0
\$\begingroup\$

I have several transformers for audio amps, with single thiner primary and dual larger area secondary windings, for lower voltage higher current. They are toroidal, all by different manufacturers, but all producing audible noise, and i thought something is wrong with their construction. So i put off a part of insulation to look at windings - they were not sticked by epoxy - and that was a source of sound most probably. But then i was surprised.

All of transformers expose similar winding style - fatter copper wire (2 secondaries were winded inside, while much thiner primary mains winding was winded outside, and its turns are wrapped kinda randomly, awkwardly. I could see it on all my devices. I thought that thin wire should be wrapped directly on the toroid code, to save space and to make accurate round turns. and secondary should be turned ourside, to "catch" all EMI, emitted from inner primary, so reduce magnetic HUM leak.

Am i right? Or why several transformers have mains primary coil outside secondary? Is order of any matter and is it worth rewinding and sticking windings with epoxy?

\$\endgroup\$
  • \$\begingroup\$ The transformers are on an iron or ferrite core which will tend to concentrate all of the magnetic flux in the center, so there's no benefit to "catching" any of it by putting the secondary outside the primary. \$\endgroup\$ – C_Elegans Dec 3 '18 at 0:17
  • \$\begingroup\$ Ideally the coils in any Transformer are going to be mixed up so that the linkage between the magnetic flux changes induces the most electromotive force in the secondary coil. The primary is wrapped around the outside because it is completely isolated and therefore is just a 120 volts flowing through a conductor which generates heat and it needs to be cooled. But that is really only concerning power Transformers. in a speaker cabinet the power to the voice coil is coming mainly from the amplifier and the voice coil is likely 8 ohms so it's voltage is small, the secondary provides that voltage \$\endgroup\$ – Danny Sebahar Dec 3 '18 at 1:01
0
\$\begingroup\$

Transformers can create noise directly. The magnetic fields at high power levels exert forces on the wires and the cores causing mechanical vibrations. This noise does not come out of the speaker, it comes directly from the cores. It sounds like metal vibrating. Arcing in the transformers can also cause noise, which you might see if you run it in a dark room with the covers taken off. Since you tried many cores, I would probably rule the arcing out. If the noise is heard through the speaker, and there is no arcing, than your problem is probably not the transformers.

\$\endgroup\$
  • \$\begingroup\$ The noise goes from the transformer itself, not from the speakers. Also noise goes almost without load (15W), despite my power transformer is designed to feed 2*TDA7294 and visually(its core is D=900,d=500,h=500mm) is quite capable of doing 200-300W. But wires are not glued with epoxy and i can move copper wires by my finger a bit. Is it worth rewinding and epoxying? \$\endgroup\$ – xakepp35 Dec 6 '18 at 10:29
  • \$\begingroup\$ First make sure that there is no loose metal in the mounting hardware, casing, etc. I don't like the idea of the thinner wire spanning over any gaps in the thicker wire such that it has a span in mid air. That seems like a bad design if it does this. Can you share a picture? \$\endgroup\$ – skeptonomicon Dec 7 '18 at 15:33
-1
\$\begingroup\$

I think you are mistaken on how current flows in the presence of a changing magnetic field. It's perpendicular to the magnetic field lines and orthogonal I guess in a 3rd dimension 2 the force in Newton's that is imposed as a reaction (which causes actuators in an electroc motor to spin, ie, a current going straightforward produces a mechanical force going to the right and a magnetic field line going up and around you in a Halo, not in front of you. It sounds like you are thinking that a toroidal Transformer is creating a magnetic field that is pointed out the hole in the middle of it but the magnetic field is actually pointed along the body of the toroid and the primary and secondary windings are wrapped around it because the toroid is ferromagnetic and increases the magnetic flux linkage. The mechanical force is what is pointing out of the toroid's empty Center. In order to get the most linkage between the induced voltage in the secondary caused by the current flowing through the primary they need to be wrapped is closely together as possible and are frequently intertwined separated only by the shielding on The wire.

