# How much power to drive this actuator?

I have a linear actuator that I need to design a circuit for, where an Arduino is going to control it.

Here's the datasheet (pdf) for the actuator: https://www.red-magnetics.com/en/product-groups/linear-solenoids/cylinder-solenoids/its-lz-2560-z/?pdf=1

It's the first time I'm using an actuator so I'm not sure how to read this datasheet which doesn't seem to have much information. So I have to choose a power supply and an appropriate MOSFET for this, but what would be the power rating?

I have the 6V version of the actuator, and it says in the datasheet that at 100% duty cycle, it uses 100 W of power. Would the power supply need to be able to deliver 100/6 = 16,7 A?

Related question: Linear solenoid specifications

• Poor datasheet, it would be clearer if it specified coil resistance. I would guess that 6V was for 100% duty cycle. But I have seen electromagnets where the rated voltage was at 50% duty cycle. Jan 12 at 14:11
• The data sheet is wrong BTW. Use common sense. Do you think that 100% duty would equate to 10 watts when 10% duty corresponds with 100 watts? Jan 12 at 14:21
• @Andyaka I was also confused by this. Jan 12 at 14:24
• @Andyaka the best I could think of for that table is “Thermal duty cycle” - 10 W constant, defrayed for higher powers over the times given. It’s disappointing that they’ve not provided a DC resistance so we can determine an electrical duty to hit these power levels. Jan 12 at 14:25

As AndyAka has pointed out - thank you very much - the table does not refere to electical power.

Therefore, you need to find the DC-Resistance of the Actuator from some other source.

Once you have determined it, you can use the Driver-Stage found below. Just design the 6V source, Q1 and D1 to whatever nominal current is required.

Here you have: Full Driver stage with 100% test-point coverage during developement and production.

simulate this circuit – Schematic created using CircuitLab

R1 Sets the Default-state for the Actuator to "OFF" if Arduino is in RESET or POWER-UP.

R2 Limits the peak drive current into the FET Gate to 5V/1K = 5mA.. Could go down to 333Ohms to get faster turn on time.

R3 Prevents resonance between C1 and the Actuator Coil during switching. Make sure, it can handle the power impulse when charging/discharging C1.

C1 Supplys current during switching events so the DC/DC or LDO does not have to do all the heavy lifting. Can be reduced to 33uF i guess.

NOTE: The C1/R3 components are optional. I would include generic footprints in the layout, and check if they are really required during testing.

D1 Prevents negative voltage spikes during switch-off of the actuator. Use any low Reverse-Voltage Diode. Make sure the reverse voltage stays low even for high current surges.

Q1 Use a N-Channel FET to switch. use some type with low RDSon which can handle the power dissipation for approx. 3A cont.

NOTE: While using a N-Channel FET will give you a lower voltage drop across the switch and therefore more voltage across the actuator which results in higher drive strength, FETs are more "sensitive" than good old NPN-Transistors. To make the circuit more reliable in case increased power-loss is no concern, i would use a NPN.

If you use a NPN, you could use 7V nom. to drive the Actuator. The 0V7 drop across the NPN will then be "compensated".

LDO/DC-DC Convert your input power to the &V required to drive the actuator. Design for approx. 2A cont. current. Make sure, it can handle the initial current surge when switching without dropping in voltage to much. This increases switch-on time. Make sure the bypass capacitors do not resonate with the Actuator coil.

R7/C2 Forms a RC-Snubber across Q1 if ringing is a problem. Can be unpopulated if not.

D2/D3 These TVS-Diode are just for good measure, as your actuator is connected via a Terminal-Block and external cabeling. Place close to the Terminal block and use any "reasonable sized" VBr > 9V device.

R4 Is a dummy footprint which can be populated with a 0-Ohm jumper during production. Is used for in-series measurements during developement or can be populated with a resistor in case there are resonance problems.

TP1/TP5 Can be used to verify the "driver side" during developement.

TP2/TP3 can be used to validate the power supply and switch performance during developement.

TP6 can be used to verfiy the "buffer" performance during developement.

NOTE: You can use R3 for in-series measurements during developement.

TP7 Can be used to verfiy the power source performance.

R5 Can be used for in-series measurements of the LDO/DC-DC input during developement. Can be replaced with a 0-Ohm jumper during production.

R6 Can be used for in-series measurement of D1 during developement. Can be repalced with a 0-Ohm link during production.

TP8/TP9 Can be used to validate the RC-Snubber.

NOTE: Can use R7 for in-series measurements.

NOTE: You can use the screw-terminals for voltage measurements across the actuator during developement.

• I would make R3 = 0. Add a separate snubber if the ringing is excessive. Jan 12 at 14:16
• @electronicstudent did you notice the complete error in the duty % row? Jan 12 at 14:19
• If it's not electrical power, what power do you think it is? Mechanical work done? As discussed in comments under Andy's answer, the table makes sense if it's instantaneous power during the ON part of a duty cycle, with bursts of power levels above what it can handle for continuous-ON operation. Jan 13 at 16:39
• Hmmmh this means the on-resistance of the FET is roughly the same as the resistance of the actuator. Also a lot of power is lost in the FET this way. Are you using a sufficient Gate-Source Voltage? If you are using a 5V0 GPIO to drive it: Are the resistors okay? Could you please measure the Gate-Source Voltage and the Resistance of the actuator? Jan 19 at 16:28
• If you source 12V across the FET and the actuator in series and the voltage across the actuator is ~6V, your FET must have a somewhat equal resistance to the FET. You could use a different FET (You dont need this type of bulky FET - use one with smaller VGs-On) or you could use a second small signal one to drive the Gate of the power one from the 12V directly Jan 23 at 13:59

I wouldn't use that solenoid on the basis of the information in the data sheet (unless you can find a better data sheet of get it corrected): -

After all, why at 100% duty would it take 10 watts whilst at 10% duty it takes 100 watts. I expect that the figures need reversing but, that's quite a big assumption given how poor the data sheet generally is.

I have the 6V version of the actuator, and it says in the datasheet that at 100% duty cycle, it uses 100 W of power.

No it doesn't and, that's therein lies the problem.

• I expect that the figures need reversing My guess is the main problem is attributing the power values with Drain, and dissipation tolerated would be more helpful - they all average out to 10 W. Jan 12 at 14:42
• That doesn't make sense because then you'd have to reverse the max duty seconds figures as well. Jan 12 at 16:51
• ? At the end of one cycle of 20 seconds, it would have absorbed 1.0*10*20 = 0.5*20*20 = 0.25*40*20 = 0.1*100*20 = 200 J. How exactly should reverse be interpreted? In row? In column? Jan 12 at 18:58
• I think interpreting "Power drain" as "maximum permitted coil power" makes a lot of sense: I read it as "infinite time at 10W, a maximum of 150s at 20W, requiring cooldown period of the same time afterwards, a maximum of 75s at 40W, requiring a cooldown period thrice the active period afterwards or a maximum of 30s at 100W, requiring a cooldown nine times the active period". Nothing needs to be reversed for this interpretation. Jan 13 at 10:21
• @MichaelKarcher: Yeah, max power during the on part of a duty cycle, not over the whole cycle. That's why there's a max duration specified for power levels higher than the 10W it can sustain indefinitely for continuous operation. Seems clear enough, although speccing only in Watts, not also volts and amps for the different power levels, seems unclear. Like will 6V push enough current through it for the 10W continuous or the 100W short-burst, or does it depend on whether the solenoid can physically move... Jan 13 at 16:37