I'm playing with a relay on a mega shield. The relay is non-latching, so I have assumed that it doesn't matter which way around the coil is connected. I'm beginning to think this may be a mistake.
When I've worked with relays in the past the coil is energized whichever way it's connected, is this always the case?
But some relays have a built in reverse biased diode, so if you wire them wrong then this diode is shorted out. These are quire rare however.
That was my first thought too - and considering my usual ability to come up with "gee, I haven't seen those before" in my stuff, that may be a consideration. The coil itself should not matter, however, if there is a diode as Grumpy_Mike says, that would change how you hook it up. Some relays like reed relays may care, but not the normal relay.
I think I've got an answer. It appears that the relay I'm working with does have a polarized coil.
Here's the relay....
and here's the datasheet.....
Look half way down the 2nd page, on the right hand side and you can see the schematic shows a '+' next to one of the relay terminals (monostable version), this suggests that it should be positive! No mention of any internal diodes, but if this relay is from the same family as a latching version then maybe the internal mechanism shows some similarities with the latching one.
Normally if a relay coil comes with a diode it will be plainly stated in the brochure and data sheet. The diagram you refer to is only showing a DC power source, not the polarity of the coil.
On the data sheet I find this reference:
Operate time typ. 1ms, max. 3ms
Release time without diode in parallel typ. 1ms, max. 3ms with diode in parallel typ. 3ms, max. 5ms
This information would not be present if there was a diode inside the enclosure.
I have hacked the shield and swapped the wiring on the relay coil - it now works fine.
The datasheets are a bit confusing, the top level datasheet for for the IM family of relays clearly mentions polarity, but the datasheet for IM-C relays is a lot less clear - the only indication is a + and - symbol on the schematic.
Yeah, some data sheets (especially foreign ones for stuff on ebay etc.) are seriously lacking in useful information (or lacking all together) which makes it tough to get things working as expected when the information you need is simply not available. I ran into that recently with a battery isolation relay for my truck - relay has some smarts in it and connects/disconnects based on time and battery voltage. It also has another control terminal on it labeled "boost" where when you connect it to ground, it will connect the batteries together so you can boost a "dead" battery. What they don't mention in the data sheet for it is that if either battery is below 9v, it will not connect them. That little tidbit is missing from the data sheet (it also will not connect the batteries in the normal configuration if either battery is below 9 volts). Happened to find that sort of mentioned somewhere else in a forum and verified it myself. Details details who needs details?
Coils themselves aren't polarized, the question is the circuitry they're used in. Some relays switch depending on polarization, most do not. Those bare relays that don't rely on polarization should still in practice be polarized in wiring because of the flyback diodes that should be present.
You may find some polarized relays but most are not.
Some older relays used an internal magnet to aid armature movement, therefore the direction of current flow through the coil was important.