The MONGREL (supersym II)

[ostripper]

Originally Posted by csy

Hi OS


Also, the overlay drawing implies the output inductor/resistor assembly is being done differently?

L1/r98 are the same just horizontal inductor (16 turns/16ga.- 20mm X 20mm=3.31034uH air core inductance. resistor goes inside.
OS

[ostripper]

Very strange , while working on my APT1 , I came across the schematic for the Luxman M120a. Very similar to the APT(attachment 1) , but more refined. I liked the use of the FET's at the input and simulated for both the bipolar and some J201 FET's.

There are plenty of VERY good reviews of this group of amps - M-117, M-03, M-05, MX-100 ..
Luxman C120A/M120A how good? - AudioKarma.org Home Audio Stereo Discussion Forums

and many , many more..

They say it has killer bass , smooth airy soundstage. The best thing I can see is that they use the 2sa970/2sc2240, which is a dead match for my a992/ksc1845's. Upon my first simulation it worked very good , in fact , it matched the spec (attachment 2) at .013-.015% all the way from 20hz - 30khz. I tried the simulation with both 2sj201 FET's and the BJT's. With the BJT's the THD dropped to .005% even @ 30khz/200w.

I could not leave well enough alone , so I added a better CCS , 2 real compensation capacitors (2X 47pF). Another little addition:
http://peufeu.free.fr/audio/memory/memory-5-vas.html

The "magic" resistor (another one!!) This halved the 5th and 7th harmonics which correlated with the French audiophiles listening tests here:
Memory Distortion Philosophies - Part 6 : Moment of Truth

Also in (attachment 3) R6 is the adjustable CCS and R24 is the gain adjust for the level shifter. 50R for R6 gives 1.3ma per input device and 60R for R24 balances the LTP at low freq. (attachment 5). Once you do this at 100hz the input stage only get more balanced as you increase in frequency...finally getting within 3uA at 20Khz.

I like the symasym and the DX style topology but for my first voltage board to use with the PB250 , I have decided on the GX (attachment 3/4). You can't go wrong with something so simple that has the strange tendency to have rock solid THD all across the audio band and even into the ultrasonics. Mr. Holman (APT1 designer) said that he found the most linear voltage stage and THEN applied negative feedback to it. I think the Luxman beats his amp hands down even as they are essentially the same.

This amp is not PPM (who gives a poop? ) , but has the strangest THD characteristics and a lovely harmonic structure (better than the symasym).

I am a little apprehensive going out on a limb like this with a new topology, but I have actually built the APT back in the "frugalamp" days.. it did not blow up or oscillate . This luxman seems to be the next logical step and it uses very similar components. The "GX1.2VB_complete.zip" will be finished tonight and actually built in september.

As far as sharing my files , I am pissed... SOME PEOPLE try to download the whole website , change their IP address after being banned , ACTUALLY port scan me to see if I get madder.. argggg. Download a few , even a hundred BUT if you get too greedy and try for the whole 40 gigabytes... You will be banned forever !!!
Just download 2/3 at a time , another 2/ 3 , do it all night for all I care , just don't DDOS me.

OS

[Salas]

There is a low output heavily biased one that I have heard good words about by a mate. LUXMAN | PRODUCTS | INTEGRATED AMPLIFIER L-590AII
Any special info on that one BTW?

[keantoken]

OS, be bold. Nothing wrong with going out on a limb... There was apprehension about my headphone amp but it's stable and playing beautifully into my 8inchers! It has not blown up despite being abused with 4 ohm speakers.

I was nearly sure the funky compensation would not work... But it does. Try getting 46db OLG at 100KHz! (on that note, look at C104a and R111a, and notice Cdoms are only 5pF... This is very much like what I use in my headphone amp and have simulated on power amps before)

With the bold approach however, you've got to go the whole hog. Be thorough. If there's a way to make it work, be sure to find it. It may be demanding, but so are most things worthwhile in life.

The main reason it's discouraged I think is that if you screw up, there aren't many around that can help you, few people go down the rabbit hole. That's not really a problem on DIYAudio, right?

