SEPIC converter with "easy" feedback loop?

[Flyback]

Hello,

Please could you help to define what circuit elements are (and which are not) in the feedback control loop of this SEPIC LED driver…………….

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

V(in) is fixed at 6V.

Input current is regulated by the loop.
Output voltage could be anything between 5V and 40V.
Input power is limited to 7W

The converter is a SEPIC , but I believe that the actual converter topology could have been anything, and the control loop would not change…is this true?

I mean, the converter’s power stage and output capacitors simply are not in the feedback loop…do you agree?
The feedback loop simply involves the input capacitors..is this true?

 

[crutschow]

The feedback compensation components would appear to be R20 and C10, and possibly R22, R23, and C11.

That circuit is very strange as it appears to regulate the input current, not the LED output current as normally is done. If you regulate the input current, the LED current will vary with the input supply voltage, which is not what you usually want.

 

[Flyback]

If you regulate the input current, the LED current will vary with the input supply voltage, which is not what you usually want.

Absolutely right.
Though our input voltage is a battery , and its pretty solid at 6V.
If it varies more , the micro can adjust the input current as required, but its not really necessary for the purpose

The feedback compensation components would appear to be R20 and C10, and possibly R22, R23, and C11.

...you appear to be acknowledging that the closed feedback loop does not involve the sepic power stage...and thus the feedback loop maths will be nothing like SEPIC feedback loop maths?

 

[crutschow]

Absolutely right.
Though our input voltage is a battery , and its pretty solid at 6V.
If it varies more , the micro can adjust the input current as required, but its not really necessary for the purpose

...you appear to be acknowledging that the closed feedback loop does not involve the sepic power stage...and thus the feedback loop maths will be nothing like SEPIC feedback loop maths?

So why are you regulating the input current rather than the output current?

The only other feedback compensation elements I see are R4 and C1.

 

[Flyback]

R4 and C1 are just a filter for the current ramp signal, they are not part of the feedback loop.....there time constant is smal enough for them not too affect the small signal transfer function.

So why are you regulating the input current rather than the output current?

Because the luminaire may be 5V or 40V, and we have to regulate the power so that the battery doesnt drain down too quickly.
If we regulated output current to , say 100mA, then that may be ok for a 5V load, but too much current output for a 40V load...therefore , we regulate the input current, and thus we have a known battery duration, no matter what luminaire voltage is used.

Are you acknowledging the massive fact that the feedback loop here is independent of the topology used?

 

[crutschow]

I'm only saying what I believe to be the feedback compensation components. I can't answer as to whether they're independent of topology (but I don't think they are).

 

[Flyback]

completely removing the output caps doesnt make it go unstable.......for a sepic , to be able to do that is unthinkable.........i think only the primary of the sepic inductor appears in the feedbback loop...do you agree?

 

[crutschow]

Since the feedback is coming from the input, I believe your statement is correct.

 

[ronsimpson]

This question was asked last week.

I (more or less) agree the sepic is not in the loop.
Normally when you say the output can be 5V to 40V, the gain is different at 5 verses 40V. You may need to build this thing and speed up the loop until it almost oscillates and see if it is more stable at 5V or 40V. (also at high/low input voltage) You need to make it stable at the worst case.

Because this supply is very strange, you need to build it and see what happens. Thinking about it has not got you very far yet so just do it.

 

[Flyback]

Thanks, this sepic has been built and is working well.
I am just wondering now though, the design appears excellent, as no electrolytics were needed...the input capacitance is 66uF total, but because its always less then 9V input, these are just 16V rated ceramics.
It seems to be the perfect converter, not saddled with the instability worries of a sepic, and not needing any output capacitance, let alone electrolytic output capacitance.

As far is the control loop is concerned, the converter might just as wel be a boost converter....it doesnt care.

It appears that regulating the input current gives this great advantage in terms of stability , and i mean stability obtainable even with no output capacitance.....that is excellent, because usually one has to have big 'lytic cans to ensure stability.

 

[ronsimpson]

When you test the stability, check on a new and old battery.
Old batteries have high resistance and appear less like a capacitor.

Your PWM is current mode and thus has good stability.

I usually try to run LED PWMs fast (1mhz) in part to keep the Cs and Ls small. Ceramic caps do not have some of the problems of Ele. caps.