DR build humming

[Randall]

I just finished my 65 Deluxe Reverb build. Shook a couple of bugs out, but I do have two things I could use some insight on. First, it pops loudly when switched to standby. It's wired as the original AB763 schematic I believe, with the only difference I am using the filter supply board from the original re-issue. It is almost the same except in fact for the node on the hot side of the standby switch, (which as been discussed earlier in another thread) is using different value caps with an odd totem pole to simulate an electrically equivalent circuit. The re-issue didn't pop. Here is the re-issue schematic:

http://ampwares.com/schematics/65_Deluxe_Reverb_RI.pdf

The second thing is I have a hum, 60 or 120 Hz probably. I methodically pulled tubes one at a time until I am down to just the rectifier and 6V6s. I grounded B+, screen supply, and PI plates on one 18 gauge wire from the filter supply board to the star 1" from the PT, along with PT ct and line in ground wire. For the preamp plates supply I ran another 18 g wire to about the mid point on the chassis close to the reverb trafo. I have a beefy copper buss with the board grounds and control pots grounded at the end of the chassis next to the normal input. Outputs have 1 ohm resistors from cathode to chassis. No filament center tap, so 100 ohm resistors from each leg to chassis at pilot lamp. 18 g solid core, tightly twisted elevated pairs to each socket. What am I missing?

Oh, it doesn't pop with just the outputs installed.

http://www.turretboards.com/layouts_...verb_ab868.gif

[Randall]

Today I traced the hum down I think. I discovered by lifting the PI plate node at the 100K/82K junction the hum will reduce to about half to an acceptable level, I tried lifting the other nodes, and it made no difference. But I don't understand it. With no tube in the PI socket, how can that feed pull any current? I measured and verified the .1 blocking caps to the output grids are not leaking. All trafos and filter board are from the original re-issue. OT grounded at output jack. Why on earth is this humming? It's not a bad hum, but it is a hum, and I'd like to get it as quiet as possible. Can any of you smarter guys help me out?

[JoeM]

Randall, I'm not quite sure I follow your grounding scheme based on your description. Can you draw up a diagram? I wouldn't ground the PI node with the B+ CT and screens. Drawing up the grounds may help you better see the problem.

[Randall]

Thanks JoeM. I'm following the original schematic I posted. The filter supply pc board has B+, screens[B], and PI plates[C] grounds tied together. This wire goes to PT CT and line ground. The board has a separate ground wire for the preamp plates[D], which runs to mid chassis. Preamp cathodes and control pots ground via bus to input side of chassis.

[Chuck H]

The second thing is I have a hum, 60 or 120 Hz probably. I methodically pulled tubes one at a time until I am down to just the rectifier and 6V6s.

And with only the rectifier and power tubes the hum is still present?

I can't say about the "pop". I suspect that a cap must be charging in a way that can couple to the signal chain. May be a ground issue.

Have you taken voltage readings throughout the amp? Both in standby and off standby?

There must be something off with the wiring. And exactly what that is isn't clear by the symptoms.

[Enzo]

I could be way off base, but...

If you get hum with only the outputs and the rectifier, AND you get a loud pop when standby switches, pull your power tubes and power up. Not expecting it to, but does it hum now, or pop? But assuming not, Now go to each power tube socket and verify with a meter that you have B+ on BOTH pins 3 and 4 of each power tube socket. How much ripple is on those? And also, verify the bias voltage on each pin 5, but more important, is there any ripple there? And since this is a build, with power off, verify each pin 8 is actually going to ground. Or 1 ohm, if you added those.

And try this: Pull just one 6V6 - how much hum? Now replace it and pull the other, and how much hum? it should be about the same with one or the other. If they are different, that is a clue.

Verify both pins 1 and 6 of the phase inverter have B+.

Hum is pretty much always going to be 60Hz or 120Hz, they are the same note, but an octave apart. But they are not pure sine waves, so the overtones can make it hard to tell by ear. They come from different causes, so we do need to know if it is 60Hz or 120Hz. Do you use a scope?

