If the DC offset is <1V (which it typically is) then placing a few diode drops in series with the supply will effectively block the DC.
My only concern would be with the magnetic energy stored in the core (current and voltage are 90 deg out of phase - but I expect that the diodes take care of that as well.
My only concern would be with the magnetic energy stored in the core (current and voltage are 90 deg out of phase - but I expect that the diodes take care of that as well.
Almost to much info to process!
Bare; Just relocating the toroid would be solve the noise problem completely. I would still have short signal cables, and be able to see the panel LED's. There are 16 conductors that would have to be extended......Nope.
Bonsai;
Re "stop agonizing" Thank you for the 'boot to the rear!' I have two 20,000 uF 40V caps and a heavy bridge. I will build a test set-up, this weekend. Do you think those caps would give me an indication? (total series capacitance would be 10,000 uF. If I see results I would by more caps.
Regarding your custom made toroid, you mentioned epoxy core and 'expose resin' for the rest of it. This was a solution I was hoping for in the beginning of this discussion. The windings appear to have little coating other than a plastic tape wrap, although the core seems to be epoxy filled. Are you dead set against some kind of treatment of the windings at this point?
wg ski; I bought this amp knowing that Home HI FI was not it's application, but I have never let common sense interfere with my projects. Being a 'do it yourself' forum, I am trying to tame this beast for my system goal; To produce a HI FI system that can produce Rock Concert system levels. To be honest, after slowing down the fans, I did not expect the transformer noise to be so prominent.
BTW, would you not agree that bridging the krells would reduce their damping factor, among other detriments to performance?
Andrew T; Thank you for info re power supply. You explain that (understandably), the 'punch' comes from the caps in the amp's power supply and that the internal bridge only conducts when these caps need to be 'topped' up. ( You mentioned 10% of each (half) cycle of the line frequency). I can see that I am being somewhat compulsive about my hesitation to insert something in series with the line side of the transformer, but still....on long sustained bass, could this not be a (minor) factor?
djk; I am surprised to see this circuit in the Bryston, Is this just to reduce transformer noise? I don't understand the crown circuit. It looks to me like a soft start circuit but I don't see the DC blocking (caps). Is that a Bryston 4B?
And finally, Maty; What is the value of R1 in your circuit, and does it not tend to defeat the purpose of the DC blocking?
Thanks to every one for your time, Peter in Hillsdale, Ontario, Canada
Bare; Just relocating the toroid would be solve the noise problem completely. I would still have short signal cables, and be able to see the panel LED's. There are 16 conductors that would have to be extended......Nope.
Bonsai;
Re "stop agonizing" Thank you for the 'boot to the rear!' I have two 20,000 uF 40V caps and a heavy bridge. I will build a test set-up, this weekend. Do you think those caps would give me an indication? (total series capacitance would be 10,000 uF. If I see results I would by more caps.
Regarding your custom made toroid, you mentioned epoxy core and 'expose resin' for the rest of it. This was a solution I was hoping for in the beginning of this discussion. The windings appear to have little coating other than a plastic tape wrap, although the core seems to be epoxy filled. Are you dead set against some kind of treatment of the windings at this point?
wg ski; I bought this amp knowing that Home HI FI was not it's application, but I have never let common sense interfere with my projects. Being a 'do it yourself' forum, I am trying to tame this beast for my system goal; To produce a HI FI system that can produce Rock Concert system levels. To be honest, after slowing down the fans, I did not expect the transformer noise to be so prominent.
BTW, would you not agree that bridging the krells would reduce their damping factor, among other detriments to performance?
Andrew T; Thank you for info re power supply. You explain that (understandably), the 'punch' comes from the caps in the amp's power supply and that the internal bridge only conducts when these caps need to be 'topped' up. ( You mentioned 10% of each (half) cycle of the line frequency). I can see that I am being somewhat compulsive about my hesitation to insert something in series with the line side of the transformer, but still....on long sustained bass, could this not be a (minor) factor?
djk; I am surprised to see this circuit in the Bryston, Is this just to reduce transformer noise? I don't understand the crown circuit. It looks to me like a soft start circuit but I don't see the DC blocking (caps). Is that a Bryston 4B?
And finally, Maty; What is the value of R1 in your circuit, and does it not tend to defeat the purpose of the DC blocking?
Thanks to every one for your time, Peter in Hillsdale, Ontario, Canada
Well, he lives in Canada. Maybe the simple AVA's solution (in box)... without capacitors:
AVA HumDinger DC Blocker Available Now! $140
But with my big 18000 microF 105ºC EPCOS I have much less riple! They are expensive (€12.5 +VAT each capacitor).
My main is very very very dirty, with "tones" of RFI/EMI, riple and DC. And with voltage near 240V !!! (Spain: 230V/50Hz).
I have these boxes in my two systems:
DCB (prototype, with EPCOS 15000 microF 85ºC) + DCB => AV Yamaha RV-V2700
DCB x2 => AV Marantz SR4500, PC (with USB DAC -ODAC) and ifi iCAN
With DCB 18000 microF 105ºC the sound is better than DCB 15000 microF 85ºC (aka prototype).
AVA HumDinger DC Blocker Available Now! $140
But with my big 18000 microF 105ºC EPCOS I have much less riple! They are expensive (€12.5 +VAT each capacitor).
