The Very Best Amplifier I Have Ever Heard!!!! - Page 212 - 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 12th January 2013, 12:13 PM   #2111
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
I have reexamined the schematics and simulation for this amp. I thought I understood it years ago but I only had a basic understanding. I didn't know what to think when Nagys kept saying it needs to be built exactly like the schematic. Now I realize there is plenty of truth to that.

This amp is designed for RF headroom (high slew rate) and phase linearity. I know this because the VAS and LTP are heavily degenerated. This allows compensation to be small, which loads the frontend less. The tradeoff is that DC gain of the frontend is very low, and audio band THD is increased. This may not be so bad, since the distortion is strongly H2 dominant due to the VAS Early effects.

This amp is pretty close to the edge WRT stability. Anyone designing a new PCB should know how to test for stability into capacitive loads. The transconductance of the FETs plays a major role in stability into capacitive loads, and so taking phase lead compensation from the output rather than the VAS could result in catastrophic instability depending on output loading, especially because of the very small L/R output network.

The output stage does not have good PSRR at RF; no output stage does. Because the LFET AC characteristics are important for the stability compensation of the amp, the rail response will be part of the output stage behavior and need to be exceedingly inert. I would not just slap films and large lytics onto the board else you risk oscillation. The 47uF decoupling lytics are important. I believe they provide damping for the rails. If a low-ESR lytic were used here it may cause instability. I would be interested to see the datasheet and ESR spec of the cap on the original Goldmund. For a given PCB layout, an appropriate RC snubber could be estimated for the rails that would hopefully dampen them well. I don't think it is enough to simply "pay attention" to decoupling of the output stage. One needs to measure and make sure the rails are inert or risk oscillation.

If this is done properly it seems to me that Goldmund could be unconditionally stable, as long as you leave on the output zobel. Of course it will ring into capacitive loads, but this is not a sign of instability, rather it's a sign of health because the amp is able to emulate a high-Q inductor. If this is a problem, the speaker cable should have its own customized snubber network, because while a stable amplifier's output inductance is either benign or with a lower limit, load reactance can vary widely.

Because the distortions of lateral FETs, FET input stages and this specific VAS are rather benign, the large RF headroom principle works well for this amp. I expect the next step if we wanted to improve the design would be to lower distortion while preserving the AC behavior.
  Reply With Quote
Old 12th January 2013, 05:58 PM   #2112
diyAudio Member
 
Join Date: Mar 2008
Quote:
Originally Posted by keantoken View Post
I would not just slap films and large lytics onto the board else you risk oscillation. The 47uF decoupling lytics are important. I believe they provide damping for the rails. If a low-ESR lytic were used here it may cause instability. I would be interested to see the datasheet and ESR spec of the cap on the original Goldmund.
Here are the specifications on the Philips 47uF/100V capacitors used in the original MM9.2. They are of a long life type for industrial applications. Not a low ESR type.
Attached Images
File Type: jpg Philips.jpg (184.1 KB, 744 views)
  Reply With Quote
Old 12th January 2013, 06:55 PM   #2113
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
Thanks. That is like I thought.

In addition to a good ESR value, the damping cap needs to have a non-inductive path from the FET drains to output snubber ground. This may be difficult without a ground plane.

It seems whoever designed this amp either knew what they were doing, or made a very lucky guess. It would seem in some ways reliability was sacrificed for sonics, for instance with offset stability, and IIRC the drivers ran a bit hot?

Also, there is no need to turn around the protection zeners in the original schematic. They will still work this way. It seems they were oriented this way to reverse bias the 1N4148 to reduce the total capacitance of the limiter, and this makes sense for preserving the input impedance of the driver stage. I'm not sure whether the difference is really significant but it still functions as a protector AFAIK.
  Reply With Quote
Old 12th January 2013, 09:33 PM   #2114
diyAudio Member
 
Join Date: Mar 2008
Quote:
Originally Posted by keantoken View Post

It seems whoever designed this amp either knew what they were doing, or made a very lucky guess. It would seem in some ways reliability was sacrificed for sonics, for instance with offset stability, and IIRC the drivers ran a bit hot?
At high output levels the VAS transistors runs at around 600 mW. The original MM9.2 design uses one clip-on heatsink for each VAS transistor, but the heatsinks are small. The driver transistors does not have this problem.
  Reply With Quote
Old 12th January 2013, 10:32 PM   #2115
diyAudio Member
 
Join Date: Sep 2009
Location: Vietnam
Blog Entries: 2
Send a message via Yahoo to quanghao Send a message via Skype™ to quanghao
Hi tommy1000!

