Driving solenoid valves

[DSGarcia]

I would like to drive some solenoid valves (about 500mA at24VDC) with a PIC. I know I could use a relay, but is there some other device I should use and how do I add the proper curcuit protections for noise and short-circuit protection? Would a polyfuse protect the device against over current (and short circuits)? I am trying to minimize physical space but still have a reliable long-life circuit.
Thanks,
Dale

 

[jpanhalt]

I would consider a transistor (2N2222) or mosfet switch on the low side of the solenoid with a protective diode across the solenoid coil. John

Edit: That would be a logic-level mosfet.

 

[Krumlink]

A 2N2222 can only drive 600ma max correct? I would be worried about spikes when the solenoid kicks in. A MOSFET would definately do the job, If you have any available, I would pick a IRF540N or a IRF630N, just becase they can handle general power apps and are cheap. If you only have 2N2222's make sure the spike doesnt kill them (poof).

 

[Boncuk]

A 24VDC solenoid normally causes a spike of 120V when deenergized. Use a fast rectifier diode which can withstand 400V. (UF5404 or similar)

Further try to employ a driving transistor with higher current capability. For the 2N2222 you are already at 83.33% of its nominal power. Use any BD type transistor or even better use a MosFet.

 

[jpanhalt]

A 2N2222 can only drive 600ma max correct? I would be worried about spikes when the solenoid kicks in. A MOSFET would definately do the job, If you have any available, I would pick a IRF540N or a IRF630N, just becase they can handle general power apps and are cheap. If you only have 2N2222's make sure the spike doesnt kill them (poof).

Can those Mosfets be driven effectively with a gate voltage of 5 V max from the PIC?

 

[DSGarcia]

Thanks for the suggestions everyone. There is no requirement to use a part that I may already have in stock. I will order whatever is best for the application. (I need to build several.)

I would like to handle solenoids up to 1.2A @ 24VDC if possible. Duty cycle is limited to 10 seconds per minute at the most.

I also need short-circuit circuit protection. The power supply is capable of supplying 100 watts. Is a polyfuse fast enough to prevent component damage?

Another requirement would be that I would rather use a part that does NOT require a heat sink for this application. For the final version of the circuit, I will fill the enclosure with potting compound (the kind that is thermally conductive). I am leaning toward a small aluminum box.

Thanks,
Dale

 

[jpanhalt]

I would like to handle solenoids up to 1.2A @ 24VDC if possible. Duty cycle is limited to 10 seconds per minute at the most.

I also need short-circuit circuit protection. The power supply is capable of supplying 100 watts. Is a polyfuse fast enough to prevent component damage?

With that higher current, I would recommend either a darlington power transistor (like the TIP series) or logic level mosfet. If this application is automotive or aircraft with 24 V systems, I would give myself some cushion with the Vds rating and go with something greater than 30 V. Digikey lists a few logic mosfets higher Vds ratings. In fact, I would probably go with the higher rating anyway.

Could you describe a little more what you mean by short circuit protection? 100W is only about 4 amps at 24V and short-circuit protection is often built into the power supply, not the load. The two classes of devices I mentioned can handle well in excess of 4 A for a short period, so I imagine a simple fuse would work. John

 

[DSGarcia]

Could you describe a little more what you mean by short circuit protection? 100W is only about 4 amps at 24V and short-circuit protection is often built into the power supply, not the load. The two classes of devices I mentioned can handle well in excess of 4 A for a short period, so I imagine a simple fuse would work. John

The system will have a 100 watt regulated power supply and it does have short-circuit protection. My concern is if one of the solenoid lines were to become shorted or perhaps partially shorted by a wire rubbing bare on a piece of metal, I do not want the momentary high current to blow out whatever device is driving the solenoid. If the device can handle the current, then that is all I was asking. The power supply can take care of itself.

Thanks,
Dale

 

[jpanhalt]

OK. Let's look at the possibilities. If the wire on the ground side of the solenoid rubs, controller is low-side, so it has zero volts in. The solenoid would pull in, however.

If the high-side rubs, you would depend on the power supply, or you could insert a fuse. The problem with inserting a fuse is you create another high and low side. Low side is protected, but high-side is not, and so forth.

The microcontroller should be protected by its power supply.

That leaves the "wire" from the mcu to the base/gate of whatever transistor you are using. I assumed they would be on the same board. You certainly do not want to run a long wire to the gate of a mosfet.

In conclusion, I don't see a significant risk in harming the mcu or the switching transistor by a short.

Now, you might want to consider whether you want to design a system to keep the solenoid from pulling in for more than a set period, which would indicate a short on its low side, but that is a more complex project.

John

 

[DSGarcia]

John,
Thanks for the information; I am satisfied as long as a short will not blow out the solenoid driving device. However, your comments touch on an interesting point and this next question is a bit off topic.

In my application I source the 24VDC to the solenoid rather than sink it. I understand the difference between how you do sourcing and sinking, but could you explain why one should sink rather than source power to a device and the advantages? Also, same question about sinking advantages except for switches and sensor contacts.
Thanks,
Dale

 

[jpanhalt]

This is the extent of my bench-level knowledge of it. Some of the trained EE's will probably have a lot more to add.

If you look at N-channel devices, such as the transistors and mosfets discussed above, the device turns on when the base/gate is at a certain potential relative to the emitter/source. By putting that device on the bottom (sink), the reference point is ground and the driving potential is readily known and produced.

Now, if you put the same controlling device on top as a source, the reference point is no longer simply ground, but is the junction between the emitter/source and the top of the controlled device. As the controlled device (like your solenioid) is turned on, its voltage drop will approach the voltage available, and the voltage at the junction between the emitter/gate and the controlled device will approach the voltage of the top rail. That is, in the case of a transistor it will be just the diode drop lower; in the case of the mosfet, it will depend on the current and the IR drop though the mosfet. In either case, your base/gate drive will have to provide a voltage that is sufficiently higher than that voltage to keep the transistor/mosfet turned on. That is particularly hard to do with mosfets, because you want something like 10V Vgs, which often means that you need to drive the gate above the highest voltage (top rail) you have available. That can be done, of course, with specialized high-side drivers, but for your application, it was simplier to use the conventional (i.e, bottom-side) control. An alternative is to use P-channel devices.

Hope that helps. John