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.

When a Luxman clone is not a clone
When a Luxman clone is not a clone
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 2018, 06:33 PM   #11
hpavictor is offline hpavictor  Romania
diyAudio Member
 
hpavictor's Avatar
 
Join Date: Feb 2006
Location: Bucharest-Romania
Quote:
Originally Posted by carlmart View Post
For a long time I have been listening to a friend of mine's Luxman 5M20 power amp. A fantastic amplifier designed in the late '70s and still considered one of the best Luxman made. It can easily stand its own on any comparison with a more modern amp in audio quality and else.

Some months ago I decided to start my research to see if it was possible to clone it, but all the active parts are extinct now. So the question was: is it possible to use the Luxman as an inspiration, using modern active transistors?

It's mi belief that you can call something a clone when it's exactly alike, which in this case was not possible. So I decided to call this enterprise "The Shadow". And the first design was to be the Shadow 5M20.

As a start what I wanted to do was to simulate the Shadow 5M20 in Ltspice.

The first thing to do was to find capable replacements for each transistor. So I asked Keantoken to help me navigate this sea and see where we could get to.

The input dual fet was a very delicate point, and looking around on what was available I found the 2N5566 to be a suitable choice. I bought a pair. An LSK389 would certainly work too, but at the time it was hard to find. I didn't want to start this project until I got the dual fet.

So my modeling is all based on the 2N5566 at the input. Of course I tried other duals and even singles to see if they made any difference. But the simulation result changes were minimal.

You are free to look for the original Luxman 5M20 schematics. I will only put here the Shadow's.

Let's put the final version I got to, and then explain where it departs from the original design, first to model the electrical characteristics of the original drivers, and second to tame a glitch on the LTP stage.

So now let's see what are the first comments I get

Carlos
I like this project !
  Reply With Quote
Old 16th February 2018, 07:15 PM   #12
carlmart is offline carlmart  Brazil
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
Hi Ronaldo,

That's why I asked what models might be missing. I think I will write a txt file with all the models and zip it all together with the asc files.

About the 270p caps on the base of the drivers, they were Keantoken's suggestion. The idea was to make the circuit mimic the original 5M20 as much as possible. The transistor they used for the drivers were the 2SC1161A/2SA653A (unavailable), with higher capacitance than newer substitute. So he suggested adding two caps: base-ground and base-emitter.

Both caps and several other changes from the original, I intend to add them later on the solder side during the listening tests.

As you see, I didn't include the base-emitter cap as it affected the simulation quite badly. The other cap doesn't seem to affect things much.

The reason why I tried separating the drivers emitter resistors junction from the output , along with 10uF bypass cap was to correct a glitch I had on the sinusoidal signal on the LTP out. If you return things to the way they originally were, you will see what I'm talking about.

Another thing that I added to the original design was C212, the 2.2p feedback cap from VAS-out to the LTP fet. This was to improve the square wave response, which I will upload the asc of too.

About the regulated low current supplies, I am not sure if I will use the original regulator. But that would probably be done during listening tests. I do not know enough of LTSpice to add the supplies on the simulations.
  Reply With Quote
Old 16th February 2018, 07:18 PM   #13
carlmart is offline carlmart  Brazil
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
Oh, about my objective: to make a very good sounding amplifier that gets close to the original Luxman sound.

Luckily I have my friend's amp to make that comparison!

The additional changes I made will be listened to on that comparison.
  Reply With Quote
Old 16th February 2018, 07:45 PM   #14
carlmart is offline carlmart  Brazil
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
OK, now I put the models on the file.

Does the sim work?
Attached Files
File Type: asc Shadow.5M21.main.x32.asc (21.0 KB, 19 views)
  Reply With Quote
Old 16th February 2018, 08:14 PM   #15
carlmart is offline carlmart  Brazil
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
New asc files, with the models txt to include on folder.
Attached Files
File Type: asc Shadow.5M21.main.x32.AC.asc (17.9 KB, 19 views)
File Type: asc Shadow.5M21.main.x32.SW.asc (17.9 KB, 17 views)
File Type: txt Carlmart-Models.txt (3.1 KB, 19 views)
  Reply With Quote
Old 16th February 2018, 08:23 PM   #16
PB2 is offline PB2  United States
diyAudio Member
 
