SET Amplifier ground loop hum

Beast_of_burden

Active Member
Joined
Jun 14, 2011
Messages
291
Points
43
Location
Bangalore
Hello Gents,

Just completed putting together a JE Labs design 2A3 SET with help from HFV member Radium. Amp is making music but there is a really loud 50Hz hum that is simply unacceptable. This design has been implemented countless times with inaudible hum.

I need help from an expert in Bangalore who can troubleshoot and fix the ground loop hum. I have to confess this is way beyond my pay grade. The amp was initially implemented with Star ground but this is not recommended by the designer so have now implemented sectional grounding atleast for the inputs. When I say inputs it's the RCA, Volume potentiometer and cathode resistor ground leads. Hum remains. It's loud and independent of position of potentiometers. It's loud enough that I can hear it in the next room 20 feet away.

The PS capacitor leads have all been grounded to the chassis at a single point. I lifted this and the hum disappeared but it also stopped playing music. I don't know what this means.

Hari has suggested replacing the paralleled 100uf electrolytic caps for the cathode resistor with MPP or MKP capacitors. I am trying to source those locally but it's hard during these times.

I live in South Bangalore, please help. I've attached the schematic. I can also provide pictures of the amplifier internals, but really what I am looking for is an expert who can "read" the schematic, take some measurements and suggest changes to the grounding scheme.

Signed
Frustrated with that old demon - Hum
 

Attachments

  • JELsc-ST2A3DX.jpg
    JELsc-ST2A3DX.jpg
    47.4 KB · Views: 41
Is the hum from the speakers or trafo? If from the speakers, is it there even when there is no load on the inputs? Does it vary with volume?
 
Is the hum from the speakers or trafo? If from the speakers, is it there even when there is no load on the inputs? Does it vary with volume?
It's there on the speakers. I have measured the frequency and level using an app. See attached.Screenshot_20200702-220545.png

It's there even when inputs are disconnected.

No variation with volume.
 
Nice to see triodes.Thought they had, gracefully, accepted their obsolescence like many other things from my generation! Anyway, here's a checklist that might help:
1) Are the input grounds connected to the chassis? Disconnect the RCA inputs from the chassis and check. If you must shield the input connector, use a shielded RCA wire and connect the shield to the ground and not the sleeve.
2) It looks as if the wires that supply the filament current are not twisted together. The current is AC at 50Hz. Check if the hum remains after you have moved the filament supply as far away from the outputs as possible.
3) Now, on to minor tweaks. When implementing a star ground, we used to connect all grounds to a single nut by means of a crimped ring-terminal; the nut was then connected to the chassis (a ground lifter might be useful in some situations) along with the earth wire from mains--you are using a three-pin plug I hope? Do not attempt a double-isolated build with triodes in a metal case!

Cheers,

~HP
 
A quick edit-- Please disregard my suggestion for a shielded connector. It does not apply to this situation. Disconnect the RCA connectors from the chassis as suggested and test for hum.
Cheers,
~HP
 
A quick edit-- Please disregard my suggestion for a shielded connector. It does not apply to this situation. Disconnect the RCA connectors from the chassis as suggested and test for hum.
Cheers,
~HP
Thanks for responding. RCA inputs are isolated from chassis, however RCA ground leads are connected to the chassis ground. RCA input is connect to volume potentiometer and the ground leads from the Volume potentiometer are independently connected to chassis ground.

Tested by moving filament supplies close to the output tubes but hum remained.

Today, I changed the grounding from star ground to sections and then connecting each section to the chassis ground. I was hoping this would make a difference but it didn't.

Forum member Hari has suggested changing the paralleled electrolytic capacitors for the cathode resistor to MKP or MPP type - I will try and procure those capacitors tomorrow.

The hum is being amplified by the output stage for sure. Thanks for your help.
 
Disconnect the RCA grounds from the chassis is what I meant! The ring of the RCA connector goes one end of the potentiometer (done); the wiper goes to the grid (done). Connect the 2 resistors (1K and 22K), the free end of the potentiometer, and the 100microF/100V capacitor to the free lug on the potentiometer. This is your ground, to which you should connect the sleeve from the RCA connector with a wire.
Cheers,

~HP
 
Last edited:
Disconnect the RCA grounds from the chassis is what I meant! The ring of the RCA connector goes one end of the potentiometer (done); the wiper goes to the grid (done). Connect the 2 resistors (1K and 22K), the free end of the potentiometer, and the 100microF/100V capacitor to the free lug on the potentiometer. This is your ground, to which you should connect the sleeve from the RCA connector with a wire.
Cheers,

~HP
I disconnected RCA ground and the hum got much louder. I will check the other connections. Hum is not dependent on volume level though.

Earth and ground are not the same. The chassis is supposed to be earthed for safety; it is not meant to carry a return current unless there is a fault in the circuit. Check this PDF: http://hifisonix.com/wordpress/wp-c...4/ilimzns-Excellent-Posts-on-Ground-Loops.pdf
Also, please be careful when working with high B+ devices.
Cheers,
~HP
I have understood the part about the mains ground being connected to the chassis otherwise called earth for safety reasons. I will connect the mains earth at a separate spot on the chassis. Currently all the sections and the mains earth are connected to the same spot on the chassis.

When I disconnected the PS capacitor ground leads from the chassis the noise disappeared. But then I also lost music don't quite know what to make of this. The whole of the input signal is being conducted through the chassis?

