I've got a small PIC-based TTL circuit device, powered by a 4.5V coin cell, that needs to drive a solenoid which requires 12V. Obviously my logic level device can't directly power that so I could either replace the cell with something stronger, or use a boost converter. Is there another option I'm missing?

This seems like it would be a very common problem so I don't want to reinvent the wheel here. What's the most common way to power higher voltage devices like motors, solenoids, etc from logic level circuits?

Thanks in advance.

  • 1
    \$\begingroup\$ Please see additions to my answer. Please tell us more about solenoid, battery, operate time, cycle time, cycles needed, cycle life, ... EVERYTHING.||We can do (almost) anything. We just need to know which version of anything is actually wanted:-) Which is why I am ALWAYS asking for a complete and competent problem definition. \$\endgroup\$
    – Russell McMahon
    Nov 1 '11 at 12:40
  • \$\begingroup\$ Re Perhaps the 9V is my best route - 1 x AA would do 3 x AA marvellous. BUT you need to give us a goo feel for what you really want. What are the solenoid specs (operate current, hold current, hold time) how often used. What are size constraints on battery etc. Why . How long mut battery last? What is acceptable boost cost? Do you need battery voltage for other ccts? What else does it power? etc. One off or 10 or 100 or 10000 or If you want a real answer then we need to know enough of the real problem to help properly. Full spec gets full answer in one pass ;-). \$\endgroup\$
    – Russell McMahon
    Nov 1 '11 at 21:14
  • \$\begingroup\$ @RussellMcMahon roger that. Still working on it but will update this thread when available. Truly appreciate all the feedback. \$\endgroup\$
    – NateFisher
    Nov 1 '11 at 22:38

There are many boost converter ICs available (as you probably know).

One which is low cost available and effective is the ancient but still worthwhile MC34063. If it's a one off and you want maximum performance and don't mind paying a few dollars for the IC there a better ICs than this. But for even modest volume this IC does well in many cases and is about as low cost as any and is very flexible and easy to use.

Disadvantages include

  • Not vastly efficient at low Vin -
    but that's not a problem here as you'd run it only when needed and for brief periods.

  • 100 kHz (or slightly mote) operation.
    That's not terrible but means the inductors are somewhat larger than with 500 kHz or 1 MHz or higher devices.

  • Not totally tiny. DIP-8 or SO8 packages.
    Only a problem if you want super miniature. SO8 is usually acceptable.


  • Extremely flexible topology. Has a driver with both ends available that can support buck, boost, buck-boost, SEPIC CUK and more.

  • Low parts count For the circuit you want it needs 1 inductor, 1 diode, 1 capacitor, 3 resistors (2 to set voltage) + a Vin and Vout filter capacitor. You can hardly get less parts (some include diode or use sunchronous switching).

  • Low cost.
    About 50 cents US in 1's. About 20 cents in volume. I get them for about 10 cents in Asia.

Here is a datasheet for the ST version - most major IC makers make them still.

See fig 13 for a stepup circuit. The 0.22 ohm resistor is an optional current limit sense resistor.

enter image description here

More after tea ... :-).


A question has been raised re required power and energy levels. I assumed that as you are using such a small battery you would be wanting to pulse the solenoid and not to hold it in. As we don't know the application this assumption is just an assumption.

A look at Digikey's site shows the samllest solenoids they sell have continuous operating powers in the 2 to 3 Watt range. You will be able to buy smaller and a custome one could be ,ade that was MUCH small power (and performance) wise.

If you assumed you wanted to pulse the solenoid to open a latch or release a rope or spring or ping a bell, then say a 0.1 second excitation pulse may be good enough. Assumptions can be adjusted. Say 3 Watts x 0.1 second = 0.3 Watt second or 300 mWs. Allow say 60% overal conversion efficiency so say we need about 500 mWs. As the battery cannot supply that instantly a capacitor must be used to store the energy. For charging purposes that's 500 mW for 1 second or 250 mW for 2 seconds or 50 mW for 10 seconds etc.

Battery type is unknown. 4.5V sounds like it MIGHT be 3 x Alkaline. so 3.xv would be close to true.

