EEVblog 1726 - REPAIR of an EV Charger

[EEVblog]

Repairing a failed EV charger DC-DC converter had an unusual twist.

[Kleinstein]

With so little usage time (could still be relatively long connected to mains), I am a bit surprised that the electrolytic capacitor failed.
2 capacitors are not just for redundency, but sometimes also to avoid a large from factor (e.g. parts hight).

There could well have been another defects, like a ceramic capacitor that shorted out from mechanical stress or the SMPS chip going bad from an overvoltage spike.

I am a bit surprized that the supply did not run with only 1 of the capacitors. With a near zero output voltage a single capacitor should still be enough to avoid fast spikes triggering the feedback. There may have been another defect (like cold solder joint, broken trace or via) that accidently got fixed with changing the capacitors.

[Per Hansson]

Kleinstein: The cap failed because of physical damage: something pierced a small hole in the vent.
(To me when I watched the video it looked exactly like the pencil marks that are commonly used in manufacturing to keep track of things until Dave pointed it out at the end).
About why the failure caused the unit to not work: looks like the two output caps are separated by an inductor, so without one of the caps in-circuit there was not enough energy to ride through the AC waveform?

[SiliconWizard]

I wonder what could have poked a hole like this, especially given that the hole is small and the vent seems to only be bent very slightly close to the hole. I doubt you can do that with a pencil, but worth a try.

[golden_labels]

Maybe the pick-and-place machine’s grabber malfunctioned? They push THT parts into PCB at insane speeds and the positioning of the hole seems to be matching how components are being held during insertion.

I can’t imagine many other explanations. Discarded parts, salvaged from the dumpster and reused? An unhappy or bored worker poking holes in capacitors? The hole looks like something that goes inwards, not outwards. So not like it would be blown from the inside.

[thm_w]

https://skytech.ir/DownLoad/File/896_dk1203.pdf

The failed cap is the one right after the diode, before the inductor.
TL431 should be after the inductor, fb coming from the good cap.

So as mentioned surprised that it stops working with 800 -> 400uF on the output, especially with no load.

[EEVblog]

Kleinstein: The cap failed because of physical damage: something pierced a small hole in the vent.
(To me when I watched the video it looked exactly like the pencil marks that are commonly used in manufacturing to keep track of things until Dave pointed it out at the end).

That's why I wen the entire teardown video and this video before finally noticing it. Very common to have marks on top of a cap to indicate tested or whatever.