I have a fix for this, but I'm trying to understand exactly what is causing it.
The amp is a Headphone amp (though it would certainly drive speakers), Hybrid, a Triode strapped C3M Tube stage feeding into a Mosfet source follower, the tubes are mounted to the top of the case, with shielded cable running to the Anode Grid and Cathode.
I'm posting here since the noise is something todo with the tube stage
Basically when I power the amp on sometimes I get a 100+mV 60Hz hum out of one or both channels, If I lift the top of the amp (where the tubes are attached), it modulates the hum, completely eliminating it in some positions, and I can always bring it back by picking particular angles for the top of the case. If I take a lead connected to my ground point and ground the shield of the cable running to the tube at the end nearest the tube, the noise instantly vanishes.
I don't believe it's a bad connection, or anything to do with moving the connections. The tube Heater is DC with < 1mV of ripple, There is no AC anywhere near the right hand side of the case where the right channel is.
I haven't had a chance to run a permanent wire to ground from the shield to verify that fixes it.
My best guess at this point is, that the aluminum case top is not grounded because of it's coating, and acting as as antenna, but the size of the noise has me skeptical of that explanation, so I'm just wondering if anyone else has experienced something similar, and has a better explanation.
B+ is 150V, the Anode current is ~10mA, there is a 2K2 grid stopper not depicted in the diagram because it's mounted on the incorrectly labelled tube socket.
On the C3M connection 1 is Anode, 2 is Grid, 3 is Cathode
And the Case for illustration purposes
And the internal layout, Pre boards are at the top on the outsides, Disconnected orange wires are for the tube heaters.
The amp is a Headphone amp (though it would certainly drive speakers), Hybrid, a Triode strapped C3M Tube stage feeding into a Mosfet source follower, the tubes are mounted to the top of the case, with shielded cable running to the Anode Grid and Cathode.
I'm posting here since the noise is something todo with the tube stage
Basically when I power the amp on sometimes I get a 100+mV 60Hz hum out of one or both channels, If I lift the top of the amp (where the tubes are attached), it modulates the hum, completely eliminating it in some positions, and I can always bring it back by picking particular angles for the top of the case. If I take a lead connected to my ground point and ground the shield of the cable running to the tube at the end nearest the tube, the noise instantly vanishes.
I don't believe it's a bad connection, or anything to do with moving the connections. The tube Heater is DC with < 1mV of ripple, There is no AC anywhere near the right hand side of the case where the right channel is.
I haven't had a chance to run a permanent wire to ground from the shield to verify that fixes it.
My best guess at this point is, that the aluminum case top is not grounded because of it's coating, and acting as as antenna, but the size of the noise has me skeptical of that explanation, so I'm just wondering if anyone else has experienced something similar, and has a better explanation.
B+ is 150V, the Anode current is ~10mA, there is a 2K2 grid stopper not depicted in the diagram because it's mounted on the incorrectly labelled tube socket.
On the C3M connection 1 is Anode, 2 is Grid, 3 is Cathode
And the Case for illustration purposes
And the internal layout, Pre boards are at the top on the outsides, Disconnected orange wires are for the tube heaters.
Last edited:
What are you using for tube heater power?
I had trouble with hum in a tube amp until I changed from an E-core to a toroid transformer for the 6.3v tube heaters.
Yes, anodization insulates.
Try - on the cover, scrape away circles around each mounting screw holes on both sides just big enough for the screw heads and on the chassis as well using a dremel tool?
Usually 60 cycle hum is magnetically produced by a transformer though. Toroid types produce the lease amount.
I had trouble with hum in a tube amp until I changed from an E-core to a toroid transformer for the 6.3v tube heaters.
Yes, anodization insulates.
Try - on the cover, scrape away circles around each mounting screw holes on both sides just big enough for the screw heads and on the chassis as well using a dremel tool?
Usually 60 cycle hum is magnetically produced by a transformer though. Toroid types produce the lease amount.
It's the toroid, the amp has basically 2 power supplies, the HV 150V one which has the too large Hammond transformer, and a 20V DC supply, which is driven off the 18V toroid.
I intended to try this yesterday, and didn't have time to mess with it, what I could do was try and connect ground to the top, without scaping tings I'm not sure I was getting a great connection, but it did greatly reduce the noise on my scope (though not as much as grounding the shield) which is where my theory originally came from.
This is my assumption, both supplies a re full wave rectified, so any noise from the outputs ought to be 120Hz (though I've measured the ripple on both and it's sub mV even under load), and this is obviously 60Hz.
The AC signal is run down the left of the case to the power switches.
I guess one experiment I can do it make up some extensions and move the transformer out of the case to see if that makes a difference, I could also I guess find some shielded wire for the AC run.
But I assume you concur that it's most likely the case or the shield picking it up.
Sounds like a ground loop from what you have written. A ground loop can also happen in the heater circuit and be audible.
That was also a thought, but I don't see how moving the top of the case is modulating it.
And I've tried running additional grounds from various points in the circuit, back to the central ground point, and the only ground that fixes it is grounding the unattached shield from the signal wires to the tube. Or as mentioned above grounding the top of the case.
I certainly wouldn't rule it out, but I'm at a loss on how to determine what it is and how to fix it if it is.
OK I believe I've resolved this, I believe the issue was that the metal base of the C3M was contacting the screws that were mounting the socket, in turn the screws from the isolation plate were coupling that to the top of the case, and the entire thing was picking up noise from the transformer or possibly just conducting noise from case ground.
The case isn't connected to anything in the C3M so it had to be RF.
As a quick test I just changed the mounting screws to be nylon, and that resolves the issue.
Now that doesn't really explain why grounding the disconnected shield on the tube wires made any difference, and I'm open to suggestions why that might have been.
The case isn't connected to anything in the C3M so it had to be RF.
As a quick test I just changed the mounting screws to be nylon, and that resolves the issue.
Now that doesn't really explain why grounding the disconnected shield on the tube wires made any difference, and I'm open to suggestions why that might have been.
Last edited:
Another possibility is RF oscillation that causes excess HT current that causes hum. But first I'd check whether your grounding is really making contact (rather than just looking as though it does). Serrated washers are your friend. For the RF, 10nF down to chassis from HT. Your photograph shows lots of wiring, so that 10nF needs to be a local 10nF at the gain stage.
When I put the ground screw in the case I was concerned with the anodizing, so I did check most of the panels, and they were all connected to ground if I checked inside an un anodized screw thread. I didn't check the top plate, because it wasn't in place at the time and without screws it's isolated, I'll fix that.
My ground point at the moment is at the input jack, I did experiment with it when the amp was on veroboard out of the case, and it was the least noisy point.
The HV supply is grounded at the gain section, and I could certainly add an AC ground there.
I'll also rerun the shielded wires and ground the shields I didn't do that because while the shielding did make a small difference grounding it didn't, but since it didn't hurt I might as well.
The amp is very quiet at this point, but it's not a huge effort to fix some possible issues, I do have a very slight 60Hz hum on the left channel, I can only hear withe VERY sensitive headphones and it's slight enough that if I heard it on both I'd think I was hallucinating it (and I can't isolate it on my scope). This at least isn't a surprise, since the AC line to the power switch runs down the left side of the case, and it's likely the cause I'll try moving it/shielding it and see if it has an impact.
Last edited: