Well, Yesterday I modified it again and I am happy to report I finally nailed it. My amplifier is now quiet to an acceptable level!
After two years, it was about time. And wow did I learn a lot in the process.
I'll borrow my friend's camera tomorrow to take good quality picture. Though it's not much of a MSSH anymore.
Originally Posted by swek
I have read a bit about your build of the SSMH and it seems you've had a great deal of trouble, I'm sorry to hear that. What are the key issues that you needed to resolve? You are the only one I've seen with cages around your tubes, Why is that? Why does it not go well with sitting on top of your DAC?
The reason for wanting to power it from a lower supply voltage is only because I had found a 30V supply readily available as well. I do believe you're right in that it's asking for trouble to box a switched supply within the amp. In the meantime I found wall warts available in a range of wattages as well, so I'm gonna go for that anyway.
What is the benefit of more voltage in this application? Does it reduce noise?
My big question now is: Will a 313mA (15W) power supply will suffice? I know the suggested one is around 387mA IIRC. Has anybody measured the actual current draw of the amp? I have simulated the quiescent current to about 151mA/channel. Which only leaves me with 28,65mA to spare if the supply delivers exactly to spec up to its worst case OL protection. If I find one I'll make sure to go with a beefier one though.
The only problem I've been fighting ever since day one was noise. For me this project started when I salvaged a transformer from an old stereo. The transformer has many taps, with the outmost set offering a nice 27-0-27 volts. Rectified, it gave me well enough tension for a linear regulated 48 V rail. When I figured that out, I went "Challenge accepted!". The goal of this project was to learn, not to have a final product.
The problem is, going this route means introducing mains AC into the enclosure, and bathing the whole thing in electromagnetic interference from the transformer. Those two things together means I'm shooting myself in the foot from the start. Even a perfectly quiet linear PSU will have its output corrupted if the wires pass too close to any wire carrying AC. I learned the hard way how important the layout of the internals is. Once you mix AC and DC together, building the amp becomes a lot more than replicating a schematic. Oh and grounding scheme is the most voodoo thing ever.
The cages around my tubes are there because the tubes are so close to the transformer, there was some magnetic coupling between them and the transformer. They picked up a 60hz wave along with all the nastiness of AC. It was ugly. The cages are ferrous and grounded, allowing them to efficiently shield the tubes from electromagnetic fields. Oh and not to mention I keep the amp on my desk surrounded with wifi, bluetooth, cellphone, wireless mouse... Unshielded, the tubes picked up the modulation of these signal.
As for the DAC, I suspect there was some coupling between the circuit of the amplifier and the toroidal transformer inside the DAC. But that's almost expected if I stack them directly on each others. There is no problem if they are side by side.