DIY filtered powerblock

[sometrolls]

Found a design on the web that basically consists of a box with a couple of cheap RF filters in series and an outlet. Are 2 poor quality filters better than 1 good one, or 3 cheap ones better than 2 etc?

Also any simple changes that could be made for a significant improvement?

www.triode-systems.com/modules/wfsection/article.php?articleid=4

www.triode-systems.com/modules/wfsection/circuits/chassis-filter-layout.png

www.triode-systems.com/modules/wfsection/build-photos/filters/mains-filter-large.jpg

The above version cold be knocked up for £20, so suggestions and general comments are invited

[rickcr42]

Are 2 poor quality filters better than 1 good one, or 3 cheap ones better than 2 etc?

I don't think you can find a "poor quality" line filter to be honest.

Any line filter is a good line filter and that especially so for anything digital such as you CD player,DVD player or DACs.Because they operate in the mhz range they are particularly suceptable to RFI

go for it would be my recommendation but leave out the POS/sonic butcher Varistor unless you feel you need it.Those things ARE audible and every single time they get a surge degrade which means the longer you use it the worse it will sound unless you have a regular replacement schedukle similiar to smoke detector battery replacement schedules

[rickcr42]

BTW-check out this one also

www.hagtech.com/pdf/conditioner.pdf

[sometrolls]

Thanks, thats a neat solution he has there.

The choice of filters is quite large, for example Rapid have 7 chassis mounted models ranging in price from £5.50 to £21, so if bought in multiples this difference will add up.

I recall the note on the X-PSU project, saying that the inlet filter was removed from the updated design

[rickcr42]

I recall the note on the X-PSU project, saying that the inlet filter was removed from the updated design

could have been a cost cutting measure,could have been about the claim some make that RFI filters tamp down the dynamics which is something I once beleived until I realised it was the RF artifacts coloring the sonics I was hearing,something that adds content to the upper registers and while hardly accurate (being no part of the original signal) added a certain amount of "zing" which a damn lot of folks seem to like calling it high end accuracy.

It is not until you sit and listen for extended periods and listen to the music instead of the electronics that you come to appreciate just what actual music sounds like-relaxed,flowing not etched,in your face exciting.
The former allows long sessions with cans,the latter is headache territory if you listen too long and is somewhat alleviated by using a cross feed network but rather than for the intended purpose-moving the image forward,adding a natural L/R sonic bleed through-is instead used as no more than a low pass filter to get rid of all the crap at the top end.

You only truly hear what a circuit is about if you can get it to a point where that circuit is totally isolated from any external influences other than the music signal input.All else is unintended and should be eliminated when possible unless the cure is worse than the disease though even here balance is everything

For instance-using a decoupling network or output decoupling series resistor on a feedback amp (anything using an opamp or looped opamp/buffer circuit) WILL ron some of the ultimate available output current but WILL eliminate any possibility of RFI entering the output then being fed back to the high Z of the input through the feedback connection where it can and will be treated as a signal.
Since this new "signal" will not also be present on the "+" input of the opamp you will not get any cancellation (opamp differential input cancels out any COMMON MODE signals so it amplifies only single mode signals,why we use feedback )

I will take a slight current reduction over potentially destructive signals entering my amps being a realist while the "purist" can have his perfect world/pefect environment/perfectly operating on paper amps.My amps are friggin' bear traps and if you want in you better bring a freakin' RF battering ram ;D

For reference/your reading pleasure :

the "why" :

www.jensentransformers.com/an/an001.pdf


the $5 commercial version ($10 stereo and still a bargain)

www.jensentransformers.com/datashts/oli3.pdf


EMI/RFI IEC mains inlets,RC networks on all circuit inputs,decoupling networks on the outputs of all feedback amps (not needed for open loop amps ),and finally LC filters at all power supply pins.

BOMB PROOF AUDIO !


.......but not Nuke proof,those big time EM pulses take out anything solid state so if you want music in a post nuclear blast world you need a big fkn gas operated Honda Generator and tube amps

[rickcr42]

Interesting "add-on" read,one that show a good example of a "parallel filter" :

www.tnt-audio.com/clinica/mains_e.html

www.tnt-audio.com/gif/m-filter.jpg


thwe first time I ever came across this type of "mains crap eliminator" was the "tweaks" section of Magnan Audio around 7 years ago

www.magnan.com/pages/tweaks/powerline.html

one thing that has always stuck with me is the potential for shock (physical and emotional ) this device has if you pull it from the wall outlet once installed.

Having no "on-off" or discharge mechanism the caps stay at full potential so when you suddenly remove them from the wall you can get a nasty arc of electricity that even if it does not harm can scare the shit out of those without strong nerves

here is a commercial version :

www.russandrews.com/lookup/1/region/UK/currency/GBP/customer_id/PAA2710061406437EGLMUKBZUCFYRFRJ/product-The-Silencer-1403.htm

[sometrolls2]

I took the easy approach: single inlet filtered in box, and distribution block wired into it. The scratch build seemed a hassle and filters in series questionable.

Bits list from Rapid( add VAT):
Schaffner FN9222 10A version £6.50 code 26-4465
75 X 50 X 25 BLACK ABS BOX 74p 30-0500

plus IEC lead and distribution box, and cable clamp from spares, but might cost another £3.50-£5 for low cost versions

The benefits to sound were as expected for my CD player: general increase in weightiness plus more detail, as per the reviews of power leads/filters etc. Bit too much bass for some electronic beats via heaphones for my liking, but thats a taste thing, but objectively an improvement.

On X-Cans a clearly negative response, a plain nasty addition.

Then tried the thing on other household goods: similar results with sound on both PC and TV, with some improvement to the displays also. Minimal impact on PC overclocking, which is similar to OCZ filtered power cable with graphics cards. But big benefit with digital TV set top box, more stations detected and picture improved, and when combined with the TV, producing most impresssive results for peanuts. It allowed me to alter the contrast and brightness( as there was more in the image to play with), producing a major step up for my cheapo TV.

Generally filters are listed in 3 grades: normal, high performance and medical. The medical seem less available than in the past, so were irrelevant. The Schaffner was amongst the cheapest high performance, and cheaper with inlet than without, so seemed a good buy. There are models that cost 3 times as much, perhaps it is worth the extra, but I doubt it. The current rating for the filter may matter, with higher rated filters applying less colouring to the sound. Obviously cables and block can be endlessly upgraded.

[rickcr42]

goes against convention and will void any warrantee but if you use a commercial filter strip for audio get rid of the transient suppressors !

They ARE audible and more so as they degrade which is every single time a high voltage transient hits and they try to clamp it.

second.If variable levels of quality are a problem (consumer grade/Industrial Grade/Hospital Grade) there is no reason the device can not be opened up and marginal parts replaced with better graded parts,especially the caps (I would leave any inductors in place).
Just use AC Mains rated parts or you will be looking for trouble