PB2's Avatar
 
Join Date: Sep 2004
Location: North East
Quote:
Originally Posted by forr View Post
They buffer the inputs of the VAS transistors. Known as doing a lot of good. Such buffers are found in Self's single ended VAS, and in Cordell and Groner's differential VAS.
I knew the purpose, I was simply commenting on the placement in the schematic,
a matter of style but they look like they are trying to be a diff pair. Purely a visual thing.
They are beta multipliers, obviously.
  Reply With Quote
Old 16th February 2018, 08:26 PM   #17
PB2 is offline PB2  United States
diyAudio Member
 
PB2's Avatar
 
Join Date: Sep 2004
Location: North East
Quote:
Originally Posted by r_merola View Post
Carlos,

What you did with the Chinese board you order to adapt this circuit?

What are your project objectives?
Regards from SP - Brazil
Ronaldo
Do you have a link to the Chinese board?

Did you consider Sanken's in the output stage?
  Reply With Quote
Old 16th February 2018, 11:46 PM   #18
PB2 is offline PB2  United States
diyAudio Member
 
PB2's Avatar
 
Join Date: Sep 2004
Location: North East
Might be interesting to try a Tiger.01/Bryston output stage with gain and then the
boosted front end would not be needed. Just as an alternative.
  Reply With Quote
Old 16th February 2018, 11:50 PM   #19
carlmart is offline carlmart  Brazil
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
Yes, I forgot to elaborate on that.

Those Chinese boards looked like a good option for the prototype stage, as it would easy up part of the process.

Those boards were designed for a Goldmund "semi-clone", as it used bipolar outputs instead of Mosfets. When I saw the circuit I had the idea to use them for the Shadow project.

Pcb mono placa amplificador goldmund 2scC5200 saida ajustavel Uma discreta 2sa1943 tubo goldmund amplificador traseiro Frete Gratis em Amplificador de Eletronicos no AliExpress.com | Alibaba Group

I bought a pair of boards and I'm working on how to adapt them. There are two Shadow stages missing on the Goldmund semi-clone: the LTP-VAS buffer and the pre-driver. It should be easy to add two small piggy-backs for them.

But I think there other things to be considered after trying the prototype, like the influence a correctly designed pcb has on the audio quality.

On the prototype I should be using the MJLs for the output. Sankens are not easy to source, and it seems there are many fakes around. It's quite likely I will buy the output transistors directly from On Semi, if they still sell direct. Just to be sure.

Last edited by carlmart; 16th February 2018 at 11:52 PM.
  Reply With Quote
Old 17th February 2018, 10:59 AM   #20
r_merola is offline r_merola  Brazil
diyAudio Member
 
Join Date: Apr 2009
Carlos

If you love the sonic signature of this old amplifier, you have the following options:

1- Order a old sample of this product in good working conditions (Best Option, expensive and risk);

2- Make PCB using exactly the same layout and parts. Use also exactly the same circuit structure, transformer type and so on. It will be expensive and difficult as parts are old and rare.

3- Make a new amplifier project with new components and circuits correction. I am sure you will get a best amplifier. It was also done with Goldmund clone and there are lots of solutions (some goods, other very questionable - it will depend of designer skills and dedication of course).

Simulation is a good way to start, but bench tests and audible test will define what you need. If you change amplifier compensation as you did in your simulation file you got a different sonic behavior for sure - and it becomes a new project. It must be done in sonic tests when you listen results.

You can read in HTForum all design criteria I used in my actual amplifier using this same topology. It is a more than 10 years old project.

You know I love this circuit arrangement and I prefer the sonic signature of Mark Levinson's amplifiers with the exactly the same topology but with best circuits solutions.

Good lock
Ronaldo
  Reply With Quote

Reply


When a Luxman clone is not a cloneHide 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
Luxman clone carlmart Solid State 6 18th February 2017 09:23 AM
FS: Drivers for JBL 250Ti clone (LE14, 108H clone, 104H-2 GordonW Swap Meet 10 20th May 2014 08:19 PM
clone Luxman B-1000f monoblock amp space2000 Solid State 7 18th May 2008 07:40 PM
Luxman A-522 Rega Clone speed problem kffern Analogue Source 3 2nd October 2007 12:01 PM
Luxman m120a clone zerohead_ak47 Solid State 0 21st May 2007 07:09 PM


New To Site? Need Help?

All times are GMT. The time now is 06:22 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki