D C Blocking on mains supply

[firedog]

Hi-

Just installed this: the Avahifi Humdinger , which is a DC Blocker. 

 

Costs $125 US Plus shipping. It's small and not heavy, so shipping is reasonable. Shipping for me was $35US for International, but I'm not in Australia. 

 

Worked like a charm. 

[mwhouston]

Guru: nice build. Mine combines DC blocker and EMI\RFI filter. I fuse (10A slow blow) and earth the small metal enclosure and add a ferrite choke on the power cord. Of six systems these DC blockers have been tried in only one said he could hear a difference. All the others claim instant improvement, generally in the sound stage.

 

retro-thermionic DC Duffer

[Guest guru]

Guru: nice build. Mine combines DC blocker and EMI\RFI filter. I fuse (10A slow blow) and earth the small metal enclosure and add a ferrite choke on the power cord. Of six systems these DC blockers have been tried in only one said he could hear a difference. All the others claim instant improvement, generally in the sound stage.

 

retro-thermionic DC Duffer

I've made quite a few for clients and all have found them to work very well. We will be using them at the show in October so you can have a closer look. We do them in all different sizes now , 1,2 ,3 or 4 outlets and we are getting some custom chassis made up for extra bling, the Japanese supplier is most accomodating to shapes and anodising colours.

Regards, g.

[valvedude]

warning - do not attempt this without a very competent knowledge of what is involved and the safety required.

I have been interested in the blocking of dc voltage coming in on the ac mains line for some time and after a discussion with a client who is the head of foxtel engineering regarding devices which contaminate the mains supply entering our houses, I decided to build a very simple outboard dc blocking distribution box for exactly that purpose. the schematic was courtesy of rod elliots excellent website and the components all courtesy of rs components so nothing to exotic, iec input from furutech and mains sockets from hpm. all blocking modules were assembled and slowly brought up to 240 volts using a variac as exploding capacitors are best not experienced and everything worked as hoped and after final assembly, rechecked again and soak tested under load. all good.

2 types of use were envisaged, for visual components and for audio.

firstly, dc voltage on the ac line is very common these days as components using switch mode power supplies have a degree of cause and even more recently, solar panelling and the associated inverters. dc on the ac line has a tendency to cause transformers to run hotter and noisier being the main issue.

on the visual side when inserting this device in line with the mains supply to the plasma we have at home showed a quite remarkable improvement in picture colour, stability of image, white noise level or blackness depth and the amount of heat generated from a 5yr old NEC50 plasma. can't watch anything now without it being in line. the depth of the black's reminded me of the best pioneer plasma's and I would not be surprised if their reputation was based on an internal dc blocking device as the effect on the black's and colour intensities was certainly similar.

as for sound reproduction on a modest system, the results were likewise positive with a quieter background noise being most evident and with transformers running considerably cooler.

more listening to be done but at this stage, all very positive.

cheers, g.

why not us a good isolation transformer DC on mains new too me i was under the impression it was alternating current 

[Guest guru]

why not us a good isolation transformer DC on mains new too me i was under the impression it was alternating current

I have a very good isolation transformer but never used it. After further testing in other systems and extensive listening is that I would never install a system without these cd blockers in place. the qualitative improvement is very substantial. D C voltage is generated on the mains Ac line by numerous devices these days as more solar inverters are installed, more switch mode power supplies, more electronic transformers for halogen lighting. It's one of those things you don't realise it's there or what its effect is until you get rid of it.

[andyr]

I have a very good isolation transformer but never used it. After further testing in other systems and extensive listening is that I would never install a system without these cd blockers in place. the qualitative improvement is very substantial. D C voltage is generated on the mains Ac line by numerous devices these days as more solar inverters are installed, more switch mode power supplies, more electronic transformers for halogen lighting. It's one of those things you don't realise it's there or what its effect is until you get rid of it.

 

I absolutely agree with you, g.

 

But as I understand it, DC doesn't pass through a 1:1 isolating transformer either - so why should a cap/diode DC blocker deliver any better result?