I also just completed amplifier similar to this amplifier, invite you to view images!

637 Amplifiers

Thanks
__________________
Group buy DAC-END R (ES9018) full assembled board
update 06/14/2013 :http://www.audiodesignguide.com/DAC32/index2.html
  Reply With Quote
Old 13th January 2013, 02:54 PM   #2116
diyAudio Member
 
Join Date: Jan 2011
Location: Fosser, Aurskog-Holand, Akershus, Norway.
Quote:
Originally Posted by alex mm View Post
Sooner or later you will end up with GOLDMUND

I am about to. Soon.
Buying cabinet from Italy.
Click the image to open in full size.
Without those handles, I think.

But they deliver every part of theese cabinets as spare parts.
__________________
Sooner or later you end up with TANDBERG
  Reply With Quote
Old 13th January 2013, 03:10 PM   #2117
BMW850 is offline BMW850  Netherlands
diyAudio Member
 
BMW850's Avatar
 
Join Date: Aug 2004
Location: North East of the Netherlands
Default Almost identical

The Cabinet is almost identical

Regards,
Rudy
Attached Images
File Type: jpg goldmundmimesis92front.jpg (104.7 KB, 622 views)
File Type: jpg mimesis92datasheet.jpg (116.8 KB, 617 views)
__________________
Itís not what you know, but who you know ďand where they areĒ

Last edited by BMW850; 13th January 2013 at 03:12 PM.
  Reply With Quote
Old 13th January 2013, 03:14 PM   #2118
diyAudio Member
 
Join Date: Jan 2011
Location: Fosser, Aurskog-Holand, Akershus, Norway.
Quote:
Originally Posted by AndrewT View Post
A teaser:

Are the lower devices shown inverted?
Not really, but I have seen this in many libraries, where they draw theese devices both this way, and with the gates the opposite way.

Nice having fun with theese .
__________________
Sooner or later you end up with TANDBERG
  Reply With Quote
Old 13th January 2013, 03:22 PM   #2119
diyAudio Member
 
Join Date: Jan 2011
Location: Fosser, Aurskog-Holand, Akershus, Norway.
Quote:
Originally Posted by tommy1000 View Post
At high output levels the VAS transistors runs at around 600 mW. The original MM9.2 design uses one clip-on heatsink for each VAS transistor, but the heatsinks are small. The driver transistors does not have this problem.
I also see it nessesary to add heatsinks for the VAS.
Theese will be made of some 1mm Coppersheet where I will make a transistor/copper sandwich of it all. Could ofcourse make some clips, and put on individually, bud the sandwich is so much easier to make.
__________________
Sooner or later you end up with TANDBERG
  Reply With Quote
Old 17th January 2013, 05:51 PM   #2120
diyAudio Member
 
Join Date: Jan 2011
Location: Fosser, Aurskog-Holand, Akershus, Norway.
My goodness what trouble one gets in.

Can anyone tell me where to get the 220uF 250V capaqcitors, with the quite narrow 16mm diameter there is left space for on the PCB?

Or better: What voltage do You guys have over theese capacitors?
I am building with 2X55VAC for supply, wich I have calculated to a voltage just over 155V over those. Can anyone confirm?
__________________
Sooner or later you end up with TANDBERG
  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
Has anyone heard of the Pioneer SA-7300 amplifier made back in the 70's? pkgum Solid State 17 14th November 2013 03:24 PM
anyone heard of these? beauistheman Multi-Way 12 22nd September 2008 07:49 AM
ANA Audio Research APA 8 Power Amplifier. Heard of it? Junior Solid State 0 1st August 2007 01:56 AM
Anyone ever heard of Staticjeep Music 0 9th March 2006 03:15 AM
Anyone heard of Jam-Tech True Fidelity Amplifier? tschanrm Class D 1 17th November 2005 02:10 PM


New To Site? Need Help?

All times are GMT. The time now is 04:12 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