• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

how to get rid of transformer hum

Status
Not open for further replies.
ok then, its a buzz, not a hom. Coming out of the speakers. I can hear the main transformer buzzing as well, 220W cast in toroid (mechanically buzzing that is). Even when there is no load. It went quiet, when I put it on the floor, but it buzzes when its on a metal plate. Even when there is nothing connected to the outputs. All I found is DC on the AC line
http://www.diyaudio.com/forums/solid-state/39895-transformer-making-noise-how-cure-3.html

Here is some PCB board on ebay, there is a schematic with it too (image attached). I guess its worth it (the ebay thing), there was a humbuster 3 that did the same thing for USD299 and same ingredients.

http://www.ebay.com.au/itm/PCB-for-...251217?hash=item1ead0bcf51:g:VecAAOxydgZTIInx

would a shield help? I tried earlier, put a metal beaker over it, did not help.
It draws 1.7 W or about 3 VA on idle, open outlets
 

Attachments

Last edited:

spec

Well-Known Member
Most Helpful Member
ok then, its a buzz, not a hom. Coming out of the speakers. I can hear the main transformer buzzing as well, 220W cast in toroid (mechanically buzzing that is). Even when there is no load. It went quiet, when I put it on the floor, but it buzzes when its on a metal plate. Even when there is nothing connected to the outputs. All I found is DC on the AC line
http://www.diyaudio.com/forums/solid-state/39895-transformer-making-noise-how-cure-3.html

Here is some PCB board on ebay, there is a schematic with it too (image attached). I guess its worth it (the ebay thing), there was a humbuster 3 that did the same thing for USD299 and same ingredients.

http://www.ebay.com.au/itm/PCB-for-...251217?hash=item1ead0bcf51:g:VecAAOxydgZTIInx

would a shield help? I tried earlier, put a metal beaker over it, did not help.
It draws 1.7 W or about 3 VA on idle, open outlets
Hi CS,

The circuit does not make sense, to me anyway.

You are not by any chance operating a 120V equipment on 240V are you?

Or perhaps the equipment has a voltage adjustment switch/link that is not set to the appropriate position to suit your mains supply voltage.

spec
 

alec_t

Well-Known Member
Most Helpful Member
I can hear the main transformer buzzing as well, 220W cast in toroid (mechanically buzzing that is). Even when there is no load. It went quiet, when I put it on the floor, but it buzzes when its on a metal plate.
Mount it on a heat-proof rubber or cork layer. If there is a mounting bolt passing through the toroid then the bolt must be insulated from metalwork so as to prevent it acting as a shorted turn of the transformer.
 

JimB

Super Moderator
Most Helpful Member
If the local mains supply has a net DC component, this can lead to magnetic saturation of the transformer core and high current peaks on the peaks of the supply voltage.

Also consider that the windings can move slightly due to passing a current in a magnetic field, hence low frequency vibration.

Also there is magnetostriction, where the transformer core changes size slightly as the magnetic field changes.
Look here: http://hyperphysics.phy-astr.gsu.edu/hbase/Solids/magstrict.html

JimB
 

ronsimpson

Well-Known Member
Most Helpful Member
If the local mains supply has a net DC component
Think of all the billions of transformers sitting on the mains, each will short out the DC.
If there is DC it would have to be injected in my house or from house to pole. My house is transformer isolated from the mains.
If there was DC then the core will draw more current (on the + and less on the - of the sign wave{or the reverse}) and thus eat up the DC.
 

JimB

Super Moderator
Most Helpful Member
Think of all the billions of transformers sitting on the mains, each will short out the DC.
I was thinking in terms of some iffy cheap power supplies or some other thing using half wave rectification and hence distorting one particular half cycle.

If the two half cycles, positive and negative, of the AC waveform are not identical, there will be a net DC component.

JimB
 

MrAl

Well-Known Member
Most Helpful Member
ok then, its a buzz, not a hom. Coming out of the speakers. I can hear the main transformer buzzing as well, 220W cast in toroid (mechanically buzzing that is). Even when there is no load. It went quiet, when I put it on the floor, but it buzzes when its on a metal plate. Even when there is nothing connected to the outputs. All I found is DC on the AC line
http://www.diyaudio.com/forums/solid-state/39895-transformer-making-noise-how-cure-3.html

Here is some PCB board on ebay, there is a schematic with it too (image attached). I guess its worth it (the ebay thing), there was a humbuster 3 that did the same thing for USD299 and same ingredients.

http://www.ebay.com.au/itm/PCB-for-...251217?hash=item1ead0bcf51:g:VecAAOxydgZTIInx

would a shield help? I tried earlier, put a metal beaker over it, did not help.
It draws 1.7 W or about 3 VA on idle, open outlets
Hi

"it buzzes when its on a metal plate"

Den dont do dat :)

Seriously, why put it on a metal plate?
If the metal contains iron it will be pulled by the magnetic field and will probably vibrate at 120Hz with a 60Hz line. It may not be much but it might be enough to make a buzz. So dont put it near anything metal if that metal contains some iron.