 

I recently made up several 'Isotran' units for myself, using a transformer customised for me by SES, here in Melbourne.  I could lend you one if you are interested in seeing what it does, compared to your DC blockers, on some source device or a TV which pulls <= 150ma @ 240v.  (The transformer is rated to 250ma @ 240v - ample for the sort of draw which my preamp, motor speed controller or active XOs draw.)

 

You would need to pay return postage, though.  And why am I interested in spending my money to send it up to you? ... to get a comparison against a different method of blocking DC!

 

 

Regards,

 

Andy

[Guest guru]

Thanks Andy, I have never tried the isolation transformer I have so it was not a question of one better than the other. Sorry if I gave that impression. As for 250 ma, I use these blockers on everything including power amps, digital equipment, pre's, large plasma's. I just make sure there is one module for each individual outlet. If we do the Melbourne show you would be welcome to check them out as the entire system would be running off them even though monkey coffins at the end of a 3.5 watt stereo amp isn't your kind of sound going on what you listen too. It will be vinyl though so some redeeming feature.

Cheers,g.

[andyr]

Thanks Andy, I have never tried the isolation transformer I have so it was not a question of one better than the other. Sorry if I gave that impression. As for 250 ma, I use these blockers on everything including power amps, digital equipment, pre's, large plasma's. I just make sure there is one module for each individual outlet. If we do the Melbourne show you would be welcome to check them out as the entire system would be running off them even though monkey coffins at the end of a 3.5 watt stereo amp isn't your kind of sound going on what you listen too. It will be vinyl though so some redeeming feature.

Cheers,g.

 

Thanks, g.  You have just enunciated one advantage of the DC blockers ... they don't have a current limitation (so you can use them on power amps)!   Which means I had better build me a couple.

 

Regards,

 

Andy

[Guest guru]

You will be very pleased with your efforts if you do.

[Newman]

Remember that the mains transformer in your gear IS a mains DC blocker!!   :blink:

 

As Elliott clearly states, the only objective of this kit is to help if you are having occasional mechanical rattling or 'growling' issues with your bigass 300+VA mains transformer, due to DC on the mains wiring.

 

Nice thread though, good implementation, well done guru.

[Guest guru]

having reread mr. elliots excellent write up, I concur that is what he said and as he only used a "cro" as a means of evaluating the device, perhaps he should have put it into a circuit and listened on an audio system to see if there was a sonic benefit to the implementation.

[andyr]

having reread mr. elliots excellent write up, I concur that is what he said and as he only used a "cro" as a means of evaluating the device, perhaps he should have put it into a circuit and listened on an audio system to see if there was a sonic benefit to the implementation.

 

I also believe listening is the only true test, g (not sure that Newman will agree, though!  ) but, much as I think Rod Elliott is to be lauded for his designs - I've been using his active XO boards for 15 years - he is definitely of an engineering bent (ie. measurement is the key).  In one of his articles he states categorically that there is no difference in the sound of (expensive) 'boutique' caps. :lol:

 

 

Regards,

 

Andy

Edited July 24, 2014 by andyr

[TMM]

The visible effect of Mains DC is vibrating transformers.

I think you have been mislead. DC will not cause transformers to vibrate. What causes enclosures to 'hum' is the enclosure being resonant at 50Hz or some multiple of 50Hz. The mains does not remain completely stable at exactly 50Hz - it varies above and below ever so slightly. That is why enclosure hum comes and goes instead of occurring at a steady volume all the time.

DC can only cause additional heat to be dissipated in the primary winding of the transformer but "significant" would be a gross overstatement. The AC heat losses will dominate.

 

Thanks, g. You have just enunciated one advantage of the DC blockers ... they don't have a current limitation (so you can use them on power amps)! Which means I had better build me a couple.

Regards,

Andy

They certainly do have a current limitation. That being the ripple current rating of the capacitors at 50Hz. If you push them past their ratings that will suffer a premature death (possibly with a bang) and the load will increase on the diodes considerably. The diodes should be specced so that should the capacitors go open circuit they do not fail as well. Edited July 24, 2014 by TMM

[andyr]

They certainly do have a current limitation. That being the ripple current rating of the capacitors at 50Hz.

