Homemade solenoid

[Technician1002]

also thought it would be quite fun to use a big solenoid with a blade attached inside a chamber to rupture a burst disk

use an air cylinder... that will work better

At high activation speeds, energy is lost to eddy currents unless there is some sort of barrier to the current. This arrangement and this loss is demonstrated very clearly in another high impulse current application, the coin shrinker and can crusher. This is a higher power application of the impulse. Lower power won't crush the cylinder, but energy loss is very real.
[youtube][/youtube]
http://members.iinet.net.au/~pterren/misc1.htm

[jimmy101]

There are many issues with trying to create a fast electric solenoid. Lets cover a few. Some solutions are mentioned but without details.

1. Inductance. A coil of wire is an inductor that will resist step changes in current. A high voltage to force the issue is a solution. To prevent the current from rising too far over time, a capacitor that discharges over the same time make a nice fast risetime in the coil. As you add turns to increase the power, you also add resistance and inductance that in turn slow it down.

AlphaOmega will definitely want to find a decent sim program for the coil. There's a pretty good one used by the coil gun folks called "Barry's RLC simulation".

The coil + cap have to be tuned (designed) as a set. For a particular coil there is an optimal cap and vice versa. Once you choose one part then for the other "bigger is better" isn't true.

I kind of doubt a solenoid will have enough ooomph .

You could try the photoflash board from a disposable camera. They run off an AA battery and the cap stores about 5J of energy (120uF at 300V). I would WAG the expected efficiency of a solenoid at a couple percent. Not sure if ~0.1J is enough to work a valve. Larger caps are a possibility but that will change the needed coil design. (For comparison, a 0.177 BB from a cheapo BB guns is about 1.5J.)

In the design phase you'll probably want to decide how fast the piston needs to move, which gives the time the coil is energized. Then design the coil + cap combination so the RLC time constant is similar to the desired coil energization time.

[Technician1002]

There are many issues with trying to create a fast electric solenoid. Lets cover a few. Some solutions are mentioned but without details.

1. Inductance. A coil of wire is an inductor that will resist step changes in current. A high voltage to force the issue is a solution. To prevent the current from rising too far over time, a capacitor that discharges over the same time make a nice fast risetime in the coil. As you add turns to increase the power, you also add resistance and inductance that in turn slow it down.

AlphaOmega will definitely want to find a decent sim program for the coil. There's a pretty good one used by the coil gun folks called "Barry's RLC simulation".

The coil + cap have to be tuned (designed) as a set. For a particular coil there is an optimal cap and vice versa. Once you choose one part then for the other "bigger is better" isn't true.

I kind of doubt a solenoid will have enough ooomph .

You could try the photoflash board from a disposable camera. They run off an AA battery and the cap stores about 5J of energy (120uF at 300V). I would WAG the expected efficiency of a solenoid at a couple percent. Not sure if ~0.1J is enough to work a valve. Larger caps are a possibility but that will change the needed coil design. (For comparison, a 0.177 BB from a cheapo BB guns is about 1.5J.)

In the design phase you'll probably want to decide how fast the piston needs to move, which gives the time the coil is energized. Then design the coil + cap combination so the RLC time constant is similar to the desired coil energization time.

In this application another factor enters that make the simple RCL too simplistic. In RCL, the energy remains in the components. In a valve some energy is transferred to mechanical motion of the core. Because of this the energy time constant is no longer valid. I don't have the answer and would tackle it with a trial an error approach monitoring the current during the actuation to get real world results. In reality the motion creates a back EMF force just like any electric motor accelerating. Initial energy consumption is large resulting in high current. The current drops off as the plunger moves in so a higher than predicted voltage is present for the pull-in stroke. When the core stops due to end of travel the back emf drops again.

I agree for the fastest operating times a tuned circuit is best, but the energy will need to provide energy for motion, high initial force, and the cap discharge at the end of travel. Circuit protection should be included as a reversal of polarity is most likely on the coil.

The basics are simple enough, but modeling the entire event is a complex model.

The idea was fantastic. Good find.

[jimmy101]

In this application another factor enters that make the simple RCL too simplistic. In RCL, the energy remains in the components. In a valve some energy is transferred to mechanical motion of the core. Because of this the energy time constant is no longer valid.

