Activating a 24V Relay

[NMP]

Hi all,

I have several Inputs that the voltage can vary from 60 to 90V.
To activate a 24V Relay im thinking of a 24V zenner in parallel with the Relay with a resistor in series to drop the voltage, but probably it will heat up a bit.

Any suggestions?

[Rerouter]

As your range is 60-90V, get the datasheet for your relays and work out the trigger and hold currents, and go from there,

I would say skip a direct zener approach, if you had to, a zener regulator with a transistor.

If you had a large number of relays, it may be worth considering a proper stepdown supply.

[mikerj]

What is the resistance of your 24v relay coil?

[NMP]

Thank you for the replies

Normally the Voltage is 70V.
A zenner with a transistor sounds good

Its this Relay
https://www.mouser.com/ds/2/316/mech_eng_ldp-908288.pdf

The coil resistance is 2.880 ohm.

[Seekonk]

Use a switching power supply.  I buy these 100-240VAC supplies to 12V 1.25A for a little more than a buck shipped with an aluminum case. They for sure have 24V similarly priced.  I regularly use these at 50V DC.  Your operating range would be easy for them.  Oversize these for current because their current capacity drops at these voltages.  There is a quick mod for any of these wall wart type supplies that won't start at lower voltages.

Go to a thrift store and pick up an old HP printer pack that is 31V and try it.

[Zero999]

Hi all,

I have several Inputs that the voltage can vary from 60 to 90V.
To activate a 24V Relay im thinking of a 24V zenner in parallel with the Relay with a resistor in series to drop the voltage, but probably it will heat up a bit.

Any suggestions?

You've over-complicating this considerably. A simple resistor will work much better than a zener diode!

According to the data sheet, the relay's maximum voltage is the nominal, multiplied by 1.3 and the minimum, 0.75 times the nominal. You need it to work from 60V to 90V, a ratio of 1.5, between the maximum and minimum, yet the relay will work over a ratio of 1.3/0.75 = 1.73.
https://www.mouser.com/ds/2/316/mech_eng_ldp-908288.pdf

The relay has a coil resistance of 2880 Ohms, so a 5k5 to 6k7 resistor will suffice. Use a 6k2 resistor and it will take into account that the coil resistance has a tolerance of 10%. When the supply voltage is 90V, the power dissipation in the resistor will be 0.62W, so use a 1W reistor and there will be plenty of headroom.

[NMP]

Hero999 you are so right.. I read the datasheet and i thought it where 2 ohm, not almost 3Kohm! that whould give way too much power to dissipate.

Thank you all for the help!

[JS]

More details of the whole device would be nice to help better, if you have power available other than the input signals I would use that and the input just as signal.

If going for the resistor approach, you coyld add a parallel RC in series with the coil so you hold with a lower current, long live the coil. Other more refined option is to build a CC source, two or three diodes, one transistor and two resistors would do the trick, this could allow much wider range of inputs and still be safe for the relay.

JS

[jitter]

Agreed!
Dissipating .62 W to operate a .2 W coil seems a huge waste of energy (relative, of course), especially if that relay must remain actuated for long periods of time.

[ajb]

You probably have a transistor to drive the relay, the coil is an inductor (obviously), and you'll need a flyback diode anyway, so all that's needed is to PWM the transistor.  Voila, deducted unregulated step-down converter with zero additional BOM cost.  If you wanted to get fancy you could monitor the supply voltage and adjust the PWM duty cycle to keep coil current roughly constant, but with a range of 60-90V that's probably not necessary.

[Zero999]

I agree that simply adding a resistor would be a waste of power, but that might not be an issue and it does result in the lowest possible BoM cost.

Yes, if a microcontroller is being used, then PWM would be a good idea. It would be the most efficient solution and wouldn't increase the BoM cost. The only downsides are switching noise and more code.

If you want a constant current source and a transistor, with a well-regulated control voltage is already being used. All that's needed is an emitter resistor and the current will be constant. The main downside is the power dissipation in the transistor will be very high, so perhaps it's not such a good idea.