> If the power amp is switched on, spiking noise shows up on the supply rails of the preamp.
That is a function of how much resistance is in the line from the hi-fi shelf back to the generator. (Mostly in your walls and the line to the street.)
Transformers won't fix it.
Its interesting that you have said this.
We have a 100 Amp panel, in our late 1950's house,
but an electrician is coming in tomorrow morning,
and we are going to discuss installing a new 200 Amp panel.
I just hope the resistance issue is not between our panel and the hydro pole.
However, a few years ago , I had a new 6 awg earth ground wire installed
from the panel to the copper water pipe as it comes into our house.
But our hydro earth remains quite noisy.
After tinkering around with various power supply designs,
I've come to the conclusion that using a power transformer
that has low capacitive coupling between the primaries and secondaries helps
reduce spiking noise on the supply rails.
In this aspect, an E core , with dual bobbins, measures much better than a high quality toroid.
Also, the Fo Felix EMI filter, being offered on this website, certainly helps reduce spiking noise.
I recommend building one and giving it a test.
But as you've said, the spiking noise is a function of resistance between the street and the audio rack.
.
How did you measure the 'spiking noise', or was that conclusion from listening tests?After tinkering around with various power supply designs, I've come to the conclusion that using a power transformer
that has low capacitive coupling between the primaries and secondaries helps
reduce spiking noise on the supply rails.
...Also, the Fo Felix EMI filter, being offered on this website, certainly helps reduce spiking noise.
It would be good to identify how mains borne noise can get through to your amplifier's output signal, as that type of noise transfer is usually eliminated by good amplifier and audio system design.
An engineer (PhD) where I was employed used to say you could reduce the impedance of the AC line by putting a capacitor across it - much like you do with a DC supply output. The context was inrush current transient testing of switching power supplies, connected to the AC precisely at the 90 deg phase angle mark.
Same point where any full wave bridge connected to a lot of capacitive filtering draws the current spikes, which I assume is what is meant by "spiking noise".
Of course, the type of capacitor used matters and its voltage rating. I've only done this with 1 uf 400V film caps... Pretty sure a circuit simulation is in order, to see what steady state current is for different larger values of capacitance.
Perhaps it's not really practical, considering the amplifier capacitors @ maybe 10,000 uf (even reflected through the power transformer ratio) is what the AC capacitor would be discharging its stored current into.
Same point where any full wave bridge connected to a lot of capacitive filtering draws the current spikes, which I assume is what is meant by "spiking noise".
Of course, the type of capacitor used matters and its voltage rating. I've only done this with 1 uf 400V film caps... Pretty sure a circuit simulation is in order, to see what steady state current is for different larger values of capacitance.
Perhaps it's not really practical, considering the amplifier capacitors @ maybe 10,000 uf (even reflected through the power transformer ratio) is what the AC capacitor would be discharging its stored current into.
You can connect capacitance across the mains AC, but you need to appreciate why you are doing it and whether it is allowed by your electricity and safety codes.
One typical application is to modify the power factor of the site when there is a substantial inductive load (like a warehouse with lots of old-style fluorescent lights). The capacitor has to be X rated, and suitably protected for over-current.
One issue is that any such capacitor can draw in large current levels and spikes from outside your site (eg. house), which can then make you local supply even noisier because the wires coming to your house are conducting noise spikes that otherwise wouldn't be coming in with the same magnitude.
If your house has a noisy mains, then you need to identify where the noise is coming from, and what it is. It may be your own lights or washing machine or ...., in which case the cause needs to be modified first and foremost.
One typical application is to modify the power factor of the site when there is a substantial inductive load (like a warehouse with lots of old-style fluorescent lights). The capacitor has to be X rated, and suitably protected for over-current.
One issue is that any such capacitor can draw in large current levels and spikes from outside your site (eg. house), which can then make you local supply even noisier because the wires coming to your house are conducting noise spikes that otherwise wouldn't be coming in with the same magnitude.
If your house has a noisy mains, then you need to identify where the noise is coming from, and what it is. It may be your own lights or washing machine or ...., in which case the cause needs to be modified first and foremost.