I am working on an application where the standard R/C timing 
forovercurrent protection in a device is not sufficient.... For those 
really interested we're talking about a L6206, driving a load where 
there is an internal capacitor.

What happens is that the L6206's overcurrent protection kicks in during the initial capacitor inrush.  The amount/duration of the current is not 
sufficient to damage the L6206 (it's within the ratings), but it is enough to trigger the overcurrent for long enough that the device shuts 
down and the load never fully powers up.   I can tweak the R/C values and get it to power up, but then short circuits will occasionally kill 
the device, due to the short cycle.   And so on.

My thought is to put an external circuit between the overcurrent detect output and the enable pin which allows me to more finely adjust what I 
want the overvoltage circuit to do.  The goal is to be able to delay the shutdown by around 5us or so (adjustable) and then keep it off for a few 
seconds.  I was thinking about the 8 pin pics - however, with a 500ns instruction time, this sure doesn't look reasonably likely to work for 
me - especially if I want to control both OCD output/input pairs.   I was also looking a bit at the AVR equivalent since it looks like the 
instruction time could be as low as 125ns, but still, I'm not sure that's fast enough.   There is also the possibility of using an external 
XTAL, but I'd like to avoid that.

Because this is for a design which is fairly late in it's design cycle (somehow we didn't catch this until "beta"), I don't have a lot of space 
- and I'd like a small solution.   I was wondering if there was something else I missed...  I will be playing a bit with a 556-like 
circuit, but intuitively it feels that this would have similar issues to the R/C which won't work for me.

-forrest



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