1940's GE YGA-4 Audio Oscillator, My Frankenstein Preamp

Thanks again for all your help...  Here's a rough mock up of where I'm at.  Obviously my mistake was in picking off the B+ from after the 15K on the 3rd channel instead of using the "open" and available wire just dangling in there. 

The effect was a grungy mellow overdrive on Channel 2.  Chan 1 and 3 don't have that going on.  My initial guess was that since I was supplying 2 1/2 tubes off of one branch of my power supply I was really starving the 2nd channel for current, and/or the voltage was too low...  Happy accident if I can control it and not burn something up... 

The thing I really need to do is take some measurements so I know where I'm at before I go changing anything...  My fear is I won't get back to this once I wire it up correctly.

millzners said:

Thanks again for all your help...  Here's a rough mock up of where I'm at.  Obviously my mistake was in picking off the B+ from after the 15K on the 3rd channel instead of using the "open" and available wire just dangling in there. 

The effect was a grungy mellow overdrive on Channel 2.  Chan 1 and 3 don't have that going on.  My initial guess was that since I was supplying 2 1/2 tubes off of one branch of my power supply I was really starving the 2nd channel for current, and/or the voltage was too low...  Happy accident if I can control it and not burn something up... 

The thing I really need to do is take some measurements so I know where I'm at before I go changing anything...  My fear is I won't get back to this once I wire it up correctly.

Click to expand...

I imagine that 15K plate resistor for ch3 may have gotten a bit warm when wired that way.  Part of the mellow distortion you describe is likely from lowering the B+, so that aspect can be built in and controlled, probably best to pick one channel to do that on.

If taking the ch2 feed from the "open" feed I would remove the 47K and just wire directly to the 4.7K (not the PS 4.7K) decoupling resistor for ch2.  That 47K is a 1/2W and even with modest 4 or 5ma draw from ch2 it could dissipate over 1W and drop over 200V of supply.  I'm also inclined to say lower the 4.7K PS resistor to something like 2.2K or possibly just use a 5K rheostat there for B+ control on ch 3.

Best to get full set of voltage measurements first - everything is guessing until then.   

I was wrong about where I was picking off that B+ for channel 2, it was after the 36k resistor not after the 15k.  I'll update the schematic shortly...

Here's the readings:
Channel 2 (the one with the distortion I like):
Stage 1
Vp = 20.5V
Vc = 1.0V
Stage 2
Vp = 42V
Vc = 0.645V

Channel 3
Stage 1
Vp = 49V
Vc = 1.68V
Stage 2
Vp = 201V
Vc = 7.1V

Voltage after the 36k in question = 82.5V (voltage supplying the NYD circuit)
Vcc supplying channels 1 and 2 after the 4.7k PS resistor = 272V

No wonder channel 2 got all distorted, it was only being supplied by 82.5V to start...  Good to know for when I try and recreate this, which I'll do only for channel 2.  It'll be a nice touch, I'll definitely use a pot as a rheostat in there, a 2W 35k pot is available and should be ok...

I fixed the B+ to channel 2 and it's back to normal.  I also rerouted all the grounds from channel 3 to a star ground, but there was still -40dB of 120Hz buzz. 

I moved on to the cathode bypass caps, wired up two switches and tried them out...  It was hard to tell if there was a difference in frequency response, because the first thing that happened was upon switching to either 10uF or 1000uF on the first stage knocks the buzz completely out -- I get more gain and no noise with a bypass cap on the first stage...  The 2nd stage switches also offer a little more gain but have no effect on the buzz.

I'm not sure why the bypass cap on the first stage makes such a dramatic improvement to the buzz, but if that buzz is sitting on the plate then I believe the cap would shunt most of it to ground...  I will try some things (more filter caps, moving grounds, etc.) but I may permanently wire the 10uF in place and then add a 100 and 1000 to the switch, thus meaning I always have a bypass cap in there even with the switch off.

The only real problem is that Channel 3 has kind of a slight hissing noise on it when I play through it, it's not there unless you send a signal thru it and the signal comes out sounding great but with kind of a 4k shhhh sound on it...  Makes me think one or both of the tubes are bad.

Does turning the interstage volume control down on ch3 make the buzz go away or at least noticeably attenuate it?  Also try grounding the grid of V1.  If that kills it may be pointing to the input transformer.  How close to PT is it?  Try rotating it and listen for change in buzz level.

