Question about relay voltage

[zmwworm]

I bought some relays to use with some roll bar lights on my truck. I tested them with a 12V battery to make sure that they would work, and started soldering everything together when I noticed two different voltages on the label. The label is as follows:

896H-1CH-D1
Coil:12VDC
50A/30A
14VDC

I don't know what the two different amperages are for, but I assume that the "Coil" rating means that the relay is capable of switching 14V but the coil should only be run with 12V. My truck puts out 14.6V. So what can I do? I can use up to a 100 ohm resistor in series with the relay and it still works, would this keep the relay from getting fried (or will it have no effect?)

[dyarker]

Here's the spec sheet:
http://www.wesgarde.com/Product%20Library/896H.pdf

The two contact current ratings are for the normally open (NO) and normally closed (NC) contacts.

Max continous coil voltage is 120% of rated voltage (14.4V), so a 10 to 18 Ohm resistor might be a good idea.

Cheers,

[zmwworm]

Thanks a ton, that spec sheet is really helpful. I knew that resistors were current-limiters, but I didn't know if they would protect something from voltage. Anyway, after reading your post I bought some 68 ohm resistors from Radioshack and we'll see how it works. I used 68 ohms because I want the most protection but I also want the relay to work with a worn out battery, 100 ohms might have been too high.

[dyarker]

You've already tested with 100 Ohm, and bought 68 Ohm, okay.

I think with a resistor value that high the relay may possibly drop-out momentarily when you hit a pot hole. If this never happens, okay; if it does the problem is not enough current to hold the contacts against the G-forces. The cure it to lower the resistance.

[MrAl]

Hi there,

One question i would ask is does that relay already have a coil flyback
diode or do you have to add your own?

[dyarker]

If the part number in the first post is the whole part number, then there is no diode. The ones with "-001" or "-002" after the "D1" have built in diode.

zmwworm,

I think where Mr Al is headed is that you should add a diode (normally reverse biased) across the coil. Almost always a good idea, absolutely essential if switching the relay with a transistor.

[haklesup]

I think where Mr Al is headed is that you should add a diode (normally reverse biased) across the coil. Almost always a good idea, absolutely essential if switching the relay with a transistor.

True, Almost always a good idea. In this case since an automotive power system is inherently noisy and lighting is generally immune to these kind of voltage transients, I would say it is unnessary unless there are sensitive electronic loads whose own power supplies are insufficient to filter noise (and any automotive radio or amp should be able to). I don't use diodes on my lighting load switch and the only problem I recall is that an older CD changer would reset when toggling the fog lamps. You'r call, not needed but could help.

A series resistor to lower coil voltage is not necessary and in fact will lower the overall reliability by adding a component which can fail easier than a relay coil (more self heating and less protective package). The Coil Ratings table in the spec sheet clearly allows 120% coil V at continuous duty.

[MrAl]

Hi again,

Yes i was thinking of a transistor as one possible turn on device
which needs protecting with a diode, but just about any other
device upstream might also be better off if there is a catch diode.
This is even true for a switch because the switch can arc over with
1000v spike and possibly pit the contacts each time. Also, without
a diode when the circuit opens up (to turn off) and the voltage
jumps up, the coil wire enamel has to handle a higher voltage
than it should have to. If this eventually breaks down the insulation
then the relay is a pile of junk after so many operations.
Then there is the EMI too as hackle mentioned. The diode lessens
this problem too.

I dont think i ever ran a dc relay coil without a diode, for the above
reasons.

[zmwworm]

A series resistor to lower coil voltage is not necessary and in fact will lower the overall reliability by adding a component which can fail easier than a relay coil (more self heating and less protective package). The Coil Ratings table in the spec sheet clearly allows 120% coil V at continuous duty.

Less protective packaging shouldn't be an issue, I soldered and shrink-wrapped the resistors right to the wires, which probably makes the other aspect worse, but I didn't even think about the fact that resistors dissipate heat until I was in the middle of using electrical tape over the shrinkwrapping. Do you think that this could be an issue?

Hi again,

Yes i was thinking of a transistor as one possible turn on device
which needs protecting with a diode, but just about any other
device upstream might also be better off if there is a catch diode.
This is even true for a switch because the switch can arc over with
1000v spike and possibly pit the contacts each time. Also, without
a diode when the circuit opens up (to turn off) and the voltage
jumps up, the coil wire enamel has to handle a higher voltage
than it should have to. If this eventually breaks down the insulation
then the relay is a pile of junk after so many operations.
Then there is the EMI too as hackle mentioned. The diode lessens
this problem too.

