DC filter

The diodes do not turn on because with the presence of AC line voltage the capacitors are effectively short circuits... i suspect that an offset over 2V would probably turn on the diodes, but rarely are there offsets this big. If for some reason the offset is quite large, just string extra diodes in series.


vbd said:

:confused:
Let me think twice :
First, let's us assume DC only :
- as soon as the DC voltage on the upper mains terminal is greater than two diode drops (say 2 V with big rectifiers), D1 and D4 will conduct and bypass the capacitors...
- If the DC is reverse, i.e. negative on the upper terminal, D2 and D4 will be direct polarized and then bypass again the capacitors...
Then AC only : If I consider the bottom terminal as a reference, the voltage at the top terminal will change from +325V to -325V peak, assuming a 230V mains (european practice...). The diodes will be polarized as soon as the voltage across them is greater than two diode drops, but now they are shunted by the impedance of the capacitor. If the capacitor is big enough, this low impedance will keep the diodes from conducting...
Now let's us consider DC + AC :
-The AC component will pass through the capacitors, if they are large enough, but the DC component will be blocked by them, as they have a very high impedance and will take the less impedent path at zero frequency, through the diodes, except the first 2 V...
Maybe I missed a point and I would be happy that someone shows me where... A spice simulation showed the same blindness as me...
I fully agree with you that the diodes can be non-conducting under normal circumstances, i.e. no DC, but DC seems not to be blocked !
 
I'm with you Randy; I was thinking also that the DC offset to be removed is less than a couple of volts. I think I've read that it doesn't take much very DC at all to get the toroids to "sing".

At this point, I'd like to read what the original designers of this circuit had in mind. Can any folks in Europe chase down the magazine this came from? As I stated in the schematic I posted, I first saw mention of this approach in a netnews posting in the rec.high-end.audio group in '94 (back in the good ole days, they used to have good stuff and discussions in some of the netnews group; the signal to nosie ratio was much better in my opinion :rolleyes: ). Anyway, the poster also mentioned a "Danish HiFi magazine".

Your mission, should you choose to accept it ..... :)

Michael
 
I must admit that if I met the "singing toroid" problem, I had no idea of the DC values that could cause this problem... I thought that a quite high DC value, say several volts, was in cause... a value that this circuit could not suppress !
If you say that only 1 to 2V are enough, and I do believe you ;), the diodes will then suffice, with additional diodes if needed, as suggested by R. McAnally... and I fully agree for the effect of the circuit on the first 2 volts !
 
This is true, I just want to clarify a little. The diodes also protect the relatively low voltage caps from forward voltage too. If the current draw is high (such as power up) the caps may exhibit a large voltage drop, possibly much higher than the 16v rating suggested. The diodes are there mainly to protect the capacitors and nothing else.

Maybe using back-to-back zeners would be a good idea since they can be selected to turn on at various voltages and possibly block more DC. Correct me if I'm wrong but maybe this could work
 
The one and only
Joined 2001
Paid Member
They had better be some pretty big Zeners :)

I use Thermistors in series with my AC line, and I find
that they are fairly good at suppressing the DC hum
problem by limiting the current through the primary,
which is usually 1/2 ohm or less. The Thermistor settles
into a value of an ohm or two, and the saturation current
goes way down.

If you don't like the resistance of the Thermistor, then
you can bypass it with electrolytics, but for Class A amps
you simply figure the Thermistor drop into the equations,
as the draw is pretty constant.
 
Hi all,

finished my DC- filter yesterday, it didn´t explode nor did it smell funny when I connected it to the mains. (Never trust something that works at once!)
Now you can all imagine I´m very anxious to test it but like I mentioned yesterday my power amp has given up it´s humming, it´s as quiet as it has never been before! So I´ll guess I´ll have to wait untill some neighbour turn on this specific appliance again......:cool:

I´ll let you hear when this happens......... and when it never happens, it still was worth the work.

William
 
re: Nelson's last post

Hmmm....

I use a high power, low value resistor (shorted by a relay after 7 seconds) as my soft-start circuit.

I've noticed that when the powerline conditions are such that my transformer (big avel linberg 1200VA) buzzes, it doesn't buzz until the soft-start resistor is shorted.

I may play with Nelson's thermistor idea to see what happens.

Stay tuned ....

Michael
 
michael,


same for me. I use a 110 ohm resistor (2 x 220/ 25 watt) for about 5 seconds to start the amp softly (2 x 500VA) and also note that the buzzing starts after the shorting of the resistor.

A thermistor would be a much simpler solution but I don´t know what the extra resistance will do to an amp that´s more class AB than A.

william
 
seems to be a lot of talk regards to the diodes...well if u had any more than a few tens mV DC u'd sure notice it. The transformers in our systems and most other appliances hooked up around the house present a very low impedance to DC. Only a few mV dc offset will create currents in the ampere range on really large trannies. So the diodes actually do block DC because there should not be high DC voltages present. They also ensure the voltage over the caps are never over a volt or so. (That's why the caps are only rated 25V.)
 
I'm still wondering why the electrolytic capacitors don't blow in the DC-filter as suggested. Depending on which AC mains wire is live and which is zero there must always be one capacitor installed in the wrong direction, that means it is polarized wrong!
And my experience with wrong polarized capacitors is a big bang and an awful smell.
So if anyone can explain it to me...
 
Since the caps are basically shorts with AC, there is no reverse voltage under normal situations. Doesn't matter which lead is "Hot" or not, the caps don't know or even care for that matter :cool:... read the previous posts which explain all of this many times.

leroy said:
I'm still wondering why the electrolytic capacitors don't blow in the DC-filter as suggested. Depending on which AC mains wire is live and which is zero there must always be one capacitor installed in the wrong direction, that means it is polarized wrong!
And my experience with wrong polarized capacitors is a big bang and an awful smell.
So if anyone can explain it to me...
 
Re: Re: one version of DC blocking circuit ....

vbd said:

Hi,
I' m afraid this circuit has a DC path ACROSS the capacitors via the diodes and will block no DC at all... The surge & HF supression part will probably be the only cause of improvement you can expect from it !

If you would expect several volts of DC on the AC line cord, your reply would be correct.

But the DC that you can expect on the AC line cord is on the order of millivolts up to say a couple of hundred millivolts (extreme case). That's all a modern torodial transformer need to 'saturate' and that's when it starts to hum). Theese are the DC levels that the capacitor is supposed to block and at theese levels the diods are far from active. The diods are there to protect the capacitor.

:)
 
Just built two of them...

...and they work. My toroids are finally quiet! Haven't had time for a serious listen yet, but first impression is that the sound is just as good as before, just more stillness when not playing music.

(I used two 35V/10000µF caps back-to-back on only one line, and not a single cap on both lines, which basically does the same thing, but saves two diodes.)
 
DC filter for mains supply

Back to basics
A mains transformer is an inductive device, and couples the input to output via magetic fields - no voltage is induced in the secondary if the there is no change over time - the formaula dv/dt applies

Therefore, mains DC (if present) cannot be transmitted - it is far more likely to be either harmonics or earth return hum

Coupling mains via electrolytic capacitors is as likely to induce earth leakage and exasperate the problem

Roland
 
Member
Joined 2002
Paid Member
Plitron transformers has written a white paper on transformer noise and it's causes. It may be on their web site.

This mechanical noise is related to many factors such as flux densitiy, core type and winding of the core bands and the copper windings being exactly balanced. AC quaility has been getting bad all over the world lately. You can actually see it on an AP THD+N measurement. Harmonics are present DC offset and even power company transmitted signals.