If there is no buzz on ch2 it would appear that it may not be a main PS filtering issue.  Have you wired up ch1 yet?  I would have guessed it to be more likely for filtering hum issues as shown. 

Hissing problem may be solder joint issue.  You can take a chopstick and prod wires and joints to expose them while the circuit is on.

Great advice on troubleshooting.  The Channel 3 buzz is almost entirely 1st stage, which also coincides with the cathode bypass switch on the same stage.  I think I have a good idea of where that's coming from.  The little bit of buzz on Channel 1 is almost entirely 2nd stage, and looking over the schematic it's obvious that's the least filtered power.  Furthermore after turning the power to unit off, there's still a buzz there.  So there's likely some coupling going on between the AC mains and the output transformer, despite Channel 1 being the furthest from the power supply... 

And upon another quick run through, I have to say the NYD channel, channel 2, is the best so far.  Not only is it the quietest and has the most gain, it sounds the best.  I'm tempted to re-bias Channel 3 to the NYD schematic now, since I'll be all up in there anyway...  They'll still be unique for a lot of other reasons not the least of which is the tubes.

Furthermore after turning the power to unit off, there's still a buzz there.  So there's likely some coupling going on between the AC mains and the output transformer, despite Channel 1 being the furthest from the power supply... 

Click to expand...

Does buzz fade & disappear as B+ dies down or constant?    I've had 60Hz hum issues strictly from AC filament bleed/balance problem - so it would start immediately at power up - well before tubes started conducting current - and stopped immediately when AC was shut off.

Yeah oddly enough, the buzz on Channel 1 went away when I plugged the unit in to a different AC outlet.  I've been plugging it in to this surge protector power strip b/c it's right there in the open, and when I plugged the unit into a UPS "surge" out, no buzz...  Oh well, I'm happy...

Before I figured that out I did some significant changes to Channel 1, thinking the issue had to be power filtering, or something to do with how the B+ was feeding through the transformer.  So I changed it around (I'll update the schematic tomorrow), and there was no effect on the buzz.  Turn the power off, same buzz.  Unplug the AC plug, buzz goes away, plug it back in (unit powered off) buzz comes back...  Wierd, but not crazy.

So I'm down to one last issue before moving on to getting the input xfo's installed, and that being the buzz on channel 3 without the 1st stage bypass cap.  If I can't get to the bottom of it by bedtime tomorrow, I'll hard wire the cap and move on. 

before moving on to getting the input xfo's installed

Click to expand...

Just curious,  when testing did you have the input iron connected with alligator clips? or not connected at all?

If not connected, were the 1M grid leak resistors and jack tip leads soldered in place?

Reason I ask is that any bit of stray wire, long test jumper wires or clipped resistor lead hanging off the socket grid pin can easily induce buzz in a circuit. 

No input transformers are in the circuit yet, they're still in bubble wrap.  They'll get installed this weekend.

All my channels have their TS connectors dangling out of the front panel, from grid to tip is about 10 to 12 inches, not an ideal way to be testing things but I haven't put any thought into laying out the front panel yet.  I assume they are all acting like antennas right now, so my goal is just to get things as good as I can before I start drilling holes. 

I think it's going to be hard to rule out this being the root of the problem with Channel 3, considering the 1st stage is the source of the buzz, so it may be time to start laying out the front panel.  The challenge is everything from fuse to outputs have to be located on the front panel, because of the way the chasis is put together.  It's going to require some thought...

The bypass caps do have an effect on frequency response, which is disappointing in a way b/c I really wanted to just cut and run on the idea...  But I've decided to hunker down and battle the buzz.

It's entirely in the first stage, starts slowly at power up, and gets going around 10 seconds after cold start.  It stops immediately upon hitting the off switch.  Seems to me, with my very limited tube experience, to be coming in through the grid of the first stage...  I'll throw everything I've got at it, let's see what happens.

Just to stay on same page - is this predominantly 60Hz or 120Hz type hum/buzz?  and not electrostatic buzz as would be heard if an input cable shield weren't grounded?

Have you differentiated 1st stage origin?  Grounding of interstage vol pot vs grounding of V1 grid?


