered version - not tube - of the inverted clone.

I look forward to that... my chips are still in there little plastic bags

"War is failure, not a strategy."

I heard that on the radio recently too... i'm not exactly positive who said it, by i think it was Jacques Chirac (sp?). (and i have to agree with the sentiment)

Nelson Mandala also had some not-nice things to say about W.

dave

[dimitri]

Hi, Joe
--------
... include the tube stage is not practicable if you look closely at the circuit. The buffer (unity gain) has no way of easily accepting feedback.
--------
You can reconnect from the follower to common cathode stage

-----------
I am loath to use a triode gain stage as there is already too much feedback - I want to go LESS
not more. Also I've avoided 'warming up' the sound - and adding a stage that increases 2nd
harmonic may well sweeten things up (and other tricks). Actually lessening feedback did the
job for me and got the BALANCED result I was seeking.
-------------
I could not get how you make LESS feedback, than in typical application from data sheet. What feedback makes wrong to you (and your gainclone)?
It 'warming up' the sound was not your aim why have you used the tube? For marketing purposes, just to add that nice word in your ad?

[ThorstenL]

Hi Joe,

(Or should that "Hey Joe"?)

> I am loath to use a triode gain stage as there
> is already too much feedback - I want to go LESS
> not more.

Actually, I feel the reverse is true. Around the Power Op-Amp you need a lot of negative feedback. Anyway, views vary.

The inverted 'clone circuit that Thorsten
supplied has served the DIY fraternity well. If
I may, and hopefully Thorsten won't mind too much,
I am going to post a very reconsidered version
- not tube - of the inverted clone.

Hell, the infamous "inverted gainclone" was simply the result of someone on the Gainclone board asking "how do I build an inverting Amp". I also kept a lot of considered "belt & braces" in there, which I personally would do without and designed in the volume control as fundamentally neccesary and stabilising element.

There are many things I'd optimise on that puppy myself. It was never meant as a suggestion of "crushing superiority - maximum tweaked out and freaked out eschaton of chip amp's". It was simply a quick slap-dash for people who wanted to try the difference between inverting and non inverting Amp's.

Tweak away to your hearts content.

Sayonara

[halojoy]

Originally posted by Kuei Yang Wang
Hi Joe,
(Or should that be "Hey Joe"?)
---------
There are many things I'd optimise on that puppy myself.
------
It was simply a quick slap-dash for people who wanted to try the difference between inverting and non inverting Amp's.
Tweak away to your hearts content.

Sayonara

That was what I thought.
The hype of your Gainclone is not much to bother with.
This wouldn't not be your own first choice of gear.
Probably not even the 10th (tenth).

So, now all guys out there ..
you are free to jump on the next "hype of the day"

/halo - sittin' on a fence - watching the boys jumping around ....

------------------------------------------------------------------
Sittin'On A Fence (Jagger/Richards) Rolling Stones

Since I was young I've been very hard to please
And I don't know wrong from right
But there is one thing I could never understand
Some of the sick things that a girl does to a man, so

I'm just sittin' on a fence
You can say I got no sense
Trying to make up my mind
Really is too horrifying
So I'm sittin on a fence

All of my friends at school grew up and settled down
And they mortgaged up their lives
One things not said too much, but I think it's true
They just get married cause there's nothing else to do

The day can come when you get old and sick and tired of life
You just never realize
Maybe the choice you made wasn't really right
But you go out and you don't come back at night, so

I'm just sittin' on a fence
You can say I got no sense
Trying to make up my mind
Really is too horrifying
So I'm sittin on a fence

[Joe Rasmussen]

Hi Guys

A number of things you guys have queried:

Kuei,

You are quite correct, I have read about a good
number of tweaks, but in making the JLTi I didn't
consider it another tweak but basically set out a
whole program development - a quite serious under-
taking, not to say time consuming. It lead to
some things I would not mind passing on to the
DIY fraternity.

About Thorsten's 'suggested' way of inverting
the 'clone (which he argues that it is no longer
a clone), I don't even know if he ever actually
did put one together? Pretty sure he didn't do at
the time... it was, as Allen Wright would say, not
a "paint by the numbers" project. And I DO agree
that inverted circuits sound better - when S/N is
no probs.

Don't prejudge... wait and see what I will offer.


halojoy,

> The hype of your Gainclone is not much to bother
with. This wouldn't not be your own first choice
of gear. Probably not even the 10th (tenth).

Six months ago I would have agreed with you. Those
who know me will tell you I am not, basically
speaking, into feedback (series or shunt) amps.

Oh BTW, JLTi stands for "Just Listen To It" .

Why? It's a challenge to keep an open mind. :-)

It challenged me!

I make something called an Isolated Loop Amplifier
which took about nine years to develop to the
stage I am now. It's a tube amp that is completely
isolated in an AC away from the power supply, it
literally 'floats' and amplifies the signal as a
contimuous loop - break the loop and no sound. It
is AC referenced to ground but and AC signal can
be injected into the loop because it IS grounded
with respect to DC - hard to grasp I know. As it
amplifies only as a loop, it doesn't need a phase
splitter and it only works in Class A as anything
that interupts the loop means NO sound. So when it
clips, it clips in Class A, just as SE does, but
it isn't SE (the one thing everyone agrees with)
but is it PP? I am being attacked by some of the
new "religious right" - which I humourously refer
to the more blinkered of the SET brigade.

The "RR" says it is PP, so what? How does it
sound, like SE? Like PP? Actually it has the
better properties of both, but it SOUNDS more
like an SE than PP.

