> Is that the circuit and code that checks to see if the circuit is self
> or USB powered?  I left the circuit out because this project will always
> be self powered, and I think I "tied off" the code to always indicate
> self powered.  I'll check.
iirc the standard code has two checks, one for "bus power/self power" and one for "is the bus attatched" (this saves power trying to enumerate to USB when there is nothing on the other end)


> > also i have discovered that putting a drain resistor on the 
> > main power rails considerablly improves reliability of startup.
> 
> This improves regulator stability or uC reset/startup?
the board i had this issue on was bus powered, the symptoms were it would start fine if left disconnected for a while but a quick plug and unplug would mean it would almost always fail to come up. 

> 
> 

> > finally i see while you've put in the low value, low esr 
> > close to the chip part of decoupling you don't have any large 
> > cap on the power rails, a quick search for the regulator you 
> > are using says that it should have a 22uf cap on its output.
> 
> I've fixed that also.  From the list of changes section, here is what
> has been corrected so far on the board:
> 
> VBUS and VUSB no longer connected.
> VBUS has 100K pulldown
> VUSB has .47uF cap
> capacitor added to regulator, across Vout and GND
right, i'm running pretty low on ideas then :(



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