A while ago, my roomate broke his laptop. It wouldn’t charge, but it wasn’t something obvious like a busted charger or anything like that. So, I helped him move all of his important files to a new machine, but I haven’t had time to look at the old one for a while. Now that I’m starting to settle in at my new job, I have a bit more free time at the end of the day when I get to play with projects.
So, I opened it up. The first thing I noticed when working towards the insides was the corrosion on a cable that goes to the front panel LEDs. It probably was just some liquid that spilled through the front of the computer and got in the connector. In any case, the cable is completely fried without hope of salvage. On the plus side, it’s not really an important thing (other than indicating if caps lock is on). I removed it, and cleaned the connector. Somehow, when I plugged it in, I was able to get it to boot.
At some point pretty soon after booting, the charging symbol went away while I was using it. My power supply has a smart trip and will turn off when there is a short, and it kept tripping. By this point, the battery had charged up a bit, so I was able to continue using it, but I thought this was strange. However, after a reboot, it was no longer able to start, even from the battery, which still had a charge. My theory here is that the cap was on the verge of blowing, and was allowing trickle charge onto the battery, but when I went to draw more current, it just popped. So, we get to go back inside and look for more issues.
The next place to look was at the charging dock. From a brief search online, this model frequently has issues with the filter caps near the power supply shorting out and causing issues with running off charger, while battery power still works fine. For this machine, however, it could not boot off battery either, so maybe there was some deeper issue.
Just by looking at the board, it’s easy to see there was bad caps near the power supply. There was ash and one of the components was blown to smithereens. So, I removed all of the caps on the nearby traces that were shorted.
After taking a better look at all the nearby parts, it seems that the +19V pad and the power supply to the MOSFETs are buffered with a few caps, but if all of the parts on that pad are caps, then no power will get across and be able to go through the transistors. So, there has to be some kind of connection. I’m not sure what part should go on there, but it has to be a diode of some kind, acting as a reverse bias protection if you were to plug in the wrong charger or something like that. Since I know I won’t plug the wrong thing in (just watch me), I just replaced the unknown diode with a piece of wire. And Voila! It boots and charges just like normal. To recap: three filter caps (I replaced with .22uF 0805 25V) and a diode on the 4 pads connecting the +19V pad to the MOSFET drains.
I definitely was a little uncomfortable doing this repair without a schematic, but by using common sense and taking a look at the visible traces (very lucky that all the relevant ones are on the top layer of the PCB) I was able to work it out backwards. However, in my search for a drawing, I did discover something interesting. For more complicated electronics, it’s becoming increasingly common to use a type of interactive schematic called a “boardview”. You can familiarize yourself with them by just watching some Louis Rossman videos online. It’s the sort of schematic he has there where you can just click parts and it will show you what parts are connected and is sort of a cross reference between a drawing, silkscreen, and parts list.
A quick search for N550J schematics quickly turns up the boardview, but it’s encrypted. I tried to use Paul Daniel’s OpenBoardView software to view it, but the encrypted files can’t be opened until it’s, well, unencrypted. Asus used an interesting encryption method for this, relying on a specific special key to protect the information. If you’re more interested in this, you can find the support ticket on OpenBoardView’s Github page with instructions for decrypting the key used on the boardview, which you can pipe the file into, and then take the resulting file and view. I don’t have much interest in cybersecurity, but I did think it was interesting to learn about how companies protect their sensitive data and how there are many methods to reverse-engineering encrypted data.
So, if you’re having the same issue as I was, I hope I was able to solve it. Chances are, you have something similar and just by taking a meter and checking around the power circuitry, you can pretty quickly find the issue. If you have a deeper issue or some other problem somewhere else on the board, I would recommend to check out OBV and the community that works to support it, as well as to learn a little bit about how modern “service engineering” would be done if you were to do it professionally.