How to clean the DC at mains. And ripple too.

0 Members and 1 Guest are viewing this topic. Read 13267 times.

maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #40 on: 22 Feb 2018, 04:16 pm »
Four generic? cheap capacitors: 47,000uF, 85 ºC, 16V


maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #41 on: 22 Feb 2018, 06:31 pm »
I have a thread about DC Blocking (and others about DAC, amps, audio links...) in other forums. I just updated it and the first image is from the inside of my DCBx2 = DC blocking turbo (two PCB "in series").

DCBx2 aka turbo






DCB + DCB prototype with EPCOS




DC blocker turbo + DC blocker, both with electrolytic capacitors EPCOS 105°C




DCBx4 = DC & Ripple Blocker x4 ME aka Maty Edition

























« Last Edit: 22 Feb 2018, 11:03 pm by maty »

maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #42 on: 23 Feb 2018, 10:17 pm »
Interesting. MATLAB software.

MathWorks DC Blocker

Quote

Use DC Blocker to Remove DC Component of Signal

In the model, run the simulation. The spectrum of the input signal shows tones at 150 Hz and 250 Hz and a significant (0 dBW) DC component.


Select the DC Blocker block by double-clicking on it and change the algorithm type from IIR to Subtract mean. Rerun the simulation. The spectral output from the DC Blocker shows that the Subtract mean estimation method results in a DC component of less than −100 dBW.



Quote
DC Blocker with Fixed Point Data

Constellation diagrams and spectrum analyzers are used to show the improvements from the DC Blocker.

[BIG IMG] https://es.mathworks.com/help/examples/dsp/win64/xxdcBlockerCIC_model_diagram.png

Look at the spectrum of the noisy signal with the DC offset. Notice that the signal has a peak at 0 Hz.


Observe the spectrum of the noisy signal after the DC Blocker removes the offset. The spectral peak at 0 Hz has been removed.



WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #43 on: 23 Feb 2018, 10:49 pm »

 


   
« Last Edit: 24 Feb 2018, 02:52 am by WireNut »

maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #44 on: 24 Feb 2018, 08:53 am »
Oh my god, 16 diodes! And EPCOS caps too  :P

Link? Easy: https://www.google.com/search?q=dc+blocker+red -> Images

DCT-03 The DC trap, DC filter, high-end style, heavy duty

-> http://sjostromaudio.com/pages/index.php/hifi-projects/187-dct03-the-dc-trap-high-end-style-heavy-duty


Quote
Interesting features:

    Room for up to 6 x 68000 uF snap-in high performance caps.
    Four DC blocking voltages.
    Very thick copper traces, virtually impossible to burn off.
    Gold plated pads.

    
Quote
Operating voltage:    0-240 VAC, 50-60 Hz
Max DC voltage:    1, 2, 3 or 4 V
Max inrush current:    Approx 120-200 A peak depending of chosen diode type
Max continuous current:    See datasheet for used capacitors
Dimensions:    81.3 (3.2") x 192.4 (7,58") mm

[BIG IMG] http://sjostromaudio.com/hifi_pics/hifi_100pr/dct03r0schema_p1color.png

Quote
The DC trap makes sure that no DC slips through. The primary winding is AC coupled with help from six big electrolytic capacitors. The inrush current will go through the diodes when the capacitors have been charged over 1 or 4 volts and the normal current will only go through the capacitors.

Quote
Circuit description

This design is very simple, just a couple of diodes and capacitors. The purpose of the diodes is to limit the DC voltage across the caps, especially at high load and at power on. The pcb contains of 16 diodes where there are two in parallel for each direction. This is for increasing the current capacity to the double. Normally you should not connect diodes in parallel in they not are on the same die but in this case I think it will work fine since they will have the same temperature and will most likely not continuously carry any larger currents. I recommend that you solder all 16 diodes. The only disadvantage is a slightly higher cost. Don't try to desolder. You will only damage the pcb.