 

Aah, would you please elucidate, TMM.

 

What is - or how do we find out - the ripple current rating of the capacitors at 50Hz?

 

And how do we select electrolytics which have the necessary ripple current rating for a (power amp) current draw?

 

 

Regards,

 

Andy

[TMM]

Ripple current is how much AC current a capacitor can handle before it gets too hot and fails. It is main killer of aluminium electrolytic capacitors. Film caps can handle much higher ripple currents than aluminium electrolytic caps but obviously if you need 10,000uF you aren't going to be using film caps unless you have deep pockets and want to make something very large.

The ripple current should be specified in the manufacturers datasheet for the capacitor. If the manufacturer doesn't provide the data for ripple current then assume it is embarrassingly low. It is usually specified at 120Hz in milliamps. At 50Hz the ripple current rating will be approximately half of what it is specified for at 120Hz.

If you have a 500Watt amplifier, assuming it is 50% efficient the draw from the mains will be 1000W. You should therefore use capacitors with a 1000W/240V = 4.16A ripple current

You can parallel multiple capacitors together to handle higher ripple currents. Two capacitors with a 2.5A rating each in parallel are like one capacitor with a 5A rating.

If in doubt, i would advise not to build one.

[Guest guru]

the caps I use have a ripple current rating of 3000mA, they are Panasonic 63v/4700uF, diodes are Vishay 1N5404, caps are $7.66 plus gst each , not cheap, not expensive, diodes made in Taiwan , not China, caps from Malaysia. hope this helps.

[andyr]

the caps I use have a ripple current rating of 3000mA, they are Panasonic 63v/4700uF, diodes are Vishay 1N5404, caps are $7.66 plus gst each , not cheap, not expensive, diodes made in Taiwan , not China, caps from Malaysia. hope this helps.

 

Hi guru,

 

(I tried to send this as a PM but it seems your Inbox is full. )

 

Can you tell me where you bought those 4,700uV/63v Panasonic caps?  I had a search but couldn't find (on places like Element14) which specific cap you are using.

 

Also, I'd appreciate your opinion on:

1. Given TMM's comment that ripple current is the thing that ages electrolytics, wouldn't it be better to use 2 x 2,200uF caps in parallel, in place of each 4,700uF?
2. If so, would /50v rating be OK? (Most of the 2,200uF caps I found were only /50v.)
3. Or even 3 x 2,200uF in parallel?
4. I normally use Nichicon KG Gold Tune caps for power supplies and these I can buy in 100v (for 2,200uF), for USD6.95 - whaddya think of these?

I plan to build 2 of these DC blockers, 1 for each amp case which I have (behind each Maggie); there are 3 amplifier modules in each case, totalling 250w into 8 ohms.  I already have the diodes.

 

 

Thanks,

 

Andy

Edited July 25, 2014 by andyr

[Guest guru]

Andy, will address the inbox. Regarding the supplier, rs components. Like you, I like the way mr Elliot looks at things and supports overkill and erring on the side of caution. I considered the ripple aspect of the capacitors in the choice made and the capabilities of the diodes in their choice. Touch wood, not a single issue so far and so far 800 hours under soak test. I guess what I'm saying is if I have something that works well, I tend to stick with it so I would be reluctant to make changes. Success gives you confidence but I still check each component before use and use a variac to bring each module up to voltage and again once the unit is wired in situ. I really like what these things do and I think any component would benefit. If you look at the first image you would count 4 modules, 1 for each outlet, this is the only way I would use them , 1 module, 1 component, for $20 a module, why would you do it any other way.

Regards, g.