I was going through your pictures, OS. I think I went through several hundred files, just looking at thumbnails on the HTML's. I noticed someone downloaded your entire PDF folder. Sheesh! At least make an appointment!

- keantoken

[ostripper]

Originally Posted by Salas

There is a low output heavily biased one that I have heard good words about by a mate. LUXMAN | PRODUCTS | INTEGRATED AMPLIFIER L-590AII
Any special info on that one BTW?

Salas , the L-590A looks just like the L507u (attachment) the power supply , amp boards everything. I have the schema for the 505u and it is a 100w class AB with the same topology as the M120a. It seems Luxman stuck with what worked ... and sounded good.

Since the amps boards look EXACTLY the same , I wouldn't be surprised if luxman just use the same board with lateral mosfets and Bias them into class A. I did this in my simulation with IRFP240/9240's outputs and matched that .03/.005% spec exactly.

I am going to have 2 X the new luxman 509u in my PB250/GX combo powered by my 1KVA+ 30lb. toriod. So for 1/20 the price of the real thing I get twice the OHMMPH!!

Hey , keen... this is why I am going modular. I can unscrew 6 euroblock terminals, bring the voltage amp to the bench , hook it temporarily to a PB60 test jig with junk heatsink ... scope it , probe it ,alter it if needed. 7-15$ per voltage board... can't be beat!!

As far as downloading whole PDF collections , I set limits and even banned a whole range of IP adresses (they came from asia).
OS

[Salas]

Cheers. Your hunch they just tweaked the output to a new class A model seems very plausible. I think I remember they used to have bias button A/AB in some of their heavyweights in the 80's.

P.S. I drooled over that one as a teenager http://www.thevintageknob.org/luxman-L-550.html

[ostripper]

Originally Posted by keantoken

I was nearly sure the funky compensation would not work... But it does. Try getting 46db OLG at 100KHz! (on that note, look at C104a and R111a, and notice Cdoms are only 5pF... This is very much like what I use in my headphone amp and have simulated on power amps before)


I was going through your pictures, OS. I think I went through several hundred files, just looking at thumbnails on the HTML's. I noticed someone downloaded your entire PDF folder. Sheesh! At least make an appointment!

- keantoken

I got 36 db at 100k with 5pF. This is the only simulation I can go that low with except for Roenders RMI folded cascode amp.

The main reason it's discouraged I think is that if you screw up, there aren't many around that can help you, few people go down the rabbit hole. That's not really a problem on DIYAudio, right?

Most just build a DX ,self amp or something generic. Been there .. done that, I want audiophile DIY. This is a sure bet. I am amazed how the sim matched the manufacturers spec perfectly. I also simulated with the A970/c2240 models , same results.

OS

[ostripper]

Originally Posted by Salas

Cheers. Your hunch they just tweaked the output to a new class A model seems very plausible. I think I remember they used to have bias button A/AB in some of their heavyweights in the 80's.

P.S. I drooled over that one as a teenager Luxman L-550 on thevintageknob.org

That is one strange amp (below) , wonder what happens when the heatpipes wear out ? They do have a lifespan , similar to the new processor (cpu) heatpipe versions versions. I opt for very large convective cooling.


I have a MUCH larger power supply than those amps ... so I might just exceed the "drooled over" amp. I never gave 2 thoughts to class A after having the room heater Genesis stealth .. as I have said before , a symasym sounds just as good (I A/B'ed them -same loudspeakers) and does not make you SWEAT!!

OS

[Salas]

More than it had a full but open & silky sound with a TT I could not know back then. Surely I knew that I could not afford it on summer job savings though, and I got a small Nikko.

[Salas]

Originally Posted by ostripper

I never gave 2 thoughts to class A after having the room heater Genesis stealth .. as I have said before , a symasym sounds just as good (I A/B'ed them -same loudspeakers) and does not make you SWEAT!!

OS

Did you ever listen to Pass X line? Or had one from the F DIY line by any chance? Those are interesting examples of the hot persuasion to lend an ear.