Hi!
First of all: real nice work creating the emon! I just love the concept of open source, and you seem to have pulled it off beautifully!
I'm but worried/confused about the safety of the current transformer SCT-013-000, though.
Measurement instruments are apparently classified in IEC 61010 (IEC 60664?) according to their resistance to flash-over etc. at voltage spikes. Instruments that are connected before the main breaker in e.g. a distribution board must reach CAT IV standards; if it is connected after the main breaker it must reach CAT III.
Now things get a bit unclear to me, but according to this site: http://www.ni.com/white-paper/5019/en this would mean the instrument must be able to withstand a voltage spike of at least 4000 V.
The datasheet of the SCT-013-000 specifies its dielectric strength to be 1000 VAC (1400 V peak?): http://shop.openenergymonitor.com/100a-max-clip-on-current-sensor-ct/. This would make this current transformer unsuitable for anything other than low-voltage use (not suitable for measurements on mains voltage at all).
On the other hand, Seedstudio classifies the same CT to 6000 VAC: http://www.seeedstudio.com/depot/noninvasive-ac-current-sensor-100a-max-...
Also, I did find this datasheet of a different CT which says: "When used on an insulated primary cable, the voltage category is improved according to the cable insulation category": http://www.pqmeterstore.com/crm_uploads/brultech_split-200.pdf. Of course then one would have to take creepage and clearing distances into account when mounting the CT.
Can you clear this up for me?
Re: SCT-013-000 dielectric strength and safety
"Instruments that are connected before the main breaker..."
Whether that applies to a CT or not would depend on how pedantic your definition of 'connected' is. surely.
Forgetting regulations and just being practical, in my house the meter tails run very close together at one point. The distance between L and N conductors is less than the distance between the L conductor and my CT windings. So if there was ever a flash which the meter tail insulation couldn't contain it would favour the path to the other conductor rather than the CT. Although I've no doubt the monitoring kit might also take a small hit.
Unless the system was being worked on, though, there would be no human contact with their the CT or its output.
P.
Re: SCT-013-000 dielectric strength and safety
I think a little common sense is called for here. As Paul points out, it needs a breakdown of the primary insulation before there is any risk of danger. Used as intended, there is no risk of galvanic contact between primary and secondary. Were the c.t. to be mounted on a busbar, for which purpose it is clearly not designed, then things would be different, because the primary insulation then becomes an unknown number of layers of tape that are probably OK until scuffed or cut. (Look carefully at the YHDC c.t. and you'll see what I mean).
Re: SCT-013-000 dielectric strength and safety
Ok, so the conclusion is that the SCT-013-000 is safe (i.e., used "with common sense" on an insulated wire it conforms to CAT III)?
What about the different ratings here and on Seedstudio? That seems a bit weird, don't you think?
Re: SCT-013-000 dielectric strength and safety
As Seed can't even get the ratio correct, I think I'd prefer to rely on the manufacturer's data. (Especially as it agrees with my measurements, and I've measured 6 SCT-013-000s now).