Re: I wouldn't try the circuit found/posted by Stabist....
Hmm - I know I have problems with my english - but - aren't on "my" picture two caps in series and each has one diode in paralel - and they are connected in opposite way?? And I think that's what it says in the text by audionutter ...
BUT OFCOURSE - I'M NOT THE AUTHOR of the drawing and have not try it out in praxis ...
audionutter said:
Hmm - I know I have problems with my english - but - aren't on "my" picture two caps in series and each has one diode in paralel - and they are connected in opposite way?? And I think that's what it says in the text by audionutter ...
BUT OFCOURSE - I'M NOT THE AUTHOR of the drawing and have not try it out in praxis ...
If a transformer hums it is due to that either some loops of the winding has come loose or worse that the core is not tightly wound. Since there must be a vibration caused by alternating currents, I have a little hard to understand why blocking DC can cure. The only cure to my knowledge is to vacuum impregnate thetransformer. Simply take the transformer out and go to a shop where they rewind electric motors take away the outer plastic wrapping and let the good guys treat it with their magic laquer.
The transformer will be as quiet as a sleeping rock. If you continue to let it hum, there is a risk of shorting a few turns this will lead to heat build up and further deterioration.
If you don't have shop nearby you can use any low viscosity epoxi and pot the transformer yourself. But that is messy but with slow hardening and low temperatures the result will be decent. Vacuum is neccessary to get into the underlaying windings. I vaccuumed some high voltage transformers and every cubicmillimetre was filled with epoxi.
Andreas W
The transformer will be as quiet as a sleeping rock. If you continue to let it hum, there is a risk of shorting a few turns this will lead to heat build up and further deterioration.
If you don't have shop nearby you can use any low viscosity epoxi and pot the transformer yourself. But that is messy but with slow hardening and low temperatures the result will be decent. Vacuum is neccessary to get into the underlaying windings. I vaccuumed some high voltage transformers and every cubicmillimetre was filled with epoxi.
Andreas W
The schematic was wrong.
You need a bipolar cap or two back to back polar caps. The pair of caps then have a bypass using a single or double ( same direction) diode to let excess voltage pass the caps & then you must bypass the caps in the other direction for reverse current flow. Do not connect the junction between the caps to the diodes!!!
The second harmonic content in the mains supply causes vibration in windings plates or even the whole transformer or what it is connected to. The second harmonic can be roughly compared to a true sine wave plus a small dc offset. The caps remove the dc offset and make the distorted sine wave look more like a pure sine wave. There's lots of other harmonics and spikes & every other crap in the mains supply but short of regenerating that with your own alternator you stuck with it.
You need a bipolar cap or two back to back polar caps. The pair of caps then have a bypass using a single or double ( same direction) diode to let excess voltage pass the caps & then you must bypass the caps in the other direction for reverse current flow. Do not connect the junction between the caps to the diodes!!!
The second harmonic content in the mains supply causes vibration in windings plates or even the whole transformer or what it is connected to. The second harmonic can be roughly compared to a true sine wave plus a small dc offset. The caps remove the dc offset and make the distorted sine wave look more like a pure sine wave. There's lots of other harmonics and spikes & every other crap in the mains supply but short of regenerating that with your own alternator you stuck with it.
Hi everyone,
Thanks for all the input. I have tried both schematics that have been mentioned here by Stabist and Byteboy. The first one I tried was the same one as Byteboy had posted (which I found in the earlier thread link that he has also posted) but that didn't seem to make any difference to the noise. I wondered if somehow the diodes are defeating the object/purpose of the filter.
Next I had tried the second one that Stabist had posted (which I also found on DIYparadise) and this did work. I also had reservations about the caps not having a failsafe but I couldn't work out how to filter DC with a failsafe in place. What I also found was the effectiveness of the filter does vary, probably depending on the amount of DC current present within the AC lines. Cos when I plugged it into one of the ring circuit which also includes the fridge, computer, etc the filter's effect was almost nil whilst when I plugged it into another socket the effect was noticeable. By noticeable I mean it reduces it by a measurable level using an SPL meter.
The power amp that gave me the humming/buzzing noise is a spare power amp, so I have not used it much. What I found very effective was when I connected the amps up to an isolation transformer which I was using for something else, that killed the hum/buzz completely. What I may try next is to epoxy the toroidal transformer windings as has been suggested by Andreas W cos this will stop the buzz/hum even if any DC is present. Then that way I won't have to worry about a failsafe for the capacitor in my existing filter.
Thanks for all the input. I have tried both schematics that have been mentioned here by Stabist and Byteboy. The first one I tried was the same one as Byteboy had posted (which I found in the earlier thread link that he has also posted) but that didn't seem to make any difference to the noise. I wondered if somehow the diodes are defeating the object/purpose of the filter.
Next I had tried the second one that Stabist had posted (which I also found on DIYparadise) and this did work. I also had reservations about the caps not having a failsafe but I couldn't work out how to filter DC with a failsafe in place. What I also found was the effectiveness of the filter does vary, probably depending on the amount of DC current present within the AC lines. Cos when I plugged it into one of the ring circuit which also includes the fridge, computer, etc the filter's effect was almost nil whilst when I plugged it into another socket the effect was noticeable. By noticeable I mean it reduces it by a measurable level using an SPL meter.
