Designing a boost converter

[0xC6]

Hey, I'm trying to desgin a boost converter to boost the voltage from 6 18650 cells in parallel to 5v and up to 6,5A out.
I'm relatively new to designing electronics but I do know the principle of how a boost converter works, however I'm having a bit of trouble deciding on a regulator for it.
I have looked into the MC34063A but I don't know if there are better alternatives for regulators?
If you could give me some suggestions/alternatives that'd be greatly appreciated!

[Buriedcode]

He said parallel, so the input would be 4.2V down to ~3.1V (3.0 cut-off is generally safe).  Although givne possible issues with cell balancing I would recommend that the OP does indeed use the cells in series and use a buck converter.   Having them in parallel, and for 6.5A out, assuming a 90% efficiency, the input power would be (5V*6.5A)/ 0.9 = ~ 36W.  With an average input voltage of 3.7V thats roughly 10A in, which is quite a lot to design a boost converter for.

OTOH, as spec assumed/suggested, if you have the cells in series, your input range is from ~18v - 25V.  Average 3.7V* 6 = 22.2V.  With an input power of 36W, that gives an input of 36/22.2 = ~1.6A which is much easier to deal with, but will beyond the capability of the old MC34063A.  Even with an external switch, its efficiency would be pretty poo.  There are much "better" devices out there (by better, I mean, more expensive, but higher efficiency, higher switching speed, which means small inductor/caps and more features like protection).

Since national semi were taken over by Texas Instruments, you can use their "bench designer" http://www.ti.com/design-tools/overview.html  I'm not a fan of it myself but it can at least show you options.

A 40-50W buck converter designed for 24V in and 5v out could work, but the market is flooded with modules like this, with few actual specs posted, so they could be awful or work jsut fine, I don't know.

[spec]

He said parallel, so the input would be 4.2V down to ~3.1V (3.0 cut-off is generally safe).

Arrraagh. Thanks will delete my post.

[Buriedcode]

He said parallel, so the input would be 4.2V down to ~3.1V (3.0 cut-off is generally safe).

Arrraagh. Thanks will delete my post.

No!! As I said, it is far better for series connection, so your post was perfectly valid - I wasn't having a go 

[Kasper]

Perhaps a little of both would be best.  3P2S (3 parallel 2 series) and a buck converter.

[hsn93]

He said parallel, so the input would be 4.2V down to ~3.1V (3.0 cut-off is generally safe).

Arrraagh. Thanks will delete my post.

unfortunately, i think its nice to edit it and say this post was wrong or put strikeline on it.. its very helpful for us to learn from others..