I am gonna have to read up on that document.
 
In its present configuration, your chassis is part of the signal chain which it not ideal. Disconnect everything from the chassis and test the circuit, without power, on an insulated surface. With an ohmmeter, check all connections to ground, i.e., check that every node in the circuit that should be at ground potential is connected--you should see near identical resistances very close to zero when you measure from one leg of a component to another. Ideally, all interconnect wires in point-to-point systems should be the same thickness, which will allow you to subtract wire-resistance from your measurement. Once you have verified that the circuit is properly wired, power up and check if the hum remains; use an incandescent bulb in series.

Just some advice from an old timer: If you enjoy working with valves and other high B+ devices, invest in a good RCCB your room. An isolation transformer is also useful but they are expensive and will become dead weight once you've got the hang of working with high voltages.
Cheers,
~HP
 
It's been a deeply frustrating experience and I have been fighting the urge to abandon and look elsewhere. But the amplifier sounds superb when the music is playing, superb balance top to bottom and refined. Very encouraging, if this music playing quality is retained and the hum is fixed this amplifier is a keeper.

I read the grounding scheme document a number of times and implemented some of the recommendations except the 10 ohm resistor between SG and PG. There was a slight reduction in volume of hum by 2 db.

Now, the amp will be shipped to Chennai and and someone vastly more experienced and the right tools will troubleshoot the 50hz hum.

Edit: I sat down to test a couple of songs and I am still here listening to the Wall side 1 connected to an iPhone. Amplifier sounds absolutely gorgeous, praying the hum issue is resolved.
 
Last edited:
As discussed, the hum is due to power supply ripple voltage which finds way to the B+ supply. The only way to reduce this is to replace all the Electrolytic capacitors in your power supply with WIMA MK4 type of capacitors of the same value. Also the Electrolytic bypass capacitor across your cathode by-pass resistor needs to be finally replaced with WIMA DC link capacitors. These are expensive mods and cannot be done in a day or two. These components are available in mouser.in Hope this helps.
 
The filament power supply for 2A3, is that A.C or D.C ?

Ideally, look at D.C to reduce hum from this area. The 2A3 is notorious with A.C as far as I know.
 
The filament power supply for 2A3, is that A.C or D.C ?

Ideally, look at D.C to reduce hum from this area. The 2A3 is notorious with A.C as far as I know.
The beauty of 2A3 tube lies in the AC filament supply. Convert that to DC and all magic gets lost and they sound like normal amplifier.

Due this hum issue I am not considering v2A3 tube in my DIY design. But I love AC filament output tubes. I will be designing with IDH tubes with AC heating
 
@Beast_of_burden see Figure 5 on this page. It's not difficult to implement and is quite cheap to buy the parts (easily available at local electronic part stores). It completely cured ground loop issues I've had in my preamps and amps.
 
Hi,
Nice build. I suppose it must be above the chassis. but from picture can not confirm. Have you used choke in power supply ?
Regards
 
The filament power supply for 2A3, is that A.C or D.C ?

Ideally, look at D.C to reduce hum from this area. The 2A3 is notorious with A.C as far as I know.
The consensus is that 2A3 is one tube that can be desinged with no hum with AC heating. I can believe this because a pair of Wright monoblocks I had were dead silent.

What brand 5U4 and 6SN7GTB tubes are you using?

Does the amp has biasing option to be done every time you change tubes ?

I had similar problem with my SE 45 tube amp which uses one 6SN7gtb each channel. My biasing efforts would minimize it but slight hum was always there and it was annoying.

I changed my premium chrome topped GE 6SN7gtb tubes to RCA black plates and the hum vanished. ( More money doesn’t not necessarily equate to better Sonics in Audio !)

This nasty hum is a 50hz / 60hz hum Which comes from woofers and stays put irrespective of volume. It does not increase or decrease with volume but stays put, right from powering on. Interestingly it’s completely inaudible on my 100dB FR but its there on the 97dB woofers.

Changing driver stage tubes made it vanish.
Biasing is set by a hum pot. The nasty hum is exactly like you describe its through the woofer and independant of volume control. The hum takes about 15 seconds to manifest itself, it appears to be happening once the capacitors are fully charged and then its a jackhammer. I am actually using a sovtek 5Y3 for the 5UG. The 6SN7GTB is a new production Tung Sol. The Wright monoblocks had new production EH 6sn7 and it was a superb amplifier, just saying new production 6sn7 tubes are quite good. All tubes in this JE labs design are new production and yet it sounds awesome, except for the hum ofcourse. The 2A3 tubes are either Sovtek mono plate or Shuguang 2A3C.

@Beast_of_burden see Figure 5 on this page. It's not difficult to implement and is quite cheap to buy the parts (easily available at local electronic part stores). It completely cured ground loop issues I've had in my preamps and amps.
Looking into these grounding schemes, it will take time to implement. Taking the help of an expert as I am not comfortable working with high voltages beyond a point. Parts like the WIMA MKP caps arent cheap and not easy to source either, will keep y'all posted on the progress.
 
In the schematic, heater for 6SN7 is center tap and it is grounded.you can remove this from the ground and referenced it to DC voltage.
more details see below website

Star ground can be connected to chassis through 100ohms resistor parallel with .1uf X2 capacitor.
 
The Marantz PM7000N offers big, spacious and insightful sound, class-leading clarity and a solid streaming platform in a award winning package.
Back
Top