[andyr]

Andy, will address the inbox. Regarding the supplier, rs components. Like you, I like the way mr Elliot looks at things and supports overkill and erring on the side of caution. I considered the ripple aspect of the capacitors in the choice made and the capabilities of the diodes in their choice. Touch wood, not a single issue so far and so far 800 hours under soak test. I guess what I'm saying is if I have something that works well, I tend to stick with it so I would be reluctant to make changes. Success gives you confidence but I still check each component before use and use a variac to bring each module up to voltage and again once the unit is wired in situ. I really like what these things do and I think any component would benefit. If you look at the first image you would count 4 modules, 1 for each outlet, this is the only way I would use them , 1 module, 1 component, for $20 a module, why would you do it any other way.

Regards, g.

 

Thank you very much, g.

 

I bought my diodes from RS - so I assume they are the same as wot you bought!

 

OK, you bought the Panasonic caps from RS; I have the latest paper catalogue, so it is easy to flip through Section 16 ... but I couldn't find any 4700uF/63v Panasonic caps?   So I'm wondering what their part no. is?

 

Irrespective, whilst I appreciate the philosophy of "if you're onto a good thing, stick to it" ... what do you think of the concept of using 2 x 2200uF caps in place of 1 x 4700uF cap?  Given that this improves the ripple current aspect?

 

And is 6600uF better than 4700uF?

 

 

Regards,

 

Andy

Edited July 25, 2014 by andyr

[Guest guru]

RS COMPONENTS 127-824

can't answer the other questions, haven't seen any schematics suggesting alternatives other than using smaller voltage caps and single diodes which I made up and they worked but this combination sounds much better by being much quieter giving a much more dynamic presentation. at the moment I hardwire the components together but we have started work on a high quality PCB to reduce the fabrication time.

[andyr]

RS COMPONENTS 127-824

can't answer the other questions, haven't seen any schematics suggesting alternatives other than using smaller voltage caps and single diodes which I made up and they worked but this combination sounds much better by being much quieter giving a much more dynamic presentation. at the moment I hardwire the components together but we have started work on a high quality PCB to reduce the fabrication time.

 

OK, thank you, g.

 

I am going to hard-wire the components together (as wire can carry much more current than any PCB trace).

 

Given your input that you had used lower-voltage caps - and this didn't sound as good - I will use the 2,200uF/100v Nichicon caps and use 3 in parallel (for ripple advantage).  Then I'll send it up to you, so you can decide whether this arrangement sounds better.

 

 

Regards,

 

Andy

Edited July 25, 2014 by andyr

[Number 9]

Thank you very much, g.

 

I bought my diodes from RS - so I assume they are the same as wot you bought!

 

OK, you bought the Panasonic caps from RS; I have the latest paper catalogue, so it is easy to flip through Section 16 ... but I couldn't find any 4700uF/63v Panasonic caps?   So I'm wondering what their part no. is?

 

Irrespective, whilst I appreciate the philosophy of "if you're onto a good thing, stick to it" ... what do you think of the concept of using 2 x 2200uF caps in place of 1 x 4700uF cap?  Given that this improves the ripple current aspect?

 

And is 6600uF better than 4700uF?

 

 

Regards,

 

Andy

i'd suggest looking at the US made Mallory/ Cornell Dubilier HES series. They are made for high power Strobe light applictions and are particularly good in terms of ripple current.

Edited August 6, 2014 by Number 9

[andyr]

i'd suggest looking at the US made Mallory/ Cornell Dubilier HES series. They are made for high power Strobe light applictions and are particularly good in terms of ripple current.

 

Thanks, Steve - I will investigate.  First impressions are, though ... that they are very large!

 

Regards,

 

Andy

[Rhythm Willie]

I have a bit of mechanical noise from the 400va toroids in  my "Aussie Amp" monos.

 

I'm thinking of getting this PCB ,( I already have the parts) and installing them inside each amp.

 

http://sjostromaudio.com/pages/index.php/hifi-projects/109-dct02-the-dc-trap-high-end-style#.U-ION2MzfbY

 

About $30 for the 2 PCB's.

Edited August 6, 2014 by Rhythm Willie

[Guest guru]

A question likely to go unanswered by anyone here but dc blocking devices do work and they have a very positive effect on the sound of systems I have installed them but then everyone's ears are differently shaped.