Question Regarding Source of Hum

[LOUDthud]

Grounding is a complex subject. Star grounding will work theoretically as long as there are no extra paths to ground and there arent any shared wires. Star grounding usually falls short because there are grounds at the input and speaker jacks and a stack of ground lugs on a transformer bolt is far from an ideal star.

The fixes I have suggested are a way to band-aide your amp to the point where the hum is acceptable. Many people will disagree with what I have said. Everone it seems has their own way of doing it, some better than others. I for one never ground the HV center tap at a transformer bolt. YMMV

You can try a separate wire for each filter cap, it may work.

As far as hum goes, the 120Hz buzz type hum comes from the pulses of current that flow from the transformer, through the rectifier, through the first filter cap and back to the transformer. These pulses have a peak current roughly 3 to 5 times the DC current that the amp draws. So part of the path for these pulses flows from the star point at the transformer bolt through the wire that goes to the ground of the filter caps. Since there is a small resistance to that wire, there will be a small voltage drop across the wire of a couple of millivolts. You may be able to measure a small voltage drop across that wire with your DVM if it can measure below 1 millivolt. No big deal you say? It's those few millivolts that are getting into the signal path. When you turn off the amp with the standby switch, you interupt those pulses. You should hear the hum go away immediately when you hear the switch click off and for a second or two the hum will be gone.

[AJD]

Very interesting stuff.

When you say that:

When you turn off the amp with the standby switch, you interupt those pulses. You should hear the hum go away immediately when you hear the switch click off and for a second or two the hum will be gone.

do you mean the few seconds that you can still hear sound between the time that you turn the amp to standby and the time when all current finally ceases to flow a few seconds later?

When you say that

Since there is a small resistance to that wire, there will be a small voltage drop across the wire of a couple of millivolts.

would a larger wire be of any value here or would the difference in voltage drop be negligible?

Pardon my web-illiteracy but what is YMMV?

Thanks very much for the tips. That is a lot of food for thought and it looks like I have a few days of experimenting ahead of me. I have to go out of town this evening but I will report my results as soon as I get back and try out the suggested mods.

[AJD]

As far as hum goes, the 120Hz buzz type hum comes from the pulses of current that flow from the transformer, through the rectifier, through the first filter cap and back to the transformer.

Would a larger filter cap in the first position or a choke be of any value here?

[Andy Le Blanc]

Something I noticed on the schematic, you should put a grid stop on one of
the 6v6. Or between the socket pins in your dress, I've seen that as an issue
in many older scheme, and in my own builds, doubling the number of power
tubes does a couple things. It makes twice the output for the same signal in,
but the transconductance is doubled while the plate resistance and effective
load are halved. Also the current demand is doubled even though the voltage
remains the same. So a couple things could be going on. The doubling of
transconductance could magnify the differences between the tubes,
you can buffer this a little with grid stops, or the current demands might be stressing
you PT.

[LOUDthud]

do you mean the few seconds that you can still hear sound between the time that you turn the amp to standby and the time when all current finally ceases to flow a few seconds later?

Yes, exactly. As the B+ dies you can still hear sound for a second or two.

would a larger wire be of any value here or would the difference in voltage drop be negligible?

That would help but having a separate wire for the first filter cap would help more. Since you have all the filter caps connected to the same wire, the buzz gets on the ground of all the filter caps. Once it's on the ground side, the caps conduct it to the B+ side and some of it will get to the plates of the preamp tubes. Normally those resistors on the B+ side of the filter caps do a pretty good job of attenuating any hum or buzz on the B+ and prevent it from getting to the preamp tubes. By connecting the grounds of the preamp filter(s) to a quiet ground, (a separate wire to the star) you let those resistors do their job.

YMMV= Your Mileage May Vary. It was a disclaimer that used to run at the end of car commercials in the US. By that I meant that there are many people that ground the HV center tap to a transformer bolt and get acceptable results.

Would a larger filter cap in the first position or a choke be of any value here?

A larger filter cap makes the current pulses higher in peak amplitude but shorter time wise. It won't fix the ground problem here and could make it worse. A choke will really reduce ripple on the B+ but won't make much of an improvement until the common ground wire is fixed.

[AJD]

Hi Andy and LOUDthud,
I only have a second here, but I wanted to thank you very much for your input. I'll be doing some experiments when I get back into town and will report the results. I am eager to get to the suggested mods. While I think that the amp is 98% or 99% perfect as is, getting this last 1 or 2% is where the real challenge is......
AJD

[AJD]

Oh yeah Andy, any suggestion of what size grid stopper resistor to use given my schematic? Thanks, AJD

[Andy Le Blanc]

1k to 2.2k to start with, between the g1 pins on the tube sockets, so one tube
gets a straight connection and the other if feed through the resistor, you see
this hook up often in older twin scheme. Or wire it like a regular p-p with
grid resistor on each grid.

[AJD]

Well, this was an interesting day. One by one I tried out all of the various modifications/suggestions offered up by everyone here. The result........nada. I was surprised to say the least. I figured at least one of the leads would pan out significantly. After the first mod (altering the star grounding setup) I excitedly fired it up and....nothing. WTF? After way longer than I care to admit I traced the problem to a broken wire under the circuit board. It seems that the solid wire that I'm using does not like tight 90 degree bends at all. Given that I had only 1/4" beneath the board I had little option but to bend the wire at a 90 degree angle. This sort of thing happened to me once before and so I decided to put and end to it by removing any and all connections that had been made underneath the circuit board and moved them so that all connections are now made from the top of the board. It is not as clean looking as before but it is easier to work on now. Finally got it all working again and ran through the remaining suggestions to no avail. I eventually tried a choke and that actually made things worse, to my surprise. Eventually I ended up changing the 68k input resistor and, more importantly I think, the shielded input wire. That actually made a difference. It sounds a bit quieter now but as it is 2:38 a.m. here it would be a bit impolitic to plug in and crank it up at the moment but I'll give it a test tomorrow. Thanks to everyone for all the help and suggestions. I'll report back after giving it a test tomorrow..........

[Structo]

Yes on the solid wire.

I suppose if it is an amp that you have built a lot and know the circuit inside and out with no tweaking solid wire would be ok but after suffering a similar thing on an amp I only use stranded.

Sometimes on the heater wires I'll use solid because they twist up so nice.

I had a wire that must have gotten bent too much back and forth.

But it still looked ok, that is, I think the break was just inside the jacket.

I spent countless hours troubleshooting that amp and only when I was tearing it completely apart to rebuild it did the broken wire turn up.

So now I only use the Teflon coated multi-strand wire. Love it.
Not so much stripping it because it is too slippery for strippers but I juse use a small razor blade and it's fine.

[AJD]

I had a wire that must have gotten bent too much back and forth.

But it still looked ok, that is, I think the break was just inside the jacket.

I spent countless hours troubleshooting that amp and only when I was tearing it completely apart to rebuild it did the broken wire turn up.

Yep. Same deal here. Lesson learned. I really do like how clean the solid wire can be made to look due to it's rigidity, but I guess I'll have to get over that........

[M Fowler]

The solid wire is very useful and I am just starting to use more of it in my builds. I have found that it can get nicked when trimming, break with bends and when using needle nosed pliers can break from the pliers. I have learned to be more gentle. I hate the stranded wire for switches or tube sockets.

Mark