Electronic – Using a relay switch to switch power to solenoid

Presumably there is a regulator between the 12V supply and the arduino. As long at the 12V supply remains high enough for the regulator to function, it will keep the voltage to the arduino constant.

However, regulators aren't perfect. In particular, they can only compensate for input power variations up to some frequency. Noise above that will be passed to the output. If you think the input power might be noisy enough so that the regulator can't block it all, put a small filter on the input of the regulator. Most regulators will work OK up to a few 10s of KHz. You didn't say how much current the arduino draws, but you have plenty of voltage headroom. A 10 Ohm resistor will only drop 1V at 100mA, which still leaves plenty for the regulator to work with. That followed by a 10uF ceramic cap to ground right in front of the regulator will form a low pass filter with 1.6KHz rolloff. That should be good enough to keep the input of the regulator down to frequencies it can deal with.

Of course the best way to handle noise is to not make it in the first place. You absolutely need a diode or some kind of snubber to catch the inductive kickback when the solenoid is turned off. The main reason is to keep from frying whatever is switching the solenoid, but a secondary effect will be to decrease overall noise.

The following summary is based on the supplied table and the derivation is explained below. The table is based on a common solenoid frame which has various coils available. Wire sizes range from 27 awg (27g) to 37 awg (37g) with number of turns and coil resistance varying with wire size.
"Power" for the purposes of solenoid activation force is taken as ampere-turns in the solenoid coil (see below):

• At 12V with the specified 34 gauge winding you can power the solenoid on for up to slightly over 5 seconds and then need a 15 second pause to achieve a 25% duty cycle.

• You could run the 34g solenoid on up to 20V for up to 2 seconds and achieve more power than you can get at 12V, or ...

• You could run a 32g version of the solenoid at 12V for up to 2 seconds to achieve more power than you can with the 34g solenoid. With your desired 100 mS on you can apply 12V for 100 mS and then repeat again after 300 mS and repeat that cycle indefinitely.

• For maximum power with a 34g coil and 100 mS on you could run at 20V and 100 mS on followed by 300 mS minimum off. The 12V/32g and 20V/34g arrangements produce the same solenoid power)

• The design limit is to not exceed an average dissipation of 3 Watts, with an instantaneous dissipation of up to 30 Watts for 2 seconds allowed, followed in all cases by a long enough pause to reduce the average to 3 Watts.

The following is derived by rearranging the data supplied and applies to all coils:

• ton max = (60 / Watts_on) seconds

  (Watts max = 30)

• Tcycle >= (Watts_on x t_on_actual) / 3 seconds

The table is saying that to get maximum allowed power for up to 2 seconds (and your 100 mS is a subset of this) you can operate the solenoid at 30 Watts at 20 degrees C, with 847 amp turns of drive.

No other condition provides more power and no higher instantaneous power level is shown as allowed.

A.t = Amp.turns = Amps in coil x number of turns.

The 34 awg entry says DC resistance is 13.1 ohms,
so at 12 V you'd get I = V/R = 12/13.1
= 0.916A.

Turns for the 34g version are shown in the table to be 567
so Amp turns would be 0.916A x 567t = 520 A.t.
This is below the allowed 847 At for 2 seconds.
It is about equal to the allowed 536 At for 5 seconds.

At 12V Wattage = V^2/R
= 12^2/13.1
!= 11 Watts
which is well below the allowed maximum.

This means that if you want genuine maximum power for the solenoid frame provided, based on table data, you can either

• Run the 34g solenoid at 20V

or

• Run a 32g version of the solenoid at 12V. (Actually 11.9)