This article is currently an experimental machine translation and may contain errors. If anything is unclear, please refer to the original Chinese version. I am continuously working to improve the translation.
Over a year and one month later, due to the current absurdly high hardware prices, I’ve become even more reluctant to throw away my long-forgotten Orange Pi PC 2 gathering dust in the corner. After all, even the tiny Orange Pi Zero Plus, which uses the same Allwinner H5 SoC, now costs more than what I originally paid for this board.
Luckily, I recently upgraded my toolkit—new soldering iron, adjustable power supply, multimeter, and all that fun stuff—so I figured, why not give it another shot? :P
You can read my previous repair attempt here.
0x04 Removing the AMS-1117
Nothing fancy here—just went in with the soldering iron and chiseled it off.
0x05 Measuring the Load
Turns out the chip’s output was indeed shorted to ground. The AMS-1117’s output was directly shorted to GND.
0x06 Pray to the gods that the SoC isn’t fried
0x06 Applying Voltage Manually to Locate the Fault
I used the adjustable power supply to manually apply 3.3V to the original Vout pad. The current immediately shot up to 4A.
Since the short circuit point should be drawing a lot of current, it would get very hot—literally.
And sure enough, one spot on the board was burning hot to the touch. At first I thought it was a blown transistor, so I desoldered it.
But the short was still there. After removing two nearby capacitors, I finally found the real culprit—capacitor U40 had shorted!
Picture
0x07 Replacing Components
Once I removed the blown capacitor, the short was gone. I had a spare AMS-1117 on hand, so I replaced it. The 3.3V rail was now generating correctly.
The transistor was labeled A1SHB, so I ordered a replacement from Taobao.
The other two capacitors were just for filtering, according to the official schematic. Their pads were already damaged anyway, so I figured skipping them probably wouldn’t be a big deal. (?)
0x08 Testing
Once the A1SHB arrived, I soldered it back into place.
On power-up, the TTL serial console worked and I could SSH in—but the CPU got extremely hot, hitting 94°C right after boot.
I tried going into armbian-config to adjust the CPU frequency, but the board suddenly powered off.
0x09 Measuring Again
I measured the resistance between the A1SHB’s drain and ground—got 0 ohms.
Applied 3.3V again, and saw 2.5A current draw. This time, the SoC itself got scorching hot.
Checking the schematic, I realized the CPU does require the VCC-IO power rail. Looks like it’s confirmed—the SoC is shorted.
Power schematic
Well, that’s it. No point in continuing.
0x0A Summary and Analysis
Theoretically, I did fix it the first time—but apparently the SoC was already shorted.
Maybe when the capacitor blew, it took the SoC with it?
Or maybe after I powered it without the filtering capacitors, the unstable voltage from the AMS-1117 fried the SoC?
In the future, I should definitely be more careful during repairs.
This article is licensed under the CC BY-NC-SA 4.0 license.
Author: lyc8503, Article link: https://blog.lyc8503.net/en/post/orange-pi-repair-2/
If this article was helpful or interesting to you, consider buy me a coffee¬_¬
Feel free to comment in English below o/