High-End Regulated Buffered Inverted GC

[carlosfm]

I will send you a new board:

Still the LM3875?
Still NI?
Still those widely separated PSU caps?
Still no buffer?
Still unregulated PSU?

Nothing new then...
I think I'll scavenge your amp and put my own stuff inside.
Thanks for the attention, anyway.

Gotta go, this site is too slooooow now, and time to go to bed here.

 

[Peter Daniel]

Still the LM3875?
Still NI?
Still those widely separated PSU caps?
Still no buffer?
Still unregulated PSU?

As my friend says, you don't touch a proven design.

 

[jam]

If that's the case we will still be making fire by rubbing two sticks togther..........................😉 . I thought you were out of here several posts ago.

 

[Peter Daniel]

I thought you were out of here several posts ago.

But I also said, we aiming to please😉

 

[fdegrove]

Hi,

As my friend says, you don't touch a proven design.

Especially when there's no design to touch in the first place...

Cheers,😉

 

[Peter Daniel]

Hi,



Especially when there's no design to touch in the first place...

And how do you know that?

 

[fdegrove]

Hi,

Still the LM3875?
Still NI?
Still those widely separated PSU caps?
Still no buffer?
Still unregulated PSU?

LM chip, nothing to design there.
NI or I, nothing to design either.
Widely seperated PSU caps, now that's DESIGN, right?
no buffer? no buffer.
Unregulated PSU? Designed that part? Congratulations.

As my friend says, you don't touch a proven design.

So, your friend must be a very wise man...he knew right away there was no design to change.

Now if someone feels like doing something more challenging, how about designing one of these:

Cheers, 😉

 

[jam]

Touché

 

[Peter Daniel]

LM chip, nothing to design there.
NI or I, nothing to design either.
Widely seperated PSU caps, now that's DESIGN, right?
no buffer? no buffer.
Unregulated PSU? Designed that part? Congratulations.
So, your friend must be a very wise man...he knew right away there was no design to change.

But all those are design choices, wouldn't you agree?

Maybe my friend is not very wise man, but at least my design choices don't bother him as much as they bother you.

Why not?😉
http://www.diyaudio.com/forums/attachment.php?s=&postid=422262

 

[fdegrove]

Hi,

but at least my design choices don't bother him as much as they bother you.

From what I recall your "design choice" was to go for the IGC....
After testing the field that turned out to be a bad "design choice" and 20 or so amplifiers were called back to turn them into NIGC amps.

How many of these "design choices" were there to make?

Thank god it wasn't two dozen of them...

Of course it doesn't bother me that someone claims for about a year or so that IGC is the way to "design" with GCs...
Nah, surely no one's fallen into that mousetrap....

Cheers,😉

 

[millwood]

Of course it doesn't bother me that someone claims for about a year or so that IGC is the way to "design" with GCs...
Nah, surely no one's fallen into that mousetrap....

Cheers,😉

I know, 🙂

I have been advocating NI designs in those "dark" days but was told my ears apparently were golden enough. You got people mathematically proving why the I topology is that much superior to the NI topology, and how much nicer the I topology sounded over the NI topology, etc.

that's, of course, someone figured out that it is better for one's check books to advocate the sudden virtures of NI topology.

then all in a sudden NI is all the rage. I always wondered why those advocants of I topology are the same people who are advocating precisely the opposite.

 

[Peter Daniel]

From what I recall your "design choice" was to go for the IGC....
After testing the field that turned out to be a bad "design choice" and 20 or so amplifiers were called back to turn them into NIGC amps.

How many of these "design choices" were there to make?

Thank god it wasn't two dozen of them...

Of course it doesn't bother me that someone claims for about a year or so that IGC is the way to "design" with GCs...
Nah, surely no one's fallen into that mousetrap....

Cheers,😉

Well Frank,

It seems like you are a puppet in someone else's bad game. To talk about GCs, you should know a bit more of the previous developments.

When I got into that, I had no clue which config is good or bad. Everybody was following Thorsten's design so I tried it as well. It seemed to be better than most offerings around, so I decided to go with it, although I already modified it slightly. At that time you can say that it was not my design choice, I was following the trend.

