yeah okay, your right about the processor brownout. so i will resign to using the personal effort of checking my polarity… oh well. one thing to think of would be adding jumpers to bypass the diode, but if you think that would take too much space, i will not press this issue. this sortof started as a rant after i fried my board.
By the way, when is this “next run”?
Hey Jim,
According to the LM2937 datasheet its “Ideally suited for automotive applications, the LM2937 will protect itself and any load circuitry from reverse battery connections”. Nice feature, if it really works.
national.com/mpf/LM/LM2937.html Look at the actual datasheet.
Kinda makes you go "Hmmm. . . ", and wonder what going on then.
The next run will be in a few months or so.
The regulator may very well protect itself from reverse polarity, but the caps will not. The bypass caps can short when reverse polarity is applied. They can do this without even looking bad.
Good point! Wasn’t looking outside the box, so to speak.
That is why you are the Head Honcho!
where are these bypass caps? are they possible to replace? i was thinking that i could replace the caps and the atmel IC and maybe get my board running again.
They are everywhere… There is a .1uF on the input, there is a 22uF on the output which is the most suspect, and there are lots of other .1uF caps sprinkled all over the board. Check out the schematic.
the 0.1uF are probably not damaged as they are ceramic mono and not polarized.
the 22uF, either tantalum or electrolytic, is quite susceptible to being destroyed by reverse polarity. C12 is the most likely culprit.
all the caps are SMD, so it could be pretty hard to replace. it may be possible to run it without the cap, i could just de-solder it and see what happens. i would guess the caps are for removing ripple, and i am running off of a battery. i could be wrong, so i’l wait until someone smarter checks my theory. one other idea: C12 is on the output end of the regulator, so it would probably be protected by the regulator, if it works the way it should. i would think it would be C13 that would be most likely to be damaged.
one thing to note is that the power LED does work. perhaps it is a burned out trace running to the IC, or some sort of ‘temporary burnout’ ?
Um, does C12 get hot when the power is applied? Be careful… if it is the failure item it WILL burn you.
if a capacitor is damage it is probably not a dead short but enough to trick the regulator into current limit mode thus the output voltage drops. normally this kind of cap failure results in the cap burning up, literally, however the regulator is protecting itself by limiting current which in turn keeps the cap from getting hot enough to let the smoke out. it’s like a catch-22 thing.
green LEDs take about 2V to turn on… the micro can’t do much at that level… it’s internal brown-out circuit is probably holding it in reset.
sorta forgot about this thread, i ended up ordering a new board and that’s the end of that, but does this mean that my Atmel IC is probably not damaged?
Guys, no need to flame each other.
If reverse protection was used, could I suggest a P-Channel series MOSFET with it’s gate connected to the BAT-? This would be a slightly more expensive option, but would only drop, at most, 0.1V.
Just my suggestion.