Thru holes in the pcb are too small for anything but this 0.5mm enameled wire. I think I got just enough to do both channels.
for now I've soldered a pair of MLCC 20pF caps across the plate-grid and grid-cathode on the input side of both tubes at the socket pins, and I've soldered a pair of 1K8 resistors to the inputs on the PCB and soldered the shielded cable from the RCA jacks to those. There is a reduction in that noise, and it didn't have a noticeable attenuation on the overall sound, I still had the volume control at the same point. Not sure if it's rolling off the highs at the correct point any more, but it sounded right. When I get the time, I'll use some of that enameled wire to keep the signal from the tracks in the PCB and send it straight to the 1K8 resistor that's soldered to the socket pin.
for now I've soldered a pair of MLCC 20pF caps across the plate-grid and grid-cathode on the input side of both tubes at the socket pins, and I've soldered a pair of 1K8 resistors to the inputs on the PCB and soldered the shielded cable from the RCA jacks to those. There is a reduction in that noise, and it didn't have a noticeable attenuation on the overall sound, I still had the volume control at the same point. Not sure if it's rolling off the highs at the correct point any more, but it sounded right. When I get the time, I'll use some of that enameled wire to keep the signal from the tracks in the PCB and send it straight to the 1K8 resistor that's soldered to the socket pin.
Add a 0.01uF to 0.1uF ceramic disk capacitor directly from each RCA jack shell (GND) connection to the pre-amp chassis, keep it very short.. This has helped a lot in my RF rich environment a mile from a bunch of AM transmitters + my local wifi and cell phone.
And one other thought, if the RCA cables use a spiral wrap shield rather than a woven mesh you might want to replace. Spiral wrapped shields aren't particularly good at shielding RF. Check with KAB or elsewhere for better cable.
Last edited:
is this the schematic?
if so, adjust volume control to noisest place, and temporarily install a 10K potentiomer between the 51K and the shielded wire/volume control, (one connection to the wiper, the other to one end of the potentiometer) Adjust to best noise noise floor, then turn off the unit, measure the resistance of the test potentiometer and replace with a resistor close to that size.
Adding a significant amount of series resistance between the cartridge and the input tube grid will appreciably increase the pre-amp noise floor due to johnson noise generated by the series resistor.
Ferrite beads are another option placed on the leads from the RCA jack to the board.
Ferrite beads are another option placed on the leads from the RCA jack to the board.
I suspected as much. I'm using shielded cable for the inputs, and I'd have to wrap that around a ferrite bead, yes?
Ferrite beads are tiny. You could use one on each conductor from the input RCA socket,
if the shield will fit through it (doubtful).
But a 0.01uF ceramic capacitor from the RCA input socket ground terminal to the chassis would also be advised,
and works a different way from the ferrite bead. Using both is ok.
if the shield will fit through it (doubtful).
But a 0.01uF ceramic capacitor from the RCA input socket ground terminal to the chassis would also be advised,
and works a different way from the ferrite bead. Using both is ok.
Johnson noise is not an issue when sizing a dc coupling resistor. That is for grid stops.
Also, if you are having Johnson noise issues with these small resistor values you are not using the correct types of resistors.
I would rip the volume control out of the front and stick it in the output circuit where it belongs. Only an idiot designs a circuit where the volume control is in the front.
Last edited:
And "thisusername" I suggest you go read about Johnson noise, any resistance above a few hundred ohms will generate significant noise, values much beyond 1K will contribute significant noise. All resistors generate Johnson noise until absolute zero (0°K), what you are talking about is excess noise which varies from resistor type to type.
The "DC coupling resistor" (AKA grid resistor) also serves as a cartridge load resistor has a specified value for MM types of typically 47K. I assumed you were talking about a grid stopper resistor.
The "DC coupling resistor" (AKA grid resistor) also serves as a cartridge load resistor has a specified value for MM types of typically 47K. I assumed you were talking about a grid stopper resistor.
you need about 120-150ohms loading on it, Then you would treat is as a voltage source of 0 ohms for dc coupling, and add the input impedance in series as your coupling resistor, or capacitor couple with the approximate load line which it would be somewhere around 1uF
Since the step-up transformer has voltage gain, the secondary loading could be somewhat critical.
The cartridge loading may have to be distributed between the cartridge input and the secondary.
Or some actually do leave the step-up secondary open circuit, and have all of the
cartridge loading right at the input. I tend to think that's the better way.
The cartridge loading may have to be distributed between the cartridge input and the secondary.
Or some actually do leave the step-up secondary open circuit, and have all of the
cartridge loading right at the input. I tend to think that's the better way.
If you build with the wrong resistors, I can see Johnson noise, but we have resistors now less than 200 ppm so a lot of that theory doesn't apply anymore
I think it might help everyone get on the same page if "coupling resistor" were defined. But you're incorrect about resistor noise.
All good fortune,
Chris
All good fortune,
Chris
That's absolutely not true, Johnson noise is the unavoidable byproduct of electrons in motion in the resistive material and is a function only of resistance and temperature. You are correct about excess noise which has been improved substantially in modern resistors.
https://en.wikipedia.org/wiki/Johnson-Nyquist_noise