Should I keep these capacitors on my IEC socket?

[bikehorn]

Today I salvaged an IEC power socket from the power supply of an ancient(~10 years old) workstation computer, and it appears to have some line filtration in place. looks like a cap each from live to ground and neutral to ground, plus what might be a MOV across active and neutral, which also has a resistor in parallel with it. All wrapped in heatshrink. I've been looking around for information on the effectiveness/safety of this method, and on Rod Elliot's site he talks about it briefly:

In some cases, a capacitor may be used between live and earth (particularly in the US). This is especially common in some guitar amplifiers, and the capacitor is likely to fail - especially at 220 or 240 volts AC, since DC capacitors are commonly used. In some cases, a capacitor may be used from both active and neutral to earth. This is an extremely dangerous practice, and is illegal in many countries. Generally, I do not recommend or condone the use of capacitors from any mains connection to safety or chassis earth. Indeed, under some circumstances these caps can cause residual current devices (RCDs - safety switches) to trip. The use of any capacitor between mains and the chassis places the user at risk of electric shock if the chassis is not connected to safety earth.

but he also says:

Input mains filters can remove either form of high frequency noise component to some degree, and large spikes can be removed using Metal Oxide Varistors (MOVs) that effectively short circuit the noise pulse, reducing it to a level that is (hopefully) inaudible.

I'm building a small Class D amplifier and the area I will be using it in will not have exceptional power quality, in fact it will be shared with such things as computers a refrigerator, so if possible i'd like to keep the filter. I am given to thinking that it should be alright since it was CSA and UL passed but I thought it best to ask anyway.

Sorry about the bad picture quality

 

[Eva]

Those capacitors between mains wires and earth (metal case) are required for EMI filtering, and there are special approved types (called "Y2") intended to be very reliable in this application.

 

[wine&dine]

The more I read from Elliot, the less I think of him.

"capacitor is likely to fail ... since DC capacitors are commonly used" - what utter BS. We're not talking about electrolytics, and other than that there is no "DC cap". Yes, using caps with insufficient rating would be dangerous; this is why there are "X' (live to neutral) and "Y" (either to ground) rated capacitors. See http://my.execpc.com/~endlr/line-filter.html for a good writeup,

"a capacitor may be used from both active and neutral to earth. This is an extremely dangerous practice, and is illegal in many countries" - he's even topping the previous nonsense. Open ANY computer PSU and you'll find one L-N (X) and two L-Gnd,N-Gnd (Y) caps.

The residual or leakage current cannot trip a GFI (RCD) if the capacitor conforms to regulations. The AC current has to be limited anyway to safe levels for those cases where ground is not grounded and someone touches the case. Have you ever noticed a slight tingling when touching a notebook computer on a charger without a ground connection? That's why they increasingly come with grounded mains cables now. Regulations typically limit the residual current to 3.5 mA (medical: 0.5 mA), a sensitive GFI/RCD trips at 30 mA.

With the typical Y capacitor value of 2.2nF, the maximum leakage current at 230VAC/50 Hz (assuming a low-impedance ground path, the worst case) would be less than 0.2 mA anyway. At 4.7nf+20%, 230V+25%, 50 Hz+10% we're talking about one-half a milliamp. See also http://www.ce-mag.com/archive/2001/janfeb/Georgerian36.html

 

[richie00boy]

I'm surprised what Rod has written there, yes it's not very good information.

Please can you give me the link to the exact article(s) that information came from and I'll see if he wants to revise them. (I'm a moderator on his forum.)

 

[pinkmouse]

But at least he is erring on the side of safety, which is a good thing.

 

[bikehorn]

I'm surprised what Rod has written there, yes it's not very good information.

Please can you give me the link to the exact article(s) that information came from and I'll see if he wants to revise them. (I'm a moderator on his forum.)

thanks everyone, i guess the caps and MOV are staying...

as for the articles:

http://sound.westhost.com/psu-wiring.htm

http://sound.westhost.com/power-supplies.htm