Hi All,
I found a statement on one of the sites I am looking into that makes
sense, but I am not sure if it is completely correct:
 
"It should be noted that when there is a light load the circuit can slip
into discontinuous mode, where the inductor becomes fully discharged of
it's current each cycle.  When a load is reapplied the inductor needs to
recharge, and so the transistor's duty cycle increases pulling the
inductor towards ground, and because of the increased duty cycle Vout
decreases when we really want it to increase.  This causes an
instability, which is well known for boost converters, and not a problem
with buck converters. 

One way to combat this instability is to choose a large enough inductor
so that the ripple current is greater than twice the minimum load
current.  When this condition is met then the inductor is always in
continuous mode."

Should it read:

"One way to combat this instability is to choose a large enough inductor
so that the ripple current is lower than twice the minimum load current.
When this condition is met then the inductor is always in continuous
mode."

This value looks like a empirical value, which will give you very good
operation, you would trim the inductor size if physical size was an
issue but the last part of the statement doesn't make sense. I use
0.4Iin which is very close to the value they mention. When looking at my
simulations and even looking at the calculations for minimum inductor di
= 2* Iin, the inductor should be chosen to minimise di as much as
possible for the design.
Any comments welcomed, as I am slightly confused.

Best Regards
            Luis
  

-----Original Message-----
From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf
Of Apptech
Sent: 05 December 2009 09:59
To: Microcontroller discussion list - Public.
Subject: Re: [EE]:: Boost Converter

Luis

I wasn't at all annoyed.
Just suggesting that knowing how the components behave is fundmental to
all 
that follows.
You now seem to have mostly made the jump to light speed :-).

10 kHz is VERY low for a modern converter.
Only run there if there is special need.
Core will be much larger than usually needed.
100 kHz is a sensible modern minimum.
Using a powdered iron core from eg Micrometals (Gargoyle knows) would be

cheap and tolerably good. Beware of clone limitations of Micrometals 
products.  All nominally equally specd powdered cores are not created
equal.

Continuous mode is generally better - but the equations end up
surprisingly 
complex - not hard or inobvious - just with many terms.,

For lossless approximation the duty cycle is essentially in the inverse 
ratio of Vin and Vout.

Skimming your design procedure it looks OK enough (E&OE YMMV ...)

Building a somewhat lower powered version to start is probably wise. ~60

Watt's is enough to get good magic smoke when things gang aglae.

What controller IC do you intend to use?


          Russell 

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist