A question about MOVs / Varistors.

[McBryce]

Hi,
As I usually only work on battery powered systems up to 48V, it's rare that I have to deal with varistors. Today I had a control unit (fish tank pump / filter controller) to fix which has 6x 230VAC outputs. Each output had a varistor across the switch contacts of the ouput relay. They are Epcos S20K275. All have infinitely high resistance when measured out of circuit, but at 230VAC they all allow the full 230V through, overriding the relays function. Removing them solves the issue, so obviously they're somehow damaged. My questions:

1 - How can I reliably check this type of varistor out of circuit to know whether it's broken (without applying the 230V)

2 - Only 3 of the outputs have ever been used, but all 6 varistors seem to be damaged? Was the device possibly hit by lightning, or what could have caused this?

Thanks in advance for any tips.

McBryce.

[Seekonk]

I used to manufacture snubber devices.  Sometimes after potting they would mix up the batches so I would have to identify what was what.  For movs I had a low current voltage multiplier and looked for the voltage knee at about 1-2ma.  For a 275 mov that would easily be 400V.  There aren't really solution below that voltage.  The voltage can change with a large surge, going up as well as down.  You don't say what these are connected to.  Capacitance can be several thousand pf and you may be seeing capacitance effects if these are on the switches.   

[McBryce]

The MOVs are across the common and normally open contacts of a small relay (Tyco - 24V coil, switching 230VAC). They switch AC motors (pumps and the drum of a drum filter) depending on water levels. The effect after failure is quite capacitor-like, maybe I'll connect one to an LC or ESR meter and see what they read

McBryce.

[T3sl4co1l]

You say they're letting AC through.  How much?  Did you measure?

[McBryce]

Here's the schematic of an output. The 230VAC is measuring correctly. At the output terminals it reads 230VAC when the relay is actuated and 190VAC when the relay is turned off.

McBryce.

[Seekonk]

If there is only a meter on the output I can easily see that much voltage being read by the meter.  As I thought the problem isn't real.  If it makes you sleep easier, place the MOV across the output.  A much better place for it anyway.

[McBryce]

This isn't a design I'm doing, it's a commercial device I'm repairing. Moving the MOV somewhere else isn't an option. With a working MOV the voltage across the ouputs is about 14VAC when the relay is open.

McBryce.