Well, I have not problem to desolder the little EPCOS because they are not glued to PCB in the prototype (the news are glued) and... Aleksandar think in all:


PCB for modified Sulzer regulator – v.2

https://www.atlhifi.com/shop/bare-pcb/pcb-for-modified-sulzer-regulator-v-2/

Quote
PCB for the schematic is with dimensions 163×70 mm.

PCB thickness: FR4 1.55 mm. Copper thickness: 35um (1oz). Surface finish – Nickel Gold (ENIG).

It is one-layer design with solder mask coating on bottom layer for best durability. Silk screen with parts layout is placed on the top of the PCB.

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #45 on: 24 Feb 2018, 09:10 pm »
 :D Yes.  What do you think of it?   I'm buying 2 boards..... I'll be using them on 120vac lines inside 2 DIY power conditioners.
             That's the plan anyway. Could use your thought's.... I'm a bit green on DIY but learning.


maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #46 on: 24 Feb 2018, 09:36 pm »
I think you must draw a scheme of your system.

Power conditioners? Which? I only know one (and expensive) that has a DC Blocker but only as an option!

The more information we collect, less chance of error or spending too much.

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #47 on: 24 Feb 2018, 09:56 pm »
They will be pretty much like the pic below but with DCT-03 before 3 or 4 filters in each box similar to the photo below.
1 unit will likely be four 3 amp filters, the second unit will probably be 6 and 10 amp units.
The filters are already built.

System is bi-amped.

Turntable
Phono stage
S.S. Preamp
Tube Buffer
Marchand XM-44 active crossover
Amp-1 (highs)
Amp-2 (lows)

CD Transport
Dac
2nd Tube Buffer after the DAC, before the preamp, maybe, not sure yet but I already have 2 tube buffers.

I'll drawl a picture later if I can but that's pretty much it.



My gut feeling is the DCT-03 is likely overkill, but I can built it as I get the funds.


Not mine. Gordy built this. RIP.




Example only:







maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #48 on: 25 Feb 2018, 11:49 am »
What Power conditioners do you have?

Maybe a DCB + DCB aka DCBx2, aka turbo with Mundorf.

[IMG] http://maty.galeon.com/WP-imagenes/hum/DC-blocking-turbo-inside.jpg

Telstar

  • Jr. Member
  • Posts: 280
Re: How to clean the DC at mains. And ripple too.
« Reply #49 on: 11 Mar 2018, 02:18 am »
maty, do you remember Felicia?
http://www.audiocircle.com/index.php?topic=18441.0

That's a really good power conditioner.

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #50 on: 11 Mar 2018, 04:31 am »

What Power conditioners do you have?


Several Felix with DC Blockers. Might build the Felicia some day, maybe. IDK.


maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #51 on: 11 Mar 2018, 11:07 am »
Felix conditioners and DC blockers.

http://www.audiocircle.com/index.php?topic=150697.0

Better in white your diagram:

Quote
Lastest revision.

4 Felix, 19 amps, in an old Adcom 535 chassis.


They are only DC blockers. Good idea the star ground.

I think was better idea ONE DC blocker or maybe two and four outputs. Only RF/EMI filters, ferrites and isolated or balanced transformers clean in both sides, so it makes sense to use several. But not with DC Blockers.
« Last Edit: 11 Mar 2018, 05:26 pm by maty »

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #52 on: 11 Mar 2018, 04:37 pm »

They are only DC blockers. Good idea the star ground.

I think was better idea ONE DC blocker or maybe two and four outputs. Only RF/EMI filters, ferrites and isolated or balanced transformers clean in both sides, so it makes sense to use several. But not with DC Blockers.


Hi maty,

I'm not sure I understand correctly :scratch:
My drawing shows a DC blocker on the right, 4 "Felix" 120vac Filters (Occam's design) and 4 outlets.

Are you saying a better design would be to use an RFI/EMI IEC Inlet Filter, a ferrite on the power cord, and a isolation transformer?

Something like this maybe but with 2-120vac outlets?





 

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #53 on: 11 Mar 2018, 04:55 pm »



maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #54 on: 11 Mar 2018, 05:26 pm »
I am sorry, you are right. I need coffeeeee.

