Infallible capacitor arrangement

[pj]

In the standards for intrinsic safety (60079-11) it talks about using capacitors in an infallible arrangement. Does anyone know what an infallible arrangement looks like, would it be two capacitors in series? Quote from standard below.

Thanks

Either of the two series capacitors in an infallible arrangement of blocking capacitors shall be considered as being capable of failing to short or open circuit.

[xrunner]

See if this document helps -

[floobydust]

The infallible capacitor is two in series, so that you have coverage if one shorts. Typically used going to an antenna/bias-tee or fuel tank probe to block hazardous DC from the IS section. A fuel tank can have large static electricity charges so a dissipating resistor would also be used.
The caps must pass the hi-pot test 6.3.13 of 500VRMS, be low enough capacitance value (one part) for spark energy with safety factor 1.5, and meet the separation distances across one capacitor. If the product is mains-powered the hi-pot requirement goes up.
So they are a large parts. I've used 500V ceramic caps but these are a bear for cellphone modems in antenna circuits, you have to use a microwave-spec'd part I find.
Certifier will apply 500VAC across one cap for 1 minute I recall, and 1,500VAC to GND.

[coppercone2]

Fuel tanks certainly pose interesting problems.

[floobydust]

I didn't know static charges build up on refined fuels (surface) such as gasoline, diesel, LNG. Thought they were non-polar molecules and not triboelectric but this is not the case. An electrode in a fuel tank, say a capacitance probe does get zaps like ESD during tank filling or  slosh so clamp diodes are needed. A 1MEG bleed resistor dissipates any charge. I did see another commercial design use GDT but thought that's a bad idea to hard discharge the electrode.

[Smokey]

How is there not a Pope joke here yet?

[pj]

Thanks for the info this really helps. My proposed device only has a voltage of 7.5v so hopefully the caps won't end up being too big!

Basically what I'm trying to do is limit the current that can flow into an external antenna under fault conditions, zone 0 ex ia. I have the system working at the moment with a current limiting resistor in series with the antenna but this is pretty hacky. I need to figure out these blocking capacitor arrangements.

[floobydust]

For a cellphone modem in zone 0 I've used two series 27pF 500V and there was a stripline to help cope with the bias-tee over all the bands.  Johanson 501S42E270JV4 size 1111/2828 metric and I might have made special smaller pads if the spacing was slightly low.

Unfortunately, the standard says it's to pass a hi-pot test as if it's mains powered.
I had a battery powered product and complained "there's no 500V anywhere" and the certifier said they follow the standard and it also ensures the spacings/clearances are decent. So I had to use the 500V caps.

[pj]

For a cellphone modem in zone 0 I've used two series 27pF 500V and there was a stripline to help cope with the bias-tee over all the bands.  Johanson 501S42E270JV4 size 1111/2828 metric and I might have made special smaller pads if the spacing was slightly low.

Unfortunately, the standard says it's to pass a hi-pot test as if it's mains powered.
I had a battery powered product and complained "there's no 500V anywhere" and the certifier said they follow the standard and it also ensures the spacings/clearances are decent. So I had to use the 500V caps.

Thanks again that's insane, the interpretation of the standards seems to be a bit of a black art. I will ask my certifier if they will require the hi-pot test and see what they say.