Pioneer SX-6000 clone

[Pacorl]

Dear gents,

in the last few months I have been attracted by vintage amplifier designs. I have read here a couple of threads by forum member Lineup describing the simulation process for a Leak Delta 70 and a Dynaco ST-80.

Using similar devices I have tried to simulate the power stage from a Pioneer SX-6000 receiver , but simulation result is quite bad. The reason for choosing this design is that I have a power supply capable of supply 54V that matches the required value for the original model. my first attempt was to clone a Sony STR-6120 but it appeared to be very complex to my level of expertise ( close to zero ).

I have seen that STV-3 is an obsolete device , and tried to use 3 x 1N4148 instead. Tried to use a VBE multiplier before with worst results.

I would appreciate some guidance for my next steps.





thank you

 

[east electronics]

your approach seems wrong to me
Amplifier like that is way to generic and may suffer from a number of issues
Given as a fact that none of the transistors you used are the originals expect then the results to be alike ...

Japanese was Masters in parts choice so if you had the models available there could be a chance that you had better results but no more than a few clicks .

What you stated as power trafo available is not really clear and you need to define that if its a 27-0-27 or a54 CT ( center tap ) trafo that makes ideal for a DX or a P3a amplifier ( given also as a fact that the power of the trafo is enough )

Both DX and P3a are trusted designs, well proven for years with plenty of support and documentation. Both will produce above 60W at this power supply and for a home use amplifier a 200W trafo will be close to ideal for 8R load .

Remember also that the sweet tone of this style of vintage amplifiers is a product of the actual mistakes these amps had in the tone controls ,the usual way too much effective loudness, and the general too much gain ...Even though sweet none of this things are real ...meaning also that this sound is produced by the total product not the amplifier board on its own .

So now you may choose how you would like to go ahead with this
Kindest regards
Sakis

 

[Pacorl]

hi Sakis, Thank You for taking time for helping me on this topic.

In fact my trafo was 27-0-27 , and you have to believe me when I say that I ve read a bunch of times your threads about P3a ( i.e. p3A comparison table) but at the end of the day I got lost around part choice discussions, I also believed 27 volts was a little bit insufficient.

I have 4 amplifies at home (all of them LTP in the input). In fact the most expensive one features many refinements that can be found in a modern design, but I dont like the sound. On the other hand I could fix a vintage radio incorporating a topology like the one I am trying to clone , and the sound was amazing.

I had read very good reviews about vintage clones using modern devices , in another forum In spanish.

Most likely I will revisit P3a or Carlos' DxBlamEs.
Best regards.

 

[Pacorl]

hi Sakis, Thank You for taking time for helping me on this topic.

In fact my trafo was 27-0-27 , and you have to believe me when I say that I ve read a bunch of times your threads about P3a ( i.e. p3A comparison table) but at the end of the day I got lost around part choice discussions, I also believed 27 volts was a little bit insufficient.

I have 4 amplifies at home (all of them LTP in the input). In fact the most expensive one features many refinements that can be found in a modern design, but I dont like the sound. On the other hand I could fix a vintage radio incorporating a topology like the one I am trying to clone , and the sound was amazing.

I had read very good reviews about vintage clones using modern devices , in another forum In spanish.

Most likely I will revisit P3a or Carlos' DxBlamEs.
Best regards.

 

[nironiro]

Have you consider Apex AX6, here on DIY Audio forum?

http://www.diyaudio.com/forums/solid-state/236256-retro-amp-50w-single-supply.html

Retro, no LTP input, quasi complementary output, single supply, cap output...

 

[Pacorl]

Hi nironiro

I reviewed this thread several months ago, thank You for reminding me.. Worth to read again.

 

[sgrossklass]

I presume things will work an awful lot better if R14, R19, R20 are connected properly. Are you used to another schematic style by any chance? Here it's lines crossing = intersection without connection, big dot = connection.

Still, there's a lot of reasons not to go for this topology nowadays:
* A 1000 µF at the output actually is pretty skimpy by hi-fi standards, even for an 8 ohm load. 1500 µF would be preferred, possibly even 2200 µF (or 3300 µF in 4 ohm territory). Can you tell why AC-coupled outputs went out of fashion?
* You'd be using lower emitter resistor values in the output stage these days. The output stage could also do with a Baxandal diode if you want to kepp things quasicomp.
* I'm also quite certain that an electrolytic alone does not make for a good output inductor.
* A single-ended input tends to generate a lot more distortion than an LTP job.
* And yes, I'd be going back to the Vbe multiplier since it gives you a lot more flexibility, also makes thermal coupling easier.

 

[Pacorl]

Hi sgrossklass,

I really appreciated your kind response due to the many clues provided. I will check again the simulation so that the junctions are placed properly , I obviously know the difference between an intersection and a junction.

Will provide the same tag for the components as shown in the original schematic ,and include a Vbe multiplier.

Regarding your question why output cap went out of fashion, I like your socractic approach.

Looks like I am expected to revisit some of the articles by JLH about hi-fi evolution and find the reason why. In the meantime will find easily what a Baxandall diode is and redo simulation accordingly ( I tried that previously ,no significant improvement recorded) .

Thank you again, I will post the results after adjusting some resistor values as suggested.

 

[Pacorl]

After following Sgrossklass´ recommendations , better results are recorded.

-I had to decrease the amplitude in the function generator down to 750 mVp
-I also had to increase the value of R5
-THD now is 0.067% and power is 26W.

I´d appreciate some additional comments

best regards