Hum sometimes couples into audio circuits so you've got to find out where it enters the system if that happens too. Some of the old tube amps used to have an adjustment that you could turn to reduce the hum. The hum often comes from the line frequency coupling into the input of the amp.
 
its on a metal plate, since the amplifier sits in a metal housing. The mechanical hum of the transformer is not an issue, have to be pretty close to hear it. Its a solid sate amplifier I built from scratch, so no tubes and it always was 240V. The buzz couples into the audio circuit. Since I guess the whole housing gets some current induced. The transformer sits on some rubber disks. Its cast in, in a plastic case. So the trough bolt non conductive would help?
For trying how far I have to move the transformer, I first put it straight onto a metal plate and then put some non metal things underneath to lift it up. That will not help much as the transformer will move closer to the lid, once in an enclosure
Will changing the angle help, its a circular transformer?
Moving the transformer into a separate housing and feed DC into the amps would be one idea. But then may as well start again, use switch mode power supplies.
ok, bought Gauss meter
for the schematic, thats what other threads say, capacitors and diosed anti parallel. There may be DC bits, with all the inverters, dimmers and switch mode power supplies. I may as well use some in my amps, less trouble
 
Last edited:

dr pepper

Well-Known Member
Most Helpful Member
Torroids usually come with 2 rubber washers to insulate vibrations, if you dont have these then hum might well be an issue.
The windings are usually tied real tight with some fibreglass string then varnished & baked, if this wasnt done right that will cause a buzz, not a low hum more of a modulated 50 or 60 hz zzzzz.
Also if anything conductive goes around the transformer and through the centre that will be a shorted turn, if you connected the clamp washer to the chassis somehow this would be the same, small voltages at very large current would flow.
 
no, just a through bolt in the centre and nothing metal around. Sits on 4 rubber washers actially. And some V shaped metal washer inbetween. Maybe get rid of this one?
See when the gauss meter arrives.
I remember a while ago, I put some metal cage on top, and also lifted the transformer up a bit, still the same buzz coming out od the speakers. Have to try again. Will buy some 5W resistors today as load to make it more audible for testing.
 
Last edited:

spec

Well-Known Member
Most Helpful Member
no, just a through bolt in the centre and nothing around. See when the gauss meter arrives.
I remember a while ago, I put some metal cage on top, and also lifted the transformer up a bit, still the same buzz coming out od the speakers. Have to try again. Will buy some 5W resistors today as load to make it more audible for testing.
Is your amplifier class D?

spec
 

dr pepper

Well-Known Member
Most Helpful Member
Maybe I misunderstood here.
If you are getting buzz from the speakers then that is either just mains hum due to unsufficient smoothing, magnetically induced currents in wiring/components being amplified, or the dreaded earth loop problem that can be a bit of a git to get rid of.
I was under the impression the noise was physically comming from the trans itself.
Moving the trans if poss might take that issue away, or putting a ally shield over it, if you did you need to make sure theres no electrical circuit from one side to the other as I mentioned before.
Your gauss meter will show some fields, these will be leakage fields and will probably be difficult to locate, torroids being a closed magnetic circuit with no gaps leak very little, but still maybe enough to cause hum if it gets into the i/p side of the amp.
Does the amp use a star earth layout, and is the driver board seperate to the o/p devices.
 

spec

Well-Known Member
Most Helpful Member
Maybe I misunderstood here.
If you are getting buzz from the speakers then that is either just mains hum due to unsufficient smoothing, magnetically induced currents in wiring/components being amplified, or the dreaded earth loop problem that can be a bit of a git to get rid of.
I was under the impression the noise was physically comming from the trans itself.
Moving the trans if poss might take that issue away, or putting a ally shield over it, if you did you need to make sure theres no electrical circuit from one side to the other as I mentioned before.
Your gauss meter will show some fields, these will be leakage fields and will probably be difficult to locate, torroids being a closed magnetic circuit with no gaps leak very little, but still maybe enough to cause hum if it gets into the i/p side of the amp.
Does the amp use a star earth layout, and is the driver board seperate to the o/p devices.
Or is there both a physical noise from the transformer and hum from the speakers.

