a wonderfull sziklai - Page 26 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 5th November 2012, 01:38 PM   #251
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
Default purely dc coupled cfb version

Hello ,
I changed the vfb in a purely cfb dc direct coupled .No capacitors in the signal ..The node between R2 and R3 is virtual ground so there is no cap also in the feedback.
I will listen in the evening . 6mV dc output .If I like it, I will try to throw away the current mirror and the emitter follower to the vas .
The dc offset is also stable at different temperatures and also with/without short to gnd input .

Click the image to open in full size.
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."

Last edited by catalin; 5th November 2012 at 02:04 PM.
  Reply With Quote
Old 5th November 2012, 02:10 PM   #252
Piersma is offline Piersma  Netherlands
diyAudio Member
 
Piersma's Avatar
 
Join Date: Jan 2003
Location: Netherlands
Default Emitter follower

Probably unwise to lose the the emitter follower preceding the CFP.
More non-linear loading of the VAS will be the panalty of removing the follower.
Real life implementation of the circuit would probably be a challange, especially when it comes to the correct temperature tracking of the follower and the CFP combination.

Nice threat!

regards,
Piersma
  Reply With Quote
Old 5th November 2012, 02:55 PM   #253
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by Piersma View Post
Probably unwise to lose the the emitter follower preceding the CFP.
More non-linear loading of the VAS will be the panalty of removing the follower.
Real life implementation of the circuit would probably be a challange, especially when it comes to the correct temperature tracking of the follower and the CFP combination.

Nice threat!

regards,
Piersma
Hello Piersma, I was not talking about the ef preceding vas. I was talking about the ef between ltp and vas.If I will remove the current mirror in a future I don t see a good reason to keep the ef .
Thank you!
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."

Last edited by catalin; 5th November 2012 at 02:57 PM.
  Reply With Quote
Old 5th November 2012, 07:54 PM   #254
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
This last version seems to be the one with the deeepest detail and clarity in highs and mids
And a lot of powerfull low frequency .
Can somebody simulate the open loop gain/phase ?
Thank you !
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."
  Reply With Quote
Old 6th November 2012, 09:02 AM   #255
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
R22=10k seems a bit high. Does your sim show a Vdrop ~7V?

That Vdrop will change slightly with signal and result in a significant variation in Vce for Q16

I have been using 1k0 to 2k0 here.
Some use 0r0 and then complain the transistor blows up.

Using 1k0 limits the worst case dissipation in Q16 to ~220mW when Vsupply = +-42Vdc.
__________________
regards Andrew T.

Last edited by AndrewT; 6th November 2012 at 09:06 AM.
  Reply With Quote
Old 6th November 2012, 09:37 AM   #256
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by AndrewT View Post
R22=10k seems a bit high. Does your sim show a Vdrop ~7V?

That Vdrop will change slightly with signal and result in a significant variation in Vce for Q16

I have been using 1k0 to 2k0 here.
Some use 0r0 and then complain the transistor blows up.

Using 1k0 limits the worst case dissipation in Q16 to ~220mW when Vsupply = +-42Vdc.
Hello Andrew ,
There are 2 reasons for that value of R22 .
Q16 is a ef for the signal ,but :
Q16+R23 is creating a current source for the collector load of Q16 .So the current in load will be the same always and independent by the load value .So even if we increase the load the ef will do the same job .

One of reasons is that in transient mode the maximum current in Q16 and also Q6 (if we don't have R22) will be more than the maximum acceptable of the transistors beacuse the current is not limited in Q16.
In this case the ef will be distroyed .
But if we introduce R22 we introduce a current limit through Q16 of about = (V-) / R22 ~4mA .So the current limit is about 4mA but the current through ef will be decided by the 1kohm resistor from emitter of Q16 .Threfore the current is ~0.8mA .Now from the loadR22 point of view we have a constant current source (0.8mA) + current limitter (4mA) .
The key here is to calculate the resistor to have a voltage drop less than (rail voltage - ) - 3volts .Vce is varying from 30Volts only with maximum +/-0.3Volt so there are no issues .The constant current is helping a lot .

