• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

DIY: Mains Line/Noise/EMI Filtering

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
4,308
Likes
9,126
Location
Gold Coast, Queensland, Australia
#21
but in several places adding a DC blocking filter did fix some transformer hum. I've forgotten the details, but IIRC the magnitude of common-mode offset needed to induce hum in a toroidal transformer was a lot less than I expected (led to me doing a little more research to find out why).
I've got an oldie but a goodie- 'squeaky clean' line filter on my bench feed to run all my test gear- in fact everything (32 outlets) runs through it and it has DC blocking in the form of diodes/caps along with a bunch of other filters all in a heavy steel case with potted filters etc. Pretty sure there are 2 banks of three series connected diodes IIRC when I pulled it apart to replace a few neon indicators many years ago. I think there was a 1.5v differential between input and output voltage. It's riveted together so not fun to dismantle.

Don't have a schematic, but it is very effective at killing toroid hum/buzz and even is effective on some marginally noisy EI transformer primaries.

Our mains waveform here is terrible looking on my DSO, makes me think we should all be locally regenerating our own.
 

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
4,303
Likes
6,619
Location
Monument, CO
#22
I've got an oldie but a goodie- 'squeaky clean' line filter on my bench feed to run all my test gear- in fact everything (32 outlets) runs through it and it has DC blocking in the form of diodes/caps along with a bunch of other filters all in a heavy steel case with potted filters etc. Pretty sure there are 2 banks of three series connected diodes IIRC when I pulled it apart to replace a few neon indicators many years ago. I think there was a 1.5v differential between input and output voltage. It's riveted together so not fun to dismantle.

Don't have a schematic, but it is very effective at killing toroid hum/buzz and even is effective on some marginally noisy EI transformer primaries.

Our mains waveform here is terrible looking on my DSO, makes me think we should all be locally regenerating our own.
I want one! :)

The usual scheme is caps to block DC with diodes across (shunting) them to keep the voltage across them small under transient conditions. That way, you can use large (high-value) caps with modest voltage ratings.

I think the problem with toroids is that it is very difficult to gap them in manufacturing like normal EI transformers making them extremely prone to saturation with small DC offsets applied. Toroids better confine the EM field to reduce noise in packed electronic components but they do have their drawbacks.
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,073
Likes
1,026
Location
Iasi, RO
#23
"[...] A number of potential sources of DC current injection have been identified, particularly the use of high frequency ballasts within fluorescent lighting loads, and switch mode power supplies for PCs. Additionally there are no limits for DC harmonics defined under EN61000-3-2, and additionally items taking less than 75W do not have any harmonic limits defined at all (exceptions under clause 7). Hence for - some household equipment, such as mobile phone chargers, it is not known whether any DC current component is produced. It is therefore recommended that a selection of these devices are tested to measure and attempt to quantify any DC current components. [...]"
 

graz_lag

Major Contributor
Joined
Oct 13, 2018
Messages
1,296
Likes
1,452
Location
Le Mans, France
#24
True symmetrically balanced isolation transformer ...

I would like to report some details abt. my most useful upgrade on the main line so far by means of symmetrically balanced isolation transformers.
I decided to develop this upgrade after testing several Schaffner' filters with the intent to reduce the AC noise and electromagnetic interferences. These were so-and-so effective, but I felt the dynamic was also affected in some way. :(

So, after testing the 1st prototype, I have then build four units in total :
transformer #1 for the sources of both Hi-Fi and HT setups
transformer #2 for the Hi-Fi setup's amplifier
transformer #3 for the HT setup's amplifier
transformer #4 for the HT's projector

All are based on transformers with 2x screen magnetic shielding, so that the highest reduction on AC noise, which is the cause of ground loops, can be secured.
Each unit is configured for 230VAC (France) true balanced mode operation, so a 1:1 isolation transformer with a precisely placed center tap on the transformer’s secondary.
The incoming 230VAC (115V on the L terminal and 0V on the N and G ground) is split into halves on the transformer’s output.
The "new" AC line has then 115V on the L terminal and 115V on the N terminal, all that when referenced to the new center-tapped G ground, which remains at 0VAC.
The opposite polarity between L and N is due to completely cancel all common mode noise from the incoming AC line.

I paid from 120 EUR to 250 EUR including shipment (2015) for each transformer from a French manufacturer, plus : the aluminum enclosures, the IEC sockets, wirings, circuit breakers on the outgoing, silent blocks, minor stuffs here and there, say from 300 EUR to 450 EUR depending the size each finished unit ... including the fancy voltage indicators ...
I have tested the integration of filters (Schaffner #2070) but I prefer with no filter.

Manufacturers like Balanced Power Technologies, Bryston, Keces, ect., sell these units for thousands of $ ... :facepalm:

The same principle can be applied to any incoming VAC, i.e. in the case of 120VAC one gets +60VAC and -60VAC on the outgoing section.

Easy, funny, but most of all, a "straight-to-the-problem" solution with an incredible value for money ! :p

PS The most difficult task is the sourcing for the transformers, as these "symmetrically balanced isolation" ones are not very common within the manufacturer standard catalogs.
 
Last edited:

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,073
Likes
1,026
Location
Iasi, RO
#25
Some pictures will really help. The price you paid is a really low indeed, I paid about 100 EUR myself for my isolation transformer (250 VA, 150 Watts). However, we could both purchase few 1000 EUR power cables anytime. :)

Is there's a way for you to safely measure your mains input, in front of the isolation transformer, then measured again after the transformer? This will help in understanding what this isolation transformer resolves for your audio setup. In my case, I have a good AC main at home, so my isolation transformer did absolutely nothing to my AC sinewave (same THD, same sine etc.), instead did resolved a tiny hum-noise caused my a ground-loop (same results could be achieved in a cheaper way).

