the ncp349 chip used for overvoltage protection on the dc power input for the beaglebone seems to do a good job of providing overvoltage protection. Even looking at it’s datasheet though, I’m not completely sure if it will protect the board against reverse polarity. (I’m not out to overtly or purposely try this but I too have occasionally made mistakes…)
so should the beagle bone white/black be protected from application of reverse polarity being applied, i.e. powering it with -5v instead of +5v, by the ncp349 (or might it be a good idea to add a reverse polarity diode across DC in that will pop the power supply & save the board.
Full wave bridge rectifier, if they make 'me small enough would be the cat’s meow. Worked we’ll on an old project back in the op amp days.
are you saying that the ncp349 CONTAINS a full wave bridge rectifier? I certainly didn’t see that in the data sheet. Note the point is protection from overvoltage and possibly, thus the question, polarity reversal which may or may not be the same as severe under voltage.
It will not protect against reverse polarity.A full wave rectifier will cause a voltage drop and then you o longer have 5V. Yes I could have added some reverse protection, but I must admit out of 200,00 boards shipped I have sen about 10 cases of reverse polarity. So adding he cost to all boards for that feature does not make a lot of sense. I removed the NCP349 on the BBB…Out of the 150,000 of those shipped, I think we maybe had 3 cases of reverse polarity.
Thanks Gerald. You make a good point about the returns you see. It would be interesting to know if those numbers are due to many if not most powering boards from usb due to it’s immense ubiquity and popularity. While I generally try to be careful in getting things right including checking things multiple ways with VOM prior to, during and, after assembly I can say it has happened where a power cord on some item has been assembled or hooked up improperly such that polarity ends up reversed.
I’m aware of the voltage drop issue with a rectifier which is why I was thinking single diode placed across the input (basicly as a crowbar to the supply) as the point is protection but again with the rma levels seen I can see how that dictates economics.
Actually, USB power is not all that common. A lot of people have a cape they are using or Ethernet, so that is typically more than a USB port can supply. So we see more DC power applications. Also, a lot of the USB power ends up going to the FTDI device on the board as it is activated when connected ot the USB port.