Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

I'm in search of a class G or H amplifier design
I'm in search of a class G or H amplifier design
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 16th February 2019, 05:15 AM   #41
apexaudio is offline apexaudio  Serbia
diyAudio Member
 
apexaudio's Avatar
 
Join Date: Feb 2010
Location: Serbia
Quote:
Originally Posted by kct View Post
Click the image to open in full size.

Apex, I was wondering, is there a specific reason or benefit why the input and output connections of your HX11 amp are located in close proximity to another ?
This PCB is not my design, it is designed by diyaudio member "willy"
  Reply With Quote
Old 17th February 2019, 12:31 AM   #42
suzyj is offline suzyj  Australia
diyAudio Member
 
suzyj's Avatar
 
Join Date: Jan 2006
Location: Western Australia
I don’t understand the motivation behind a minimalist class G amp. By it’s very nature an amp of this type is going to be a relatively complex beast, with multitapped (or multiple) transformers, multiple rectifiers and smoothing networks etc. Plus there’s the additional speed and OLG needed to ensure distortion products are well suppressed as the amp switches to the higher supplies. So IME an amp of this type with 13 transistors and a single sided PCB is only going to result in frustration and mediocre performance.
__________________
http://www.suzyj.net
  Reply With Quote
Old 17th February 2019, 03:20 AM   #43
ostripper is offline ostripper  United States
diyAudio Member
 
ostripper's Avatar
 
Join Date: May 2008
Location: Smoky Mountains , Tennessee
Quote:
Originally Posted by suzyj View Post
I don’t understand the motivation behind a minimalist class G amp. By it’s very nature an amp of this type is going to be a relatively complex beast, with multitapped (or multiple) transformers, multiple rectifiers and smoothing networks etc. Plus there’s the additional speed and OLG needed to ensure distortion products are well suppressed as the amp switches to the higher supplies. So IME an amp of this type with 13 transistors and a single sided PCB is only going to result in frustration and mediocre performance.

It would be a good sub amp.

With a high quality SMPS for both low/high rails and a better design , I intend to go "audiophile" !!
8 transistors - that is my OPS and 12 for the IPS. (20)

Yes , I agree ... 14db 20Khz extra gain margin is needed for just the AB inner pair. One needs to negate the 2 switch - on points of the class G. I eliminated that , too.
My outer pairs always conduct 20ma minimum.
Crest uses a D-pac dual soft switching schottkey with- out any calculated
snubber network. And they switch , as well. This gives them .1% (full power 1k thd).


20 devices is not "complicated". Without class H switching , better snubbed
low rail diodes , and the "bad a$$" CFA stage .... 30ppm at full power.
My simulated Triple EF3 gives 20'ish PPM @ 20K.


I simulated that NAD , really good. But it switched ... I could not track 200Khz class H. But , it is the best OEM class H I reverse engineered yet.
I would welcome any criticism of this design. I have actually built and
used most of it , except for the CFP trackers (Crest).
PS - output/driver devices will be faster Sanken's. Maybe even MT-200's !

I looked at Self's class G kit (intensely) , in my opinion it IS that sort of "minimalist" design as you speak of now. None of the class H extra distortion
mechanism's are addressed. Layout and basic circuit is better , but not acceptable to me.



OS
Attached Images
File Type: jpg greenamp.jpg (299.6 KB, 82 views)
__________________
Wanderer - GREENAMP , less is MORE.
https://www.diyaudio.com/forums/soli...ss-output.html

Last edited by ostripper; 17th February 2019 at 03:26 AM.
  Reply With Quote
Old 17th February 2019, 05:11 AM   #44
suzyj is offline suzyj  Australia
diyAudio Member
 
suzyj's Avatar
 
Join Date: Jan 2006
Location: Western Australia
Havent really looked at the input bit, but the output stage looks very well thought out. have you got any waveforms showing the voltage on the output stage collectors? It’d be interesting to see how the 22p and 150p networks dictate rise/fall times of the supplies.
__________________
http://www.suzyj.net
  Reply With Quote
Old 17th February 2019, 07:10 PM   #45
ostripper is offline ostripper  United States
diyAudio Member
 
ostripper's Avatar
 
Join Date: May 2008
Location: Smoky Mountains , Tennessee
Quote:
Originally Posted by suzyj View Post
Havent really looked at the input bit, but the output stage looks very well thought out. have you got any waveforms showing the voltage on the output stage collectors? It’d be interesting to see how the 22p and 150p networks dictate rise/fall times of the supplies.

C111/112 allows me the "shape" rise/ fall with the bypass of diode Ct/driver Ccb.
In the plot (below 2) is the R116 current (pink-no switching) , red and blue are the class H trackings.
I would most likely have to "tune" C111/112 and R121 to the specific devices I
end up using. Most models (even slow mje15032/33) work perfectly.
C111 even mitigates most of the reverse leakage "glitch' of D103 ,and R121 even
mitigates the much larger switching of the low rail diodes.
Using Infineon soft switching schottkey models for D103 and D106 - no "glitch' at all.
D106 will be a Infinion dual parallel 6A D-pak schottkey - like Crest uses.

Q111 refines impedance and voltage to give a superior reference. Reduces
D103 glitch , as well.


The result is worst case "difficult tracking' becomes quite easy (below 1) clipped 1k
with 200khz carrier .
The CFA's 200V/uS slewrate is also "easy" to negotiate. My goal was to
match or exceed my standard Sanken based EF3. The EF3 could slew to
almost 400V/uS with 60mhz Sankens ... could not match that .


But most is overkill. Also, Relatively simple for this level of performance.

Lots of pretty LED's ... too. Clips like a Valve ,as well.
Edit - full power 20Khz @ 42ppm. The EF3 was 20K@ 20PPM. The class G adds 20ppm , early effect +
imp. changes on the floating rails. But I was comparing this to a 10 device EF3 simulation.

OS
Attached Images
File Type: jpg trackinggood.jpg (300.9 KB, 53 views)
File Type: jpg greenamp class H.jpg (409.2 KB, 50 views)
__________________
Wanderer - GREENAMP , less is MORE.
https://www.diyaudio.com/forums/soli...ss-output.html

Last edited by ostripper; 17th February 2019 at 07:20 PM.
  Reply With Quote

Reply


I'm in search of a class G or H amplifier designHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
class d amplifier design issues sweetperfume Class D 9 3rd August 2014 01:42 PM
Help with Class D amplifier design Bass Audio Speakers Class D 5 1st May 2011 04:48 PM
Can I have some help with my class A amplifier design? GregH2 Solid State 30 3rd March 2011 11:23 AM
10w class D amplifier design yasar262 Class D 12 10th April 2010 10:50 AM
Looking for a CLASS D amplifier design . darkm4n Class D 4 19th October 2005 11:49 PM


New To Site? Need Help?

All times are GMT. The time now is 02:20 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki