Relay issue

[lowell]

I am adding a relay to an amp. It's a 9v relay here is the datasheet.

http://www.mouser.com/ds/2/315/mech_...2y-1076009.pdf

The relay will not switch. I've tried 2 different relays...same model. They do switch with a 9v battery.

Relay coil current is 22.2ma per the datasheet.

Here's how I have it wired.

+31vDC - 1k 1w - relay coil - ground

31 - 9 = 22v
22v/.022 = 1k resistor

The voltage on top of the coil is 31v when the other end of the coil is lifted. When grounding the other end of the coil (low side switching) the coil voltage is + 9.15v.

So I'm quite confused as to why it isn't working. My calculations seem correct. Help!

[eschertron]

Your math seems good. I'm wondering if the 'holding' current is not much lower than the 'turn on' current. Any specs on that?

edit: OK, so I looked at the data sheet. all the current, resistance, etc., looks pretty much in line with 22mA and 9v for all conditions. The data sheet says it will take 200% of rated voltage (indefinitely?). so downsize your resistor until it pulls in. Or use a regulator to get your control voltage. I'm still thinking the 1k resistor is 'choking' the relay.

[The Dude]

How are you measuring the 9.15V- to ground or directly across the coil? Your math does look good. I'm wondering if your low side switching has some resistance or voltage drop that you are not counting in the calculation. If you haven't already, measure voltage directly across the coil.

[lowell]

Thanks.

I'm measuring across the coil. It's the DARNDEST THING!!!

I have the flyback diode in place. Anode to low side. I cut it out but same results. Baffling.

[eschertron]

with the 9V battery, did you have any issue with matching polarity of the relay coil to battery? Some relays/bases have a snubber diode built in and require correct polarity to work.

[g1]

Good point. The schematic shows pin1 +, pin 16 negative.

[The Dude]

I don't know what's in that relay, but if there were a built in snubber and the wrong polarity voltage was supplied, wouldn't you read a diode voltage drop across the coil instead of 9.15V?

[lowell]

Agreed dude. The chinaman is not the issue here.

But seriously... I have the +v at pin 1. So it is backwards from how they show it on the datasheet. It'd be quite odd if it works but I'll swap it tomorrow.

[loudthud]

Some relays use low cost metals in the magnetic path which become magnetized the first time the relay is tested at the factory. They call this a "feature" because it takes less Voltage to pull in the relay when the same polarity is used as when the magnetic path was magnetized. This comes from the old school "If you can't fix it, feature it."

[lowell]

It was the POLARITY! Interesting. I guess current has to flow the correct way to cause the magnetic field to work correctly. Thanks all!

[nickb]

It was the POLARITY! Interesting. I guess current has to flow the correct way to cause the magnetic field to work correctly. Thanks all!

Well that's one to watch. I see the pin out diagram does show the polarity on the coil terminals. I guess it must have a 'helper' bias magnet inside.

[R.G.]

Well that's one to watch. I see the pin out diagram does show the polarity on the coil terminals. I guess it must have a 'helper' bias magnet inside.

Maybe. It could also be as simple as having an internal flyback catch diode. Many small relays do have these diodes. It's odd that it is not shown on the datasheet if so. @lowell: it would be interesting to see if your multimeter says there's a different coil resistance one way versus the other. I would expect a diode to be cheaper than a small, precision magnet.

[Jazz P Bass]

This from Omron: "A Relay equipped with a diode that absorbs coil counter-electromotive force will not operate properly if the positive and negative voltages applied to the coil are connected incorrectly.".

I believe the 'bar' on the relay indicates that it is a polarized coil.

[lowell]

Oddly enough it reads the same in both directions. Resistance as well as semi. 410ohms/517mv.

[Enzo]

Do this experiment:
take a plain old 1N4007 diode, and clip a 410 ohm resistor in parallel with it. Oh 390 ohms is a close standard value. Now try measuring the diode. I think you will find the low resistance obscures the diode function.


In my days working on pinball machines, we drove the solenoids - the coils that worked the moving parts to slap the ball - with DC voltage and transistors. Those solenoid coils would destroy a transistor in an instant, were it not for a reverse diode across each coil. COnsidering a coil might be 550 turns of #24 wire, and so very low resistance, I had to convince them they would never detect a shorted diode with an ohm meter.