It's big, it's heavy and it's expensive

(Unlike the JLTi).

AND NO FEEDBACK!

So, down here in Oz the battle lines are drawn.
And I say I don't care. Hey, I like SETs
too. I like other things too!!!!

Re 'clones:

The fact is that here is something that was
brought to my attention and I was dragged into it
by a friend... and one thing led to another. If
you get it right (buffer the input) this thing can
sound awfully good, I mean REALLY awfully good.

Surprised me? Betcha!

But then I'm no "RR" - I can explore new things.

Dimitri,

>I could not get how you make LESS feedback, than in typical application from data sheet. What
feedback makes wrong to you (and your gainclone)?

I will cover that soon. Just as a foretaste, I use
1 Meg feedback resistors in the JLTi.

But for starters, my CD (and SACD) player only
outputs 0.7V RMS, so I simply apply LESS feedback
to get more gain. What output does the DACkit
have, about 1V? Could use more gain. Thorsten's
circuit drops the gain down to 17dB when pot is
in the mid position, is this good? Me thinks... I
think you know what I think.

>It 'warming up' the sound was not your aim why have you used the tube? For marketing purposes, just to add that nice word in your ad?

The answer is simple: IT MAKES IT SOUND BETTER!!!
Over the years I have worked on tube equipment
used in studios, even designed and made from
scratch like the JLTi, and mixing tubes with SS
can make a huge difference, not just 'warming up'
but adds a 'fleshing out' of tonal colours that
is more listenable, yet NOT less detailed, but
often MAKES the detail - listen to a line feed in
a top notch studio and you might just freak - more
acceptable to the ear (phsyco acoustic?). Tubes
can also add 'drive' ... and so much more. It's
about making music.

You may wish to read the following article, even
though it's a work in progress:

http://members.ozemail.com.au/~joera...gainclones.htm

That's it for now, gotta go cut the grass and got
friends coming over here for a BBQ. Yesterdays'
weather was 35C, the two days before that 42C and
45C respectively. Wonder if I have enough beer?

Regards

Joe Rasmussen

www.vacuumstate.com & http://members.ozemail.com.au/~joeras

[Joe Rasmussen]

Oops!

My turn to make a corrction.

The line that read:

"It is AC referenced to ground but AC signal
can be injected into the loop because it IS
grounded with respect to DC - hard to grasp I
know."

Should have read:

"It is NOT AC referenced to ground but
and AC signal can be injected into the loop
because it IS grounded with respect to DC
- hard to grasp I know."

Joe

[ThorstenL]

Hi,

About Thorsten's 'suggested' way of inverting
the 'clone (which he argues that it is no longer
a clone), I don't even know if he ever actually
did put one together?

I know for a fact (as evil alter ego I should) that Thorsten did spend about four years in the 1980's buiding for fun various versions of active speakers using the A2030 (pretty much a close copy of the TDA2030). Things played with then included noninverting/inverting, current/voltage feed, feedback around drivers, bridging with current dumper transistors and a bunch of other things that I can't remember as they resulted in little audible differences. Oh yes, I also worked the few stabilising methodes to avoid the common RL output combo.

I messed again with chip amp's in the middle of the 1990's (must have been more or less around the time as Final & Kimura started on this).

My KEY point with the IVGC was to note that common style Op-Amp's ALLWAYS sound better inverted and the worse the Op-Amp, the bigger the effect. Most power op-amps including the LM1875/3875 are quite dreadful when compared to their best low level companions, so any help they can get is welcome.

In the end I do agree with Nelson, namely that a Valve Amp is a Valve Amp is a Valve Amp is a Valve Amp. And nothing else is.

Still, given the effort and easy experimentation I do very sincerely and honestly recommend messing aboiut with the better chip amps' to each and all, done right they sound better than they have any right to and in monetary terms building halve dozend different versions leaves still change enough for a few CD's and a player for a big one ($ 1,000).

Sayonara

[Joe Rasmussen]

Hi Kuei

>I know for a fact (as evil alter ego I should)
that Thorsten did spend about four years in the
1980's buiding for fun various versions of active
speakers using the A2030 (pretty much a close
copy of the TDA2030).

Right, and my impression was that the circuit
he submitted recently,and if I remember right,
it was the 3875, was not one he just done using
that exact chip (and he did remark on the other
chips as well) - it was certainly based on his
past experience, as you bear out.

> My KEY point with the IVGC was to note that
common style Op-Amp's ALLWAYS sound better
inverted

Agreed. It also have applied to some discrete
component circuitry as well.

> and the worse the Op-Amp, the bigger the effect.

Can't speak for experience on that, except that
inverted seems to be ALWAYS to be better - S/N
allowing.

> Most power op-amps including the LM1875/3875
are quite dreadful when compared to their best
low level companions, so any help they can get
is welcome.

Not everyone agrees with that... a little while
ago I would readily have wanted to agree.

Again may I reiterate my reason for making the
JLTi ("Just Listen To It" because I knew there
would those starting out as sceptical as I was
and you still are) was in the end NOT DIY, but
rather there are those out there NOT doing DIY
and this may be the product for them... and gulp,
one that I can happily say is capable of High-End.
I suspect you won't agree, but you haven't heard
the JLTi yet, have you?

It seems that the name JLTi is going to be very
apt. :-)

I have already demoed the JLTi to a number of
real hot-head audiophiles and not only did the
sound surprise them, it quite well knocked them
for six (a cricket term). I am not the least
surprised, but I am repeating myself.

Regards

Joe