Holy crap, it works!

[lmolter]

Sorry about the confusion. Clear explanations aren't my forte. So... All the pot grounds are bussed and ground to the brass plate. There is also a ground wire soldered to the brass plate that goes to a tranny bolt. On the eyelet board, all the grounds are bussed and a single ground wire comes off and is attached to the same bolt as the brass plate ground on the tranny bolt. The filter cap grounds for the preamp and pi stage gets tied there, too.

Regarding the shielding... I don't have any more 270K carbon comps lying around, so I would have to go to Radio Shack and get metal film resistors. At this point I would be diluting the gene pool, so to speak, because I wanted to have no metal film resistors in the signal paths. I think I'll try desoldering the two on the eyelet board. Oh boy...

Tonight there is too much on the tele, so it's a weekend project. I'm in no hurry because I don't have a cab or speakers yet. But, I have no car for Saturday, so there's nothing else to do but work on the amp.

Thanks for the advice and I'll post after I make the corrections.

[lmolter]

I found 2 270K 1/4W carbon resistors and installed them on the pots. I was able to remove the old resistors and the wire successfully. I installed a short length of shielded wire but the hum is still there. The volume controls affect the loudness of the hum, so it has to start in the previous stage, I'm guessing.

I'll poke around the first stage tomorrow.

L

[loudthud]

Hard to follow how you have the ground connected but you might find this usefull:

http://www.el34world.com/charts/grounds.htm

The main thing that makes this grounding scheme work is the break in the grounds of the filter caps, marked 'F' on the diagram. The ground for the power tubes has quite a bit of noise on it and you don't want to couple that noise into the preamp ground.

When looking at the signals on the two output tube plates, the signal should look pretty much the same. The transformer couples the signal to all windings. Even if you removed one output tube, the plate will still have a signal on it upsidedown compared to the other tube. If you have 1 ohm resistors in the cathodes of the output tubes, there you will see clipped waveforms.

[lmolter]

Again, sorry for the confusion. I used that diagram originally during my final hookup, but deviated slightly. This morning, I made mods so that my grounding scheme followed the diagram as close as possible. However, there's still hum. If I pull the first preamp, the amp is quiet up to max volume on the pots. So, in desperation, I rewired the input jacks and moved the heater wires away from the signal wires (as best I could). Same. Hum increases with volume settings. Both bright and normal, with shorted inputs (through a 68K). Time to cut the grass.

L

[clyde]

Are the pots isolated from the chassis? If not, and they have a common buss wire soldered to them, I would think that might create a ground loop.

[lmolter]

Well... the pots are mounted to a brass plate, and the whole plate thing is mounted to the chassis via the controls' hex nuts. There is no buss wire across all the backs. However, all the pot terminals that are grounded are bussed together and tied to the common ground on the eyelet board. For experimentation's sake, I taped a buss wire to all the pot bodies and tied it to ground. No difference. I thought I would try it with tape before I sanded each pot, moved wires around, soldered the buss wire, and burned the nearby wires with the soldering iron <g>.

For fun again, I replaced the 12AY7 with an AX7 and the hum was reduced. I would have thought it would have increased due to the higher gain of the AX7. But... this is my first build and I don't know alot yet. I want the AY7 in V1 because I want this amp to be a fairly accurate replica.

I am getting frustrated, though. What I'm fearing is that I'm going to have to redo all the grounds, but there may not be enough lead length on some of my ground wires to move some ground points around.

L

[loudthud]

I re-read this thread and did'nt see any mention of a center tap for the heater winding or the so called virtual center tap using a pair of 100 ohm resistors. The heater supply can inject hum into the preamp if there is no means to ground it. You should measure about 3.2VAC on either wire to ground.

[lmolter]

Oops. My bad. CT it is. In this regard, is it bad that the HV CT and heaters CT get grounded to the same point as the eyelet board ground and preamp filter ground? The power tubes and their filter caps get grounded elsewhere.

L

[loudthud]

The heater winding CT can go just about anywhere, just so long as it's grounded. I think your problem might be the preamp ground being the same place as the HV CT. An important clue is what happends when you put the amp in standby mode. Does the hum instantly go away when you hear the switch click or does it slowly fade out? If it goes away instantly, you are picking up the switching noise from the rectifier, usually at the HV CT ground point. If it fades away for a second or two, you are picking up the raw 60HZ from the heaters, transformer coupling, or some other way. If you can see the hum on the output of the amp or at the volume control with your scope, is it a 60Hz distorted sine wave (sounds like a low organ note) or 120Hz little spikes or sawtooth (sounds like a buzz)?