But I can claim that among some choices that were made, MUR860 were my design choice. When you check D-K, chances are they are out of stock on those. Other design choice was using Caddocks in a feedback path. Not so long ago, all major parts distributors were out of 22k value on that resistor. Same was for Rikens, which are also my design choice. All major distributors were short of 680ohm value and when I was talking with Parts Connexion they were very surprised that suddenly they sold all 300 of them.

My design choice is using insulated chip package, as previously it was uninsulated that was most popular. My design choice is using Cardas wire for umbilical cord, as to me this delivers the best sound signature.

My design choice was to go with NI configuration, in spite of what everybody else in diy circles was claiming. Fedde was actually faster by a day or so😉. Before that you could see only single cases were NI config was used and everybody was wondering why 47Labs GC is not using one.

I don't care if you call it field testing, or cutting the losses, it is still a marketing decision and it is still a design choice that we went with. Don't blame me for that, as I wasn't the only one deciding here.

All my other design choices stayed pretty much the same.

However those are only few parts, that are nothing to brag about. My main design choices are reflected in the chassis design and layout and this accounts for at least 20% of the sound signature produced by the amp (sorry Carlos).

So, as I stated already, I was not claiming that IGC was the only way to go, I was merely repeating someone else's words (much as you do here, now).

I still wonder why are you suddenly became so interested in GC amps and ask all those questions, are you bored with your tubes or you just seeking cheap thrills only?

 

[fdegrove]

Hi,

It seems like you are a puppet in someone else's bad game.

Perhaps you're not aware of the game you're in but I wouldn't even think of calling it a game...
I know I've never been anyone's puppet and never will be.
That's just for the record.

To talk about GCs, you should know a bit more of the previous developments.

Well, I've read you're entire post and there's nothing there I didn't know already.

Nevertheless:

At that time you can say that it was not my design choice, I was following the trend.

Surely I'm not the only one recalling you claiming that you compared both configurations, IGC and NIGC, and stating clearly your preference for the IGC.

But I can claim that among some choices that were made, MUR860 were my design choice.

Again, I recall you stating that the MUR860s varied in sound depending where they were sourced from.
Not a very wise "design choice" to go into production with something as variable as this it seems.

My design choice was using Caddocks in a feedback path. Not so long ago, all major parts distributors were out of 22k value on that resistor.

And?

All major distributors were short of 680ohm value and when I was talking with Parts Connexion they were very surprised that suddenly they sold all 300 of them.

So, what are you telling us here? 300 resistors of the same value, that's peanuts anyway...

My design choice is using Cardas wire for umbilical cord, as to me this provides the best sound signature.

Tells the story of the power supply loud and clear to me but who needs a "steenking consulant" anyway.

Before that you could see only single cases were NI config was used and everybody was wondering why 47Labs GC is not using one.

No problem...Time will tell....Maybe they'll recall all of their sales as well?

Don't blame me for that, as I wasn't the only one deciding here.

I don't and I won't, somebody may though.

My main design choices are reflected in the chassis design and layout and this accounts for at least 20% of the sound signature produced by the amp (sorry Carlos).

Tubes are supposedly at least a factor of ten more microphonic than a chip should be....
When 20% of the total effort goes into countermeasures against microphonic effects you'd better come up with something efficient or lose the idea entirely IMHO....

Layout?
Did I hear someone talking about widely spaced filtercaps?
Must have been a "design choice" as well...

So, as I stated already, I was not claiming that IGC was the only way to go, I was merely repeating someone else's words (much as you do here, now).

Indeed, I'm mostly repeating what you've said before.

I still wonder why are you suddenly became so interested in GC amps and ask all those questions, are you bored with your tubes or do you seek cheap thrills only?

So you keep asking.
Yet I've already said before that I take no particular interest in GCs as such, I'm not bored with tubes and no, I don't seek cheap thrills either.
Then again, this is not about me.

Cheers,😉

 

[Peter Daniel]

I'm not as dedicated as you, so I only use simplified quotes:

>>Perhaps you're not aware of the game you're in but I wouldn't even think of calling it a game...
I know I've never been anyone's puppet and never will be.
That's just for the record.<<

Puppet usually doesn't know...

>>Well, I've read you're entire post and there's nothing there I didn't know already.<<

So what?

