ferrite core transformer

[yadnesh256]

Hello sir,
My name is Yadnesh Samant. Currently i m making BE project on Isolated SMPS .

we need a FERRITE CORE TRANSFORMER with following specifications:-

Shape=>E-I core (ferrite - core transformer)
Vin= 7.5 volts to 8 volts
Vout1= 7.5 volts
Vout2=5 volts
frequency=52Khz
Nprimary= 17 turns
Nsecondary=17 turns
Nauxillary-secondary=12 turns ( 1 primary and 2 secondary)

BUT WE ARE NT GETTING IDEA ABOUT HOW MANY BOBBIN PINS AND CORE AREA SELECTION..... I DNT KNW HOW TO CALCULATE THESE..
CAN NE 1 HELP ME OUT WITH THIS.... I M NT ABLE TO FIND APPROACH ON INTERNET...
IT WILL BE KIND OF U.

 

[andre_luis]

Yadnesh ,

You forgot to specify circuit topology.
Take a look here :




+++

 

[dustmote]

Wow, I must say andre_teprom, that is one of the handiest websites pertaining to SMPS design that I have ever seen!

The only thing that is a little strange at first notice is their terminology regarding certain forward converter topologies. In most of the "classical" books and references I have perused, the term "Push-Pull" generally refers to a transformer isolated forward converter layout where there is a center tap on the primary winding of the main transformer.... This effectively turns the primary winding into two logically distinct windings--IE they are carrying only half of the total power flowing through the converter each....

Or maybe my understanding of the basics is out of wack? But I don't see any topology employing a 'center-tapped' primary winding on that website...

Curious........

 

[andre_luis]

I also didn´t see the PUSH-PULL topology, wich is very commonly used.
However, I already performed a calculation to one of these topology that worked fine, and this is the reason I sugewsted this page.


+++

 

[yadnesh256]

i m using buck converter.... where after switch (transistor) i m using transformer for isolation...it is not a exact forward converter....
my Vinmin=10V, Vinmax=40V, switching frequency=52Khz, Vout=5V, Iout=1A

I am using TL594 as a feedback pwm controller.... i am not able to decide bobbin geometry, core area of ferrite transformer...or can i use any other transformer...

PLZ help...

- - - Updated - - -

i asked a dealer of ferrite transformer...they are asking me about bobbin pins and core area... i m not able to decide these parameters...can any 1 help...
Regards,
Yadnesh

 

[FvM]

I presume you are able to count the number of required bobbin pins yourself. Usually, primary and secondary are using opposite rows to have sufficient clearance according to intended isolation strength.

The essential point is calculation of core area. It's based windings voltage integral and core flux according to acceptable core losses, a flux up to 200 mT is suitable for 50 kHz and ususal power ferrites.

 

[szhighstar]

please advise current(Irms), so that you can decide which core and bobbin you can select, thanks!

 

[yadnesh256]

no.... i am not able to calculate the bobbin pins, core area,
i need to design the whole ferrite transformer....

current through input transformer is 1A...

i m attaching my circuit diagram...plz take a look at it

 

[dustmote]

In that circuit diagram, It appears you are using a 47 Ohm Pull UP resistor to Vin+ attached to the gate of the switching transistor???

What is the idea here? I thought that one generally used a pull down resistor to ground in order to discharge to gate when gate drive voltage was removed so as to aid in quickly and completely 'turning off' the transistor?

Or am I confused about it all?

--------------------
I have embarrassed myself with my ignorance in asking a question that should have been trivial if I had done more reading... Of course it is a pull up resistor to Vin+, it's a PNP MOSFET and they are driven low at gate in order to "turn on" rather than the NPN which is driven high at gate to "turn on". I am still very new at this and without any schooling. It seems that I so often encounter the use of NPN Mosfets rather than their PNP counterparts that I was ignorant of this plain and simple fact.

Alas, maybe somebody can learn from my mistake. That is what these forum boards are for anyway I suppose :grin:

 

[FvM]

Or am I confused about it all?

I think so. Th eresistor does exactly what you are discussing, quickly discharging th eswitch transistor. Notice that's a PNP transistor.

There are however various faults involved with the schematic.
- diode/ inductor series connection. A second diode is missing here
- output voltage measurement/optocoupler circuit

 

[mtwieg]

Yeah it looks like you're trying to implement some sort of forward converter, but it's missing a lot of necessary parts. On top of missing the freewheeling diode on the secondary, it has no means of circulating the primary magnetizing current. Also I can't see the purpose of the other secondary winding, it doesn't seem to have any useful function.

For core parameter calculation, you just want to pick a maximum flux swing in the core (200mT is a pretty safe start), then use the volt time product to go from there:
V*t=N*dB*A. Where V is the applied voltage, t is the duration of the applied voltage, N is number of turns, dB is flux swing (in tesla), and A is core area (in square meters).

 

[dustmote]

Also, here is a very handy link about transformer design in SMPS....
**broken link removed**