I wouldn't replace any of them that aren't problematic...and I'm a expert level board repair tech. These capacitors are very difficult to replace and without the proper board repair skills AND the right equipment, the results of this attempted repair often result in more NEW problems than are actually resolved.

This should never happen, unless the fuse powering the bridge rectifier is substantially over fused, it would blow immediately upon this type failure.

Don't take this number at face value, this number represents an average life expectancy of the part in a powered on state at a specific temperature before it has any degradation. Unless you owned the game from new and installed a run hour meter, the machines actual power on hours cannot be determined with any degree of accuracy.

This statement tells me you have never removed a "snap cap" from a double sided board. I have 30+ years in board repair and can tell you for a fact they are not the same...even with commercial grade equipment.

You determine this with an RMS voltmeter or an Oscilloscope to measure the amount of AC ripple on the DC voltage.

Sorry...I'm just reading this post for the first time and replying to posts in order...not trying to make this a pissing match; however your first advice to replace these capacitors to someone of unknown skill without doing any of the proper diagnostic checks will almost assurance have disastrous results...unless you work for Rottendog

In my opinion, unless you measure greater than 5% AC ripple on your DC voltage (250mV@5VDC) you should not replace the capacitor.

What does your AC ripple measure?

Good luck, if the problem goes away after reseating then you know where the issues lies. Many folks don't realize this, but the life expectancy of these headers and connectors is 50 mechanical actions before the plating begins to break down...removing and replacing a board is 2 actions as both the removal and installation are separate mechanical actions.

It's so hard to tell with all of the low quality stuff that comes out of China. I primarily stick with commercial grade equipment from Pace (made in USA), with the full expectation that I will never have to replace it.

That's because these don't fix the problem, they hide it.

Yeah I think you nailed it; I don't know what meter you are using but it's likely giving you an average DC voltage which likely has a lot of AC ripple causing voltages to drop much lower than the 4.86 you are measuring.

This is one of the best posts I read on here in a very long time, and I couldn't agree with you more.

I'm sure its not ... I would be much more interested in the 5VDC reading on the CPU board than any other measurement as this will tell you both the CPUs operating voltage, as well as the voltage loss from the PDB, as it's the voltage difference measured between this point and the +5V TP on the PDB.

IMG_1716 (resized).JPG

This is measured across C31 the red and black test clips, the large capacitor on the CPU board, one side is your 5V source through the Inductor, the other side is ground.

IMG_1713 (resized).JPG

.04V loss from the 5V test point on the PDB to the MPU is very good.

Looks like once you get C5 replaced you'll likely be in good shape, I would be interested in what the AC ripple voltage is. To take this measurement, just put your voltmeter leads between the 5V test point and ground on the MPU and switch your meter to AC volts. This would make a good comparison point to see how much this ripple voltage decreases with the new capacitor.