Secondly, here are more reasons why it's not coming from your Transformer. It's not moving physically and any decent power supply is going to be filtering that waveform. Are you talking about noise like Johnson noise which comes from the movement of electrons due to heat? Or shot noise due to the statistical probability of the quantity of electrons flowing in a current being not exactly the average but within some standard deviation around the average current, or 1/f noise that has no known source but has a frequency response that decreases with frequency or are you referring to an offset voltage or current that is not being rejected by the amplifier's CMRR or PSRR?

It sounds like you are having problems with interference not noise. Audio circuits are meticulously engineered to avoid noise and they don't generally conduct large amounts of voltage or current until it's at the speaker so you likely don't get very much noise from heat or from a large current flowing, but rather there is an interference being picked up on the signal path that is being amplified.

Transformer fluctuations can be due to Eddy currents which don't really occur in toroidal Transformers because the toroid is not a solid conductor but rather individual ferromagnetic moments suspended in a clay like resistive material. Also because the toroid is not a solid conductor it doesn't saturate, or rather, the magnetic flux is not "clipped" buy the material's permittivity.

If you're getting noise in an audio amplifier the instrument is very likely picking up interference from everywhere, same with the microphone, and mixers and good digital equipment usually have Transformer coupled outputs and inputs because that reduces the noise on the signal. if the inputs aren't coupled by Transformers there coupled by capacitors and will have some sort of common mode rejection for interference.

Those big amplifier toroidal Transformer sitting in the bottom of speakers are positioned such that their magnetic field is rotating around the toroid at the bottom of the speaker cabinet and not in parallel to the magnetic field of the voice coil magnet that is creating the sound so that they don't interfere. that principal is also adhered in the design of audio crossover circuitry (usually the only other circuit within a speaker besides the power supply) which separate the frequencies for the Tweeter speaker and the subwoofer using only inductors because capacitors are like giant antennas and will pick up interference from any direction whereas inductor is only really subject to magnetic interference because the inductor specifically is designed to oppose any changes in current. which is what interference and noise are.

Here is a photo of an audio crossover circuit you can see two large inductors one facing One Direction so that its magnetic field is perpendicular and not interfering with the magnetic field of the other inductor and the two inductors create a low pass filter and a high pass filter to send certain frequencies to each speaker and they change which frequencies go where depending on which frequencies are present which is cool.

enter image description here

The bottom right inductor is has a magnetic field that is pointed towards the camera lens and the cylindrical inductor on the bottom left has a magnetic field that it's pointing to the top of the photos brain. So they are in two different planes to avoid any interference. Since there are three spatial dimensions the Transformer that is the one you're worried about, the power transformer is in the third plane so that all three magnetic fields within the speaker cabinet are not interfering with each other. Lastly the voice coils giant magnet moves so very little that it does not contribute a magnetic interference because it's frequency is the same as the signal's.

Transformers I like the oldest electronic Component and they have been used in speakers for as long as they have existed and people understand how to make them the best they can be made so it's not your Transformer that is causing your noise. You are picking up interference somewhere along the signal path. It might be in the instrument or in the speaker cables most likely. Speaker cables for having unbalanced signals or having mismatched impedance is full all the grade the Fidelity of the speakers output. But the speaker itself is just a giant magnet with some wires wrapped around it and very likely is not the source of your noise.

\$\endgroup\$
  • \$\begingroup\$ power supply huge transformer (~250W coil) stays apart from coils/speakers/amp. i eclosed it in a separate case. coil is producing physical hum noise, under 15W load (when it powers up 2xTDA7294) when no signal is applied (signal shorted to ground). Despite 15W is a huge idle consumption (probably i have old huge el caps 33k uF, with a moderate-to-high charge leakage), that is different story. main story is - why so big transformer is producing hum and what to do with it? Is it worth rewinding/gluing in epoxy? I can move winding with my finger a little.. \$\endgroup\$ – xakepp35 Dec 6 '18 at 11:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.