# help with handwound transformer

Discussion in 'General Electronics Chat' started by lokeycmos, Jul 14, 2011.

1. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7
im in process of building this ESR meter : http://ludens.cl/Electron/esr/esr.html

i have some questions about the transformer. it is a 400:20 ratio. he specifies that he is using a EA-77-188 core. i attached a pic of the cores that i have available in my parts bins. the smallest wire i have in my arsenal is 22 guage with plastic insulation. is the core type critical? could i use one in the pic? as far as turn ratio could i do something smaller like 200:10 or even 40:2 so i can fit my wire on the core ??

thank you!

File size:
288.3 KB
Views:
65
2. ### Bernard AAC Fanatic!

Aug 7, 2008
4,310
432
It would be helpfull in estimating sizes if a scale of some sort is included with picture. Look for smaller gauge magnet wire.

3. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7
hope this helps

File size:
292.4 KB
Views:
43
File size:
290.8 KB
Views:
35
File size:
290.8 KB
Views:
33
4. ### someonesdad Senior Member

Jul 7, 2009
1,585
141
Since it's the ratio that matters, I'd bet you'd be OK using fewer turns and keeping the needed ratio. It's easy to try. As suggested, you can also use magnet wire (try e.g. All Electronics).

5. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7
one concern i have with going with smaller turns and the same ratio is that with less turns there is less impedance on the primary. would this cause more loading or am i totally wrong?

6. ### Adjuster Well-Known Member

Dec 26, 2010
2,147
300
Yes, the inductance would be less, and remember that if a coil is wound with N turns the inductance varies as N squared, not just N. You could get away with rather fewer turns if you could get hold of a core with a higher specific inductance (Al) value: L=N$^{2}$*Al

The Amidon EA-77-188 has Al=1060nH, so a 200 turn winding would come out at 200*200*1060nH or 42.4mH

https://www.amidoncorp.com/specs/2-40.pdf

7. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7
good info. is there a way to tell the AL of a core without a specsheet? right now im pulling small transformers out of SMPS's and using inductance meter to find ratio =sqr(priL/secL) hoping to get lucky and avoid winding that many turns. just for the record if i end up winding my own, could i use a ferrite rod from am radio?

8. ### Adjuster Well-Known Member

Dec 26, 2010
2,147
300
You would need to wind a test coil on to the core, and then get hold of some means of measuring inductance. One way might be to make the coil into a resonant circuit with a capacitor, and then measure the resonant frequency. This would require access to some test equipment.

It would be difficult to get 40mH out of a ferrite antenna bar without winding loads of turns: even a long-wave antenna coil would only be something like 3mH.

Also, this kind of open core would only give somewhat loose coupling between primary and secondary, but would tend to cause all kinds of stray couplings to other things - after all it's specifically designed for that.

9. ### debe Well-Known Member

Sep 21, 2010
961
186
Looking at the constructors unit & how its described I would be going for a simmilar E core ferrite T/F. Heres a pic of a few ive salvaged from various powersupplys for re use, some have an air gap which he doesnt recomend. in the pics the top 2 dont have an air gap & probably would be quite suitable. One way of dismantling these T/Fs is soak in paint stripper for 24Hrs, be prepared to crack the cores on some this way. The other way is cook them in a Microwave oven for a minute or 2 this is how i do it & works well. Warning dont use a good oven as its hard on the Magnetron I use an old oven that i picked up for free. I have used salvaged cores like these & rewound them for various low voltage SMPS projects quite succesfuly.

File size:
268.5 KB
Views:
74
10. ### SgtWookie Expert

Jul 17, 2007
22,183
1,732
I don't know how anyone would get that meter to work, as the TL06x, 7x, and 8x series can't "see" within 1.5v of +v or within 3v of -v. Since the supply is only 5v, that leaves 0.5v for the usable I/O range - however, they've biased the inputs at Vcc/2, so that's 2.5v - out of the common mode range.

The circuit is using an opamp out of its' specified operating range.

11. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7

could you elaborate on that? are you saying it wont work? this is rather disappointing. i spent a lot of time unsoldering parts from junk circuit boards.

12. ### SgtWookie Expert

Jul 17, 2007
22,183
1,732
If you look at the datasheets for the TL06x series, you will see that the common mode range is from -v + 3v to +v -1.5v. So, if your supply voltage is 5v, you only have 3v to 3.5v as the entire input and output range.

That family of opamps was intended for use with dual rail supplies.

But, the way they used the 100k resistors to bias the opamp inputs at Vcc/2 will work if a single supply is high enough in voltage. If it doesn't work with a single 5v supply, try increasing the supply voltage to 10v.

13. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7
transformer update! i was going through my box of computer power supplies unsoldering transformers. i came across one and with using my inductance meter and the formula sqr(priL/secL) it has a 18.588:1 ratio. when compared to the 400:20 in the schematic it is 371.75:20 this is where i need the experts opinion on whether it would be a good substitute.

File size:
259.8 KB
Views:
32
14. ### billbehen Active Member

May 10, 2006
39
1
Depends on the application! Is this a power supply or other power circuit, or an RF circuit, or audio isolation? Wire size/length determine DC resistance & thus power handling capability. Fewer turns or lower perm core tanslates to lower impedance, but how much you need depends on what its hooked up to....

Check the DC winding resistance and calculate I^2 * R to see if will handle the current!

15. ### lokeycmos Thread Starter Active Member

Apr 3, 2009
432
7
application is the 1st post on page 1 of this tread