Symasym - the next generation (supersym)

[ostripper]

After many E-mails and even a few download attempts at my "supersym" PCB ( It's not done yet.. have patience )
, I am starting a new thread to : 1. Share the existing amp (PCB and construction advice) , 2. Design a professional board that will maximize performance and fidelity. , and 3. develop it further with JFET inputs (anatech's suggestion)and a MOSFET OPS.

The design as it is now...


It has been the best sounding by far of my DIYA creations.
It stays biased within 2mv , any environment , drives paralleled 15" 3-ways all day at full volume .. no turn-on thump , completely stable in every way.

The base schematic is here ..

I am using C21 as my HF compensation. I have a 470K in place
of c6/r13.
Attached is the LT file to see for yourself (all models included). Make sure the "supersym .txt" in the zip file is in the same directory as the .asc file (models)..
OS

[ostripper]

I have only gotten this far on the PCB , I whether a am wondering whether a 3 X 9" with opposed output pairs would be a better
design instead of "the long way" (Attached).
OS

[forr]

Hi,
On this kind of arrangement for the low power stages, two differential stages, the second one being loaded by a current mirror, I wonder if C7 is really useful.

[destroyer X]

to avoid be running over the same old story again and again.

This is a lovely amplifier, with astounding dinamics and nice sonics, and has room to developments...not too much explored..only few guys made modifications on it.

Then you will be offering something good for us to assemble, and will be decreasing our forum "agging", as we were turning a lot of oldies doing only ancient designs.

All i could do was my old style schematics...but for sure you can do something more modern as you are younger than me and more modern inside your mind.

This one is much more modern, some folks said Sansui made into the eigthies..but i could never find that Sansui schematic that matches.

Maybe a Marantz had schematic almost alike.

I love this amplifier performance.... wonderfull gift Michael Bittner left for us.

regards,

Carlos

.................................................. ......

Nobody has made the "forum" amplifier, with a selection of subcircuits... alike to use the most brilliant ideas...input from Roender, second stage from Symassym and output from other guy.... some forum different circuit, with all our good ideas united.

[Bigun]

Originally posted by ostripper
I have only gotten this far on the PCB , I whether a am wondering whether a 3 X 9" with opposed output pairs would be a better
design instead of "the long way" (Attached).
OS

Hi OS,

This looks interesting.

my 2c: the pcb shape and output device arragement might depend on how you feel this would be heatsinked ? Some people have chasis where it's convenient not to have the opposing pairs but like to put them all along one edge.

The other thing is whether you would have a benefit if you were to route the main power busses around the opposite sides of the board, this puts all the circuitry inside the 'magnetic flux loop' created by these high currents - can you route the heavy traces along one side only so that +/- are next to each other (most elegantly done with 2-sided pcb but TGM1 did it one side since it has a wimpy current compared with your beast) ?

[ostripper]

By DX - to avoid be running over the same old story again and again.

I will be good here , Carlos.. Even as this is similar to Mike B's design , it is NOT. It is running at 77v rails , has twice the bandwidth of the "sym" ,different CCS , and a "turbo" output stage.

When the board is done it will also have a cascode and jfet inputs , I
have actually built, designed and listened to a smaller version with these mods. At this point it really IS NOT a symasym any more. Something closer to it can be found here (Copyright).
http://ampslab.com/bi240.htm The original has not been "supersized" to this point, so this thread is to explore that aspect.

By forr - I wonder if C7 is really useful.

A real good observation.... I have tried this amp with just C7 , just C8 , stays stable-sound does not change. The only benefit of the 2 Cdoms is that they load the LTP equally at low levels.

By BIGUN- The other thing is whether you would have a benefit if you were to route the main power busses around the opposite sides of the board, this puts all the circuitry inside the 'magnetic flux loop' created by these high currents - can you route the heavy traces along one side only so that +/- are next to each other (most elegantly done with 2-sided pcb but TGM1 did it one side since it has a wimpy current compared with your beast) ?

Yours is not "wimpy" seriously , you have read my mind again, while laying out the 'opposing" pair layout (attached) , I noticed all the traces were running parallel , these were not at right angles and parasitics might "erupt". Not good for an amp like this.
the board layout I already posted will fit on a 5" X 10/15" HS.
OS

[jaycee]

Make room for a cascode in the LTP - beneficial with both transistors and FETs in the front end.

Use a triple output - IMO this is the best output stage type, every time I have looked at it, it reduces THD considerably.

[Rafael L]

Hi

I am designing an amplifier with this topology, the suggestion would be in Miller compensation , schema Attached:

[Rafael L]

Clipping and square wave, 20Khz

[Rafael L]

Loop DC (without input filter) 1,7Mhz -3dB
I will take problem?