0
\$\begingroup\$

I've fried the controller on this PCB when by accident connected 24V supply in reverse polarity. The PCB includes a SS34 schottky on a positive side. However the controller still got damaged (completely dead).

I already got a replacement controller and will replace it on the board. My question is -- why the controller got damaged even though there is a supposed reverse polarity protection and how it should be protected properly?

Notes:

  • I am not the designer of linked board
  • The board includes 3 separate power inputs. 2 were connected correctly, 1 was incorrect (reverse polarity), which should be causing short through the ground I presume?
  • The SD card that was inserted into the slot also dead, but nothing else appears to be damaged
\$\endgroup\$
6
  • 1
    \$\begingroup\$ Probably because there was other wiring connected to 0V. When you connected the 24V to 0V, the pcb became the fuse. The diode had no part to play. \$\endgroup\$
    – Kartman
    Commented Feb 9, 2021 at 4:01
  • \$\begingroup\$ @Kartman thanks, I've added that SD card also got damaged, everything else connected seems to be fine. On that note, I still don't get why this caused damage, from schematic it all 3 inputs are isolated from one another (not sure actual traces as I don't have the gerber files for that board). How would connecting peripheral devices that supply no voltage then cause the controller to be damaged? Schottky is before voltage regulator, all the components are fed from voltage regulator, so even if positive lead connected to ground side that should not cause damage since there is no current flow? \$\endgroup\$
    – Alexey Kamenskiy
    Commented Feb 9, 2021 at 4:09
  • \$\begingroup\$ Take a look at the "Reverse Current Vs. Reverse Voltage" chart from the data sheet. Reverse biased diodes can still conduct current, which is why they really aren't great for polarity protection. Looks like the SS34 can conduct around 12mA when reverse biased at 24VDC, which is significant (unless I'm reading the chart wrong). Schottky is good for low forward voltage drops, but bad because they have higher reverse current leakage. I think the board designer picked a poor method of protection. \$\endgroup\$
    – Ron Beyer
    Commented Feb 9, 2021 at 4:32
  • \$\begingroup\$ @RonBeyer thanks, that makes more sense. But how would one put better protection in this case? \$\endgroup\$
    – Alexey Kamenskiy
    Commented Feb 9, 2021 at 4:36
  • 1
    \$\begingroup\$ You can add a diode across the pair of input lines and fuse the input lines so reverse polarity shorts the diode and instantly burns the fuse. Reminds you not to reverse connect without having too high a cost. \$\endgroup\$
    – K H
    Commented Feb 9, 2021 at 7:27

1 Answer 1

Reset to default
1
\$\begingroup\$

I'm trying a fuse followed by a reverse-biased diode. With reversed polarity, the diode conducts and takes out the fuse. In fact, I'm using a unipolar transorb (e.g. SMBJ24C), which will also conduct on over-voltage with the same result. In one application, I've even followed the above with a forward-biased diode - to prevent the transorb from having to discharge large input filter capacitors.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.