Diagnosing source of 100Hz hum

[timrobertson100]

Hi folks

I've recently finished wiring up a 6v6 Silver Jubilee and am having trouble diagnosing the source of a 100Hz hum - I'm tired and feel I am overlooking something obvious.
Any suggestions for where to look next, please?

With V1 and V2 pulled, the hum is present with the master at 0 and increases significantly with the master volume (perhaps loudest at 9). I can scope it on the way out of the master.
I've chop-sticked around every wire, and nothing changes it - heaters, power leads, those long runs to the power tubes etc.

I have 2 main ground points. 1 at the input for the preamp and send/return, and 2) for the PI and onwards, along with the relay power and the elevated heater - it's OK to ground the relay and elevated heater at the same spot as the first 3 B+ nodes right?
(OT secondary centre tap is grounded along with the the 6v6 cathodes)

Thanks all.

[Stevem]

If you lift the ground on the master pot and just jump it out what’s the result of that?

How much ripple voltage does your elevated heater have?

If you don’t use it what is the result of that change?

[GAStan]

I'm pretty much a noob myself so this is a question to the gurus as well as an observation. It looks like the B+ from the rectifier (pre filtered) is bundled with the post filtered B voltages. Is there enough ripple voltage to transfer inductively/capacitively between the two?

[timrobertson100]

Thank you both

From those comments:
- Scoping the elevated heater shows no ripple, but I've now referenced it to ground instead - no effect on hum.
- I've taken the master out of the circuit - no effect on hum other than volume
- separating out the b+ #1 and chopsticked around moving the b+ wires and no effect.

Additionally (and to no effect)
- I've removed the power to the relay circuit to take that out of the equation
- I've tried grounding the preamp b+ filter cap on the bus rather than the star
- removing the GNF has no effect

Current status:
With V1 and V2 removed, there is a hum that increases with master volume.
Grounding the input to the PI and it's silent. Grounding before the 22nF coupling cap (i.e. on master out) does not silence it. This seems... odd?

Thanks all

[bepone]

Hi folks

I've recently finished wiring up a 6v6 Silver Jubilee and am having trouble diagnosing the source of a 100Hz hum - I'm tired and feel I am overlooking something obvious.
Any suggestions for where to look next, please?

With V1 and V2 pulled, the hum is present with the master at 0 and increases significantly with the master volume (perhaps loudest at 9). I can scope it on the way out of the master.

100 Hz is rectifier hum, it is inserted through the output tubes or from preamp..
preliminary test - pull all the preamp tubes and check the speaker terminal with the scope, should be some minimum humm

1. check the current through each output tube, must be close, otherwise you have 100Hz on the speaker
2. check if you miswired the choke or PI Ub+ node

[bepone]

then checking the layout.. rectifier diodes are not in the good position, they are too close to output tubes, they can spread noise around them to all the sensitive wires

remove all the rectifier diodes (biggest source of hum) to the far right position, close to the first electrolytic capacitor, to make path rectifier-first elco as short as possible

[timrobertson100]

Thank you @bepone

then checking the layout.. rectifier diodes are not in the good position, they are too close to output tubes, they can spread noise around them to all the sensitive wires

remove all the rectifier diodes (biggest source of hum) to the far right position, close to the first electrolytic capacitor, to make path rectifier-first elco as short as possible

Before going for that - this thing is *silent* if I ground the input of the PI. Would that not indicate the diode location is not the problem here?

[bepone]

Thank you @bepone

Before going for that - this thing is *silent* if I ground the input of the PI. Would that not indicate the diode location is not the problem here?

ok, then ok.. problem is before the PI. Ground every stage input (grid g1) and check where noise is starting..but then problem is strange, 100Hz is rectifier noise, and in the preamp should be filtered out totally. But lets see.

Check Ub+ node on every stage, post mVpp results here. Scope x100 or x10 AC coupling, dont use x1 because scope input can be burned

[timrobertson100]

Thanks for confirming.

So I have v1 and v2 pulled, and somewhere between the output of v2, and input of v3 (PI) I get this 100Hz - so basically the tonestack.
If I ground before the coupling cap to the PI hum remains, but if I ground after the coupling cap (i.e. pin 3) it's silent. I've wondered if it's coming in through the tail side of the PI...

My scope leads are only 1:1 and rated for 300V, so I can't probe the B+ for V2 and V3. Is there a suitable capacitor value you'd suggest putting in between the probe to remove the DC so I can see the ripple, please?

Thank you again - I'm pretty stumped on this now.

[bepone]

dont measure directly Ub+ line with 1:1 then..

so grounding before input capacitor to the PI, gives humm.. to be sure that we are in line, this point -

IMG_5344 (1).jpeg

humm must dissapear. if doesnt, it is created in PI

[bepone]

My scope leads are only 1:1 and rated for 300V, so I can't probe the B+ for V2 and V3. Is there a suitable capacitor value you'd suggest putting in between the probe to remove the DC so I can see the ripple, please?

you can try with 100nF 630V at least, but i would find or borrow 1:10 probe...

So I have v1 and v2 pulled, and somewhere between the output of v2, and input of v3 (PI) I get this 100Hz - so basically the tonestack.
If I ground before the coupling cap to the PI hum remains, but if I ground after the coupling cap (i.e. pin 3) it's silent. I've wondered if it's coming in through the tail side of the PI...

Grounding after coupling cap- pin 3 is not grid, it is shutting down PI totally. Also if you ground pin 2 , grid g1, again not ok because grid is floating on +40-50V DC.. so cannot be grounded this way - directly to the 0 V

[bepone]

im back to my original thoughts, solving one by one point
first
- relocate main rectifier diodes to the far right position- for the test to not destroy the board, can be built in the air, p2p with several new diodes, and mount directly to the capacitor for main Ub+ node

[timrobertson100]

dont measure directly Ub+ line with 1:1 then..

so grounding before input capacitor to the PI, gives humm.. to be sure that we are in line, this point -
IMG_5344 (1).jpeg

humm must dissapear. if doesnt, it is created in PI

Yes, Grounding at that point and there remains a little hum (more than you’d like) - same as with master at zero. Increasing master and it gets considerably louder.

[bepone]

Yes, Grounding at that point and there remains a little hum (more than you’d like) - same as with master at zero. Increasing master and it gets considerably louder.

Ok, this is the hum what is created from PI and output stage together.. maybe the output tubes are far from matched, and this hum (if we are sepaking about 100Hz component) can be eliminated with matched tubes, better layout of rect. componets and so on.

If you open master volume, and the 100Hz gets louder, this part is produced in preamp stage.

So keep master 100% open, and start to check the preamp part, easiest way is described before, ground stage before PI- i dont have schematic in the head now but somewhere where is g1 normally connected,, cathode sits on few volts, there, not at some cathode follower
Then check output from this stage after coupling cap with your 1:1 probe, and if you see 100Hz this is not ok, and this stage doesnt have clean DC supply and it gets amplified, and send to PI, amplified more, and to speaker output.

If there is no 100Hz hum, this stage is ok. Remove ground, and go to the stage before that. GND input g1 again, and see what is going on the output from this stage after coupling cap. If there is 100Hz, it is produced in this stage and after amplified more in following stages.

When you find out where is produced, then we can discuss how to fight that problem.

[bepone]

i dont see filter capacitors and how they are connected (below the board), but connecting wires to filter caps should not be close to other sensitive wires (signal) and components (coupling capacitors are like antenna, roll of many turns of the foil, if they are close to not clean DC supply wire, there will be coupling to signal circuit )..