STK 1050II - 1080II and STK 3056 - 3076 new replacements

[Benji01]

Disclaimer: Everything that is shown in this document, you do at your own risk. I am not liable for any harm and damage that is done by following this document. Remember electricity and mains voltage and even dc voltage can kill you. So please be careful and only do it when you know what you are doing and have proper training for working on electronics.


This following document and the corresponding files are not intended for commercial use. The distribution and the commercial use of this document and the included files is strictly prohibited. The commercial production of these PCBs and the commercial production of all modules is strictly prohibited. The original schematics are owned by their respective owners and are only in this document to demonstrate how to make replacement parts for components that are hard to come by.
The following document is about the design and testing of STK replacement modules mainly of the STK1050II and the STK 3076. The STK3056 can be replaced by a STK3076.
The PCB for the STK1050II can be used without much modification to replace the STK1070II and the STK1080II.

Originally, I intended to document my development process here as a series of posts, but after realizing that there are way too many pictures and way too much text, I decided to edit all of this in a pdf that can be downloaded.

The project should have been finished a long time ago, but due to problems with my original design, I decided to rework all PCB’s to get the best possible replacement modules.

[MegaHurts]

Awesome job, thanks for sharing.
I have a R700 and R900 that I'd like to make stks for to try and get the dc offset and distortion on one channel down.
A BOM from digikey or mouser would be great if you have the time.

[shakalnokturn]

Nice work!

Along similar lines I recently completed a layout for the Technics STK8050, fake replacements have been around for about 2 decades now making any repair attempt a deception for the unaware. There are a couple other DIY layouts around but they aren't pin to pin compatible.
The main differences in design compared to yours are that the device is larger so I was able to mount the output transistors sideways and use the original STK mounting hole spacing for attachment reducing the mechanical hassle. The downside is that they need insulating when mounting which is a little fiddly.
As output transistors have their terminals facing one another I used a SOT89 as the one used for bias adjustment and thermal compensation, placed physically between the + and - collectors with a few 1pF capacitors as thermal jumpers. (I couldn't find any thermal jumpers available here.)

You'll find that your emitter resistors manage better at full power if you give their terminals larger copper areas to solder to, something my design did not have room for.
Your design may suit an aluminium IMS with TO263 transistors as output stage, that would also help emitter resistor power dissipation problems and simplify thermal compensation transistor mounting but the layout would have to be single-sided.

[Benji01]

This post has the pdf document attached. The document contains following things:

-   The comparison between the original modules and the fake ones
-   The repair of the Yamaha R700 and the first revision of the design
-    The complete building instruction for the STK10XXII replacement and the complete BOM
-   The complete building instruction for the STK3056 / STK3076 replacement and the complete BOM
-   The complete building instruction for the test PCB and the complete BOM
-   The complete test procedure of all modules including oscilloscope shots
-   The mistakes that I made
-   Thermal testing of my design
-   Included schematics (schematics with better resolution are included in the files)

[Benji01]

This post has all files attached for the replacement STK1050II, STK1070II, STK1080II. The files include:

-   Gerber files for the PCB manufacturing
-   The source files for the PCB so that you can edit them in EasyEDA
-   High-res schematic
-   DXF drawing of the mounting plate

[Benji01]

This post has all files attached for the replacement STK3056 / STK3076. The files include:

-   Gerber files for the PCB manufacturing
-   The source files for the PCB so that you can edit them in EasyEDA
-   High-res schematic
-   DXF drawing of the mounting plate

The STK3056 / STK3076 is a little modified, so that the adjustment of the DC-Offset is now possible.

[Benji01]

This post has all files attached for test PCB. That PCB is required to test all modules without installing them in a receiver or amplifier. The files include:

-   Gerber files for the PCB manufacturing
-   The source files for the PCB so that you can edit them in EasyEDA
-   High-res schematic


The files for the LT-Spice simulation of the modules is also attached in a separate folder.

[Benji01]

The PDF is way more in detail then these posts. So definitely take a look at the PDF before ordering any parts.
I hope that these files help other people to keep their beloved vintage gear alive.
The project was quite a lot of work but in return I learned a lot. So, it was definitively worth the effort.
Feel free to contact me if you have any questions about this project.
If you tried these modules and installed them in your receiver or amplifier please upload pictures here.

[coromonadalix]

wow  incredible job     KUDOS   

[Chris56000]

Thank you very much Sir – this sort of superb work is so valuable to many Members who might have valuable hi–fi pieces that may otherwise have to be needlessly scrapped for want of one these Sanken/Sanyo devices!

Above all, thank you for sharing so generously – alas to say my design skills aren't up to anything as sophisticated as this, as I was trained as a repair rather than a design engineer!

Chris Williams

[floobydust]

Great project!

Note: you can partially test STK's with a multimeter - it's usually the output transistors that fail.
On Diode-Test using the STK internal schematic there are many tests of junctions possible at the pins.
In the repair shop, I've replaced dozens of STK's in the 80's and a second test was tapping them with a screwdriver - the bonding wires would break and they would have intermittent connections inside, popping sounds etc. so we always banged on them as another way to know if they are defective.

[shakalnokturn]

Thanks Benji01 for all the documentation even if I'll probably never build one because of the machined heat spreaders I really appreciate the effort put into the design and report.
 
It's funny that I also ran into oscillation problems with my STK replacement project, i think the main reason for this is the transistors used having different hFE and fT from the original design. None of the STK DIYers around have mentioned measuring transistor gain on original modules.

As a related side anecnote to floobydust's comment:
A recent repair on a Korean built Telefunken amp. that had blown an output fuse and was showing full negative supply voltage on that output was surprisingly not due to a shorted output stage on the STK0040. (Although without the fuse it would have ended that way...)
The supply rail was also measurable on one of the input pins (-42V on pin1, IIRC) and after isolating the STK input from the rest of the circuit the voltage was still there coming from the STK itself. Yet there was no measurable blown transistor that would explain this. Only after cracking the STK open did I realise there was a short between pins 1&2, concluded to a to a case of tin whisker, gave it a little zap with the lab power supply to blow the short, it has been working normally since.

[coromonadalix]

Great project!

Note: you can partially test STK's with a multimeter - it's usually the output transistors that fail.
On Diode-Test using the STK internal schematic there are many tests of junctions possible at the pins.
In the repair shop, I've replaced dozens of STK's in the 80's and a second test was tapping them with a screwdriver - the bonding wires would break and they would have intermittent connections inside, popping sounds etc. so we always banged on them as another way to know if they are defective.

same here, changed a lot of them ...

[floobydust]

It was funny the whole shop would be listening to techs banging screwdrivers on heatsinks to flush out an intermittent STK module.

They are a great low distortion amp when they work. I think many counterfeits or low quality ones exist. Many did not look like Sanyo or Sanken quality.
Biggest problem is they had many different (audio power amp) STK's depending on VCC because the bias current is set internally by a printed resistor. OP has the trimpot, hurrah!

For testing STK0040II, STK0050II, STK0060II, STK0070II, STK0080II using a multimeter, this is my chart. The driver stage is usually STK3082.

[yarebel]

Benji01,
Thanks for the great job!
You are my hero!

Can you sell a couple of modules?