Small signal transformer for step down Mains.

Thread Starter

AlexLPD

Joined Aug 22, 2010
60
Hi to you all, Im working in a project requiring 2 things;
Step down the voltage of Mains AC 127VAC to 12VAC or so, I try to use a regular transformer, but it distort the signal ( I will be sending a small 100Khz near the zero crossing of each sine wave, and that is the signal I want to keep.

-------------------------------------------
This is the original project:
https://forum.allaboutcircuits.com/...n-on-a-led-amplify-stage.162725/#post-1434452
-------------------------------------------

So in this another question;
Members of this nice forum point me; I need a pulse transformer, or a ferrite core one, to keep the signal integrity.

So I have to make a job and one of my clients donate my an ATX power source, so I disassembly and retrieve the transformers, and the fer rite inductors, the biggest one is the one I foccus on;


This is the ATX before remove the components.



This is the small, yet the bigger transformer I can remove from the supply, I was hoping this has a 10:1 relation or so, but my measurement differ;



So, they are the inductances and the resitance of the coils, I think It 5296uH and simetrical coils to a common tap o the rigth. ( I messure from the center tap, the bigger wire in the middle, to each of the pins on the transformer)

Sadly I calculate the reactance of the primary winding and I get 1.99Ohm at 60Hz.

So I got this questions;

This small reactance on the 60Hz intended frequency of use, means I can not power this device over the 127VAC that I was needing.
Even If I try to use one of the taps at 198uH I got a Reactance of 74.97Ohm at 60Hz.
Are my calculations correct?
After I calculate the reactance, I just use the formula I = V/R to estimate the Amperage on the primary coil correct?
Since I can use this transformer directly and I just need it for signal ( 100mA or so) would be possible to add some resistor divider to attenuate the mains voltage, still getting the response?

If this is not feasible, I can try to re-wind this core, it has 11 x 11mm at the core, I get this simplificated formula of "Ampletos" site, they use the 42 as constant, I dont know from were yet.
The site: http://www.videorockola.com/tutoriales/como-calcular-el-transformador-para-su-amplificador/

Can please verify this?:

I got the core area;
1.1cm x 1.1cm = 1.21cm2.

Square the core area gives the maxium aviable power; (Ampletos assumption)
1.21 x 1.21 = 1.46Watts.

Number of turns per volt = 42 / core area (Ampletos assumption)
Nt = 42 / 1.46cm2
Nt = 28.76 turns per volt.

Since the smaller diameter wire I got here is 30AWG, I will be using this.
So; primary 127VAC x 28.76 = 3653.42 OMG ------it mus be a joke.
seconday 12VAC x 28.76 = 345.12

The main Idea is to replace the ferrite core, cheap transformer, with a more suitable one, for high frequency, Im aware my calculations may be are wrong due the fact I pull up the formulas from an over simplyfied site, and also I know it may be a comercial transfomer, But this is for me a great learning excercsie, so I like to know more and correct theory and make some practices.

Im a industrial contractor of refrigeration, I work all day with mains voltages of 220 and 480VAC, in a hundred amps ratio, so I know how to keep me safe. Still a newbie on electronics, and far more new in the Analog side.

Thanks for the insights, correction and lectures.
-Alex.
 

Thread Starter

AlexLPD

Joined Aug 22, 2010
60
You could connect the transformer to the mains via a capacitor to allow 100kHz but block 60Hz. The capacitor should be X rated (suitable for continuous connection to the mains).
Yes yesterday I was searching for a supplier of X275VAC capacitor.

In this case, the capacitor in series with the inductance of the coil, will form a voltage divider rigth?
How this is called or where I can read more about it to make some formulas?

-Alex
 

DickCappels

Joined Aug 21, 2008
10,153
Zt = ZL - Zc

I don't think I got that backwards.

The difference between 100 kHz and 60 Hz is approximately 17,000 : 1. You don't have to loose very much signal in that divider.
 

Thread Starter

AlexLPD

Joined Aug 22, 2010
60
This isolation pulse transformer I designed about 50 years ago
especially for use on 220/380 VAC grid:

View attachment 187398

Ferrite core B64290L0045X830,
R 16.0 X 9.60 X 6.30 mm, permeability μ = 4300.
Magnet wire 27 gauge (0.355 mm dia), 20 turns for 1mH.
PVC tube outer diameter = 5mm.
I think I will change for a sinusoidal wave.
The PVC in you design is for prevent the wire from touch the core?

I found core of the same dimensions, and even has 25turns of wire but it's inductance is not near 1mH. Can this due the fact of the ring is yellow? I read some were the ferrite core rings uses the colour to mark it's permeability rigth?

Alex
 

Danko

Joined Nov 22, 2017
1,829
The PVC in you design is for prevent the wire from touch the core?
Not only. PVC tubes guarantee very durable insulation between transformer windings, as required for isolation transformer.
I found core of the same dimensions, and even has 25turns of wire but it's inductance is not near 1mH. Can this due the fact of the ring is yellow? I read some were the ferrite core rings uses the colour to mark it's permeability rigth?
"Unless they are marked with the mix used to make the cores, you have no idea how they will behave. You don’t really even know if they are ferrite or powdered iron. Powdered iron cores are often painted certain colors if they are from Micrometals or Amidon, but unless you know the source don’t trust the color."
Read this article: https://www.nutsvolts.com/magazine/article/July2015_HamWorkbench

Try EMI suppression ferrite cores used on cables. They usually have high permeability.
upload_2019-10-8_0-55-27.png
 
Last edited:

Thread Starter

AlexLPD

Joined Aug 22, 2010
60
Thank you very much Danko, will make the reading and some experiments and come to report some progress.

-Alex.
 

Danko

Joined Nov 22, 2017
1,829
X-10 modules sand and receive 120khz tones on the power line. Here is an example schematic. They are using 100 ohms of resistance and a high voltage capacitor to strip the 60hz off. National has/had white papers on this subject. (X-10)
Nice, but it is transformerless device, feeding directly from mains.:(
 
Last edited:

ronsimpson

Joined Oct 7, 2019
2,986
Nice, but it is transformerless device, feeding directly from mains :(.
Yes, X-10 mostly lives on the mains. (not all do)
If you want isolation it can be done through the transformer.

If you want to do some looking, there is/was a number of ICs for "power line communications". The application notes for these parts are very helpfull.

another example:
 

Thread Starter

AlexLPD

Joined Aug 22, 2010
60
View attachment 187523
X-10 modules sand and receive 120khz tones on the power line. Here is an example schematic. They are using 100 ohms of resistance and a high voltage capacitor to strip the 60hz off. National has/had white papers on this subject. (X-10)
Thank so much for this schematic I try to search for them and star on an application Note I belive it was from On seiconductors, yet the filter they proposes were not very good, in simulation at least.

This is quite amazing !!!
-Alex.
 

Thread Starter

AlexLPD

Joined Aug 22, 2010
60
Precisely yesterday I recibe an order of inductors, coils and ferrite transformers, In soon I will make some experiments
Thanks you very much, this subject is making me learn many things.

-Alex.
 
Top