The power amp that gave me the humming/buzzing noise is a spare power amp, so I have not used it much. What I found very effective was when I connected the amps up to an isolation transformer which I was using for something else, that killed the hum/buzz completely. What I may try next is to epoxy the toroidal transformer windings as has been suggested by Andreas W cos this will stop the buzz/hum even if any DC is present. Then that way I won't have to worry about a failsafe for the capacitor in my existing filter.
stabist, let us know how you get on, OK?
byteboy, believe me, I was worried about the 240V AC too! Unfortuantely I do not have the necessary stuff/voltmeter to measure the voltage drop (if any) across the caps. What I did do is to have a listen to the amps after I had connected the filter up. They sound the same to me, but I had my reservations as to whether the caps I used (10,000 microF, 63V) is sufficient or not for the purposes of a power amp that has a 5amp fuse rating.....anyone here that can work out what is the minimum capacitor value sufficient for this?
byteboy, believe me, I was worried about the 240V AC too! Unfortuantely I do not have the necessary stuff/voltmeter to measure the voltage drop (if any) across the caps. What I did do is to have a listen to the amps after I had connected the filter up. They sound the same to me, but I had my reservations as to whether the caps I used (10,000 microF, 63V) is sufficient or not for the purposes of a power amp that has a 5amp fuse rating.....anyone here that can work out what is the minimum capacitor value sufficient for this?
Audionutter:
I was asking because I wanted you to verify that the voltage over the caps will be higher than the intended limmiting function by the diodes of max. 0.7 Volts.
I think the circuit you are using will block DC but is IMHO wrong and potentially dangerous.
The diodes can be considderd not present as far as their application for the protection of the caps is concerned.
This means you are probably using those caps at a higher voltage than their intended working voltage.
This can cause the caps to explode, so be aware!
I think the capacitance value is not so much a problem: at 5000uF (2x 10.000 uF electrolytes in series) at 50/60 Hz it would give an impedance of abt. 0.5 - 0.6 Ohms,[ Xc = 1/(2 pi*f*C) ] not too much but could be lower.
High ripple current capability would be the thing to look for.
But the working voltage (without the protection of the diodes) would have to be 240 X 1.41 = 340 Volts minimal, better use 400 Volts.
Two capacitors of 10.000 uF/400 Volts will be realy expensive!
That is why the diodes were introduced in the first place: To get away with using a (smaller and much cheaper) lower voltage capacitor!
Isolation transformer
The use of an isolation transformer is a good solution to get rid of a lot of interferences (like mentioned by Andrew T) from the powerline, not just DC.
Especially if you use one with an earthed screen between the primary and secondary windings or the type that uses two different coils on two different legs (U/I core) to prevent capacitive coupling of the primairy and secondary windings and so the passing thru of interferences from the power line.
In case of DC it is of course the isolation transformer that will be humming now! ;-)
But this can be stowed away in a closet away from the listening room/position!
Andreas W:
I think you will have gathered that we were not discussing hum introduced by mechanical defects/shortcommings of a transformer but because of a DC component present on the AC power line.
But good info on getting rid of these problems, especially the tip to go a motor shop to get transformer coils impregnated!
I was asking because I wanted you to verify that the voltage over the caps will be higher than the intended limmiting function by the diodes of max. 0.7 Volts.
I think the circuit you are using will block DC but is IMHO wrong and potentially dangerous.
The diodes can be considderd not present as far as their application for the protection of the caps is concerned.
This means you are probably using those caps at a higher voltage than their intended working voltage.
This can cause the caps to explode, so be aware!
I think the capacitance value is not so much a problem: at 5000uF (2x 10.000 uF electrolytes in series) at 50/60 Hz it would give an impedance of abt. 0.5 - 0.6 Ohms,[ Xc = 1/(2 pi*f*C) ] not too much but could be lower.
High ripple current capability would be the thing to look for.
But the working voltage (without the protection of the diodes) would have to be 240 X 1.41 = 340 Volts minimal, better use 400 Volts.
Two capacitors of 10.000 uF/400 Volts will be realy expensive!
That is why the diodes were introduced in the first place: To get away with using a (smaller and much cheaper) lower voltage capacitor!
Isolation transformer
The use of an isolation transformer is a good solution to get rid of a lot of interferences (like mentioned by Andrew T) from the powerline, not just DC.
Especially if you use one with an earthed screen between the primary and secondary windings or the type that uses two different coils on two different legs (U/I core) to prevent capacitive coupling of the primairy and secondary windings and so the passing thru of interferences from the power line.
In case of DC it is of course the isolation transformer that will be humming now! ;-)
But this can be stowed away in a closet away from the listening room/position!
Andreas W:
I think you will have gathered that we were not discussing hum introduced by mechanical defects/shortcommings of a transformer but because of a DC component present on the AC power line.
