50/60/whatever cycle hum

[engels]

I'm trying to find out how to completely get rid of the AC hum. The device I'm trying to "fix" is an instrument preamp with two 5670 tubes. It works fine and it's very quiet but there's that slight HUM even with the volumes down, with everything on zero etc. It's definitely not from the sound path and I'm trying to find out where it comes from. It's very, very deliberate - you do not pay attention to it untill you turn the power off and notice there's no background humming. I've got many other devices that make the same type of noise - for example, my AIWA stereo amp is also humming with everything down.

There's no change in the hum level when you adjust the volume, it's a constant noise.

I understand it should come from the device's ground, a ground loop maybe? Or is this some other known bug?

I'm not sure the schem is needed to answer this, but still here's the schematics and the layout.

 

[jan.didden]

It is important to be absolutely sure whether it is just hum, which is a 'round' 50 or 60Hz tone, or whether it is hum with a slight buzz, with some buzzing or rattling components. In the latter case it is most probably a ground loop or ground wiring issue. In the former case it is most probably a screening issue.

It appears that you use a plastic case. That may lead to the hum (if it is just that).

Do you think it is hum or buzz?

Jan Didden

 

[engels]

hum

First, it's the 60 or 50 cycle hum (don't remember the number), not buzz and not oscillations. I'm completely sure it's the ac hum - the sound is around G (a bit lower than 440hz which is A of the same octave - in US it's another note, here it's the European hum).

The chassis is steel. The transformers have brass covers, everything is shielded and there are no multiple ground connections to the chassis.

 

[jan.didden]

Re: hum

First, it's the 60 or 50 cycle hum (don't remember the number), not buzz and not oscillations. I'm completely sure it's the ac hum - the sound is around G (a bit lower than 440hz which is A of the same octave - in US it's another note, here it's the European hum).

The chassis is steel. The transformers have brass covers, everything is shielded and there are no multiple ground connections to the chassis.

Well, if its hum it is unlikely it is due to bad grounding anyway. Are there any other clues? Both channels exactly the same? Open or shorted input makes a difference?
Steel shielding is not very effective for 50Hz shielding. If you rotate the transformer 90 degress, does that make a difference (assuming that's possible to do)?

Jan Didden

 

[Depanatoru]

The power supply is not very good , this could cause AC hum . The resistors from the "pi" filters are too small and although capacitors are very big , without big resistors or chokes , the ripple is pretty high for a sensitive preamplifier. Of course , the best is a regulated power supply.

 

[Sheldon]

Re: hum

First, it's the 60 or 50 cycle hum (don't remember the number), not buzz and not oscillations. I'm completely sure it's the ac hum - the sound is around G (a bit lower than 440hz which is A of the same octave - in US it's another note, here it's the European hum).

The chassis is steel. The transformers have brass covers, everything is shielded and there are no multiple ground connections to the chassis.

Sounds like inductive hum. Tube heaters are DC, so that shouldn't be a problem. I'm guessing your transformer is coupling to the steel case. Try unbolting the transformer and putting a sheet of plastic about 2 or 3mm thick between it and the chassis.

Sheldon

 

[Colt45]

My cash is on the HT supply. that would make it 100/120Hz, by the way...

 

[jan.didden]

The power supply is not very good , this could cause AC hum . The resistors from the "pi" filters are too small and although capacitors are very big , without big resistors or chokes , the ripple is pretty high for a sensitive preamplifier. Of course , the best is a regulated power supply.

That's why I asked if it was hum or buzz. If its hum, not buzz, it's not the supply.

Jan Didden

 

[jan.didden]

Re: Re: hum

Sounds like inductive hum. Tube heaters are DC, so that shouldn't be a problem. I'm guessing your transformer is coupling to the steel case. Try unbolting the transformer and putting a sheet of plastic about 2 or 3mm thick between it and the chassis.

Sheldon

I agree it's probably inductive hum, but putting a sheet below the xformer won't do anything. What *can* fix it is rotating the xformer so that the field is in another direction.
If possibly, remove the xformer, and connect it temporarily with longer wires. Then move the xformer around to find the best position/direction. Is this a toroid or an EI core?

Jan Didden

 

[engels]

induction

Induction sounds interesting. I will try to disconnect the transformers. The transformers are EI and there are 3 of them as you can see on the schematics.

I've got an older photo of the thing - currently the tubes are also shielded.

 

[Richard Ellis]

Hmmmmmmmm..

First off... the center Transformer should be rotated 90 degrees as you look straight down from the top.
You can twist it OK but you will have "dummy" holes you can 'fill' with a nut/bolt that goes nowhere. The TX will need to go forward toward the tubes somewhat to clear the lip on the chassis.
And yes you will need to change the values on your PI filter on the PS.
The wildcard is the steel chassis...the interaction of magnetic fields is the one unpredictable variable & experimentation is the only way to nail it down.
____________________________________Rick...........

 

[Sheldon]

Re: Re: Re: hum

putting a sheet below the xformer won't do anything.

Sometimes it does. If the chassis is steel and the core sits on the chassis, or very close, leakage inductance can be propagated through the chassis. Lifting it can reduce this quite a bit - like as gapping a transformer.

I agree that rotating the power transformer can make a big difference. Easy to try both at the same time. By the way, if the power transformer is coupling to the OPT's the transformer closest with be much more affected - one easy diagnostic.

Richard is right too, in that experimentation is the easiest way to solve the problem. I have found it most productive to just listen while things are carefully moved around. An easy test is to take a plate of steel 3 or more mm thick and place it between components on top of the chassis while listening. Thick aluminum works for this too (yes I know it's non-magnetic, but try it, it works).

Sheldon