Hi KevinHeem,
Before you swap capacitors in your present PSU, check out the SLB psu. It utilizes an active bridge and capMx and works very well in my class A amp builds.
The SLB (Smooth Like Butter) Active Rect/CRC/Cap Mx Class A Power Supply GB
Transformer coupling. Not PSU DC rail ripple; larger capacitors most likely won't help.
I take this from years of reading John Atkinson's measurements at Stereophile, where he always analyzed that mains frequency and its odd harmonics (3rd, 5th, ...) stem from magnetic coupling with the power transformer.
Rotate power transformer, re-orient it (mounting vertically, e.g.), or try to put a steel shield around the mains transformer. (steel type has to be ferro-magnetic, so attracted by a magnet).
Hope this would help,
best regards,
Claas
P.S.: attached a picture how my M2 looks like - that helped, along with rotating the transformer for minimum noise ...
Attachments
This is what the spectrum looks like with the transformer removed:
As compared to before removing transformer:
I don't know Kevin, it looks rather strange.
I was expecting to see a linear reduction in the entire spectrum.
Your 50 Hz fundamental is lower, but its harmonics are higher or stay the same.
Similar things with your 1 kHz harmonics.
Something in your PSU is causing this big non-linearity I think.
I'm sure someone more knowledgeable than me with be able to decipher the puzzle.
What do you see if you measure just the PSU with some resistive load?
I was expecting to see a linear reduction in the entire spectrum.
Your 50 Hz fundamental is lower, but its harmonics are higher or stay the same.
Similar things with your 1 kHz harmonics.
Something in your PSU is causing this big non-linearity I think.
I'm sure someone more knowledgeable than me with be able to decipher the puzzle.
What do you see if you measure just the PSU with some resistive load?
Haven't tried that...any ideas on how much resistance I would use to load the PS?
I will, but could use some help with how that would be accomplished.
I've been trying to reduce the amp down the it's simplest form. I removed the fancy diodes I was using and replaced them with a standard rectifier block. I removed the input transformer which I built into this amp. I'll leave the transformer out of the case until there is progress on other fronts.
When I originally built the amp and on first turn on there were mistakes with one optocoupler pin not being inserted in it's socket, resulting in the source resistors getting very toasty. wondering if that event is somehow linked to what's going on now>
Hi Kevin,
if we use Ohm's law: I = V / R, then
R = V / I
So let's say:
Vsupply=24V
current through that half is for example 1A
R = 24 / 1 = 24 ohms
now power wise 24V x 1A = 24W
use a 50 watt resistor, even better on a heatsink
same goes for the negative half
use the real values of your supply and the bias current
the other question:
from the original schematic, connect the junction between R1 & R2 directly to the positive of C2.
of course you have to find a way to break the link (traces) to the buffer and from the Edcor.
might be easier to build a small cct of R1 & R2 on a piece of proto board
and connects its output to C2.
below is the schematic for your reference
keep in mind I have assembled but not tested my M2 yet, so you are far ahead of me.
if we use Ohm's law: I = V / R, then
R = V / I
So let's say:
Vsupply=24V
current through that half is for example 1A
R = 24 / 1 = 24 ohms
now power wise 24V x 1A = 24W
use a 50 watt resistor, even better on a heatsink
same goes for the negative half
use the real values of your supply and the bias current
the other question:
from the original schematic, connect the junction between R1 & R2 directly to the positive of C2.
of course you have to find a way to break the link (traces) to the buffer and from the Edcor.
might be easier to build a small cct of R1 & R2 on a piece of proto board
and connects its output to C2.
below is the schematic for your reference
keep in mind I have assembled but not tested my M2 yet, so you are far ahead of me.
Attachments
Hi All,
Just finishing a B1 Korg Pre and starting to look at maybe building a M2x to go with it.
Looking at parts, thinking transformers, Toroidy.pl have "standard", "Audio" and "supreme audio" grade toroid.
Anyone have experience using shielded/etc transformers (their "supreme audio"), they look a great idea on paper but curious about real world, they worth the extra?
Thx.
Damian.
Just finishing a B1 Korg Pre and starting to look at maybe building a M2x to go with it.
Looking at parts, thinking transformers, Toroidy.pl have "standard", "Audio" and "supreme audio" grade toroid.
Anyone have experience using shielded/etc transformers (their "supreme audio"), they look a great idea on paper but curious about real world, they worth the extra?
Thx.
Damian.
Still outside. If you look at the bottom right corner of the spectrum analyzer I put notes about the configuration.
Here's the spectrum only the right channel connected to PS:
And then both channels connected to PS again:
I added a grounding wire towards the back end of the PS board, this seemed to help some, but still not right.
Last edited:
Regarding power supply toroids, my experience with AnTek toroids was that connecting the internal electrostatic shield actually added hum to my amplifier.
A transformer with external shielding might offer some benefit but standard toroids already have pretty low leakage and EMI and are probably beyond the point of diminishing returns for a loudspeaker amplifier. I could see possible advantages to the shielding for preamps where the transformer is inside the circuit chassis.
Also, I prefer the F6 to the M2x, possibly because you can more easily increase the bias on the F6 and get it really sounding good. Other things that I think make the F6 sound better are the use of all N-channel devices for the outputs and the use of a Jensen transformer at unity gain (for push-pull polarity reversal) instead of the Edcor for voltage gain. They really are very different amps and although I enjoy the M2x input module differences, I can't seem to get past the power amp sonics. To me, the F6 just sounds better.
I’ve been using toroidy audio supreme transformers for a few years now and like them a lot. I’ve built a M2x with this transformer and i had no issues with noise pick up etc.
Is it worth it? I’m not sure. I’ve seen many great builds not using a shield but I’d try to get a shielded transformer for builds like the M2x and F6 that’s just me.
Kevin,
I can not figure that one for you but I think you are not using RV1 for what its role is - offset adjustment.
There must be a better way to suppress that 120 Hz spike.
Maybe you can try a Cap Multiplier, like in Zen V4 or FW F3.
It will eat 4 volts from your supply and it has to be done for the positive and negative rails but I think is worth a try.
I use such things with all my circuits - 3 x ACA, MoFo, Pearl I, 12AH7 pre, B1, Zen Lite headamp, E88CC SRPP headamp.
Works great.
I can not figure that one for you but I think you are not using RV1 for what its role is - offset adjustment.
There must be a better way to suppress that 120 Hz spike.
Maybe you can try a Cap Multiplier, like in Zen V4 or FW F3.
It will eat 4 volts from your supply and it has to be done for the positive and negative rails but I think is worth a try.
I use such things with all my circuits - 3 x ACA, MoFo, Pearl I, 12AH7 pre, B1, Zen Lite headamp, E88CC SRPP headamp.
Works great.