Now the second reason but for me is more important than the first one is that the AC transfer function is better with R greater .By doing this we move the dominant pole to a higher frequency which means that we can decrease the miller cap .This is one of the reasons why this amp is stable with 12pF and reach more than 100V/us SR .
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."

Last edited by catalin; 6th November 2012 at 09:41 AM.
  Reply With Quote
Old 15th November 2012, 09:41 AM   #257
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
Default Temperature and supply independent .

A major flaw to these topologies is the psrr .
Psrr was bad to the negative rail so I improved with capacitance multiplier circuit .
Another major flaw of the audio amps is that the idle current varies with current from the vas which varies with temperature.
Now from -10 to +60 degrees Celsius the vas current varies only with 10uA .

I will listen the vfb and after that the cfb mod.
The thd figures are better than previous .At 20khz from 1Watt to 100W thd <0.003% by keeping the same SR >100V/us

Click the image to open in full size.
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."
  Reply With Quote
Old 20th November 2012, 06:10 PM   #258
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
Default Joy version 5

@Jay: Hello Jay ,can you do the same simulation for all the power levels like before ?
Idle current should be 20-22mA through 0.22 ohms resistors.
Also the library is attached .
Thank you !

We can take also the negative feedback from the common point of the 330 ohms resistors if we want this ,thd will increase in this way from 0.002 to 0.02 % but the output will be a pure buffer . I want to build a pcb layout with this version v5 .
By this I will have a lot of options like a cfb amp,vfb amp and a output stage which is not in the feedback loop .
Also I want to have 2 pcb's separated ,one for the voltage gain and the other for the current gain(output stage) .
Attached Images
File Type: jpg joy_v5.JPG (92.3 KB, 429 views)
Attached Files
File Type: asc joy_sin.asc (9.1 KB, 24 views)
File Type: zip standard.zip (7.7 KB, 39 views)
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."

Last edited by catalin; 20th November 2012 at 06:27 PM.
  Reply With Quote
Old 27th November 2012, 11:16 AM   #259
catalin is offline catalin  Romania
diyAudio Member
 
Join Date: Jun 2007
Default Let's get low

Getting better thd figures with 22 ohms resistor in drivers 5171,1930 .
And also the imd figures are very low .
The current through drivers is now 20 mA and thd at 20khz with 9 harmonics at 1 to 3 watts is 0.0005 % .No tmc or other kind .
The joy here is that the final stage is biased only with 10mA -12mA .
And the version without negative feedback (nofb) over the final stage has only 0.01% thd at 20khz .
Attached Images
File Type: jpg joy5_1.JPG (145.1 KB, 346 views)
File Type: jpg joy5_1_nofb.JPG (144.6 KB, 334 views)
__________________
"please try to listen to some music through the amplifier instead. Life is so short ..."

Last edited by catalin; 27th November 2012 at 11:23 AM.
  Reply With Quote
Old 27th November 2012, 01:48 PM   #260
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
22r & 20mA gives an output Vbe of 440mV.
Is that correct? This seems far too low. I would expect 550mVbe to 600mVbe, giving a driver bias current range of approximately 25mA to 27mA

10mA to 12mA of output bias must be excluding the driver current.
The total current bias through the emitter resistor must be: driver bias + driver base current + output bias, i.e. approximately 35mA to 40mA.
Is that correct?
__________________
regards Andrew T.

Last edited by AndrewT; 27th November 2012 at 01:53 PM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sziklai output Moby Solid State 68 27th January 2011 10:39 AM
Hybrid Sziklai CFP Stee Solid State 0 27th December 2008 10:53 AM
ccs-loaded sziklai ionomolo Solid State 132 30th June 2008 06:39 AM
My DIY project - Wonderfull Loudspeakers ! Emeric Multi-Way 6 9th May 2008 02:38 AM


New To Site? Need Help?

All times are GMT. The time now is 08:55 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2