My main is very very very dirty, with "tones" of RFI/EMI, riple and DC. And with voltage near 240V !!! (Spain: 230V/50Hz).
I have these boxes in my two systems:
DCB (prototype, with EPCOS 15000 microF 85ºC) + DCB => AV Yamaha RV-V2700
DCB x2 => AV Marantz SR4500, PC (with USB DAC -ODAC) and ifi iCAN
With DCB 18000 microF 105ºC the sound is better than DCB 15000 microF 85ºC (aka prototype).
Last edited:
Aleksandar tells me:
I, maty:
DCBlocker x2 = DCB connected in series to other DCB
An externally hosted image should be here but it was not working when we last tested it.
I, maty:
DCBlocker x2 = DCB connected in series to other DCB
EPCOS 18000 microF 105 ºC -> 0.017 ohms
B41505A7189M000 - EPCOS - Electrolytic Capacitor, B41505 Series, 18000 µF, ± 20%, 35 V, 35 mm, 0.017 ohm | Farnell element14
33 // 0.017 = 0.01699 ohms
B41505A7189M000 - EPCOS - Electrolytic Capacitor, B41505 Series, 18000 µF, ± 20%, 35 V, 35 mm, 0.017 ohm | Farnell element14
33 // 0.017 = 0.01699 ohms
Last edited:
Thanks again, guys;
My uneducated thinking was:
The DC voltage can pass through the resistor. Let's guess at a 1 ohm internal resistance for the primary of this toroid. DC in the line would be divided at a ratio of 1 to 33. Why let 3 percent of the DC get to the Toroid?
BTW Bonsai; We're at page 4 of what might be a simple issue. I'm going to get out my soldering iron any time time now, and post the results. I just need a temporary containment for the caps so my listening room doesn't smell like exploded electrolytics, if something goes wrong. Also, in defense of my craziness, the amp is 5000 W, bridged into 4 ohms, my setup is 1000 W p c into 8 ohms. We watched the movie "Seven Psychopaths", a car exploded and scared the .... out of all of us! That's what I'm going for! Now my only worry is possibly cleaning up window glass, not whether my old Phase Linear will die a fiery death. The sub woofer cabinets are copies of JBL cinema subs and I used RCF drivers of very similar T/S parameters. Almost 1/4 the cost but 3 db less efficient, requiring this monster amp. I listened to some music and saw a clip light blink occasionally, that was 2 days ago and my ears are still ringing.
Thanks again, I really appreciate all this advice. I had no idea that toroid noise was a common problem. I would love to see an amp schematic with a DC blocker, built in.
My uneducated thinking was:
The DC voltage can pass through the resistor. Let's guess at a 1 ohm internal resistance for the primary of this toroid. DC in the line would be divided at a ratio of 1 to 33. Why let 3 percent of the DC get to the Toroid?
BTW Bonsai; We're at page 4 of what might be a simple issue. I'm going to get out my soldering iron any time time now, and post the results. I just need a temporary containment for the caps so my listening room doesn't smell like exploded electrolytics, if something goes wrong. Also, in defense of my craziness, the amp is 5000 W, bridged into 4 ohms, my setup is 1000 W p c into 8 ohms. We watched the movie "Seven Psychopaths", a car exploded and scared the .... out of all of us! That's what I'm going for! Now my only worry is possibly cleaning up window glass, not whether my old Phase Linear will die a fiery death. The sub woofer cabinets are copies of JBL cinema subs and I used RCF drivers of very similar T/S parameters. Almost 1/4 the cost but 3 db less efficient, requiring this monster amp. I listened to some music and saw a clip light blink occasionally, that was 2 days ago and my ears are still ringing.
Thanks again, I really appreciate all this advice. I had no idea that toroid noise was a common problem. I would love to see an amp schematic with a DC blocker, built in.
Hi djk; I went back to post #17, and I will just comment that a note I was referred to by Rod Elliot, proposed a circuit with back to back series caps. Obviously, Bryston has confirmed that parallel caps work. BTW, the 3B has four caps, all paralleled with polarity in the same direction. The 4B has caps are connected parallel with polarity reversed.
Again, I don't understand how the Crown circuit blocks DC, but it will be very simple to test; simple remove the caps from my prototype.
Again, I don't understand how the Crown circuit blocks DC, but it will be very simple to test; simple remove the caps from my prototype.
The problem that I see with "testing" is that the growl from a DC affected toroid is not continuous and varies from second to second and minute to minute.
I cannot see how one can average out the two sets of noise from two identical transformers connected at the same time, to allow comparison to show/prove that the DC blocking circuit is 100% effective, or 50% effective, or 10% effective, or not at all.
I cannot see how one can average out the two sets of noise from two identical transformers connected at the same time, to allow comparison to show/prove that the DC blocking circuit is 100% effective, or 50% effective, or 10% effective, or not at all.
Easy.
My two AV have EI transformers, not toroidal but... without DC blockers they, EI transformers, were very hot! And worse sound because they worked in saturation zone
Worse sound because:
-> https://en.wikipedia.org/wiki/Saturation_(magnetic)
My two AV have EI transformers, not toroidal but... without DC blockers they, EI transformers, were very hot! And worse sound because they worked in saturation zone
Worse sound because:
-> https://en.wikipedia.org/wiki/Saturation_(magnetic)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.