It hums...

[Phil_S]

...and I'm not figuring it out.

This 5B6 build, it hums 120Hz. It's loud. The hum is louder as I turn up the volume control. Hum is mitigated when I turn to tone pot to 10 -- that is when it meters at 2 ohms, effectively open so that it grounds what's on it. Heck, I guess that grounds what's coming from V1, so it "fixes" the hum?

Hum shows a very pronounced decrease when I pull V1 (6SC7). I replaced plain wire with shielded twisted pair on the input jacks, but that made no change.

I have thought about a ground loop, particularly the output transformer. The 0 tap goes directly to the chassis ground lug and the sleeve of the output jack does to that same ground lug. But that only makes sense if the noise remains after pulling V1, which is doesn't.

I keep thinking it is just the way those metal 6SC7's are. I've got 3 of them, and the all do the same thing. Pin 1 (shell) is grounded to the chassis via the screw that holds the socket. Should that go to the circuit ground -- is it a possible ground loop?

I checked and resoldered any suspect grounds. I checked to be sure all grounds are present and accounting for.

Lot's of pictures here. Since posting those, I have fixed the rectifier wiring error, fixed the wiring error in the PI between the .1 caps, and replaced the input wires for V1. http://home.comcast.net/~psymonds/5B6.html Today's overview is attached along with a layout.

I'd be grateful for some suggestions. Thanks.

[Milkmansound]

get that first reservoir filter cap off the ground bus - that was my initial thought when looking at your layout. Try moving it to another ground spot (heaters, or AC inlet) far away

[Phil_S]

I don't think so. Your eyes may not see at first, and I'll explain.

There is a 2 part buss joined at cap #3 to capture the power tube cathode R/C and the first 3 filter caps. The first two caps are in parallel for B+1. The 3rd cap is for the power tube screens. This is grounded to the chassis via a black wire at the SE corner of the board, from the corner turret.

Maybe I should run the power tube R/C directly to the chassis ground bolt?

There is a second and longer buss for the preamp sections. There is a black wire to the chassis ground point from filter cap 4, which is for the preamp supply.

Te chassis bolt is just southeast of V4 (6L6). There are 5 wires connecting:
1-High voltage caps and cathode R/C for the power section.
2-Filter cap 4 and the preamp ground buss.
3-0 tap from the output transformer.
4-Sleeve from the output jack.
5-PT center tap.

You may be on to something with a possible ground loop. The schematic shows the paraphase inverter grounded with the power tube cathode R/C. Looking at the picture, counting turrets from the left, go to the 9th at the top and drop down to the mid board turret at this location. This is the junction of 6800 and 270K resistors, which needs to go to ground. Do you think maybe this is better on the preamp buss?

Now, I know lots of folks want to ground the preamp buss near the input jack. I have never done it that way and it has never been a problem. This is how I was taught to do it by an old-timer. Please. let's not debate this finer point at this moment.

So, that's two ground questions.

Milkman, You sparked a couple of good thoughts. I don't want to be moving things willy-nilly.

Edit: Are you suggesting I break the ground buss between B+1 and B+2, taking each directly to the chassis bolt?

As I said, maybe, but how is it that the hum practically dies when I pull V1?

[tsutt]

I think i would ground your input shielding to the ground buss at the jacks not to the tube socket screw.

[Phil_S]

I think i would ground your input shielding to the ground buss at the jacks not to the tube socket screw.

The shield is not part of the circuit. It does not belong on the buss. It is an RF shield and it performs a similar function to the chassis. Thanks, though.

[tsutt]

Its shielding rf and bringing into your preamp tube like an antenna the way i see it. Does your ot have a dedicated ground wire for the output jack?

[guitarmike2107]

Hi Phil

on the photo I dont see a ground connection for your second filter cap, red dot with no wire going to ground buss, that would make V2 float.

On the layout I also don't get the negative feedback loop (purple wire), looks like you are grounding the hot tip of the speaker jack, maybe thats just a layout error... or my error

goodluck

[Phil_S]

guitarmike: Thank you for looking. There is a bridge between the two turrets for the 1st pair of filter caps. The layout drawing probably has some errors. In the photo, there is a purple wire between the 4 ohm tap on the swtich to the mid board turret (underneath) where the 6800 and 25R are joined. The red dot is there to remind me that there is a wire under the board from that ground to finish the NFB loop. That V2 cathode R/C is elevated by the 25R, which is grounded.

If folks will keep going, sooner or later something will pop!