RF/EMI filter and ferrites is always better for each component like your implementation.

You can easy try to connect two of the Felix filters: Felix -> Felix -> component. And verify if the sound is better or not.

If the sound is better you can add other Felix or Schaffner RF/EMI filter. Each component is different; someone only need one filter, others two or one + ferrite or two + ferrite. It is trail and error.

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #55 on: 12 Mar 2018, 01:03 am »
 Hi maty,

I did try a Felix>Felix>component> awhile back but from what I remember it didn't do much better then a signal Felix and
also I didn't want to use up my parts so fast, double parts so I stayed with just 1 Felix.

I'm interested in trying ferrite's and IEC Inlet Filters, or the chassis mount type of inlet filters.
Do you have your favorites of ferrite's or Filters that you like to use?
I think your 220vac, while I’m 120vac.

The group buy at DIYaudio isn’t going very well so far with the large DC Blocker with 6 caps and 16 diodes.
Not enough interest so far.

I think I'll look into the DC Blocker pcb’s from ATL. They look good.

« Last Edit: 12 Mar 2018, 02:52 am by WireNut »

maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #56 on: 12 Mar 2018, 09:01 am »
if with two Felix you do not notice improvement then you do not need the Felix filter an ferrite with this component.

By the way, Felix attenuation measures/graphs?

Years ago I studied the specs of the known RF/EMI filters and ferrrites to work in audio band, f < 1 Mhz. The best are Schaffner filters (better B: medical) and Würth 150 kHz ferrites. If you know of other better ones, put the link and we will check their graphs.

Quote
The group buy at DIYaudio isn’t going very well so far with the large DC Blocker with 6 caps and 16 diodes. Not enough interest so far.

If I am not wrong he is in the electronic cabins design of Volvo trucks in Sweden.

Aleksandar is a very perfectionist person and he pays a lot of attention to details. All PCBs are very well made. With diyer mentality, so that the components can be changed several times without damaging the PCBs. It has also designed a headphone amplifier of its own and is with another design for speakers. With MOSFETs and very fast.


WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #57 on: 13 Mar 2018, 02:56 am »
Hi maty,

I’d like to get your thoughts on my drawing / draft. You are more knowledgeable then me and might see mistakes or a better way of doing things.

I have one turntable, a 40 year old JVC QL-7 manual, direct drive TT-71 unit.
Ive replaced 16 electrolytic caps in it, many where leaking.

The Transformer in my Turntable is not a quite one.

While I have the table apart why not move the noisy transformer into it’s own enclosure.
My drawing below show plans to remove that noisy transformer and place it into a separate enclosure along with a DC Blocker and Felix power conditioner.

Putting a Felix after the transformer secondary's might be better but, i haven't figured that out due to the 4 secondary wire outs.


Damn, being OCD and trying to learn last for hours and hours.

Here's my draft.





BTW, Black and white does look better  :thumb:

maty

  • Full Member
  • Posts: 972
    • Nauscopio Scipiorum
Re: How to clean the DC at mains. And ripple too.
« Reply #58 on: 13 Mar 2018, 11:17 am »
If you want make very good rips with a turntable the power is more important in... the phono. Phono with linear PSU and DC at mains is a bad thing. DC Blocker or a good SMPS (they do not have problems with DC).

There are many old vinyls made with masters that have problems with DC -> hum.

Then you need to add the phono in the scheme.

You do need a RF/EMI filters with the turntable. And no DC blocker too. The bottleneck is the phono.


* It is assumed that turntable has an engine that barely varies the rotation speed.

* The electrolytic capacitors dry up over time even if they are not used. Better to change them, some quality -without being audiophiles- and 105ºC to last longer.

WireNut

Re: How to clean the DC at mains. And ripple too.
« Reply #59 on: 13 Mar 2018, 01:59 pm »
Thanks maty,

I had planned on doing the phono stage as well in another separate box but I think I'll be better served sticking to my original drawing of 1 box with a blocker and 4 Felix.
Easier, less costly, more efficient, me thinks  :D