Buck converter in continuos mode

In summary, the conversation discusses the buck converter in continuous mode and the graph for inductor current on Wikipedia. The question arises about the non-zero inductor current at t=0 and whether it is possible due to the inductor's inability to change instantaneously. The person suggests thinking about the current at some time t<0 before the switch is turned on. The graph shows steady state operation over one period of time.
  • #1
daniel_8775
2
0
Hello,

I don't quite understand the buck converter in continuos mode ... looking at the graph for inductor current on wikipedia, at t=0, there is a nonzero inductor current, Imin ... how is this possible when the inductor current cannot change instantaneously (assuming it starts at zero, then at t=0 when the switch is on, it must be zero as well). Perhaps I'm not thinking about this correctly, and I shouldn't be thinking about when the experiment "starts", maybe I should be thinking at some time t<0 it had current I_min through it before the switch came on?

Here is the graph I'm looking at: http://en.wikipedia.org/wiki/File:Buck_chronogram.png

Dan
 
Engineering news on Phys.org
  • #2
daniel_8775 said:
maybe I should be thinking at some time t<0 it had current I_min through it before the switch came on?

Yes, this is what you should think. The picture describes steady state operation. The time T is one period.
 

Similar threads

Replies
13
Views
2K
Replies
14
Views
1K
Replies
10
Views
2K
Replies
13
Views
1K
Replies
14
Views
7K
Replies
17
Views
2K
Replies
15
Views
5K
Back
Top