The 2SK1056/2SK1057/2SK1058 are all the same except breakdown V.
His model is named 2SK1056.
There is a brief bio. on Jan Didden's Linear Audio website, associated with a couple of articles I wrote, if that makes any difference to you.
But my analysis of the FET models doesn't depend on reputation. There were references and links for anyone to check and confirm.
Wow , the corporate side ...
Very interesting , what is inside these chips determines the end
behavior of the design in general.
I tried a fast LT "300V/us precision op-amp" ....
I got (below 1/2). Nice rounded fast SW's !!
On all my "chip swaps" , the clipping/stability /PM was the same.
1mhz UG /72+ PM ....
The clip is all in the hawksford-superpair (perfect - below 3).
The uninformed fear of IC instability is not an issue here , as the
IC is a unity gain buffer / V-I converter only (not in the compensation loop).
Think of this as a IC based CFA-X/VSSA. The precision manufacturing
of the IC exceeds the discrete anology. You could even have High-
Z FET inputs thrown in as a bonus.
"CFA-XIC" ha ha ....
PS - this could actually be a 76mm X 50mm IPS - tiny .. tiny !!
OS
Very interesting , what is inside these chips determines the end
behavior of the design in general.
I tried a fast LT "300V/us precision op-amp" ....
I got (below 1/2). Nice rounded fast SW's !!
On all my "chip swaps" , the clipping/stability /PM was the same.
1mhz UG /72+ PM ....
The clip is all in the hawksford-superpair (perfect - below 3).
The uninformed fear of IC instability is not an issue here , as the
IC is a unity gain buffer / V-I converter only (not in the compensation loop).
Think of this as a IC based CFA-X/VSSA. The precision manufacturing
of the IC exceeds the discrete anology. You could even have High-
Z FET inputs thrown in as a bonus.
"CFA-XIC" ha ha ....
PS - this could actually be a 76mm X 50mm IPS - tiny .. tiny !!
OS
Attachments
Last edited:
HI OS
Do you have a thread for this amplifier ?
http://www.diyaudio.com/forums/atta...slewmaster-cfa-vs-vfa-rumble-eyesee-draft.jpg
http://www.diyaudio.com/forums/atta...710-slewmaster-cfa-vs-vfa-rumble-eyeseev1.jpg
Thanks
Do you have a thread for this amplifier ?
http://www.diyaudio.com/forums/atta...slewmaster-cfa-vs-vfa-rumble-eyesee-draft.jpg
http://www.diyaudio.com/forums/atta...710-slewmaster-cfa-vs-vfa-rumble-eyeseev1.jpg
Thanks
Last edited:
Aah but simple is good! I like simple.
This simple IPS tied a simple LatFET OPS is simply too much for me to resist.
I'll just have to build it.
(below 1/2) update in schema with a few tips.
With most "cheap" dual OP-amps, 180R -220R is needed for
R6/R7. The current of Vcc/Vee varies between devices.
The ratio of R8/R9 will increase the transconductance of the IC.
The worst (5532) , needs 2X - R8=R9 to give the (below2) 50+db CLG.
Below this gain , thd20 suffers. Bode plot is a typical CFA result , 10K
at the "break-point".
Of the 10+ IC models I have tried , none have oscillated ... some have
given more CLG (almost 60db) . UG has varied from 600K to 1.2mhz.
No PCB yet , before I perfect the circuit. (run more sims)
OS
Attachments
The best thread on diyaudio.com
This is THE BEST thread on diyaudio.com and ostripper has some of the best designs I have seen. Start your own thread about who has the best spice models. Remember,
Spice models, are approximations for non-linear devices. This is a difficult task.
OSTIPPER, we need you on this thread. Your ideas and designs. You are one of the
best on diyaudio.com. Take care.
This is THE BEST thread on diyaudio.com and ostripper has some of the best designs I have seen. Start your own thread about who has the best spice models. Remember,
Spice models, are approximations for non-linear devices. This is a difficult task.
OSTIPPER, we need you on this thread. Your ideas and designs. You are one of the
best on diyaudio.com. Take care.
Fast ones. like the LT1363. (any that specify 200 -1kV/us)
I'm going to play with the V/I ratio by varying the current FB/voltage gain
ratio (on the IC part of the design).
CLG between 55-65db ( 40+db @ 20K) is the goal. Then ALL the IC's will
be PPM @ THD20.
I will try to "blur the lines" between different devices. This is one of the
reasons for good designs .... device "indifference"
.OS
Beta enhanced Hawksford. Way to go (below 1).
A. IC dissipation is lower .... not as reliant on (the IC's) transconductance.
Better stability .... not "messing" with the IC as a unity gain follower.
B. 10-12db CLG increase. Better THD20 , even with "cheep chips".
C. Better Bode (below 2) ... with 52db up to 10K and 40db @ 20k.
Just the plot tells me 25ppm thd20.
An amp like the wolverine (blameless) has 110db LF and 50+@20k ,
hence ... 5ppm. 25ppm is not bad for a "cheep chip"
PS - better chips now do <15ppm thd20. (60+db CLG)
OS
A. IC dissipation is lower .... not as reliant on (the IC's) transconductance.
Better stability .... not "messing" with the IC as a unity gain follower.
B. 10-12db CLG increase. Better THD20 , even with "cheep chips".
C. Better Bode (below 2) ... with 52db up to 10K and 40db @ 20k.
Just the plot tells me 25ppm thd20.
An amp like the wolverine (blameless) has 110db LF and 50+@20k ,
hence ... 5ppm. 25ppm is not bad for a "cheep chip"
PS - better chips now do <15ppm thd20. (60+db CLG)
OS
Attachments
Last edited:
Tried the standard CM , wilson CM , with and without the hawksford.
The superpair gives lowest THD / best saturation behavior.
I might take your advice and try a CM/superpair
Like the ancient thread .... "power amp under development"
OS
You don't need to look to far, this one is CM/superpair.
http://www.diyaudio.com/forums/solid-state/253039-unique-cfa-120-230w-amp-2.html#post3866560
Ohhhh ! a "goldmine" of ideas (I will learn).
But ,
You said it ...
Not for me
, but the masses like VSSA level ... I'm surprised theylike the "spooky" , with the "busy" PCB and 100 components.
I like it !
what a forum
OS