Not a problem! Voltage injection is usually used when you have short you are trying to locate the source of (in conjunction with a thermal cam or some alcohol on the shorted components to see which one gets hot first). I don’t believe that would help much in this case.
Gotcha. Do you happen to know what you’d call this type of “short”? I can’t wrap my mind around what would cause a negative diode value.
You cant test for shorts or diode readings with power attached. You can only test for voltage like that. If it is not showing a short with the battery disconnected then it is fine.
Learned something new! Thank you :).
Got a question. This trace is supposed to be a ground pad yet I get a diode reading of 32. I think this trace is connected to VSYS? Any ideas?
2 Likes
Thank you so much! So would the next course of action be injecting voltage at VSYS and seeing what gets hot?
You can try it and I hope it will work.
But in my case the short wasn’t producing enough heat to see where it shorted out. I ended up pulling every cap on those line. No joke: it was one of the last three caps I pulled, which was the cause of the short. 
Aw man, unfortunately that wasn’t the case for me. Removed all of the components marked in red and the short is still present. The only difference is that the diode mode reading is now 53V, no longer 32. What would you make of that?
I hope you only removed the caps one by one and put them back in place. The caps are the bypass caps for the ics. Have you removed the max on the left side of the soc?
I removed the component then tested to see if the short disappeared, never putting them back on. I’ll remove, test, and put back on the way you’re referencing from now on. What’s the reasoning behind this? Also, no, I haven’t removed the left MAX ic, only the bottom MAX. I’ll see if anything changes after removing the left MAX chip.
No difference was made upon removal the MAX chip to the left of the SOC. Though after I removed the big square MAX IC on the back, the sys rail diode reading now is 105. So it’s gone up but that still seems to low? I think it should be somewhere around 350V.
Sorry, my story to my short at the vsys line was missleading.
It is easier to pull the components, check and test and resolder them one by one. Otherwise you have to memorize where each with its different capacity belongs.
Instead of taking vague diode mode reading which will merely infer the in circuit resistance, instead, measure in resistance to begin with
Also I’m a bit lost as you mentioned in the first post
which would be normal, but you’ve since said that the caps are now [always] shorted or being dragged down, what changed between then and now?
Bent USB pins can cause a subsequent short on the SYS rail in some cases so that might be worth checking. You can somtimes track the culprit down by putting your meter in resistance and going over the areas of the board where this rails show up and see which area constitutes the lowest reading further narrowing down the potential culprit.
Ok resistance mode, will try that. So before, the caps in question were the 3 around BQ that would give me a -538 diode reading (which at the time I didn’t know what normal) so I removed the MAX IC under the SOC because I know that shorted caps near BQ see sometimes linked to that max chip. So with the PMIC chip (I think that chip is the PMIC chip) removed I started testing for voltages with the battery plugged in as well as with the battery + charger plugged in (which I think is something I shouldn’t have done with that MAX IC removed). I think that’s what perhaps caused the sys rail short because that’s all That changed. Im learning a lot from working on this particular switch :).
It is technically a PMIC, but generally when people are talking about the PMIC on Switch theyr’e reffering to the main PMIC on the otherside of the board, the biggest of the three Max IC’s
All three of these Max IC’s VIN comes from the SYS rail so they all have that in common, so if you don’t measure proper SYS rail voltage at the 2R2 coil you also won’t see it anywhere else
Hmm thinking about it perhaps what caused the change was the probing the board in diode mode while the board had power which may have killed something else and this is a secondary fault… I mean I could be wrong but it’s entirely possible. Perhaps if your resistance readings don’t shed any light then might be worth pulling the BQ IC off again just in case and just keep it off until you find the actual culprit it it isn’t the culprit itself.
best way
Thank you for the clarification! So with the battery plugged in only, the voltage at the coil matches the voltage of the battery (3.92V), which I assume is what’s supposed to happen? Now with the charger + battery, the voltage at the fuse matches the 5v charger but at the coil it’s 2.08V. Not sure of what to take from that. Also, are you suggesting that I use resistance mode and go around the sys rail looking for a low resistance value compared to the rest of the rail?
Yeah, you can think of the SYS rail kind of like regulated battery voltage but independent of one another.
Strange, maybe it’s pointing to USB or USB related circuitry (m92 etc) issues.
Correct. So if the SYS rail is shorted to ground, and let’s say it’s a 10 ohm short to ground (just for example) measured at the 2R2 coil, then you might go over to SYS TP beside the main PMIC and measure the resistance to ground there also (making sure to keep the same probe polarity as you did at the 2R2 coil), it might measure 9.98 ohms (again just for example) as it’s a lower reading your “warmer”, then you might measure this rail at the other two max IC’s and let’s say they measure 10.33ohm - this is higher, so your “colder” then say you measure the SYS rail at the 8316 IC and it measure 8.7ohm bingo, most likely candidate. Just checkout Calvins diagram for most of the places where the SYS rail shows up. Hope that makes sense.
1 Like
By polarity I assume you’re referring to the 200M, 200K, 20K, 2000, ETC modes on the multimeter? I set the DMM to 20kohms and at all of the sys rail components I get 0.10, there doesn’t seem to be any deviation unfortunately.
No, as in your probe poarity, black probe on ground and red probe at the point of interest
So as your getting 0.10 on the meter, to make it more readable and more accurate adjust the range lower, keep going until it’s at it’s most readable but still within range (this might be 200ohm on your meter… I don’t know)
There will be, hopefuly once you’ve adjusted your range it will improve accuracy, you should see minute devation on this rail across the board. If you still don’t then it’s probably your meter at fault - UT61E is my reccomendation… not need to faff about with ranges after that either.