>>Surely I'm not the only one recalling you claiming that you compared both configurations, IGC and NIGC, and stating clearly your preference for the IGC.<<

You find that post and I'll send you an amp as well (same offer as to Jocko)


>>Again, I recall you stating that the MUR860s varied in sound depending where they were sourced from.
Not a very wise "design choice" to go into production with something as variable as this it seems.<<

You recall right. The better sounding ones, from OnSemi, were never a problem to get, so what's your point?


>>So, what are you telling us here? 300 resistors of the same value, that's peanuts anyway...<<

It's only from PC, the problem is they didn't have more😉 It is not peanuts for 'exotic' brand, it seems.


>>Tells the story of the power supply loud and clear to me but who needs a "steenking consulant" anyway.<<

Doesn't tell anything to me. You must be running on a different clock.


>>No problem...Time will tell....Maybe they'll recall all of their sales as well?<<

Cheap shot as usually


>>Tubes are supposedly at least a factor of ten more microphonic than a chip should be....
When 20% of the total effort goes into countermeasures against microphonic effects you'd better come up with something efficient or lose the idea entirely IMHO....<<

YHO doesn't count here. I work with what I have, and there are no other choices. It shows your lack of experience with those issues only.

>>Layout?
Did I hear someone talking about widely spaced filtercaps?
Must have been a "design choice" as well...<<

You hear it from whom? Do you think it could be done better? Some thinking effort wouldn't hurt here. In AMP1, where one is not restricted by PCB, the caps are as close as their size permits. For some reason you failed to notice that.


>>Indeed, I'm mostly repeating what you've said before.<<

And I still don't know why.


>>Then again, this is not about me.<<

I know it's not about you. You are just a puppet, remember?

 

[geewhizbang]

I apologize for my own contributions to this argument. Please let's stop.

 

[Variac]

Carlos,

So now you have my interest. I want to at least know
how "your" amp is made. the trouble is that it is like a puzzle to figure out what you have done.


From the other thread you sent us to at first, I am to assume you ares using Pedja's design for the regulated supply?

http://users.verat.net/~pedjarogic/audio/gainclone/supplies.htm

Then under Nuuk's advice, you made the caps 100uF?

but now it's:

- Very small cap after the LM338 (I used 47nf poly), or no cap at all, if you use them very near the chips.
- From 4,700uf to 10,000uf total capacitance before the regs (I use 4,700uf on the MUR860's board and 2,200uf on the regulators' board).
- Small cap on the chip (I'd say between 22uf and 100uf).


Boy, it would sure help us who don't know much who want to hear your approach to have a clear schematic.

One of the reasons people build Peter's gainclone is that he is very clear how to make it.

Now that this thread has pages and pages of arguing, it is very difficult to understand.


Also, When you talk of a buffer, you are discussing your preamp not Pedjas buffer

 

[Nuuk]

Where's Kofi Annan when you want him? 😀

 

[azamkita]

Hi guys,

Come'on. Stop that topic.

Ok, I think my analog meter "s*crew up". LM3875 only use miliamp !!!!

As such, I think LM 317/337 is more then good as they are 1 amp regulator.

BTW, how to increase the output voltage from 18 ac to 24 ac ?.

I start with NIGC, then move to IGC then add a buffer (now is BIGC) and going to built BIGC which I think is last one.

p/s: Let see, we had so many circuits here, which is which ?.

regards

 

[carlosfm]

My design choice is using insulated chip package, as previously it was uninsulated that was most popular. My design choice is using Cardas wire for umbilical cord, as to me this delivers the best sound signature.

How funny...
I've ALWAYS used the insulated package.
Everybody (including you) advised me to use the uninsulated one, because it's better blah blah blah...
A I don't like to fiddle with insulators, spacers, etc, I always used the insulated package.
At the time, with the uninsulated package, you reported differences in sound with the insulator type, and you were fiddling with different types of heatsinks (copper, aluminium, etc) and reported listening tests.
I think you loose too much time making backward steps.

What's funny is that now it's YOU that recommend, and had the idea to use the insulated package.

For me this conversation is finished, I've said it all.
Have a nice day.

 

[carlosfm]

Let me make one thing clear, in case someone doesn't understand.
When I talk buffer, I'm talking of an input (op-amp) buffer near the chip, nothing to do with my pre.
Take a look at my pics on the first page of this thread, and BTW look where I put the PSU caps for the OPA627.
Could they be closer?😀