[lmolter]

I moved the ct grounds to another bolt. Same. The scope on the B+ feeding the two plates is clean. The volume control and the coupling caps are a different matter. There's a bumpy waveform that appears to have a frequency of 60Hz. It does go down very slowly with the standby switch is opened. Is that odd?

Starting at the plates of V1, the noise is there on both the bright and normal channels. The grids are quiet (nothing measurable even with the scope at .5mV/div). That would be expected since the jacks are shorting. Just thought they may be miswired. The plate noise is about 10mV. From there, the noise is on both sides of the .02 coupling caps (bright and normal channels) and on to the volume pots where it feeds the CF and beyond.

It seems to originate around the preamp section, but that's all I can figure out so far. If it's not on the grids, and the B+ is clean, where is it being induced? I'm stumped. No more for tonight.

L

[loudthud]

I assume you are using AC coupling on the scope when you look at the B+ for noise. Some scopes will fool you if you use DC coupling by allowing you to position the trace on screen but the scope's frontend is overloaded and you just see a flat line. Since you can see it on V1's plates you are probably doing this correctly.

If you hold the scope probe near the heater wires you should see a 60Hz wave to compare to the noise on V1 plates. Verify the noise on the plates is 60Hz not 120Hz.

There could be noise on V1's inputs, it might be too small for you to see. Remember V1 will amplify it by up to about 50 times between grid and plate.

Does the brass plate extend to the input jacks? I think you said you had a wire from the brass plate to one of the transformer bolts. Have you tried unsoldering the wire at one end?

Is there a wire from the ground side of V1's cathode resistor to the input jacks? I think you said you had the grounds on the eyelet board daisey chained. Try breaking the chain between V2's ground and the PI ground and grounding the preamp at the input jacks.

[MWJB]

"is it bad that the HV CT and heaters CT get grounded to the same point as the eyelet board ground and preamp filter ground?" Yes, it is bad. Preamp filter & cathodes ground to the input jack ground. Daisy chaining grounds on the circuit board is a bad idea.

[tubeswell]

Hi Larry

FWIW My understanding of grounding (helped much by MWJB and Enzo amongst others) is that where power supply and higher voltage and/or higher current components are grounded should be kept separate from where low voltage and/or low current components are grounded. Hence I use a split-ground system whereby:

1) the preamp filter cap ground, and all the other grounds in the pre-amp (vol and tone grounds and cathode grounds and grid load resistor grounds etc, but not anything asociated with the output tubes) are connected by individual wires to a common point (like a tag terminal), which is as close as possible to one of the input jack grounds, to which it is connected by a quite short further single common wire. I only hook up 1 ground on one of each pair of input jacks, and I have the pre-amp short ground wire going to only one of those.

2) the reservoir (input filter) cap ground and screen-node cap ground, together with the output tube cathode ground(s) and output tube grid load resistor ground(s) and the heater CT/100R-resistor-ground, HT CT, are all connected via separate wires to one of the PT bolts at the other end of the chassis (from 1 above). Until recently I grounded the AC Mains ground there as well, but since Enzo said I should ground the AC mains ground seaparetly at it own bolt, I have changed my practice.

So far all my amps have been virtually hum-free.

[lmolter]

OMG, I just posted a very length reply about what I discovered after taking the whole amp apart and inspecting my grounding scheme. I used the word "beejesus" (I won't tell you where), but I wasn't sure of the spelling, so... I went to dictionary.com, looked it up, and closed the browser. Lost everything.

So here's the synopsis of my planned changes: 8uf preamp filter cap on eyelet board will connect to board grounds and a single ground wire will connect the board to the buss wire across the control backs, and the input jack grounds. (Should I also ground the brass plate?) All these will ground somewhere near or on the input jacks. And this 8uf cap filters both the preamp and the cathode follower stage.

The 4-20uf reservoir caps in the metal doghouse can all ground at the ct and output tube cathode grounding point (at the opposite end of the chassis from the input jacks). These caps filter the output and PI stages.

Since I had to order some replacement parts, I won't have this back up and running until the end of the month. I"ll post back after I've buttoned it back up again. Hopefully no other errors will be introduced in the process.

Larry

[MWJB]

"8uf preamp filter cap on eyelet board will connect to board grounds and a single ground wire will connect the board to the buss wire across the control backs, and the input jack grounds." No! Run an individual wire from each ground point on the circuit board to your input jack ground, DO NOT daisy chain them on the board.

Brass plate is grounded by virtue of being bolted to the chassis via pots & jacks.

Only ground the main B+ and screen supply filter caps to the CT grounds, ground the PI filter cap to the buss wire.

Rewire main filters like Super Reverb buit use 2x100uf 350v (or better) caps in series, each bypassed with a 220K 2W resistor.