Boost converter

[Sponkii]

Hey..

I have some interest in building a boost converter.

My demands are following:
Input voltage 13-14.5V DC

Output voltage 19 at 5A. i do not think that a small amount of ripple is a problem.

The but!
i have never had experience with boost converters before.
i have a couple of different coils and toroids that i hope i can use.

i had an idea of using something like 80khz to drive, but i have seen in other forums that you use around 50khz in SMPS?...

what would you recommend?..

I have enclosed a design idea.
- made something like that on breadboard, but uses voltage reg on input (which get hot), it outputs from Vin-0.5 to 27.8V the transistor is BUZ20, cant remember name of diode, but it should be fine, have also tested different coils.

[Sponkii]

I have already tested without regulator on a 12v gel battery.

switching freq is around 120khz

works fine with a computer fan as load.

right now my max voltage is around 17v. 230ma

and at 12v it is 130ma.
cant remember the usage with no load at 17V.

if i post a picture of the coil can some of you possibly tell me if it suitable to use as my coil for the real booster?

[raudio1969]

I would suggest using the National Semiconductor LM3478. I used it before to boost 12 volts to 20 volts at about 5A for a former employer.

http://www.national.com/pf/LM/LM3478.html

- Richard

[Sponkii]

It looks really promising, but i would like to do it analogue.

mainly because i have made a breadboard that works, i already have the needed components, but also to have a better understanding of what is really going on..

the only thing i really need to know it which switching freq would be suitable, and which inductor i should use..

but i will remember that component as it is easy to use later on..

[mflorin]

The choice of switching frequency is a function of different factors that you want to consider in your design. Usually frequencies from 20kHz (above audio) to 1MHz or above can be used.
The thing is - the higher the frequency, the smaller the inductor (magnetic components). However, as the switching frequency increase, the switching losses becomes more and more significant....
A good compromise between switching losses and inductor size is 100kHz.

[Sponkii]

thx.. some of that i already knew..

it was more if some one could give a more specific freq range, like 40-50...

but the freq also depends on the inductor right?..

- is there a easy way to find the optimal switching freq of my inductor.
I have a scope and function generator at hand, and a good multimeter..

[mflorin]

I recommend you some books where you'll find your answers:

Fast Analytical Techniques for Electrical and Electronic Circuits - Vatche Vorperian
Switch-Mode Power Supplies Spice Simulations and Practical Designs - Christophe Basso
Fundamentals of Power Electronics (2nd edition) - Erickson & Maksimovic
Switching Power Supplies A to Z - Sanjaya Maniktala (I love the style is written)

[darkfenriz]

Using a just comparator at feedback is asking for problems IMO.
You rather need a limited gain with lowpass to create a more "analog" feedback signal.

P.S. I hope what you have done is either keep mosfet off or switch at 50 duty cycle...

[Sponkii]

Something more like this..

which kind of problems do you thin it will result?..

[Sponkii]

The reason i have added the zener (not a 6v but a 5.something Volt)
is to prevent the duty to be over 50% (actually lower with a lower value zener)

so if the feedback fails or there is overload the duty will not exceed a certain level..