Regardless, such small batteries unless speciallu made have low max mA out. 1 mA may be typical for some. 10 mA would probably be mx in most cases.

Assume 10 mA, 4V available = 40 mW operating power mex. To get 500 mWs of energy we need to charge it for 500/40 ~= 12.5 seconds. Say 15 to 20 seconds to be safe. Could work out far less.

A CR2032 cell (3V Vout) typically has 220 mAh capacity. Multiply that by 4V in this example = 880 mWh capacity. Taking the 500 mWh we need from it for the solenoid takes 500 / (880 mWh x 3600 seconds /hr) = < 0.02% of battery capacity.
Notionally you could get 6000 + solenoid pulses from that batter. In practice Murphy says "less!"

Increase the solenoid on time to one second (you'll need a suitable reservoir capacitor) and operating energy does NOT increase by a factor of 10, because hold-in current can be much less than operating current. A factor of 5 to 10 is not unknown. That needs extra electronics to get a high current pull in pulse and a lower holding current, but it is not hard to design.


A proper description of the bttery is required.
Link to data sheet would be best. 4.5V coin cell is unusual. Does it have to be that small? A single AA battery has both vastly more energy content AND much higher acceptable energy transfer rate. On Watt continuous is acceptable and you can hold a small solenoid in indefinitely at that level if you need to.

If an AA battery is too big its nasty little AAA cousin may do. These are worse in all respects, less efficient on space (but smaller) and cost about the same. Or a flat Lithim Ion or Lithium Polymer battery can be had in very small capacities and may do the job far better.


What is the task.
Is there a datasheet? IS there any sort iof spec for pull in curremt, hold in current, hold ij time, stoke or force or ...?

  • \$\begingroup\$ @NateFisher, There is one important note that was not in this answer, your coin cell probably cannot handle delivering the power to power this solenoid. We could use more information on your battery as it will probably not function either way. \$\endgroup\$
    – Kortuk
    Nov 1 '11 at 9:31
  • \$\begingroup\$ @Kortuk - Energy doable - depends on assumptions. Please see additions to the answer. We can do (almost) anything. We just need to know which version of anything is actually wanted:-) Which is why I am ALWAYS asking for a complete and competent problem definition. \$\endgroup\$
    – Russell McMahon
    Nov 1 '11 at 12:39
  • \$\begingroup\$ Sorry guys about lack of clarity, I was talking generically about the cell, I didn't have one in particular. I was really just trying to avoid a massive 9V battery, which I still need to research based on this answer. Perhaps the 9V is my best route. \$\endgroup\$
    – NateFisher
    Nov 1 '11 at 16:42

You would need to use a boost converter, or add another couple of batteries.
What power rating is the solenoid?

However, a typical coin cell (e.g CR2032) will not be able to supply enough current to drive a typical solenoid. You mention 4.5V though, so it can't be a standard lithium coin cell - what kind of battery is it?

  • \$\begingroup\$ He needs to tell us more but it may be pulsed (eg de latch something) and the hold in current can be far less than the pull in current (as I know you know). Coin cell can charge cap for a pulse. As you say, not good for continous operation with most 12V solenoids. \$\endgroup\$
    – Russell McMahon
    Nov 1 '11 at 11:05
  • \$\begingroup\$ @Russell - Good point, hopefully either the solenoid or battery (4.5V?) is suitable for the intended use. \$\endgroup\$
    – Oli Glaser
    Nov 1 '11 at 14:20
  • \$\begingroup\$ @RussellMcMahon because it's going to need to hold the solenoid (it's a valve) open for a few seconds, I think the coin cell is not the best choice. \$\endgroup\$
    – NateFisher
    Nov 1 '11 at 16:44
  • \$\begingroup\$ @NateFisher - No the coin cell almost certainly won't work for this. You will need to check the power rating of the solenoid and run some calculations. If you provide the info I requested above (e.g. solenoid datasheet, proposed battery) we can try to find a decent solution. \$\endgroup\$
    – Oli Glaser
    Nov 1 '11 at 17:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.