# Voltage regulator replacement

I have a gadget (a Sonar) that stopped working (doesn’t respond to power on, power led does not turn on) and opening it to look for the typical signs of failure I see an L7818 voltage regulator in TO-220 package that looks like it has overheated.

I’m confused about two things: First: Why would one use an 7818 circuit in a board powered by 12V DC? The board has an STM32 microcontroller and peripheral components (wifi etc) so I was expecting to see an L7805, not an 18V one. The 18V could be for powering the analog/sonar bits I suppose, but should I then not also find a 5V regulator somewhere for the logic parts? Just out of curiosity, what the likely power supply design here if input power is 12V (boat battery), and the board has a VRM requiring 20+ volts input, and having 18V as output?

Second: the product code on the VRM is a bit charred but looks like “L7818BT”. I can’t find that, but I can find an L7818CV. What is the meaning of the suffix letters in this case?

Edit: top and front view of the board. The VRM that looks overheated is the component screwed to a heat sink on the left in both pictures.

The small stacked pcb wth the large heat sink is related to the WiFi (black wire connects to a case antenna). The two large square components are both arm microcontrollers.

Update: some basic testing confirms the VRM is a red herring, or at least it's not the only issue. Applying 12V at the DC input (the 2 pin connection on the left), I can measure 12V at those solder points underneath only (see the last image). That is: there is no other place on the board that gets any voltage at all. I'd assume that means it has to be one of the very first components being bad then, or a mechanical/connection issue somewhere very near the power connector? Unfortunately the power button connection (the 2 red wires joining the PCB in a pool of yellow goo) hides a lot of the things in the area nearest to the power connection so I'm having a hard time figuring out what's going on right after the power connector.

Update2/Solution: The problem was one of the power switch wires (the one connected to the 12V end of the battery) must have come loose from the PCB, under the yellow glue. Instead of trying to repair it which would have required scraping the glue, I tried bypassing the switch and connecting 12V directly to the second power switch wire, which worked. The solution to get a working power switch was to solder a new power switch wire, directly to the +12V connector on the board (posslbly bypassing a component or two such as a diode in the process). Device works perfectly after this, including the switch.

• BT .... bit toasted ..... sorry, could not resist – jsotola Jul 28 '18 at 20:07
• Is there a boost converter somewhere? It would be a strange choice perhaps to not just have the boost converter regulate to 18V on its own, but that's the only thing I can imagine you could do to have a 7818 on a 12V board. – Hearth Jul 28 '18 at 22:08
• I think if you want to know what that BT means you will have to surf datasheets for L7818s until it shows up on a binning list. It's a linear regulator, so offhand I can't think of anything they'd need to bin other than packages and temperature ranges. If you can't find it, I'd suggest making sure the replacement has the widest temperature range you can find. You wouldn't want to run too too much current off of a linear reg, so I wonder if it's being stepped up by a boost circuit and then regulated for a specific component or a mosfet driver. Pics would help! – K H Jul 28 '18 at 23:33
• @felthry, could just be a diode pump, that would explain the regulator. – Jasen Jul 29 '18 at 0:10
• @Jasen Ah, good point. Depends on the price point of the device, I think; high-end ones would likely have a boost converter if they need it, and cheap ones might just use a Cockroft-Walton generator. – Hearth Jul 29 '18 at 0:12