Two transistor 5V switching regulator

[William At MyBlueRoom]

Found this article, looks great if it works. I'll have to prototype it in the next day or two, it's cheaper than the LM2575.

Anyone used such a circuit before? Comments?

**broken link removed**

 

[Hero999]

I've seen it though I've never needed to use it, the three transistor version with current limiting looks like it could be modified into a nice little constant current PSU for powering high current LEDs.

 

[steev]

well it may create stability problem.
bcz it uses inductor,means circuit is osilating giving less efficient output.
i still vote for LM317 or 7805

 

[Papabravo]

Are you saying or implying that all switching regulators are unstable? Switching regulators are inherently non-linear systems and our notions of what constitutes stability are different from linear systems. In a linear system the root locus technique showing that the roots of the characteristic equation are confined to the left half plane is sufficient to demonstrate stability. In a non-linear system there is no useable transfer function nor characteristic equation.

Our notion of stability, for a non-linear system, centers around the outputs being bounded for any initial condition and forcing function. This notion is easy to articulate and somewhat more difficult to apply.

 

[Paul Obrien]

https://www.fairchildsemi.com/ds/BC/BC328.pdf has the collecter current at a max of 800ma, also I find the lm2575 has a very neat support circuit making for a very compact power supply. Of course the circuit oscillates, wouldn't work if it didn't

 

[Hero999]

well it may create stability problem.
bcz it uses inductor,means circuit is osilating giving less efficient output.
i still vote for LM317 or 7805

Google about switching regulators, you'll find they're more efficient than linear regulators like the LM317 or 7805.