Brief power loss / reset

[b0nz0beavis]

My 3.5 year old DIY has held up fairly well, but it now loses power intermittently. During use it will revert to the initial test pattern and the LCD will also display the initial firmware info. The unit continues to function properly, but the reset issue recurs a short time later. It’s probably 10-20 mins between occurrences.

My amateur guess would be that it’s a regulator issue, I just need some advice/documentation on where to start testing. It’s been about 3 years since I had to diagnose and fix any issues, and I don’t remember much from the last time but it was not the same issue so not sure how much would apply anyway. Thanks!

[marqs]

It could be the main regulator (AP7361) overheating, but more likely cause is a cold joint on FPGA power/gnd pins. I’d first try resoldering the bottom GND plate of the FPGA, and if that doesn’t help, then its legs on the top side.

[b0nz0beavis]

Thanks for the reply. Haven’t had time to look yet. I hope it’s the GND plate.

Thanks for the advice!

[b0nz0beavis]

Reflow of the ground pad for the FPGA didn’t improve the situation. All of the pins on the FPGA have a super solid connection to the PCB. However, there was a single loose pin in U1 and about 12 loose ones on U3. I’ve reflowed those, so it’s off to play some games and see how long it lasts until power blinkage this time. Fingers crossed it’s a few years this time again.

EDIT: 2 hours of gaming and no dropouts yet. I’m considering this fixed, until a longer gaming session potentially proves otherwise.

[b0nz0beavis]

A few minutes after the second edit of the above post, the OSSC started losing sync again.

Since then, I’ve done a reflow of all pins on the three ICs. There is now a new problem – the OSSC does not sync at all. Have tried multiple consoles and also tried using SCART and VGA inputs. It still outputs video over HDMI and all the menus and functions appear to be working.

@Marqs – any suggestions on how I can investigate further? EDIT: actually, I just flowed the connection from FPGA to ground plane and now it’s getting stable sync again. I think an attempt I made earlier to add more solder to this connection actually caused the connection to worsen – probably bad soldering technique lead to pulling solder out of the plane. Back to extended playtesting to see if it’s a longterm fix.

[b0nz0beavis]

Solid 4-5 hours of use and no issues with sync. OSSC is definitely having improved stability.

Now need a way to permanently improve the FPGA ground connection. Will have to some day try removing the chip and adding or modifying the amount of solder on the ground pad.

[b0nz0beavis]

There is now a new issue. The OSSC appears to regularly lose the output to HDMI. It doesn’t lose sync of the incoming signal (the LCD continues to show the “262-P…”) and the unit does not reset.

What is happening now is that the red LED lights up and the display goes blank.

@Marqs – is there any way I can debug this? I’m guessing it’s something to do with a signal going to the chip next to the HDMI connector or something in that chip itself. Over the last few days as part of trying to fix the issue I’ve reflowed every component on the board so I don’t believe that it is a bad connection and I’ve checked for bridges on all the ICs. I think I will go over everything once more just to be certain.

[Author]

Posts