Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

SEPIC with regulated input current.

Status
Not open for further replies.

Flyback

Well-Known Member
Hello,
The following sepic led driver regulates its own input current.

https://i46.tinypic.com/jttquc.jpg

spec:
vin= 5 to 8v
vout = 5 to 40v
pout = 1 to 5w
has overvoltage protection.
Coupled sepic inductor.
F(sw) = 30KHz - 125KHz

Its peak current mode, but the main negative feedback loop is closed around the input current.
Therefore, from a feedback loop point of view, this is not a sepic converter. It is much easier to stabilise than a sepic converter. -you can see that because it works stable even if the output capacitor is totally removed!

So you get all the advantages of a sepic (output short cct protection, and output voltage can be lower or higher than vin), but without having the feedback loop difficulty of a sepic

So can you confirm that the feedback loop equations for this converter are nothing like sepic converter feedback equations.?
 
It depends what you want. Yes this is still a sepic converter - primarily because of the coupled inductor and the coupling capacitor and the ability to regulate the output voltage when the input is above or below it. Most sepic converters are used to regulate output voltage in which case the loop stability equations are complicated. Most LED driver sepic converters are used to regulate output current. This is an emergency lighting application, where you need to guarantee operation for a certain lifetime off, say, a battery. therefore you must guarantee a certain INPUT current, not output current. If you are regulating on input current or output current, then the loop equations are different (and possibly simpler) than those of a regulated output design, but you are not discovering anything radical here. You cannot use a circuit that regulates input current to provide a fixed output voltage, so this circuit may be easier to stabilise, but no good for the application required

In each of the architectures above, the regulation loop is different, but the guts of the circuit are essentially the same. The equations governing the main 'power' operation of the circuits are essentially the same (it is a sepic), but, yes, the loop stability equations of each will be different.

By the way, either use individual inductors, or a loosely coupled sepic transformer else the input spikes on the current might make your loop compensation complicated (hence defeating the object of your original post). Incidentally, if you remove the coupling capacitor on a sepic, what do you get? Mr Flyback, the answer might be quite close to home....! :)
 
Thanks, though this converter has a tightly coupled trafo and I haven't had stability problems...and the vout tends to get clamped at the led string voltage. I didn't choose flyback because it needs a RCD snubber, and the sepic cap is actually easier than putting in an rcd snubber, -I haven't had problems with input current spikes causing instability.

I find separate inductor (noncoupled) sepics are ok but only if operated at very high f(sw), otherwise they ring like mad (the input current oscillates at the resonant frequency of 2xL and the sepic cap)....the only way to avoid this is to have the f(sw) well above this frequency........so often you end up needing 500KHz for f(sw)
 
... also, from what I can see, the circuit has no protection against LED open circuit. If the LED stack goes open circuit the voltage will ramp up and up blowing components as it goes
 
The open circuit protection is to the right, and cant be seen. The circuit does shut down and latch off in open cct output.....the voltage doesn't go above 80V
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top