X2 caps in RF controlled mains plug top adaptors dead, gone low value.

[Chris Wilson]

I have common or garden RF controlled 13 Amp mains plug top adapters. Two have been left on powering low current devices for months and they have started to turn themselves off after some seconds or minutes. I stripped them down and the capacitive X2 mains dropper poly caps before the bridge rectifier and 5V regulator have changed value from 330nF to under 1nF. They are marked as MPX brand, which is a respected make, do these have a low finite life in such applications or are they likely fakes? I have ordered new Vishay ones from RS. This, from Googling, appears a common problem with these gizmos... Thanks.

[SeanB]

Common fault, replace with better capacitors, and look in there for the series resistor, and up it's value to a larger value, like 1k 1W fusible. As well add a small 275V MOV after the resistor to neutral, to clamp the spikes that are killing the capacitor. Resistor increases mains supply impedance so a smaller MOV can clamp the spike and still absorb the energy that is applied. Capacitor will last a long time then.

Done the same recently on the geyser timer, as the 330n capacitor had done the same thing. Was just a bit difficult to fit any thing other than the same brown metal film dipped capacitor in there, so soldered the MOV behind the PCB where there is a 2mm gap, and added a 2k2 1W fusible resistor externally to the power lead. also as a precaution changed the 470uF 50V capacitor that powers the 48V relay in it. Old OHL cap was at around 220uF and ESR was around 30R, bit high for a capacitor that purported to be "Low ESR", and rated to "105C". Used capacitors both times, again Sunday work. That ewaste pile comes in handy quite often.

[floobydust]

Mains capacitive-droppers have high ripple currents and stress the capacitor, much more so if there are transients anything with fast dV/dt.
Film capacitors in that use fail low value, the (stacked) sprayed metal end cap connections go bad. I'm not sure about wound constructions, it might be the self-healing mesh fuses open-circuiting. Another film cap failure mode is the metallization corrodes due to moisture, after years.

Vishay specifically mention some of their film caps are not suitable for series impedance applications,  others only for X2 'across the line' use. They don't say why there is a difference.

[Chris Wilson]

Common fault, replace with better capacitors, and look in there for the series resistor, and up it's value to a larger value, like 1k 1W fusible. As well add a small 275V MOV after the resistor to neutral, to clamp the spikes that are killing the capacitor. Resistor increases mains supply impedance so a smaller MOV can clamp the spike and still absorb the energy that is applied. Capacitor will last a long time then.

Done the same recently on the geyser timer, as the 330n capacitor had done the same thing. Was just a bit difficult to fit any thing other than the same brown metal film dipped capacitor in there, so soldered the MOV behind the PCB where there is a 2mm gap, and added a 2k2 1W fusible resistor externally to the power lead. also as a precaution changed the 470uF 50V capacitor that powers the 48V relay in it. Old OHL cap was at around 220uF and ESR was around 30R, bit high for a capacitor that purported to be "Low ESR", and rated to "105C". Used capacitors both times, again Sunday work. That ewaste pile comes in handy quite often.

Thanks yet again Sean, these are now working fine following your advice! I could have just bought some new ones but most probably they are all built to a minimum spec, plus I have learnt something in the process