WTF. What's going on here.

[Cassiel]

Hi, I have built a phono preamp and it has HUGE problems. The noise is killing me. OK, the final stage is a 12ax7 cathode follower; it's DC coupled and the cathode has a 120K resistor. Voltage at the grid is 100V and at the cathode 115V. Both channels show this strange reading The only clue I have is that there is 2.5 volts between grid and cathode with the preamp switched off and no voltage on the filter caps. When I pull out the power cord the voltage goes slowly back to zero when I plug it back in it slowly goes to 2.5 volts again. Can you help me? What's going on here? Here's the schematic. Those are not my readings - wish they were.

 

[costis_n]

TF that is going on there is that you have a Cathode follower biased at 115-100=15V. Seems the grounded cathode stage should be biased a little hotter by decreasing the cathode resistor, perhaps.

 

[Cassiel]

is that you have a Cathode follower biased at 115-100=15V.

That I know, the reason why I don't. There's a leak somewhere.

Edit - PT is leaking out 3 volts AC. It still doesn't explain my biasing ploblem.

 

[SY]

Cathode-heater leakage? How high do you have the heaters biased?

 

[multisync]

it's been a while since I played with tubes, but remember your 2 stages are dc coupled. The plate voltage on the 1st stage is the grid voltage on the second. The plate voltage etc on the first stage will effect the second stage operating point. I don't think the tubes operating point will lower your noise level a whole lot. Is your noise hum or hiss? If hum you can fix it, if hiss maybe not. You may wish to change your design a bit and put the high frequency cut between the first and second stage. That way it will help to cut the high frequency noise. my 2 cents, feel free to ignore it

 

[Cassiel]

I thought about that, SY. I don't how to test it. My heaters are 55 volts above ground.

Regarding the hum - first I want to fix the biasing. There's something wrong at the PT primary connections - I'm checking the ceramic caps right know. That could explain the hum issue.

 

[ruffrecords]

The problem is probably with your meter loading the points you are measuring giving completely false readings. Try measuring the dc across the 300R in the CF anode circuit. If the stage is working and pulling about 1mA idle current you should see about 300mV across that resistor.

Cheers

Ian

 

[Merlinb]

Voltage at the grid is 100V and at the cathode 115V. Both channels show this strange reading

That sounds pretty normal; a typical volt meter will pull the voltages down and give you a wacky sense of bias. You said you got -2.5V grid-cathode (correct?) which isn't unusual.

EDIT: Ian beat me to it...

The only clue I have is that there is 2.5 volts between grid and cathode with the preamp switched off

Did you mean to type that?

 

[Cassiel]

First off, thanks to those who have tried to help me.

ruffrecords - I just did that. About 1 mA (.98). So the meter reads the cathode voltage fine (120K/125V) but pulls the voltage down when I measure it at the grid?

Merlinb - "Did you mean to type that?" Yes, I did. See, I have a leaky switch (just found out) and when in off position the power trafo primary still sees around 3 volts so with the preamp switched off I still have a B+ of around 3.7V. No filament voltage though. On these conditions I read -2.5 volts between grid and cathode.

 

[DF96]

Is the CF oscillating? What type of meter are you using? A DMM should not drag the previous anode down by much.

 

[Cassiel]

I'm with you on this DF96. Seems strange to me. I'm going to install a 100 ohms resistor just in case there's oscillation. No oscilloscope here.

 

[simonov]

You should not be using 12AX7 in cathode follower configuration. In this particular situation, looking at the load line of the cathode follower it seams the biasing is very bad. The bias point is near the 0V grid curve. The grid of the follower is drawing current from the previous stage resulting in it's decreased anode voltage. On the other hand decreased anode voltage of the 1st stage will increase the cathode voltage of the follower. This is pushing the bias even higher. Eventually an equilibrium will be achieved but the follower's grid will draw current permanently. What you need is a follower built around a lower Rp tube. A fine choice is the Russian 6n6p. An even better choice would be to load the cathode follower with CCS.

 

[DF96]

Grid at 100V and cathode at 115V, if correct DC readings, mean that the 12AX7 is biased well beyond cutoff. Any grid current will be very small! However, these are exactly the sort of figures you might expect from a Class C RF oscillator. CF with a capacitive load is known to be unstable.

It is true that 12AX7 generally makes a poor CF, but if used carefully and when a spare half-12AX7 is available it should do no harm.

 

[Cassiel]

I did a little mod and now the grid is at 105 and the cathode at 120. Same 15 volts. Both channels! Same current. This is odd. I had enough for today, I will continue tomorrow.

 

[simonov]

DF96 I rechecked my drawing and I agree. Bias goes near the -1.5V grid curve which is good enough.

 

[Cassiel]

There's -1 volt grid-cathode. The DMM gives incorrect readings. Biasing is OK.
Thanks.

 

[dagerman]

From someone mostly unencumbered by theoretical knowledge, but with some of my own headscratchers still fresh in my memory....

Are you using known good tubes? I've spent the better part of a week questioning one of my amp builds before finding out that my supposedly good (used) tube was in fact a bad tube.
I also recently had one of my (new, but cheap) DMMs give weird readings. Turned out that the supposedly new battery was low on juice. But as long as your amp doesn't sound the way it should, that's not likely going to fix your problem... 8)

 

[razorrick1293]

Tubes are quite a big factor to noise, if the tube isnt bad, any chance of putting a shield on it? i've had times where thats fixed around 6dB of noise at least. is it high pitched noise? or any chance of acoustic feedback? some 12AX7s with larger plates ive found can be quite a pain with that. im sure they can even interact with output transformers sometimes...

 

[Cassiel]

is it high pitched noise? or any chance of acoustic feedback?

I have all kinds of noise. High pitched, acoustic feedback, microphonics, hum, and terrible hiss. In fact this is my second layout and I'm about to give up. I have built high gain guitar amps with very little noise. I have a star ground, DC elevated heaters and a 10 ohms resistor between audio ground and chassis. Tube shields too. B+ is very well filtered, ridiculous filtered I must say for only 3mA of audio current. The chassis 'feels' charged up. I modulate the noise as my finger travels around the chassis. No shocks but it is a strange feeling, a turmoil of electrons around the chassis surface. My finger feels it - it is a strange feeling. Can't describe it well enough. Tubes are 12ax7 Russian 'mullards'. I'm going to try different tubes and then I'm going to smash this nice big metal chassis down with a bigger hammer.

 

[HollowState]

The chassis 'feels' charged up. I modulate the noise as my finger travels around the chassis. No shocks but it is a strange feeling, a turmoil of electrons around the chassis surface. My finger feels it - it is a strange feeling. Can't describe it well enough.

You have described it very well as a matter of fact because I am familiar with that feeling. Your chassis is not at earth ground. What you feel is capacitive coupled leakage from the transformer. I was feeling this same thing on the florescent light fixtures in my basement because they were floating. Grounding the metal housings fixed it. I don't know what the electrical codes are in Spain, but your metal chassis should be grounded with a third wire for safety sake alone. Whether this will fix your other problems is unknown.