voltage dropper help

[TrickyNekro]

Hmm that's a little bit bad.

I can help you but I don't have the time to do so (I got exams in 3 weeks or so)
And it'll take me a week or so to have it ready and working.

It's basically calculations, I did give a look at MC44608, cause it's quite a popular IC, and it's not that hard.

It depends a lot on what you want (feedback or not and other stuff)
Of course, I don't know how well it may work without feedback (it may drift with time)

Still, what I can do is to help you out wherever you don't get something.
What are you studying exactly? Give a look at the schematics on the last page of MC44608 datasheet.
And let's start with what you don't understand there.

[aruna1]

what I dont understand is how to calculate number of turns in transformer (in all coils)
i found this calculator in internet

http://www.daycounter.com/Calculators/Flyback-Transformer/Flyback-Design-Calculator.phtml

problem to use this calculator is I dont know the Al values of cores. I went to the local shop,they had these ferrite transformer cores (no model numbers as always), so I cant figure out Al value

[waltr]

As I said Switching power supplies are not trivial. I made many mistakes when I first started designing them and it has taken several years to get to a point where I almost know what I'm doing.

Try to get one of the Marty Brown books I linked to above. His books are much, much more clear in explaining SMPS than any other book or app note.

I did a quick look at the typical application circuit for the MC44608 (figure 18) found in the data sheet. That circuit is fairly typical of a SMPS using a transformer. Any thing in particular you don't understand or are you just totally overwhelmed?
Try reading the data sheet piece by piece and relation each to the example circuit.

SMPS do seem to have a bit of 'black magic' in their design but can be understood.

[waltr]

what I dont understand is how to calculate number of turns in transformer (in all coils)
i found this calculator in internet

http://www.daycounter.com/Calculators/Flyback-Transformer/Flyback-Design-Calculator.phtml

problem to use this calculator is I dont know the Al values of cores. I went to the local shop,they had these ferrite transformer cores (no model numbers as always), so I cant figure out Al value

You must know the un-gapped core's AL. Does your school have an inductance meter? If so then wind 10 turns on the core and measure the inductance. Then calculate the AL.

[aruna1]

oh thats too bad, school has no inductance meter. however i ordered one from ebay about two weeks ago but it seems delayed.

until I get it;
is it possible to use dimensions of unknown core and use Al value of a known  core (from a manufacturer core list) with a similar dimension

[TrickyNekro]

problem to use this calculator is I dont know the Al values of cores. I went to the local shop,they had these ferrite transformer cores (no model numbers as always), so I cant figure out Al value

Hmm, that's not hard at all... What RL circuits are good for... Even better what RLC circuits are good more...

Wound up 50 turns and build a simple Resistor - Coil filter circuit (high pass would be better)...
Better precision can be achieved with a Resistor Capacitor and Coil band pass filter.
The lower the resistance, the more narrow the band will be. Current delivery capabilities of your function generator not be a problem.
But do NOT forget the resistor... Or else you will be shorting the function generator every time your frequency is different from the oscillation frequency.

So... you can see what the inductance is... you already know the number of turns, and then go ahead a calculate anything else...
I don't remember the formula... But it won't be hard to find... google around

SMPS do seem to have a bit of 'black magic' in their design but can be understood.

Ahh, there is no black magic, just knowing what the heck you are doing

[TrickyNekro]

You must know the un-gapped core's AL. Does your school have an inductance meter? If so then wind 10 turns on the core and measure the inductance. Then calculate the AL.

Inductance meter? Shame on you! Just a capacitor and a resistor...

Cause, his school just can't fail at having a function generator...


Edit: Uuupss... you are gonna need a scope too, but generally, where there is a function generator, there is a scope too

[aruna1]

@nitro,
Thank you for the advice, I guess I will have to wait till next week as university starts on next monday.

anyway what is your idea about using Al value from known core with similar dimension to unknown one

[aruna1]

If I can figure this out I'm definitely going to write  a tutorial on "flybacks for dummies" so dummies like me will have less hedache over these 

[TrickyNekro]

anyway what is your idea about using Al value from known core with similar dimension to unknown one

Dimensions really don't play a major role in this (they do...but)

That's cause a core with relevant permeability of 4300 will have let's say an inductance of 4mH with 48 turns
and a core with relevant permeability of 2300 will need 65 turns to have the same inductance.

Since you are switching the coil on a certain frequency, that's gonna affect the current through your coil.
And the current in your coil is gonna affect the strength of your magnetic field in the core.
Note that you can't impose a current on a coil, as you can't impose a voltage on a capacitor.
Further more the less current the less power stored in the core, and the less can be extracted from it on the other side.

So it's not all that of black magic in the end. It's what you need and what you get...
In the end, it's a golden ratio that you should keep. As from what you get the energy at the second coil is the magnetic field in the core
and depends on both the current that runs the first coil, but also the number of turns...

Best Regards, Lefteris
Greece

[aruna1]

oh i see 

[aruna1]

ok I made a flyback converter using MC44608
40KHZ speed

applied power, and booooom Ic blew 

[waltr]

applied power, and booooom Ic blew  Grin

Yep, been there.

Try again but go step by step
Start with a lower voltage to only power the MC44608 and use a resistor instead of a MOSFET.
The goal is to get the MC44608 to free run without feed-back and then measure the gate drive output with a scope. Ensure that the wave form is at the correct (calculated frequency) and the duty cycle looks correct.

Then add the MOSFET and use a resistor instead of the transformer. Ensure with a scope that the gate and Drain waveforms are correct.

Keep adding in parts and measure at each stage until the the PS is working.
Start with a high value current sense resistor, MOSFET Source to ground and keep the ground path very short and close the the IC's ground reference pins.

As you proceed keep us posted with what you find and hopefully some of us can help out.

[aruna1]

well power supply has to wait. I was able to find a 50Hz transformer small enough for the proect, so I'm working on otherparts of the project. hopefully in future I will try this again