diyAudio (
-   Power Supplies (
-   -   Help my transformer getting hot quickly (

darw82 30th December 2005 11:06 PM

Help my transformer getting hot quickly
I build a push pull converter.
(10V - 15V) to 12V. 120Watts

Everything works very well. Only the transformer. It getting hot very quickly, with only 30W load.

The transformer is
TC-42212 from Mag-Inc.

Ae=0.511 cm2
WaAc=0.756 cm4
Ve=2.77 cm3

Primary winding 4 + 4.
Secondary winding 6 + 6

What's the problem? Can anybody tell me?


Danko 30th December 2005 11:29 PM


What is the switching frequency?

Have you tried increasing the primary turns?

darw82 30th December 2005 11:54 PM

The switching frequency is 100Khz.

I have tried increasing the winding.
primary (8 + 8 )
secondary (12 + 12)

But it's still very hot with light loading.

Danko 31st December 2005 12:03 AM


What kind of wire(s) do you use?

darw82 31st December 2005 12:05 AM

Danko, thanks for your help.

The wire for primary and secondary is (0.4mm * 4)

Danko 31st December 2005 12:14 AM

The 4x0.4mm wire seems a little bit too small, for me.

I designed an SMPS, which operated from 50VDC, and the primary current was about 6-8A. A wound the primary from 17-18 piece of 0.4mm wire.

You write, that output power is 120W, from 12V. That's 10A.

Try increasing the number of the wires. One single wire's diameter is good, becouse at 100kHz the penetration depth is about 0.21mm. Skin-effect rules :-)

Eva 31st December 2005 12:37 AM

Do you mean a 100Khz oscillator, or a 100Khz effective frequency through a 200Khz oscillator?

darw82 31st December 2005 12:47 AM

The controller is SG3524. Running at 200khz.
In push pull mode, this IC will divide the frequency. so it would be 100Khz effective frequency through a 200Khz oscillator

amp_cy 31st December 2005 01:54 AM

Please supply complete code number, there should be several other characters which will inform you of the material type. If there is a J or W or an H, these stand for 6040u, 12080u and 18100u permeability respectively. This type of core are suitable for common mode chokes, broadband transformers, pulse transformers and current transformers.
If there is a R or P or an F, these stand for 2770u, 3020u and 3624u permiability respectively. This type of core is good for high frequency transformers, i.e. your application.

If you have one of the very high permiablity cores J,W or H, you have a very good chance that you are saturating the core if your ON period is long enough (what is your operating frequency?). Check the current waveforms at the switching devices, if there is a large exponentinal peak (looks like a spike), you are saturating the core if this is the case.
You could have a flux imbalance problem, are you using current mode control or a series pass capacitor on the center tap of your primary?

More data is needed to solve your problem, send the complete core code, control method and operating frequency.

P.s. check this page for data sheets on you core.

Eva 31st December 2005 03:45 AM

Calculations show 100mT peak for 12V, 6 turns and 100Khz effective (Ae=0.51cm^2). This shuld be ok for most materials so the problem may be due to flux imbalance.

All times are GMT. The time now is 12:31 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2