I dont think i ever ran a dc relay coil without a diode, for the above
reasons.

You know, I almost asked about using a diode (I had heard that they protect the circuits that drive relays), but I decided that it would be unnecessary since there were no transistors involved and none of the trucks other relays have diodes (that I know of). I forgot about the fact that I'm using switches rated for AC rather than DC, and I didn't know that a diode can also protect the relay itself, so now I feel really bad about finally cramming all those fuses and relays under my dash without diodes. I think I'll see how it holds up, I'm probably going to have a problem with wires comming loose from the relay connectors eventually, and even if I don't there's no way a radio is going to fit if I don't change something anyway, so when I tear it apart I'll add some diodes. Or maybe I'll wait till a relay or switch burns out, either way. I got three switches for $5 at Radioshack, and the relays were about $4 each at an auto parts place, not horribly expensive but not cheap either, and certainly not easy to replace.

One question, is there any reason not to use LEDs as diodes? The benefit is that you can see them flash when the relay opens so you know that everything is working, but will they eventually burn out with so much voltage? I've used a relay with an LED on it for a short time, but not long term.

[Bigglez]

Greetings Zach,

is there any reason not to use LEDs as diodes

LEDs are fragile, and can't be used in place of a power diode.
The reverse voltage rating of most LEDs is 5V, in this application
almost the full battery voltage (10 - 16V) would appear across
the LED when the coil was energized.

If you want to add an LED across the coil with a ballast resistor
for the LED you'll have a fool=proof "relay active" indicator.
Of course, you'll still need the power diode to protect the
LED from damaging reverse-emf when the relay is released.

Comments Welcome!

[zmwworm]

Greetings Zach,

is there any reason not to use LEDs as diodes

LEDs are fragile, and can't be used in place of a power diode.
The reverse voltage rating of most LEDs is 5V, in this application
almost the full battery voltage (10 - 16V) would appear across
the LED when the coil was energized.

You mean only when the coil is released, right?

If you want to add an LED across the coil with a ballast resistor
for the LED you'll have a fool=proof "relay active" indicator.
Of course, you'll still need the power diode to protect the
LED from damaging reverse-emf when the relay is released.

Okay, the LED was just an intelectual curiosity. Are power diodes something special, or can I use one of my regular diodes?

[Bigglez]

Greetings Zach,

You mean only when the coil is released, right?

Well, a PIX is worth a thousand words...

When the transistor conducts both the LED and the relay
are energized. Some energy will be stored as a magnetic field
in the relay coil. Upon removal of the current in the relay the
collapsing magnetic field will reverse the voltage across
the relay coil and force the back-emf diode to conduct.

Okay, the LED was just an intelectual curiosity. Are power diodes something special, or can I use one of my regular diodes?

The diode must dissipate the energy in the coil as heat.
So a power diode (aka a rectifier) is needed. If the relay
is small (limited energy storage) a signal diode will suffice.

Comments Welcome!

[zmwworm]

The diode must dissipate the energy in the coil as heat.
So a power diode (aka a rectifier) is needed. If the relay
is small (limited energy storage) a signal diode will suffice.

The relay coil draws 133 ma, has a resistance of 90 ohms, and is run with 14.4V. I have no idea as to what a regular Radioshack diode can handle, maybe you do? I would think that the extremely short amount of time that the diode is actually used would mean that it could take a bit more than it was designed for, but maybe I'm wrong.

[Bigglez]

Greetings Zach,

The relay coil draws 133 ma, has a resistance of 90 ohms, and is run with 14.4V.

We'd need to know the coil inductance to calculate the
stored energy. As a practical matter, as long as the
diode does not breakdown when the coil is energized,
and the can carry the discharge current of the relay
coil, almost any rectifier diode will do.

I have no idea as to what a regular Radioshack diode can handle, maybe you do?

Do you have a part number or data sheet?

I would think that the extremely short amount of time that the diode is actually used would mean that it could take a bit more than it was designed for, but maybe I'm wrong.

For solid-state electronics it is the fast rise and often short
pulses that do harm...

For any competent design none of the components
should be stressed beyond their recommended ratings.
Abs. max rating are usually given to allow designers
to be inside the safe operating area under all normal
operating conditions of their circuit.

Comments Welcome!

[dyarker]

For that relay, in a vehicle, a 1N4001 (1A, 50V PIV) should work fine. As will 1N4002 (1A, 100V PIV) or 1N4003(1A, 200V PIV).

If using solid state relay driver, a faster diode would be better. Like 1N4148 for coil currents 150mA or less. A 1N4148 would work electrically for this project too, but is not as physically rugged as 1N400x diodes.

Cheers,