SoulFetish 5/19/2017 6:40 AM
A lot of noise comming through in my phase inverter.
After powering up my build for the first time and checking my power supply and biasing for stability etc, I hooked it up to a speaker load and turned it on with the input jack shorted and volumes down. I noticed quite a bit of noise output from the amplifier. Way too much noise for me and much of it is 120Hz supply ripple coupling through. Pulling the V1 and V2 tubes had no audible effect on the noise at all. I was worried that it was my driver tube, which has it's own isolated powers supply and dedicated winding. But, after pulling the PI, the driver and output tubes are dead quiet.
Without having put a scope on the rail, the simulations indicate that it should be well filtered at the PI filter capacitor. There is a switch to activate negative feedback, but that is a howling mess. But with this much noise before even adding global feedback, and it's potential for instability, I'm not surprised at all that it's oscillating.
Any thoughts as to what maybe causing the noise to couple through the PI stage? I'll post the PI schematic in a little bit.
 
SoulFetish 5/19/2017 6:53 AM
Here is a schematic of the PI:

[IMG]http://i44.photobucket.com/albums/f15/Soul_Fetish/PI%20Stage_zpsnkohp71n.jpg~original[/IMG]
 
eschertron 5/19/2017 9:19 AM
You didn't mention if you swapped a different tube into the PI.
 
SoulFetish 5/19/2017 11:28 AM
Quote Originally Posted by eschertron View Post
You didn't mention if you swapped a different tube into the PI.
I did swap out the tube. It's a new production 12AY7, which I swapped out for a new old stock 5751, which I know to be a low noise tube. It's not the tubes. Although I'm not ruling out the tube socket, or bad solder joint. In fact I'm going to pull the power supply dropping resistors on each side of that filter cap an check capacitance and ESR.
 
g1 5/19/2017 3:43 PM
Quote Originally Posted by SoulFetish View Post
There is a switch to activate negative feedback, but that is a howling mess.
If I'm reading this correctly, it implies the OT phasing is wrong. Shouldn't account for any noise when the loop is open, but it may lead you to find you have an oscillation issue that NFB cures. At least that will be a start.
Try swapping the OT primary wires.
 
SoulFetish 5/19/2017 4:38 PM
Quote Originally Posted by g1 View Post
If I'm reading this correctly, it implies the OT phasing is wrong. Shouldn't account for any noise when the loop is open, but it may lead you to find you have an oscillation issue that NFB cures. At least that will be a start.
Try swapping the OT primary wires.
When you said, " if I'm reading this correctly", I thought you were talking about my schematic. I'm staring at ithe schematic trying to find out where the hell i drew it wrong to give that away! Thinking G1 was on the other end taking one look, leaning back in his chair like he's Cliff Clayvin, saying "oh yeah, I see it now. That's your problem right there. You got the phasing wired up backwards on your output transformer."

But, I eventually caught on and realized what you were talking about
I did catch a mistake, however. The 120K feedback resistor is wrong. Should be 150K. Too much feedback. But I doubt it was primary cause.
 
Chuck H 5/20/2017 9:12 AM
I want to see the whole amp, including the power supply, in a schematic. Just because the noise seems to coming from the PI doesn't mean there's a problem with the PI individually, schematically. It may be something connected to the PI. Having the PI schematic floating out of it's context doesn't help me to figure why removing the tube stops 120Hz noise because it can't do that on it's own.
 
g1 5/20/2017 11:49 AM
Well, that's certainly not the first time I've been called Cliff Clavin. And I've won virtually millions on Jeopardy.
I was referring to the statement I quoted.
Now I'm a bit confused, does it howl when you engage the NFB? It should reduce the volume.
 
SoulFetish 5/24/2017 2:09 PM
Quote Originally Posted by g1 View Post
Well, that's certainly not the first time I've been called Cliff Clavin. And I've won virtually millions on Jeopardy.
I was referring to the statement I quoted.
Now I'm a bit confused, does it howl when you engage the NFB? It should reduce the volume.
I use a voltage divider when the feedback is disengaged to attenuate the open loop gain and balance the two levels.