Also, RMA and ARTA with your audio devices connected directly to AC mains, then connected to isolation transformer might also help in understanding something:
- 1) Why would you need isolated balanced power for?
- 2) Any measurable improvements on the AC sinewave?
- 3) Any measurable improvements on the Hi-Fi audio devices?

Please be aware that loads for more than 50% after the isolation transformer will increase THD.
From what I know and I was able to measure myself, the real benefits of an isolation transformers are:
- reducing the possibility of getting electroshocked and die when touching only one AC wire (touching both wires will shock us anyway)
- removal of ground loops, if exists (in conjunction with ground lifting safely)
- removing DC-components from the AC mains, if exists
And low-lights, from my point of view. would be:
- added THD when output load approaches its nominal output power (sinewave will start bending and look more like DC, also some spikes could appear and transformer will start buzzing)
- coil noise under some circumstances

Also few of my thoughts here: https://www.superbestaudiofriends.o...tors-ups-yadayadayada.282/page-16#post-234579. I don't expect people to take my words for granted, mostly because my AC mains looks really good, so I don't need to improve anything, so if find a way to safely measure your mains quality and how much of your Hi-Fi noise is really related to your AC mains, then please post back.

Note: If you never measured AC sinewaves directly before, please don't try now! By safely measuring the AC lines I mean to use a scope with a differential probe without getting electroshocks into your body. Also, a safer way to do that would be using a step-down transformer connected to a regular scope could also help (cold be 230V-to-120V or simply a 230V-to-12V transformer).
 

DonH56

Technical Expert
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
4,303
Likes
6,619
Location
Monument, CO
#26
I have thought about doing something similar over the years but was always nervous about the safety aspects, and without UL certification or some such my insurance would take a pass if anything happened and they knew it was in there. The good news is that, for better or worse, except for the occasional ground loop I have never really seen problems with mains leakage or such that was not either more than adequately rejected by the components or overwhelmed by other noise and distortion sources in my system.

Technically, the few times I tried it, I saw the same problems @trl describes with transformer linearity. It worked great on preamps and low-level stuff, especially when I applied it to my TT and preamp, but the large current spikes demanded by power amplifiers would cause havoc with the output unless I oversized the transformer by at least a factor of two (I was lucky to find a big power transformer so was able to put an industrial 2 kVA unit on my old 100 W/ch stereo amplifier).
 

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,073
Likes
1,026
Location
Iasi, RO
#27
My home power lines are pretty clear and perfectly sinewaves, so my tests will be non-conclusive, I need to find a home with much noisy and ****** power line. :) The only thing I can replicate would be the hum noise when ground-loops are happening (easy to reproduce that), but other than that...nope, sorry.

I've also measured the noise between ground and null wires and when a vacuum cleaner or other appliances are connect in the same outlet or even the same circuit, then the noise increases about ten times (it becomes a problem only in case of ground loops).
 

Count Arthur

Senior Member
Joined
Jan 10, 2020
Messages
420
Likes
568
#28
I have a mains filter, based on chokes and capacitors between live and neutral. It's been a while since I did a with/without comparison, but from memory I liked the sound with it in place more than without, it's subtle, but the sound is somehow "calmer"??

I've also seen some DIY prjects using off the shelf EMC/EMI Schaffner filters: https://www.schaffner.com/products/emcemi/

 

pjug

Senior Member
Forum Donor
Joined
Feb 2, 2019
Messages
355
Likes
233
#30
What about effective DC blocking? I have DC blocking in my Squeeky clean test bench feed filter and the effect on toroids in terms of silence is easily demonstrable.

Our mains is pretty ugly looking these days.
Do you mean this silences lamination hum? I've been thinking of trying this but never got around to it. Do you recommend a certain device for CD blocking?
 

ezra_s

Member
Forum Donor
Joined
Mar 23, 2020
Messages
86
Likes
68
Location
Spain
#33
I NEED HELP PLEASE !
I recently bought a Rotel RB-990BX amplifier. It will power up, but has no output. It stays in "Protection" mode even without having a load (speakers connected) to it. No fuses have blown. ANY SUGGESTIONS ???
your cat peed on it... or what's what happened to me the only time I saw one of my receivers showing that message and shutting down.
 
Joined
Jul 6, 2020
Messages
49
Likes
25
#36
I doubt ac line filters will do much to improve thing if treated alone. Common mode noise is by definition a path through electronic equipment to any antenna or ground connection enabling it to return. If it encounters any differential impedance along its path, it will convert the noise from common mode to differential, and you will hear it. Generally you’ll need to treat all cable interfaces, not just the power mains.

I have most commonly encountered it when I have metallically interconnected devices which are powered from different ac outlets. Changing that usually solves. Minimize the number of different ac outlets powering your equipment, and don’t run it off a breaker feed that also supports noisy equipment on the home.

If that doesn’t help, a simple treatment of clamp on ferrites (as you sometime see barnacled onto power cords) for power cords and signal lines can help. Dressing cable runs to avoid pickup loops near noisy equipment also helps. Don’t run signal cables anywhere near power cables.

I have seen instances of telephone systems shutting down upon a near field user pressing the talk button of a walk-in talkie. Problem was traced to rf susceptibility of a single voltage reference on one large board. One capacitor (0.01uf ceramic as I recall) completely soloved the susceptibility.

EMI problems can be very much like squeezing a ballon...you make it quiet here, it gets noisier there. Some designs will be inherently much more susceptible than others. I think it best to first address the cleanliness of your local install environment, and only think about adding new things once you are convinced that has been optimized.

PS I ran an EMC engineering and test facility for 20 yrs. Seen some bizarre things.
 
Top Bottom