Also what is the frequency of the hum:
(1) Mains frequency
(2) Mains frequency x 2 (ripple from full wave rectifier)
(3) Mains frequency x 3 (quadrature)

I would be suspicious of the reservoir capacitors.

spec
 
class AB
mechanical noise is minimum, if I put thr trafo on the floor, its quiet. Set the trafo on some sheet metal, it starts making a buzzing noise. So I thought as simple test, when the trafo is far enough from the sheet metal, so it stops buzzing, then it would not induce that much into the housing either and subsequently into the housing and the circuits.
When I disconnect thr trafo and frrd the amp via a lab power supply, the speakers have a small pure 50 Hz hum. If I leave the trafo indide the housing and connet to 240V but not to the rectifiers/caps. Feed in from lab supply. Then there is a buzz in the speakers, the kind of sound like a step down transformer nearby. Mostly high frequency. Now I got 2 way speakers connected, it tends to come from the high frequency end. Rectifier is not connected, just to get that one eliminated as source.
I was hoping, eliminating the pysical noise (the housing stops vibrating, since it gets less magnetic field) would also eliminate the noise coming from the speakers.
Crossover is a separate board, disconnected. Has seperate power supply, disconnected.
Its starground.
I cannot see anything on the scope, but its annoying, can hear it from 2m away
 

dr pepper

Well-Known Member
Most Helpful Member
Ok I get that now.
Sounds like the noise is being induced into the amp magnetically, or possibly capacitively.
Magnetic coupling is where a conductor is acting like part of a secondary winding and the resultant hum gets amplified.
If there is a lot of hf noise on the mains line then maybe its capacitively coupled, which is unusual and would most likely require very high impedances.
What is the sheet metal, aluminium would be a good choice.
Does the trafo have a screen?, usually identified by an earth wire going into the windings.
I was thinking stray inductance/capacitance ringing was causing the issue in conjunction with the rectifiers switching, but you eliminated that one allready, and you eliminated ground loops.
You say the xover is disconnected, is the amp i/p open or did you terminate it?, if open then this probably would happen.
The only thing I can think of other than magnetically shileding the transformer which would be extreme is to RC snubber the primary of the trafo in case a smps or noisy mains device nearby is injecting ac mains modulated hf noise into the primary which is then capacitively coupled to the secondary and radiated (this can be a problem in smps transformers), have you tried powering the amp in a different room?
 
the bottom and top plate is steel, rest is alu
transformer has no shield it seems
the noise stops, when I set a jumper to 0V on the amp inlets
make a runner in 1 hr, get some power resistors
 

Attachments

spec

Well-Known Member
Most Helpful Member
class AB
mechanical noise is minimum, if I put thr trafo on the floor, its quiet. Set the trafo on some sheet metal, it starts making a buzzing noise. So I thought as simple test, when the trafo is far enough from the sheet metal, so it stops buzzing, then it would not induce that much into the housing either and subsequently into the housing and the circuits.
When I disconnect thr trafo and frrd the amp via a lab power supply, the speakers have a small pure 50 Hz hum. If I leave the trafo indide the housing and connet to 240V but not to the rectifiers/caps. Feed in from lab supply. Then there is a buzz in the speakers, the kind of sound like a step down transformer nearby. Mostly high frequency. Now I got 2 way speakers connected, it tends to come from the high frequency end. Rectifier is not connected, just to get that one eliminated as source.
I was hoping, eliminating the pysical noise (the housing stops vibrating, since it gets less magnetic field) would also eliminate the noise coming from the speakers.
Crossover is a separate board, disconnected. Has seperate power supply, disconnected.
Its starground.
I cannot see anything on the scope, but its annoying, can hear it from 2m away
What type of scope are you using? traditional or digital? What is the bandwidth of your scope?

What is your amplifier?

What are your speakers?

How long is your speaker wire?

Is the hum coming from both speakers?

Here are a few suggestions to investigate:

(1) Disconnect both speaker leads at the amplifier.

If the amplifier has a headphone socket can you hear the hum on headphones?

(2) Disconnect both speaker leads from the amplifier. Get a separate speaker chassis, and connect a 10 Ohm resistor or thereabouts, in series with the speaker chassis.
Connect short leads to the speaker chassis and connect the speaker chassis to left and right amplifier outputs in turn. Is the hum still there.

spec
 
the photo is the amplifier. Its TDA7293 monoblocks. And before that analogue crossover. So each of the monoblocks generates the buzz. No headphone socket. Tried with individual drivers and loudspeakers. Its indipendent from the loudspeaker cable length, now its 2m, before it was 6 or so.
Got some 100 Ohm resistors, connected the secondary windings of the transformer. Draws about 1.6W now, its 240 to 22V transformer. Buzz did not get worse than secondary windings open.
Have to dig,think somewhere is a 10 ohm resistor, can try this.
 
Status
Not open for further replies.

EE World Online Articles

Loading
Top