[Randall]

Thank you Enzo. I verified all your suggestions. No sound with outputs removed. About the same with either tube removed. I have a 20Mhz Leader scope, but I'm not an expert at using it. For example I have a hard time figuring frequency with it, and I don't understand why my Fluke 77 and scope give me different AC readings which you can see from the following:

6V6 one

3. Fluke: 452vdc, 0.73vac 120Hz
Oscope 2.5vac saw wave

4. Fluke 450vdc, 0.02vac
Oscope 0.05 vac sine wave

5. Fluke -38vdc, 0.015vac
Oscope 0.4vac saw

6V6 two

3. Fluke 452vdc, 0.73vac, 120Hz
Oscope 2.5vac saw

4. Fluke 450vdc, 0.02vac
Oscope 0.50vac sine

5. Fluke -38vdc, 0.05vac
Oscope 0.2vac saw

[g1]

The scope shows peak to peak voltage but the meter reads RMS. So half of peak to peak times .707 is the conversion. That is for a sine wave so your sawtooth ripple may be a little different but should be close to that. So 1 volt peak to peak on the scope will be around .35V on the meter.
Some of your measured voltages look about right for conversion, but others (pin5) look like the decimal place is wrong. Please double check your readings for pin5, they should be pretty much the same for both tubes, and scope reading should be around 2.8 times the meter reading.

[Randall]

Thank you for explaining that g-one. It's a little confusing why two expensive test instruments give different readings, but I can live with your explanation.

I double checked my readings, and I did transpose on the second pin 4. They both should be 0.05vac. I don't see how to edit it at this point. But, pin 5 readings are correct. They are different. And, the saw wave on the second 6V6, (the one with the lower pin5 V's and being fed by the first PI plate on the 82K side) is distorted on the rising edge, where the larger saw wave on the other tube is very clean. On the other side of the 220K bias splitter I measure a 1.1vac saw, and 0.3vac on my meter.

I have replaced the 82K/100K PI plate supply resistors and 0.1 coupling cap on the distorted wave side, as well as verified the value of both 220K bias supply resistors All measurements were with all tubes in place and controls to off.

[g1]

So you have different levels of ripple on pin5 and you can see a difference in the ripple waveform. Swap the 6V6's around and see if the "distorted ripple" stays with the socket or follows the tube.
Keep in mind that mismatched power tubes cause hum.

[Randall]

Ripple stays with socket, however while lifting the 0.001 cap on the PI first grid, it came apart. I replaced it. Now, and for reasons I am not grasping, the voltages on outputs pins 5 is much closer. With power on, standby off, I see clean 1.4 vac and 1.5 vac p-p bias ripple saws on respective output pin 5's and PI plates. When I switch standby on, the saw on the 82K side clips at the top, and the 100K side is a bit rounded at the bottom. I'm at a loss as to what could cause this.

I did reduce the hum by about 40% by stacking another 25/50v cap on the one that was already in the bias supply.

Another odd thing is it goes into a high freq. osc. sometimes when I probe the 82K side at PI pin 1, either side of the .1 cap or output grid, but only if my probe is on x1. Switching to x10 stops it. It does not oscillate on the 100K side at all. ???

[g1]

I see clean 1.4 vac and 1.5 vac p-p bias ripple saws on respective output pin 5's ...
I did reduce the hum by about 40% by stacking another 25/50v cap on the one that was already in the bias supply.

I take it that was just for test purposes, a 50V cap will not be safe there.
The stock cap is 100uf100V, is that what you have? If adding another 25uf made such a difference I wonder if the 100uf is just bad, the 1.5Vp-p ripple on the bias sounds a little high anyway. I would try replacing the 100uf bias cap., I think 150uf100V is often the cap of choice for these bias circuits.

[Randall]

Huh. My schematic shows a 25uF/50v, so that's what I put in. I added a second 25uF/50v to it. I found myself wishing in the last couple of days that I had a nice bench supply to replace the rectifier board with, that would have told me a lot. My next investment I suppose.

I'll try upping that cap, it sounds like that's likely the cause.

[g1]

Ok, looks like we are talking about different bias caps, you are speaking of C29, 22uf63V on the '65 re-iss schematic you linked in post #1?
I don't see 25uf50V anywhere?
Vintage Fender circuits used only 1 bias cap like C36 (100uf100V) shown down below the 5AR4 on same schem.
That (C36) is the one I suspect may be bad.

[Randall]

No, we are talking about the same cap, but different schematics. I see where the re-issue has a 100/100 cap there, which I had not noticed, but the original Deluxe Reverb-Amp AB763 schematic that I have shows a 25uF/50v. I also just realized the second older schematic I posted is not the one I am working from, it's a later one with yet another different cap value. This is the right one:
http://ampwares.com/schematics/deluxe_reverb_ab763.pdf