Need some help with RIAA Preamp

[david77]

Hi everybody.

I wanted to build this phono stage as a present for a friend:

http://www.vinylsa.co.za/pdf/A%20Sim...88%20Tubes.pdf

I have not made any changes to the circuit itself, though I'm not using the
power supply described in the paper. I decided to use a really regulated one.

Now I've run into the following problem:
With no record player connected to the input the circuit is reasonably quiet
and hum free.
But as soon as I connect a record player to the input I get a horrible hum.
I have made some pics of my scope connected to the plate of the second
triode.
The first pic shows what's going on there without a source connected.



The next picture show what happens when a 47k resistor is connected
across the input:



The scope was set to X: 10mV/Div Y: 5ms/Div

So the trace in the first pic shows about 5mVpp and the second one
about 30mVpp.
What I really don't get now is why this happens as soon as the input is
"loaded" with 47k (equivalent of cartridge).
I reckon it's got something to do with faulty grounding but I have not been
able to find my error.

Maybe someone could point me in the right direction and give me a push?

David

[Arnulf]

Originally Posted by david77

Maybe someone could point me in the right direction and give me a push?

The circuit is pretty straightforward and you have an oscilloscope handy. Measure PSU ripple (and let us know which version of PSU you chose, there appear to be two versions presented in that PDF), then observe anode and cathode voltages with 47K shunt across input - does it appear that PSU noise is injected at any of those four points ?

Remember: high voltage ! You only have to observe AC waveforms.

What is your actual grounding scheme (star ground or not) ?

Where do you take your ground from for each section ?

[kavermei]

Does it happen on both channels?

Could be a ground problem. Did you connect circuit ground to mains earth?

Also check that it is not the regulated supply that is oscillating.

Happy hunting
Kenneth

[david77]

It does happen on both cannels. Circuit ground is not connected to chassis earth.
Thanks for your replies guys.

I'm using star grounding scheme, central ground point is the negative side of the power supply.
I had time to think about it now and that might actually be my mistake.
My working theory at this moment is as follows:
1. The ripple that is visible on the scope looks like some mishmash of 50/100Hz frequencies.
2. I think I connected the RCA jack's ground directly to my star point. Not the same
wire that goes to the circuit board. Obviously that would result in a different potential
on the ground wire leading to the RCA jacks. When I connect this point to the amp's
input via 47k (or the cartridge) I inject the noisy stuff from the ground wire into the
circuit. Result: appalling hum.
I'll have to check that out tonight.

If the above turns out to be the problem I'll have to rethink my grounding system.
Maybe it would be better to put the central ground away from PSU to a point closer
to the circuit board? So all grounds would be closer to the ground potential at the
most sensitive point in the whole amp.
I would be interested in what you guys think about grounding.

The PSU is OK, there is no oscillation and only a little noise, no hum measurable.
I decided to use this circuit (sorry, only German):
Der Treffpunkt für Elektronik, Mikrocontroller, Röhre, Forum, Platinen, Schaltungen, Onlineshop

It's rather straight forward. But it has real regulation as opposed to the PSU described
in the paper. I prefer that. Additional to the regulator I put in a chain of C-R-C
filters with a calculated ripple rejection of about 90dB.

This is what it looks like on the scope:


Another thing might be inductive coupling from the mains trannies. I couldn't
use separate enclosures, so they are closer to the circuitry than I would usually
prefer. I use copper clad circuit board as a shield between PSU and signal
compartments. I might actually stick the whole circuit board (excluding tubes) into
a box made from the stuff. I have had very good results doing that before.

Well, we'll see...
David

[Arnulf]

If you have any tank capacitors in circuit (away from the PSU, usually just before anode resistors), put your ground point there.

[Bennybuis]

Hallo, i am Ben and I just build a few of those pre amps, mine are virtually noise free. Let us know how it goes, I am interested in a regulated psu as well.


Kind regards from Holland


Ben

[david77]

@Arnulf:
Yes, that's what I was thinking. I'll probably rewire the thing tonight. What else would
one do on a friday evening...

[david77]

@Ben:
Just out of curiousity: At what voltage are you running your pre amps?

David

[kevinkr]

Your star ground needs to be connected to the chassis of the pre-amp at one point otherwise differences in potential between the chassis and your circuit ground will be electrostatically coupled into the circuitry.

47K is the "recommended" cartridge load impedance, most MM cartridges have source impedances of considerably less than 1K, a 220 - 560 ohm resistor is a reasonable approximation for most MM types. High output MC are generally well under 50 ohms, and low output MC from a high of maybe 15 ohms to a low of 3 ohms or so.

Any noise level over a hundred uVrms (not mV!) at the output of a phono pre-amplifier should be a cause for major concern if you want anything approaching a reasonable SNR. (Assuming ~ 40dB of gain referenced to 1kHz)

[Bennybuis]

Hallo,

I aim for 250 Volts B+. Here are some pics of my RIAA.

Kind regards from Holland


Ben