DC Filter against humming Toroidals - Summary of the most interesting Comments

tiefbassuebertr · 2011-10-03 10:14 pm

Nico Ras · 2011-10-03 10:19 pm

And your conclusion?

tiefbassuebertr · 2011-10-03 10:38 pm

Nico Ras said:
And your conclusion?

No conclusion;
read the last by post #3 from Nelson Pass about
http://www.diyaudio.com/forums/pass-labs/21487-noise-transformer.html
This is for me personally the best exist solution - unfortunately not suited for existing devices, only for new projects.
Nevertheless I work on a circuit to determine the threshold of DC current for hum/noise at various toroidals.
To comment a short conclusion is impossible, because I still want to evaluate the overview themselves.

Bonsai · 2011-10-04 10:59 am

I did not go through each link above, so if I am repeating what you have already linked to apologies:-

Rod elliot has a great description of the whole issue around DC on the mains, and circuit to block it (similar to Bryston). Take a look on his site.

When I was still in Japan, I was playing my sustem and it started to make a terrible buzzing noise from the transformer. I discivered that it was caused by the tumble dryer. When I turned it off, the noise went away. So DC on the mains is quite an issue. I will build a mains isolator in a separate box sometime. might be interesting to actually detect the DC, and then switch in the DC isolator when needed.

AndrewT · 2011-10-04 12:30 pm

He has referenced westhost in his list.
A high proportion of those links are connected to commercial offerings.
I doubt any of these are useful to builders who want to solve the problem themselves.

Ron AKA · 2011-10-04 2:28 pm

Nico Ras said:
And your conclusion?

My conclusion is to use a standard laminated core transformer.

AndrewT · 2011-10-04 2:33 pm

They hum and buzz as well.
It's not just a toroid trait.

I have found EI to be noisier than toroid in general (by my averages of my stock and experiences).

Ron AKA · 2011-10-04 5:27 pm

Quote from Bryston:

"Because toroidal cores are so efficient, they sustain magnetic flux easily. DC offset on an AC line will eventually cause the toroidal core to go into saturation by continually driving the core into one direction. This sounds bad both literally and figuratively...An EI transformer core has unavoidable gaps. These gaps are generally bad, both for stray magnetics and for overall efficiency. But, they do help here in a passive way. An EI core can’t sustain a magnetic field for long and any additional flux is sprayed out of it’s many gaps."

Net result - EI standard transformers do not have this high susceptibility to DC.

Nico Ras · 2011-10-04 9:11 pm

AndrewT said:
They hum and buzz as well.
It's not just a toroid trait.

I have found EI to be noisier than toroid in general (by my averages of my stock and experiences).

Hi Andrew, you are right that EI cores can be noisy as well, but you will find some bad manufacturing qualities, they are either missing a few laminations, windings are not tightly wound due to incorrect slack adjustment on their winders and the core is not vacuum impregnated to fill the air gaps and keeping the laminations from vibrating. Toroids can also suffer from the same bad manufacturing practices as EI cores.

When we took over Hawkins a few years ago, I had an interesting experience. The was a black gentleman who has wound toroidal cores in the company for some 37 years - that is is job. But while he wound the core he continually tapped it with a hammer. Let me assure you he is not a scientist he never went to school.

In my varied and wide stretching engineering intellectual approach I asked him why stuff up the core like that. He looked at me and said when you learn about my job you will see I am doing the right thing.

He explained that when he taps the coil as it rotates he is arranging the magnetic particals in one diredction as well as stress relieves the material - so where is the magic, I asked him to make on core without performing this stupid ritual and made two transformers. The one he tapped was less noisy, magnitizing current was 9 mA while the untapped one was noisy with 110mA magnetising current - I was bowled over.

tiefbassuebertr · 2011-10-05 1:28 pm

........................................................................................................................................................

Quote from [url said:
http://www.dalitech.com/Resources/Measuring%20Acoustic%20Noise%20Emitted%20by%20Power%20Transformers.pdf[/url]
(If the URL don't go open, go to post #155 and #157 about
http://www.diyaudio.com/forums/solid-state/2080-dc-filter-16.html
for the completly article)

WHY AND HOW POWER TRANSFORMERS PRODUCE ACOUSTICAL NOISE ??
Transformer noise has two sources:
- winding vibrations and
- core vibrations.
The single most effective way to reduce windings noise is by having a good quality controlled winding process when assembling them.
This research focusses on the cores of normally silent transformers, which make noise under adverse mains conditions. Transformer cores can become noisy as well under specific secondary load conditions which can be translated (transformed) into the adverse mains conditions at the primary as discussed in this paper.

