| Tested old one with ohm meter which shows open. Am I getting a good test? | (Literature with new one says both sides must be energized for GFCI to work)
An ohm meter is only going to tell you if the contacts are closed. I've never heard of a GFCI breaker that won't close unenergized, but that doesn't mean there aren't any.
| If not a good test, how can I check old one?
You really need to test its ground fault detection circuitry. There are testers that do this. If you have to rig one up, a nice way to do this would be to have a box that has the ability to switch in and out a few different milliamp current levels for leakage detection. These would be connected between the hot phase (or phases) and ground (not neutral) through an appropriate resistor. The design would depend on your electrical system. If you are in Europe with 230 volt single phase, it could simply have 3 switches that each control a 115k ohm 1 watt resistor. This would give you the ability to choose 2ma, 4ma, and 6ma of leakage current. I might do it with 6 switches for 1ma increments and maybe go to a higher level of leakage just to see if its an out of spec detection. This assume your standards for leakage current are 2 to 6 milliamps as here in the USA.
Testing its overcurrent protection is harder to do and riskier. Some day I want to rig something up like that myself. But I'd never recommend that to anyone (EEs qualified to do so will know that they are).
| Also, is above described breaker available in "designed for switching"? | | Given the cost of this breaker, I was wondering how many duty cycles they | are | designed to handle when used as a switch. The recommendation is to check | them | every thirty days.
My understanding is breakers in this range can handle 1000 cycles or more. This should give you 83 years of monthly testing.
| In other words, since the old breaker probably had no more than one hundred | (100) | on/off cycles over a span of six (6) years, why did it fail?
That's unknown at this point.