[Milkmansound]

how is it that the hum practically dies when I pull V1?

not sure - in my amps I get the B+ node for V1 and the main reservoir cap as far apart as possible. The first cap grounds by the power transformer with the center tap and the safety ground from the AC inlet, and V1 and its B+ node ground at the input jack (think Fender style chassis).

Everything else can ground somewhere in between the two.

This has worked very well for me - and when I tried a universal star or bus ground I have introduced that 120 cycle hum. Without hands on, its hard for me to troubleshoot. Its easy enough to tack on some new wires going to new places to try to get the noise down - if they don't work, pull them out - if they do, go back and make them permanent.

[epis]

You do have problem with grounding scheme, resistor 25 Ohm (part of NF and PI circuit) shouldn't be grounded at preamp ground buss.

[Phil_S]

You do have problem with grounding scheme, resistor 25 Ohm (part of NF and PI circuit) shouldn't be grounded at preamp ground buss.

That is a MAJOR improvement! Thank you ever so much.

There is still hum. The tone and volume controls no longer affect the hum, nor does the hum change when I pull V1. It is dead silent when I pull V2. I think we know, this means there is a problem with V2 and it suggests the PA is OK. It is too bad this amp has V1 and V2 sharing the same B+ node. I suppose I could have avoided that, but it is too late now, as there is no easy way to do it.

I did not like the ground between the 6800 and 250K at the ground side of the power tube cathode R/C. I did that as it is the literal interpretation of the schematic and I figured that maybe Fender knew something. OTOH, maybe Fender was lucky? I moved that to the preamp buss. It is hard to tell if that made a difference.

I think there is something else about the V2 ground(s). There are only two grounds. One is the R/C for the second triode cathode. The other is the one described above between the 6800 and 250K divider. The first triode cathode is elevated at the NFB divider connection to the speaker, which I just fixed. In addition, I have the 4th filter cap (which supplies that section) on the preamp buss. I am wondering, is the specific placement on the buss an issue and or if I should just put the filter cap on the power buss?

Should I try shielded to the grids of V2?

[epis]

You're welcome Phil
There is a few things what I would do :

-ground preamp buss at input jack and remove connection to star ground
-ground output jack close to itself or use metal jack and remove connection to star ground
- ground input cable to buss instead close to first preamp tube ( as it was mentioned before)

Amp should be quiet, I'm doing it this way to all my amps and I don't have any problems with hum.
Good luck

P.S.
One more thing, replace wire going from volume pot wiper to grid of PI tube with shielded one.

[M Fowler]

I've been looking at this for the past two days and so far haven't come up with any new ideas, sorry.

Mark

[Phil_S]

I've been looking at this for the past two days and so far haven't come up with any new ideas, sorry.
Mark

Well, Mark, that's not what I hoped to hear (the not finding it part)! I appreciate the effort. It will come out in the wash sooner or later. I am of the belief that there is always an explanation and a solution.

Phil

[Phil_S]

You're welcome Phil
There is a few things what I would do :
-ground preamp buss at input jack and remove connection to star ground
-ground output jack close to itself or use metal jack and remove connection to star ground
- ground input cable to buss instead close to first preamp tube ( as it was mentioned before)
<snip>
One more thing, replace wire going from volume pot wiper to grid of PI tube with shielded one.

I appreciate all the suggestions and am willing to be open minded. However, we are going to agree to disagree on grounding the shield on the circuit ground. I think it should not be done. If there is already noise on the ground buss, introducing it to the shield will defeat the purpose of the shield. Maybe it works, but I don't see it.

I replaced the plain wire with shielded as suggested. I think it is worse, not better. I am going to pull the shielded and go back to plain.

I disconnected the preamp buss from the main ground bolt and grounded it at the other end, very near the input jacks. No change. I'm going back to the way it was, as that has always worked for me in the past. We each have experience that propels us.

I am going to try a few things, in no particular order of priority and not sure in what order I'll do them. I always make only one change and then check the amp.

-Reconnect the ground wire from the 250K/6800 to the ground end of the power tube R/C in a physically different location. I will run a separate wire to the main chassis ground, removing it from the Plate and Screen filters ground on the board. (Yeah, I know that's a 2fer.)

-Move the 4th filter cap off the preamp buss and directly to the main ground.

-Before doing any of that, I'm going to take a piece of cardboard covered with foil and place it on the open side of the chassis...just to see if it simply needs a bottom! BTDT and been amazed before.

...And I'm going to keep working on it!