There are three physical phenomena that produce noise in the magnetic core:
[1-16, Tadashi Sasaki et al, "Magnetostrictive Vibration of Electrical Steel Sheets under a Non-Sinusoidal
Magnetizing Conditions, IEEE Transactions on Magnetics, Vol. Mag-23, No. 5, September 1987, pp. 3077-3079]

1. The movement of the 90-degree Bloch walls inside the magnetic domains, frequently called Magnetoacoustic Emission (MAE); see Figure 1.
2. The rotation of the magnetic domains, that is responsible for the bulk magnetostriction; see Figure 2.
3. The Lorentz Force Acoustic Signal (LFAS) causing mechanical forces between laminations of the core; see Figure 3.

MAE occurs in the steep section of the hysteresis loop; see Figure 4. Not much sound is emitted and the bulk magnetostriction is small. The rotation of the magnetic domains is dominant near saturation in the hysteresis loop.
The magnetostriction becomes "large" and the core laminations move considerably, thus generating acoustic noise (see Figures 4 and 5).
The rattling of laminations of the core (LFAS) depends largely on the construction of the core. The EI-type cores are more prone to make noise due to their many separated pieces of lamination which mostly are only sturdy clamped at the four corners.

In toroidal cores the long role of core band is sturdy clamped everywhere due to the mechanical rolling tension and the pressure caused by the winding tension.
In general: magnetostriction, occurring near saturation of the core, is the main cause of the acoustical transformer noise, while LFAS largely depends on the construction of the core. Due to magnetostriction the core vibrates at the fundamental mains frequency and its harmonics and at core resonance frequencies. In this regard it is important to notice that a noisy transformer means that
a) the transformer is badly constructed -or-
b) that the transformer is forced to operate in a magnetic region close to or at core saturation.

The main reason why the transformer is noisy may be a combination of the above given causes.

Anyway, the device has become noisy and the amount of acoustical noise produced should be measured to determine whether or not the produced noise level is acceptable.

............................................................................................................................................................
Who can post the paper from Tadashi Sasaki, mentioned above ??

AndrewT · 2011-10-05 1:35 pm

tiefbassuebertr said:
Quote from
http://www.dalitech.com/Resources/Measuring Acoustic Noise Emitted by Power Transformers.pdf

WHY AND HOW POWER TRANSFORMERS PRODUCE ACOUSTICAL NOISE ??
Transformer noise has two sources:
- winding vibrations and
- core vibrations.

WRONG
acoustical noise from a transformer can have at least four sources.
The two vibrations listed plus overloading that creates a deep growl and flux saturation of the core due to DC effect in bad mains supplies.

No single winding nor lamination can have a deep growl resonant frequency. So there must be another mechanism producing that frequency. The whole transformer is vibrating and resonating within it's support structure.

Thanks for the link to the paper. I will read and enlighten myself.

tiefbassuebertr · 2011-10-05 8:26 pm

The torodial manufacturers actually must use this keywords by google to find the appropriate informations:
"Magnetostrictive vibration" so as "non-sinusoidal magnetizing"
I have found this e. g.:
Magnetostrictive vibration of electrical steel sheets under a non-sinusoidal magnetizing condition
IEEE Xplore - Magnetostrictive vibration of electrical steel sheets under a non-sinusoidal magnetizing condition
Abstract:
Magnetostrictive vibration of some electrical steel sheets magnetized with a pulse width modulated inverter has been presented. Magnetostrictive deformation was measured by a semiconductor strain gauge applied to specimens. The content of higher harmonics in the vibration was found tobe more than that included in the magnetic flux. High frequency magnetization superposed on that of low frequency yields a relatively large amount of the deformation in the specimen. In spite of a change of magnetizing frequency of the PWM inverter, some components of the vibration remained at particular frequencies. The cause of these high frequency vibrations was experimentally confirmed to be a shape resonance of the magnetostrictive vibration. Resonance modes were detected in a ring specimen and discussions were made with reference to theoretical mode analyses using finite element method. These results allow us to confirm that the magnetostriction in electrical steel sheets is one of the main causes of high frequency vibrations in electrical machines.

