65 Deluxe Reverb build hum

[Randall]

Hi Guys,

I yanked the guts out a 65 DR re-issue to re-wire it p to p AB763 65 Deluxe Reverb. While it works and sounds fine, I am troubled by a low level hum that I didn't notice when it was wired as modern pcb.

I methodically pulled tubes from the preamp side until I got down to just the outputs and rectifier, thereby isolating the low level hum. I tried lifting several things until I came to the PI plate supply node [C]. Lifting it at the 100k/82k junction, the hum reduced by about half to an acceptable level with no tube in the socket.

I don't understand why this could be since I don't see any way that node could be pulling any current with no tube. Verified both .1 blocking caps are not leaking.

My grounding scheme is: filter supply board has B+, , and [C] nodes tied together. That 18g wire goes to main star with PT CT and line ground. [D] goes to separate chassis ground about halfway. Board cathodes and control pots are on a hefty bus going to opposite end of chassis next to normal inputs.

schematic:
http://www.turretboards.com/layouts_sch ... _ab868.gif

[JazzGuitarGimp]

Instead of halfway, I'd try moving the gnd reference for D to the input bus gnd.

[Randall]

On another note.. How does the circuit get from 415v at the trem tube supply thru the 220K ohm resistor to the first plate and become 270v? I have 400+ v on that, plus I have 0v on the cathode on both this amp and another AB763 SR in my shop. I'm missing something.

[Firestorm]

On another note.. How does the circuit get from 415v at the trem tube supply thru the 220K ohm resistor to the first plate and become 270v? I have 400+ v on that, plus I have 0v on the cathode on both this amp and another AB763 SR in my shop. I'm missing something.

400V plate pin and 0V cathode pin? Would the missing something be a tube in that socket?

[vibratoking]

I am guessing that you have tube installed. Your measurements indicate zero plate-cathode current. So the tube is either bad or there is an open for some reason. Either the plate or cathode resistor is not fully connected at the tube or source side, or is open, or the tube pin is not connected in the socket?

You measure the same thing on a SR too? Respectfully, are you making the measurement properly? It sounds awfully coincidental that two amps have similar faults.

[Randall]

I know, it's odd. Tried different tubes. Measured everything. Re-flowed everything. Re: measuring properly, neg to chassis ground, pos to test point with Fluke 77, not much to go wrong there. It's got to be something silly that I am overlooking, because not a chance a 1965 SR and my new build have the same exact fault, if there is a fault at all. Remember, the trem is working.

[Firestorm]

Oh. Trem tube. Without the footswitch connected and turned on, the raw bias voltage shuts the tube completely off. Plate goes to the supply voltage, cathode to 0. That's normal. I wasn't paying attention.

[Randall]

There it is. Schematic shows foot switch open, and makes no mention that it must be closed to activate the tube/circuit. BUT, when the foot switch is closed, the voltages swing with the oscillator. My Fluke has an average setting, and all voltages fall within spec on this setting. But, why doesn't the schematic make any note of this? I wasted a lot of time on this, and the trem works fine. My education I suppose.

[JazzGuitarGimp]

There it is. Schematic shows foot switch open, and makes no mention that it must be closed to activate the tube/circuit. BUT, when the foot switch is closed, the voltages swing with the oscillator. My Fluke has an average setting, and all voltages fall within spec on this setting. But, why doesn't the schematic make any note of this? I wasted a lot of time on this, and the trem works fine. My education I suppose.

But did you get to the bottom of the hum issue?

[statorvane]

But, why doesn't the schematic make any note of this?

IATA standard drawing practice is to show relays, switches, etc. in a position that results in a de-energized state. I think that probably follows typical industry practice for electrical schematics.

[martin manning]

On another note.. How does the circuit get from 415v at the trem tube supply thru the 220K ohm resistor to the first plate and become 270v? I have 400+ v on that, plus I have 0v on the cathode on both this amp and another AB763 SR in my shop. I'm missing something.

There it is. Schematic shows foot switch open, and makes no mention that it must be closed to activate the tube/circuit. BUT, when the foot switch is closed, the voltages swing with the oscillator. My Fluke has an average setting, and all voltages fall within spec on this setting. But, why doesn't the schematic make any note of this? I wasted a lot of time on this, and the trem works fine. My education I suppose.

The older AB763 schematic shows the switch open (vibrato "off"), but does not show the ground connection (and that was common on old Fender schematics... it was assumed that the jacks were grounded to the chassis through their bushings). The newer AB868 schematic (linked in the first post above) shows the switch closed (vibrato "on"), and shows the ground connection on the jack. I guess Fender could have added a note saying which way is "on," but that can be discerned from the circuit. Just for fun, I looked at the current DRRI schematic: it shows the switch open (vibrato "off"), and gives voltages at a couple of nodes in both "on" and "off" states. Still no mention of which way is "on," though. I was going to say that service techs were, or were assumed to be, smarter back in the day, but it looks like they are still being given some credit ;^)

[Randall]

But yet, the reverb switch is shown open, which if I'm not mistaken, is the on position?

Anyway, regarding the hum issue. I knocked it down by about 40% by stacking a second 25/50v bias supply cap. And the 0.001 PI coupling cap came out in two pieces when I lifted it. Replaced that. Rewired the PI socket and was super careful with the lead dress. Moved the preamp supply ground to the bus as was suggested.

Tried moving the bias supply ground to the star, but made no difference, so I left it at the PT transformer bolt next to the pilot lamp. Did the same thing with the two 100 ohm filament tap resistors.

I think it may be as quiet now as it was when it was a re-issue, but to be honest I'm not really sure. I can obsess about these things, especially if it's something I built. But one thing still has me vexed. The bias supply ripple that is on the PI plates is 1.4 vac and 1.5 vac respectively, and they are both clean saw waves with standby off. When switched on, one wave is clipped at the top and the other is slightly rounded at the bottom, and I just don't know why. Tried replacing everything between the plates of the PI and the grid of the outputs except the 1.5K grid resistors. They measure OK. And swapped 12AT7.

[JazzGuitarGimp]

But, why doesn't the schematic make any note of this?

IATA standard drawing practice is to show relays, switches, etc. in a position that results in a de-energized state. I think that probably follows typical industry practice for electrical schematics.

Unless you're looking at the ADG333A quad analog switch datasheet. The four SPDT switches are shown with a data 1 at the control input, which I interpret as energized. Seems counter intuitive....

[martin manning]

But yet, the reverb switch is shown open, which if I'm not mistaken, is the on position?

No, closed (grounded) is "on."

Anyway, regarding the hum issue. I knocked it down by about 40% by stacking a second 25/50v bias supply cap.

You might try adding another 22uF cap to ground from the wiper of the bias adjust pot. The clipping/rounding of the bias supply waveform is probably a result of interaction with the charging cycle of the main filters.

[Randall]

Changing the bias cap to 100uF / 100v as has been suggested elsewhere has reduced hum to a very low level.