Step up computer power supply to 24V

[rdagger]

Does anyone know of any step-up circuits to boost a computer power supply output to 24V?

[2600]

How much current do you need?
Is there any reason you can't get a 24V power supply?

[rdagger]

How much current do you need?
Is there any reason you can't get a 24V power supply?

I not sure on the amps but probably about 1A to control a 24V solenoid.
I could buy a 24V p/s but I am trying to conserve space and I already have a slim external drive p/s that provides 5V (which I'm using for another circuit) and 12V.

[MonMotha]

1A would be pushing it, but you can grab 24V off of a computer PSU between the +12 and -12V rails.  Just be careful because the low side is no longer the same as the power common (ground).  Some computer PSUs don't really enjoy this as their topology tends to make them want everything returned via common, but many don't mind.  Check the current limit on your -12 rail (usually written on the side of the power supply).

Otherwise, you can get 2x charge pumps that'll double the 12V rail up to 24, but they're usually not good to a full amp, so you'd probably need to go with a boost switching power supply and run it off either your 5 or 12V rail.  I know I've seen them integrated onto prebuilt modules (they're also available as pretty much single chip + inductor + capacitor), as well.

[ShinAce]

If you have another spare 12 volt supply, you can daisy chain them to make 24 volts.

+ from one supply to - of the other, the remaining terminals yield 24 volts.

[2600]

I'd be careful of using the PSU power supply and the daisy chaining will only work if the supplies are isolated.  May work, you just have to be careful.


1A is what I figured.  There are things out there probably, but off hand I'm not familar witht something that will get you that much current.  You could try the charge pump or a DC-DC converter.  DC-DC converters easily have the conversion.  Then use a really BIG capacitor so that you get enough current.  Not sure if you can wait for the capacitor to charge back up.
(Derrick Renaud mentioned using a Big capacitor a couple of months ago, I forget what he paired it with so you may want to give that thread a search)

[hanelyp]

I not sure on the amps but probably about 1A to control a 24V solenoid.

How many solenoids and what resistance for the coils?  Time to get out the trust multimeter.  Divide 24V by the coil resistance, then multiply by the number of solenoids to get the amps you'll need.

Another possability, your 24V solenoid might work properly on 12V.  You can't hurt it by trying.

[rdagger]

The resistance on the solenoids are 240 ohms.  Therefore, according to hanelyp they should require .1 amp each.  Although, I spoke to an engineer and he told me that solenoids have an "active" resistance called reactance and you can't put an ohmmeter on them and simply measure resistance. Their resistance varies with the frequency of the applied signal - hence "active" resistance.
After some research, I built a voltage double circuit (see the pic).  I could probably use a better PNP but I used parts that I had on hand.  The circuit works.  My 12 volt p/s (actually 11.75V) can not drive the solenoids at 12V.  However, after hooking up the doubler circuit the solenoids works, abet slightly sluggish.   The circuit is putting out 22.75V.   I measured the current in series and each solenoids is only pulling .05 amps.  I suppose that this is misleading.  I think the solenoid probably pulls .1 A or higher when firing, but then requires less current to stay open.  Unfortunately, my meter is too slow to measure the actual current pulled during firing.
Any suggestions on how to make the circuit more efficient?

[rdagger]

I finally got the circuit working with 2 solenoids.  It turns out that C1 and C2 were the problems.  Lowering them from 100nF to 10nF instantly doubled the voltage I had available for the second solenoid.  Also increasing C3 provided an additional modest boost in performance.  I tried even larger values for C3 and C4, but there was no additional benefit.  Here is a schematic for the improved circuit.  I could have upgraded all the transistors to get an additional .5 volts, but it did not seem worth the cost.

[rdagger]

Here's a 3D rendering of how the board will look.