thank you for this advice. "photobucket" and "imageshack" are not helpful without high speed web access like in my village.
Interesting observations. The reason therefore are clearly find out by measuring voltage and current flow at the same time like the diagrams from post 39 here on this thread - go to
http://www.diyaudio.com/forums/powe...buzzing-toroid-transformers-what-right-4.html
Last edited:
is this the part I was not reading/understanding?
If the core is acting as the integrator, then it is that integrating function of the core that ADDS up the imbalance of the wave to wave flux until the reactive impedance of that flux can't resist the primary voltage as well and excess current flows around the primary.
Does that correctly explain it?
BMW850 prefer the circuit from post 159 about this thread:
http://www.diyaudio.com/forums/solid-state/2080-dc-filter-16.html
the circuit description is here on his own website:
DC-Blocker - Scintilla-buizenversterkers
This circuit isn't to realize with a bridge rectifier, only with single diodes. Each electrolytic capacitor get its own diode.
What are the advantages in opposite to the most used topology as shown in Rod Elliott's schematic in fig. 3 ?? - go to
Mains DC and Transformers
Thank you for advices.
P.S. concerning DC filter here and on other portals are a lot of threads. To find the most interesting posts, go to
http://www.diyaudio.com/forums/soli...oidals-summary-most-interesting-comments.html
Last edited:
Yes, and because you can build a high-pass filter with a capacitor and a resistor, OR with a resistor and an inductor, you can use the primary/core itself as its own DC-reject element, but you have to protect it at the same time, which calls for non-linear elements as the R part.
The properties of the core are especially valuable in this role: it isn't simply a linear inductor, it is a swinging inductor that will reduce its inductance (and raise the corner frequency) when the DC component increases.
It thus makes an adaptative filter, and unlike the C-type, it doesn't have resonant frequencies between C and the magnetizing or leakage inductances.
If it is associated with the C-filter, it will damp resonances if they become too large in amplitude
Hello, Ive just read the whole of this thread as i too have a humming torroid.
I have just run some experiments using a signal generator and the ESP circuit and hooked up a transformer so theres a load. Without the load (yes, i tried this first....) it didnt work at all. The output mirrored the input. This was all low voltage/low current as i wanted to have a play and see what was happening. I dont think this is an answer but may shed some light on the diode only circuit. Apologies to those who already know all about this and if it seems obvious. Ive found something i didnt expect and thought id share it....
With the capacitors the circuit works beautifully. Takes a few seconds to stabilise but was able to remove a large dc component. (as much of an offset that my el cheapo signal generator could give before the signal clips on its rails).
Without the caps the diodes did distort the output sine wave, gradually improving until the offset became about more than 1.1v or so at which point the output was clean and offset free.
At 1v pkpk input the distortion was massive but at 5v there was essentially a sine wave although the peaks were distorted but was proportionately less. I wonder if at full mains swing this distortion (in the eyes of crown) is considered negligable and reducing possible dc offset is considered preferable. Would this distortion be filtered out in the dc rectification smoothing caps?
I tried adding the second set of diodes and it reduced the distortion a tiny bit but nothing to write home about.
I have just run some experiments using a signal generator and the ESP circuit and hooked up a transformer so theres a load. Without the load (yes, i tried this first....) it didnt work at all. The output mirrored the input. This was all low voltage/low current as i wanted to have a play and see what was happening. I dont think this is an answer but may shed some light on the diode only circuit. Apologies to those who already know all about this and if it seems obvious. Ive found something i didnt expect and thought id share it....
With the capacitors the circuit works beautifully. Takes a few seconds to stabilise but was able to remove a large dc component. (as much of an offset that my el cheapo signal generator could give before the signal clips on its rails).
Without the caps the diodes did distort the output sine wave, gradually improving until the offset became about more than 1.1v or so at which point the output was clean and offset free.
At 1v pkpk input the distortion was massive but at 5v there was essentially a sine wave although the peaks were distorted but was proportionately less. I wonder if at full mains swing this distortion (in the eyes of crown) is considered negligable and reducing possible dc offset is considered preferable. Would this distortion be filtered out in the dc rectification smoothing caps?
I tried adding the second set of diodes and it reduced the distortion a tiny bit but nothing to write home about.
Best way to avoid hum is the use of the right transformer like Plitron - go to
Measuring Acoustic Noise Emitted by Power Transformers
But in the case of existing amp devices this is often too expensive.
There are a lot of threads to this topic - go to
DC Filter against humming Toroidals - Summary of the most interesting Comments
Post #159 under
dc filter
is the royal way from my view, as long as parts used which are adequately dimensioned.
AndrewT (post #161) don't think so.
Maybe this device is still better (go also to the attachment):
http://www.horch-gmbh.de/Produkte/Line Silencer.pdf
What is currently the best solution for implementing a DC filter in exist amps with humming toroidal transformer ?
