GM70 SE amp project

[gideon1990]

I bought that regulator v4lve lover was showing in his post for that reason, he showed me some pretty awesome measurements on those. Now I just need to find some time in the lab at work because I want to see if I can find a meter that can show the actual noise floor instead of "less than 0.2-0.3 mV". Not that it matters anymore at that point but mostly out of curiosity

And well 20V toroids can be had for little money anyway

[audiowize]

Originally Posted by Schmitz77

I wouldn't recommend high freq. switching PSU for tube amps.
They are just a cheap and ready made solution, but they contaminate the whole amp with garbage frequencies that no one with a clear mind would prefer to have in such a unit.

When I have used them and gone looking for those contaminants, I was unable to find them.

[bobikdartanan]

Rectangular pulses of switched sources have a wide spectrum and short duration, can penetrate wires and as radio interference, and cannot always be seen.
Not all interference from the 120vac power supply is visible or audible, but we connect filters to the 120vac input.

[gideon1990]

Originally Posted by audiowize

When I have used them and gone looking for those contaminants, I was unable to find them.

With a real mean well you might be in luck that they are quite decent and switch somewhere above the audio spectrum. We did run into that with an in house designed impedance analyser where the switcher we used for temperature control of the measurement cell was quite visible at specific frequencies. In the end it was fixable for us with a thick metal shield and shielded cables but I'm not sure that I would want to deal with that in a DHT amplifier

[Schmitz77]

Originally Posted by audiowize

When I have used them and gone looking for those contaminants, I was unable to find them.

We have done so much, for so long, with so little,
we are now being qualified to do anything with nothing

[Flikoman]

Great project and great OPTs.

Can you give some more info on OPTs? Specs etc?

Regards

[Rod Coleman]

Originally Posted by bobikdartanan

Rectangular pulses of switched sources have a wide spectrum and short duration, can penetrate wires and as radio interference, and cannot always be seen.
Not all interference from the 120vac power supply is visible or audible, but we connect filters to the 120vac input.

Yes, a mains→DC converter creates broadband noise that is not easy to control.
In my day job, I often have to add multiple-stage EMI filters to industrial equipment, just to reduce the SMPS noise enough to get the industrial product through the EMC standards tests.

This is not usually a problem for the DHT amplifier, if it does not use cathode bias, and it has been carefully wired.

The problem arises when there is a continuous ground connexion between the DHT amplifier and a DAC, or if the DAC is close enough to the amplifier to pick up EM interference. Broadband noise can degrade the DAC's analogue section, or the timing of input signals. It can be demodulated in current to voltage stages [I2V] if older opamps are used (before modern designs arrived with their EMI-hardening). Worst of all is EMI corruption of the Timing source - the crystal oscillator. If you have paid out extra for a low phase-noise oscillator (an excellent investment IME) the last thing you need is some clumsy power supply coming along and undoing the good work.

Think about why USB → I2S stages often come with a set of digital isolators - in order to break the USB ground connexion from noisy laptop PSU bricks.

This is the (main) reason I don't recommend mains→DC as the raw DC for use with my filament regulators. The problems arise in unexpected places.

Preamps or MC stages can also suffer, especially if NE5534 , AD797 or other older op-amps are in use.

The problems can be solved, more or less, if your design skills cover RF and EMC adequately - but the performance is usually still worse than if a good screened power toroid is used. And now that these are more widely available, and low cost (Antek AS series in the US, Canterbury Windings and Air-Link (audio grade) in the UK/Europe, Toroidy in Poland - there is little advantage from these converters.

[Rod Coleman]

The pictures in the previous post were to illustrate a couple of other points.

In the block diagram, the capacitors from input to ground, and from output to ground will form a loop from the noisiest part of the converter (Raw mains, rectified and chopped up at the switching frequency) and ac-coupling it to the output negative which is connected to the filament. This is not a very happy place to ac-connect the filament to.

The other shot is of the Leakage current specification. This is the value of ac current that might flow from mains into the output. Needless to say, it's very noisy.

If you use cathode bias, the current flows through the cathode bypass cap, or the resistor.

Note also, that the converter's output cap is now in parallel with the amp's cathode bypass cap - which is not helpful.

I'm not trying to say it can't be done - but I do question whether the convenience is a good trade off for the performance, especially if remedial work is needed

[Schmitz77]

To bring this high frequency switching noise within an analog tube amp is to deal with many problems. But people try to save a buck here and there. In the end, they want a great amp with minimal effort and cost without knowing, that those principles never have worked, especially in audio.
Reknown companies like Sony or Accuphase have shown the kind of effort it took to keep the digital noise outside of the analog circuits. Its not a small margin. I can audition it in my own system. When switching the digital gear on, analog sound becomes subtile worsening. Its very delicate to deal with that, isolation transformers can help.
I would never in this whole life come up with the idea to install such a digital contamination device in a pure analog unit to save five bucks or have a ready made solution. Because its not in the audio world. Maybe everyhwere else. And thats the problem. Because they were used so often, the AC becomes more and more contaminated with high freq. garbage. In the end, we need more to invest for pure AC supply. In a good (clean) environment this isn't necessary at all.

[gideon1990]

@Rod Coleman

What is your opinion on EMI filters in tube amps? I have not seen many people use them but every now and then you spot someone does use one.