You were right about the OT primaries. I swapped them around and no oscillations like there were previously.
I'm still getting noise in the PI. But, after fixing the primary connections and re-routing some wires I thought might be picking up noise, I plugged in my guitar to test it out anyway and I noticed my tone stack wasn't working properly.
When wiring up my board I socket-ed some of my turrets so I could breadboard and experiment with some components quickly. Like so:

[IMG]http://i44.photobucket.com/albums/f15/Soul_Fetish/Turret%20sockets_zpsfgwgkhv9.jpg~original[/IMG]


At the time, I though it was pretty clever.... This is what I think of that idea now:


[IMG]https://i.giphy.com/13NvIHnUU0PU52.gif[/IMG]

It worked out okay for the dropping resistors in the B+. But I'm not getting the solid electrical/physical connections I was hoping for everywhere else. So, I'm going to go and dust off that fail, make some big boy solder joints, and then see where I'm at with my noise.
 
SoulFetish 6/1/2017 12:49 AM
I zeroed in on the source of the noise! It wasn't my prototyping turrets experiment after all. It's a faulty tube socket! Early on, I wondered if the tube socket might be the issue. The pin tension seemed a bit to loose. But, honestly, it was way down on my list of suspicions. I was not looking forward to swapping out and replacing the tube socket at this point in the project, but it wasn't nearly the pain in the ass I thought it was going to be. The silicon dampening gasket stayed intact, and the ceramic socket fit perfectly in place of the Teflon-composite socket.

Next up was fixing the tone stack:
The backstory to this is that it sounds like total shit. I mean like not even close. Turns out that I swapped the mid and bass pots by mistake when mounting them way back in the build (and connected the ground to lug 3 instead of lug 1). Good ol' RV4 numerical designations ending 253 and 254 respectively. Took care of that as well.

Now, there's one last bit of troubleshooting that I need to get to the bottom of. With the feedback engaged, there is a HF oscillation that is actually modulated on a low frequency rate... maybe .5Hz - 1Hz. It's not overly loud but it's there, and without looking under a scope. I'm willing to be there might be inaudible HF oscillations going on.
I'm using neons to protect the output tube grids at start up and I've come across people who have complained about these oscillating and ended up removing them. I'm not sure if this is symptomatic of the same conditions they dealt with. But where should I be looking to track down the source of this? Anyone else encounter this type of HF oscillations?
 
Enzo 6/1/2017 1:28 AM
You have a scope you don't care to use? Or you have no scope? That is the tool to instantly see if oscillation is going on.

Neon bulbs can oscillate, look up relaxation oscillator, but generally you can see them glowing when they do. Easy enough to lift them out of circuit to find out.

And make sure ther are no cell phones within a few feet of the amp.
 
SoulFetish 6/1/2017 2:40 AM
Quote Originally Posted by Enzo View Post
You have a scope you don't care to use? Or you have no scope? That is the tool to instantly see if oscillation is going on.

Neon bulbs can oscillate, look up relaxation oscillator, but generally you can see them glowing when they do. Easy enough to lift them out of circuit to find out.

And make sure ther are no cell phones within a few feet of the amp.
Oh, I care to use. I have that boat anchor Tek 556. No probes. But its been sitting for years before I got it and I'm worried about just firing it up and possibly damaging it. I go back and forth about restoring it, but I don't want an oscilloscope project. I want an oscilloscope to help me with my projects. Know what I mean? I've got my eye on craigslist. I'm just cheap and broke and I obsess about finding the best deal on stuff.
 
SoulFetish 6/1/2017 4:18 AM
Quote Originally Posted by Enzo View Post
And make sure ther are no cell phones within a few feet of the amp.
Thats a solid point! I may have had my cell in my pocket. Not sure, but thanks for the reminder
 
catalin gramada 6/1/2017 4:49 AM
Late edit: didn.t saw you solve it . sorry