More URL's to this topic:

Measuring Acoustic Noise Emitted by Power Transformers
ingentaconnect Acoustic noise emitted by passive components in magnetic devices ...
AES E-Library Measuring Acoustic Noise Emitted by Power Transformers
Acoustic noise emitted by passive components in magnetic devices and design of a low-noise industrial inductor
Emerald | COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering | Acoustic noise emitted by passive components in magnetic devices and design of a low-noise industrial inductor
Analysis of Structural Deformation and Vibration of Electrical Steel Sheet by Using Magnetic Property of Magnetostriction
Advanced computer techniques in ... - Andrzej Krawczyk, Sławomir Wiak, Ivo Doležel - Google Bücher
Magnetostriction and its contribution to noise in a PWM inverter fed induction machine
Document
A Novel Method to Detect Single Phase Reactive Current under Non-Sinusoidal Voltage
A Novel Method to Detect Single Phase Reactive Current under Non-Sinusoidal Voltage-????????
Structure deformation analysis caused by magnetostriction of magnetic steel sheet using combination of Electromagnetic and Structure analysis
IEEE Xplore - Sign In
Magnetostriction and motor vibration
Magnetostriction and motor vibration - Kamimoto - 2004 - Electrical Engineering in Japan - Wiley Online Library
article overview:
A Novel Method to Detect Single Phase Reactive Current under Non-Sinusoidal Voltage-????????
"Magnetostrictive vibration" "non-sinusoidal magnetizing" - Google Scholar

Very interesting seems to be this pdf file (249 pages and many pictures):
Modification of Magnetic Properties of Non Oriented Electrical Steels by the Production of Electromagnetic Devices
http://lib.ugent.be/fulltxt/RUG01/000/811/316/RUG01-000811316_2010_0001_AC.pdf

tiefbassuebertr · 2011-10-05 9:39 pm

By post #80 about
http://www.diyaudio.com/forums/powe...buzzing-toroid-transformers-what-right-8.html
you will find all files about the various circuit topologies found by the mentioned threads from my post #1.

tiefbassuebertr · 2014-01-24 5:49 pm

tiefbassuebertr said:
........................................................................................................................................................

............................................................................................................................................................
Who can post the paper from Tadashi Sasaki, mentioned above ??

I am still looking for follow paper:
There are three physical phenomena that produce noise in the magnetic core:
[1-16, Tadashi Sasaki et al, "Magnetostrictive Vibration of Electrical Steel Sheets under a Non-Sinusoidal
Magnetizing Conditions, IEEE Transactions on Magnetics, Vol. Mag-23, No. 5, September 1987, pp. 3077-3079]

CharlieLaub · 2014-01-24 6:01 pm

tiefbassuebertr said:
I am still looking for follow paper:
There are three physical phenomena that produce noise in the magnetic core:
[1-16, Tadashi Sasaki et al, "Magnetostrictive Vibration of Electrical Steel Sheets under a Non-Sinusoidal
Magnetizing Conditions, IEEE Transactions on Magnetics, Vol. Mag-23, No. 5, September 1987, pp. 3077-3079]

I sent you a PM about this.

tiefbassuebertr · 2014-02-26 8:23 am

CharlieLaub said:
I sent you a PM about this.

thank you very much for your efforts.

Bonsai · 2014-02-26 9:51 am

Well, I am living in Taiwan now and when the washing machine goes on some cycles, all my big torroid fed power amps buzz. Wierd.

So, I will soon build a DC blocker and then share what I did with you.

tiefbassuebertr · 2015-04-26 10:55 am

on the thread
http://www.diyaudio.com/forums/powe...d-transformers-what-right-17.html#post4305992
are to find new observations concerning the remove of unwanted hum.

Frederico Acardi · 2021-01-31 3:42 pm

Thank you for this, I know it's a bit late in 2021.
Is there a definitive "best" method ?

I have read most of the post and none of them a real 100% "this is how it's done".

TonyTecson · 2022-06-15 3:33 am

any transformer that is not overloaded will run quietly...if properly designed and mounted..

you should see how the Japanese made their power traffos used in their audio amps...Luxmans, Sansuis and others...

DC Filter against humming Toroidals - Summary of the most interesting Comments

Member

Member

Member

www.hifisonix.com

R.I.P.

Member

R.I.P.

Member

Member

Member

R.I.P.

Member

Member

Member

Member

Member

www.hifisonix.com

Member

Member

Member