***I didn't catch this before and I'm not sure if it was discussed but I see on the original Osc schematic that they've tied one side of the filaments to ground.  This can create hum problems.  Remove this and wire in a hum balance : 100-250r 2W pot across the filaments, wiper to ground.  If you don't have one handy, wire in 2 100r resistors from each leg to ground and see if that helps.

I tried wiring a 110R 1W resistor to ground from each side of the heater and the tubes wouldn't turn on -- no audio and no glow.  I got worried I blew everything up so I took them out and rewired things back to the way they were, and everything works fine...

So then I did some reading:

http://www.freewebs.com/valvewizard/heater.html

Down the page there is a blurb on exactly what you told me.  I don't see why it didn't work, and I'm fairly sure I wired it correctly.  I'll try it again and report back. 

The buzz is definitely 60/120 Hz, if you turn the interstage volume pot down it goes completely away, and if you ground the grid it's still there... 

The buzz is definitely 60/120 Hz, if you turn the interstage volume pot down it goes completely away, and if you ground the grid it's still there...

Click to expand...

Ok. That seems to narrow it down.  Tried swapping tubes there?  And this is same channel(3) that hum would go away if cathode was bypassed? - still doing that?

Also note whether hum is predominantly 60 or 120Hz.  60Hz tends to point towards heater related issues.  I've  had both heater and PT field induced hum going on at same time and the tone was complex.  PS filtering 120Hz hum can have higher harmonics and appear to have higher pitch.

Not sure why you lost heater voltage.  You can try it a bit more methodically - start by just removing the ground connection.  At this point it will likely hum worse but you should still have 6.xxVAC and working tubes.  Then add resistors to ground.  Sounds like one side of the heater line got disconnected somehow.

Getting either the resistors or hum balance pot wired in will help eliminate heater issues as cause or not.  The hum balance can be more telling in that you can hear the hum level change as pot is rotated.  I've had situations where no pot was really needed and you could turn full in either direction and hardly be able to induce any hum at all.  Had another situation where you had to constantly adjust pot at each power up.

The buzz is gone.  Your advice has been really helpful, thank you.

So I figured out the reason my heater circuit didn't work with the artificial center tap was b/c when I cut the ground, I removed the return for the bottom leg of the heater circuit completely.  The 6.3V transformer output was grounded somewhere else in the chasis, and it was not directly connected to the parallel wiring for the heaters, except through the chasis. 

Once properly wired, that trick shaved off around 5dB hum from the other two channels (which is awesome) yet it didn't put a dent in the big 120Hz buzz on channel 3. 

To fix the buzz I tried two things: rotate the interstage pot 180 degrees so the wires weren't so close to the power supply caps and I also swapped the 6J5 tubes' position (same two tubes in opposite spots).

One of those two moves fixed the buzz entirely.  Channel 3 is now as quiet as the other channels, regardless of the cathode bypass caps...  I already investigated that pot last week and I moved it much further from those filter caps than it is now but at the time it had no effect.  So I'm not entirely sure I know why the buzz is gone, but I don't really care at this point.

The next step is to wire up the input transformers, find out where they'll go, and then drill some holes in the panel for all the knobs and jacks.  I still need to order that 2W 35k pot for the distortion on channel 2...

Glad to hear that Millzners!    Buzz & Hum can certainly be a beyatch.  Enjoy!

Ok, so I'm back to work on this after several months.  I ran into a chasis issue, there just wasn't enough room for 3 mic input xfo's in the old GE chasis, plus the thing was built out of 1/8" steel so drilling holes was a pain...  So I used the thing as a DI box for the last few months until I had a break through.  Last week I acquired and old 1960's RF frequency standard box which is conveniently housed in an aluminum 3U rack.  So I tore out all the inards, and have begun "racking" this thing. 

So far, holes have been drilled, and the steel skeleton that holds the tubes and parts has been moved from it's old chasis to the new one and bolted down.  I connected the DI ins and outs and it still passes audio, but there's a monstrous high frequency (HF) oscillation, which is likely a grounding issue since not everything has been wired up properly to the new chasis... 

The HF oscillation is present on all 3 channels and is terrible on the middle channel, subtle on the 3rd, and somewhere in between in the 1st... 

It's done.  It sounds fantastic.  I never knew a stock SM57 could sound this good...

millzners said:

It's done.  It sounds fantastic.  I never knew a stock SM57 could sound this good...

Click to expand...

That is awesome! and so true... ;D

congrats!