Two issues with 6G3 build--hum and low preamp plate voltage

[loudthud]

I'm looking at the historic Fender schematic and layout found here: http://www.el34world.com/charts/Sche...DELUXE_6G3.pdf

Did you use the grouning scheme of the Fender layout? An important detail not shown in the layout is grounding in the "dog house", the box where all the filter capacitors are located. Did you tie all the grounds within the dog house together there? If so, that could be the problem. The 8uF capacitor should ground only at the black wire that comes through the hole near the Normal channel Volume control. The rest of the grounds should tie to the black wire that is soldered to the chassis near the power transformer.

Edit: I didn't read through the entire post #15 above in reference to the JJ can cap. Those can caps cause grounding problems because the grounds are all tied together. One solution would be to add another cap and connect it where the 8uF is on the schematic (disconnect the can cap section) and ground it in the preamp. That should work. If you don't want to do that, some changes to the way things are groundid may be necessary. Please describe all ground connections.

[eschertron]

Unfortunately, I don't seem to be getting anywhere with finding the source of the hum. In fact, I gave up in disgust for a while and finished up another amp that came out great. Here is what I think I know so far on the problem child...

- hum is present even with no cords plugged in
- by ear and verified with a spectrum analyzer it is predominantly 120 Hz
- with the phase inverter pulled there is just an extreemely faint noise from the power supply that is well below what I cosider acceptable
- pulling the first (initial gain stage for channels 1 & 2) or second (tone stack recovery and oscillator) preamp tubes has no effect on the hum
- I have swapped about five tubes through the PI with no improvement in hum
- hum is worst with gain on both channels at zero or quite low and then will fade away once either channel's gain is raised to roughly 3/4 of full or more (at which point it gets hissy, but that's not a major concern right now
- all voltages seem to measure close to what the layout indicates they should
- the layout is pretty similar to pictures of original brown Deluxes that I've seen except that the chassis is installed tweed-style so the tubes are rotated 90 degrees around the chassis if that makes sense
- not that I could not have missed something, but all wiring appears to be corrrect as do component values
- putting a 20 uF capacitor in parallel across each of the filter caps (individually) had little or no effect on the hum
- the filter caps are a JJ 40-20-20-20 can which I expect should be better than the original design, at least in theory
- except for the hum, the amp sounds quite good

If anyone has further ideas to check, I would appreciate the suggestions.

Pete

If the area around the PI is suspect, maybe:
Put your test cap (22u) across the top of the PI (where the 100k and 82k plate resistors come together) to ground. If this quiets the amp, the filter node to the PI may be the problem. And by problem, I mean there may be a cold solder joint or something else that's preventing that filter node from working the way it should.

What I'm getting from your description is that something somewhere is allowing PS ripple into the signal path prior to the PI. Do you have a scope to put on it?

edit: have you tried disconnecting the NFB from the PI?

[pjhalliday]

Loudthud, this amp has a cap can rather than the dog house in the original so I don't think that the grounding there is an issue.

Eschertron, I tried the cap you suggested and it did not make a difference. I do not have a scope. Yes, I originally had the NFB disconnected as I was going to leave it that way, but in trying to quiet the hum hooked it up with little or no benefit.

[g1]

Loudthud, this amp has a cap can rather than the dog house in the original so I don't think that the grounding there is an issue.

See his edit to post #16, the can could still be a problem.

[Dave H]

I’m not used to Fender layout diagrams and I don’t have the amplifier. How are you supposed to know where all those bloody ground ‘triangles’ connect? Some look like they connect to the back of the pots, yikes! Do the others connect to the chassis at various places too? Yikes again.

But I do know that the usual source of 120Hz buzz is the connection between the power transformer centre tap and the first filter cap negative terminal, by ‘first filter cap’ I mean the cap with its positive terminal connected directly to the rectifier (point ‘X’ on the layout). Disregard what you think the layout is telling you if it is not as follows -

Connect the wire(s) from the transformer CT (red-yellow, green-yellow? on the layout) directly to the first filter cap negative terminal without going to chassis ground or any other ground on the way. Do not pass ‘Go’ do not collect £200. Go directly to the cap -ve terminal nowhere else or you go directly to gaol. Make your other ground connections to the first filter cap -ve terminal not the PT CT wire. There are no exceptions to this rule. It’s the law

[pjhalliday]

All--thanks for the suggestions so far. I have not yet tried separating the last filter cap from the common ground only because of lack of space and I have built essentially this same amp before using the same style can with no trouble. I'm not claiming enough experience to say that's not it or anything, just putting it off until exhausting the easier possibilities. To answer the ground layout question, I am using one of the PT screws as the common attachment. Going directly to this are earth from the power cord, the virtual CT from the heaters, the ground terminal from the cap can, the ground from pins 1&8 of the power tubes, and a connection to a ground bus. I did move the red/yellow PT CT to the cap can ground terminal per Dave's suggestion rather than taking that straight to th ecommon ground point. The ground bus has all grounds from the circuit board terminals (5 not 4 as in Fender layout due to separate bypass paths for V1a and V1b), ground connections from the vol/tone and tremolo pots, and the ground from the bias supply which is also board mounted. This is also essentially the same as I used in a previous 6G3 build which did not have issues but is easy to modify if there is something I've done that is poor practice. The jacks and pots are not isolated from the chassis for the same reason.

[silverfox]

First let me stress how new I am to the field of tube amp builds and troubleshooting. Only about two years so far with plenty of other electronic and computer troubleshooting experience in the past. However no audio. By offering suggestions I obtain knowledge and perhaps I can contribute. When I offer suggestions, being in the learning phase myself, feel free to correct and if desiring to do so, "school them dice".

"How are you supposed to know where all those bloody ground ‘triangles’ connect?"

There will usually be some sort of grouping symbol associated with grounds to keep pre amp grounds grouped together and grounds further along the circuit grouped. While all paths lead to ground nirvana, a path created between ground groups may cause a current to flow between ground points.

Perhaps this would help. Merlin (Valve Wizard) has a good free chapter on grounding here: http://www.valvewizard.co.uk/Grounding.pdf

Here is what I suggest for consideration: Disconnect portions of the circuit, by lifting a supply resistor leg, and see if the hum disappears. Perhaps start with the first gain stage?

Silverfox.

[deci belle]

Chuck said:

I don't even know what might cause high cathode volts and low plate volts with the right supply voltage and resistor values.

I found out the hard way today… non-existent preamp grounds!