Class X2 cap failure mode

[Mick Bailey]

I have an SMPS for repair that's blown the main 3A plug-top fuse three times in just a few weeks. The inlet circuit is a straightforward Delta configuration with a Class X2 cap across the 240v supply and parallel 560K FP resistor, followed by a pair of Y2 caps from each supply leg to ground. All upstream of the F1.2A internal fuse which hasn't blown.

I can't recall ever having an intermittent short with an X2 cap - when they short on 240V they blow a hole in the case and go dead short. But this is the only reason I can think of for the plug-top fuse blowing.

Anyone experience these going intermittent?

[nickb]

I have an SMPS for repair that's blown the main 3A plug-top fuse three times in just a few weeks. The inlet circuit is a straightforward Delta configuration with a Class X2 cap across the 240v supply and parallel 560K FP resistor, followed by a pair of Y2 caps from each supply leg to ground. All upstream of the F1.2A internal fuse which hasn't blown.

I can't recall ever having an intermittent short with an X2 cap - when they short on 240V they blow a hole in the case and go dead short. But this is the only reason I can think of for the plug-top fuse blowing.

Anyone experience these going intermittent?

FWIW that seems a very unlikely mode of failure.

This may not have any significance to your problem , but after I replaced the the MOSFETs on an SMPS I had issues with the fuse blowing. The problem was the original FETs were no longer available. The replacements I used outperformed the originals in every respect. It turned out that the lower Rds(on) of the replacements contributed to a pre-existing switch-on surge issue.

[Enzo]

Well, doesn't matter if they haven't. That still doesn't mean yours didn't. Parts never feel obligated to only fail in expected manners. Otherwise what would techs have to tell about after work across a couple pints?

Just to be contrary: Have you a different power cord you could attach just for troubleshooting? In case the cord assembly itself may be defective.

[g1]

My understanding is they are designed to fail open, so they prevent fires for class X, shock hazards for class Y.
But they are supposed to be self-healing, so I suppose a recurrent problem like you are having is possible.

[Mick Bailey]

I'm running the supply on a different cord to that used by the customer. In fact I've not got the unit to fail on the bench and it's only just over a year old with no previous repair. Because the internal fuse is right after the line filtering, and this fuse is lower than the plugtop fuse (almost by a factor of three), I'm suspecting if there is a fault it has to lie within the filter section.

My understanding with X2 that it relates to a specific rating, flash testing and permitted location, and that it can fail short.

Vishay defines them as "Capacitors with unlimited capacitance for use where their failure due to a short circuit would not lead to the danger of an electric shock." I've seen plenty of failures and always shorted. Sometimes with secondary damage from the superheated ejected material.

For the sake of parts costing pennies I think I'll 'shotgun' this and replace the filter components. Sometimes it can be worthwhile doing this even when the board looks fine. I've had some jobs where there's been signs of tracking on the board underneath filtering components, not visible until they've been removed.

[Enzo]

So you have not had it fail on your power cord, but it failed repeatedly on his cord?

[Mick Bailey]

Ah, sorry - just re-read my last post and didn't include the equipment had also been loaned out to another band and it had failed with their lead. I've removed the X2 cap and clipped it across my Variac set to +10% (264v) via a bulb limiter and it shorted right out. It's rated at 275v continuous, so must have been borderline on 240v.

[Mick Bailey]

Out of interest I sanded the side off the cap and there's a streak of carbonized material inside. It raises the question as to why place a class X2 cap across the supply, when a class Y2 cap in the same position would fail open?

[nickb]

Out of interest I sanded the side off the cap and there's a streak of carbonized material inside. It raises the question as to why place a class X2 cap across the supply, when a class Y2 cap in the same position would fail open?

[ATTACH=CONFIG]37286[/ATTACH]

Both X and Y are supposed to fail open. The difference is in the rating and peak impulse voltage withstand.

Nice bit of sanding BTW and clearly an inquisitive mind at work

Still, really surprised that the cap failed like that since the basic requirement is NOT to fail that way due to fire risk

[Mick Bailey]

I've seen a few shorted X2 caps over the years - nearly always as a result of a lightning surge or nearby strike, and usually with accompanying catastrophic damage. I guess there are limits to what anything will stand if the breakover voltage is exceeded. There's also a void in the middle of this particular cap - I wonder if there's a manufacturing defect?

What something is rated at and how it actually behaves becomes somewhat distorted when we take the wider context of Chinese manufacturing. 'Certified' dental X-ray machines that expose the patient to more than a lifetime's worth of radiation in a few milliseconds. Melamine in baby milk, lead in toys. Last week I spent almost three hours on a totally bizarre motorcycle fault and it turned out that after fitting five brand-new relays and still getting failures, the problem was with the relays. A post-mortem showed that every one was shorting the coil to the contacts. Made in China.