Measuring Acoustic Noise Emitted by Power Transformers
But in the case of existing amp devices this is often too expensive.
There are a lot of threads to this topic - go to
DC Filter against humming Toroidals - Summary of the most interesting Comments
Post #159 under
dc filter
is the royal way from my view, as long as parts used which are adequately dimensioned.
AndrewT (post #161) don't think so.
Maybe this device is still better (go also to the attachment):
http://www.horch-gmbh.de/Produkte/Line Silencer.pdf
What is currently the best solution for implementing a DC filter in exist amps with humming toroidal transformer ?
Attachments
Last edited:
In this case this interview with the German developer is of interest:
Interview mit einem Audioentwickler (vorsicht lang!!!, 24 kB)
or
Google Groups
As I found out, using a solid state relay to sequence on the amp will cause a huge imbalance and buzz.
)
yes, here are examples:
L C Audio Technology / DC Filter
DC filter — Audio Chews
Combined Module DC Blocker & EMI/RFI/Common Mode Filter - Assembled & tested PCB | eBay
http://www.sedlbauer.de/media/dc-netzfilter-250v-10a_313141_deutsch.pdf
DC-Filter, Gleichspannungs-Filter, Netzfilter, 230V, Hifi, Audio
A major disadvantage of all commercial units, however, is the loss of dynamism while listening tests (mostly by the use of big toroidal transformers with 1KVA and more). The reason are too small electrolytic capacitors with a small cv product (without screw terminals), i. e. very high capacity and very small outline at the same time.
I am still looking for the schematic of the unit in post #61 - go to
Variations of DC Main Filter against buzzing Toroid Transformers - what is the right?
because this approach is very different (without any loss of dynamism).
L C Audio Technology / DC Filter
DC filter — Audio Chews
Combined Module DC Blocker & EMI/RFI/Common Mode Filter - Assembled & tested PCB | eBay
http://www.sedlbauer.de/media/dc-netzfilter-250v-10a_313141_deutsch.pdf
DC-Filter, Gleichspannungs-Filter, Netzfilter, 230V, Hifi, Audio
A major disadvantage of all commercial units, however, is the loss of dynamism while listening tests (mostly by the use of big toroidal transformers with 1KVA and more). The reason are too small electrolytic capacitors with a small cv product (without screw terminals), i. e. very high capacity and very small outline at the same time.
I am still looking for the schematic of the unit in post #61 - go to
Variations of DC Main Filter against buzzing Toroid Transformers - what is the right?
because this approach is very different (without any loss of dynamism).
Last edited:
The caps do the blocking.
The rectifiers just limit the over-voltage on the caps so the do not go BOOM if something goes wrong. Use standard Si parts. Low-threshold parts will spoil the capacitor blocking.
what happens, if I remove the diodes and replace the electrolytic caps by non polar versions ? like those under
file:///C:/Users/admin/Downloads/technical_file_LNK-P5Y-8000-80.pdf
or
file:///C:/Users/admin/Downloads/technical_file_LNK-M3_1-4700-70%20_.pdf
found under
DC Link Power Capacitors Archivi - ICAR
Overview resp. line-up from ICAR:
https://www.amelec.ch/infocenter_icar.html?file=tl_files/amelec/downloads/icar/LNK_2016.pdf
Actually the best solution - isn't it ?
file:///C:/Users/admin/Downloads/technical_file_LNK-P5Y-8000-80.pdf
or
file:///C:/Users/admin/Downloads/technical_file_LNK-M3_1-4700-70%20_.pdf
found under
DC Link Power Capacitors Archivi - ICAR
Overview resp. line-up from ICAR:
https://www.amelec.ch/infocenter_icar.html?file=tl_files/amelec/downloads/icar/LNK_2016.pdf
Actually the best solution - isn't it ?
Dc blocking circuit (developed and implemented in the 60s) works very well and I have been using it for many years, since the 90s. When it comes to "dynamic reduction" i wolud say it depends on what you are looking at. It´s not that simple.
The buzzing sound from a transformer is reducing the s/n ratio quite dramatically and can be annoying. A DC blocker fix that!
If you think that the voltage drop of 1 volt out of 230volt will cause a dynamic reduction electrically i would say no. Unless you have a zero feedback design. Otherwise with a small amount of negative feedback the difference in power supply voltage is not affecting the music signal at all. IF your amplifier is that sensitive for power supply voltage, that a small supply voltage change is noticeable in sound reproduction, simply change amplifier!
The buzzing sound from a transformer is reducing the s/n ratio quite dramatically and can be annoying. A DC blocker fix that!
If you think that the voltage drop of 1 volt out of 230volt will cause a dynamic reduction electrically i would say no. Unless you have a zero feedback design. Otherwise with a small amount of negative feedback the difference in power supply voltage is not affecting the music signal at all. IF your amplifier is that sensitive for power supply voltage, that a small supply voltage change is noticeable in sound reproduction, simply change amplifier!