But good info on getting rid of these problems, especially the tip to go a motor shop to get transformer coils impregnated!
dang old thread but i was reading it and like Andreas W was actually under the impression that the problem was a mechanical or transformer defect. I thought the thread was misdirected to a DC issue. Which makes me have to ask... Audionutter, was the xformer making an audible sound with no speakers plugged in? The xformer was literally acting like a speaker itself correct? Did you find the problem?
7 years later ...
hi everyone !
i just want to thank you all, because i was in a big trouble with my studio monitors (Focal CMS65), whose internal toroidal transformers made lot of noise, making them excellent monitors for mixing or listening, but just annoying when idle ...
I've used schematic posted by byteboy in post#15 :
My monitors transformers are now perfectly silent
So thanks a lot guys !
hi everyone !
i just want to thank you all, because i was in a big trouble with my studio monitors (Focal CMS65), whose internal toroidal transformers made lot of noise, making them excellent monitors for mixing or listening, but just annoying when idle ...
I've used schematic posted by byteboy in post#15 :
An externally hosted image should be here but it was not working when we last tested it.
My monitors transformers are now perfectly silent
So thanks a lot guys !
see if you can safely measure the voltage drop across the diodes.
If the diodes are starting to turn on then the capacitors are too small.
The turn on voltage of a cold rectifier diode is a bit less than 500mVpk
If you are measuring AC voltage, then you should be less than ~350mVac
You can't use an oscilloscope, because the lines are at mains voltage.
If the diodes are starting to turn on then the capacitors are too small.
The turn on voltage of a cold rectifier diode is a bit less than 500mVpk
If you are measuring AC voltage, then you should be less than ~350mVac
You can't use an oscilloscope, because the lines are at mains voltage.
Another thing that will make a transformer buzz is an overload.
Could be your amp ha partially died and is loading the mains transformer too much or asymmetrically.
I would check your amps output transistors for failure.
Or power supply smoothing caps.
If the amp is old then they maybe need changing.
Could be your amp ha partially died and is loading the mains transformer too much or asymmetrically.
I would check your amps output transistors for failure.
Or power supply smoothing caps.
If the amp is old then they maybe need changing.
That looks good.
183mVac is a conversion to an equivalent sinewave where the peak voltage is sqrt(2)*Vac, i.e. ~259mVpk
However the current is not sinusoidal, when feeding a capacitor input filter. It is is spiky, pulses that can be anywhere from 2times the average to 10times the average.
Can anyone explain how this works? It seems to me that the capacitors are passing the mains voltage/current and so will also be passing any higher frequencies and that the diodes will only be blocking any d.c. which is less than the forward bias on two diodes in series - so less than 1.2V roughly.
So what is happening - anyone?
So what is happening - anyone?
There are a few things to watch for:
1. - Don't mount the toroidal directly to a metal chassis, it will hum,
rather lift it up by about 15mm by fitting a wood or plastic spacer under it.
2. - Don't use the mounting bolt down the centre to fasten it to the chassis either, as dangerous eddy currents can flow - I mount mine on a wooden spacer with foam weather strip glued on. I then tie the lot down with cable ties.
3. - Design your power supply with an earth lift resistor to ground - 1k 2w.
4. - To stop the diodes from causing spikes, put a resistor between the transformer and the rectifier. In a transistor power amp the value of that resistor may be 0,1 Ohm - 5 Watt. With tube power amps it could be around 1 Ohm and in preamps (transistor and tubes) some 10 Ohms - I fitted INRUSH CURRENT LIMITING THERMISTORs here, a) to act as slow turn on for the transformer, stops that massive current draw as the caps charge up and
b) as a resistance for the diodes - definitely sounds better.
5. - Put 100n caps across each diode, ala John Lynsley-Hood style.
6. - Take all speaker earths directly to the earth on the power supply board.
7. - Fit a snubber resistor & cap across the transformer live side.
8. - Fit an EMI filter on the mains inlet to your equipment.
http://novelbooks.weebly.com/uploads/1/0/3/5/10358893/9951887_orig.gif
1. - Don't mount the toroidal directly to a metal chassis, it will hum,
rather lift it up by about 15mm by fitting a wood or plastic spacer under it.
2. - Don't use the mounting bolt down the centre to fasten it to the chassis either, as dangerous eddy currents can flow - I mount mine on a wooden spacer with foam weather strip glued on. I then tie the lot down with cable ties.
3. - Design your power supply with an earth lift resistor to ground - 1k 2w.
4. - To stop the diodes from causing spikes, put a resistor between the transformer and the rectifier. In a transistor power amp the value of that resistor may be 0,1 Ohm - 5 Watt. With tube power amps it could be around 1 Ohm and in preamps (transistor and tubes) some 10 Ohms - I fitted INRUSH CURRENT LIMITING THERMISTORs here, a) to act as slow turn on for the transformer, stops that massive current draw as the caps charge up and
b) as a resistance for the diodes - definitely sounds better.
5. - Put 100n caps across each diode, ala John Lynsley-Hood style.
6. - Take all speaker earths directly to the earth on the power supply board.
7. - Fit a snubber resistor & cap across the transformer live side.
8. - Fit an EMI filter on the mains inlet to your equipment.
http://novelbooks.weebly.com/uploads/1/0/3/5/10358893/9951887_orig.gif
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.