Maybe, maybe not, depends on how the system is setup. If the coil simply accelerates the piston and the coil is off when the piston strikes the valve then the system is identical to a coil gun. RLC is still an approximation since the magnetic susceptibility of the core is changing but the RLC simulation should be close.

If the coil is still energized when the valve is hit then you have to supply the work to move the valve while the cap is still discharging.

In the first case you would probably want a fairly short RLC that gets the piston up to the correct speed. The KE of the piston does the work.

In the second case you probably want a much longer RLC since it isn't necessary to get the piston moving all that fast if it is operating the valve directly.

In either case you will definitely have to do some tweaking, so set the coil up so that turns can be added and removed fairly easy. The coils for the two designs will be completely different. The coil gun like setup might have just a couple turns of thick wire, the second design may have hundreds or thousands of turns of fine wire.

Coil gun folks sometimes wrap their coils with VHS tape. The thought is that the ferromagnetic coating on the tape acts like a metal shell and helps to focus the magnetic field. Not sure if it really helps all that much.

[Technician1002]

Psudo science galore... Scuse me while I quit laughing..

OK.. Coil guns, they have one turn in the inductor. The current discharge cycle power discharge is mostly LRC. They operate on very high voltage, The back EMF from projectile motion is insignificant. The entire cycle resembles a stalled motor. A solenoid with many more turns and operating on under 100 volts instead of several thousand more closely models an electric motor. A very low turns solenoid with a capacitive discharge will perform somewhere in between.

The tweakable coil is a very good idea if you don't have the engineering background and wish to cut and try to dial it in. I'm in agreement on that as I don't have the tools to properly model it either.

The VHS tape cause me to bust out laughing. VHS tape has a fairly wide hysteresis loop to retain a recording.

hysteresis loss causes heating. Heating is energy loss. I could see it's usefulness to reduce high frequency ringing of harmonics as a suppression device to act as ferrite beads,, but not for any other practical purpose.

http://en.wikipedia.org/wiki/Hysteresis

The width of the hole and the ability to retain a magnetic field and the strength of the field is related to the coericity or the ability to be magnetized.
http://en.wikipedia.org/wiki/Coercivity

[Ragnarok]

The thought is that the ferromagnetic coating on the tape acts like a metal shell and helps to focus the magnetic field. Not sure if it really helps all that much.

It doesn't have to be VHS tape - and indeed, it's better if it isn't. You might find this page interesting:
http://www.coilgun.eclipse.co.uk/enhanc ... ments.html

[joepage2008x2]

I thought that i invented this type of valve that works off the princepals of a coilgun, well i guess not. But a bit of advice, if its built the way i did then there is hardly any limit to pressure and only one moving part, also no wear on parts and no service ever needed.

To find my vid with the valve on just type in "homemade solenoid valve" on youtube and its by joepage2008x2.

[jackssmirkingrevenge]

I thought that i invented this type of valve that works off the princepals of a coilgun, well i guess not. But a bit of advice, if its built the way i did then there is hardly any limit to pressure and only one moving part, also no wear on parts and no service ever needed

You can embed videos:

[youtube][/youtube]

I feel your frustration with working a lot to develop something only to find out it's already been done before, the culmination of my constant flow automatic valves turned out to have been already patented in 1927

It pays to do your research well before commiting yourself to the work bench.

[jimmy101]

Psudo science galore... Scuse me while I quit laughing..

How so?

Coil gun builder's often include external ferromagnetic material. The physics is rock solid, indeed 99% of transformers use external metal to control and guide the magnetic field. Granted magnetic recording tape might not do much since the ferro layer is so thin, but it is often used by coil gunners. Feel free to argue with them.

Coil guns don't often have a single turn in the windings. That tends to give a RLC that is much too short to be useful. Coil guns (at least ones that work "well") rarely have more than a couple dozen turns and less than ten is common.

Coil gun generally have terrible efficiencies, but I suspect any solenoid setup that must give a fast movement with significant energy is going to have awful efficiency no matter how you do it. An AC or DC motor has great efficiency (typically) but I've never seen one that can accelerate anywhere near fast enough for this application. High acceleration (high speed overall) pretty much forces you into crappy performance.

[Technician1002]

Psudo science galore... Scuse me while I quit laughing..

How so?

Coil gun builder's often include external ferromagnetic material. The physics is rock solid, indeed 99% of transformers use external metal to control and guide the magnetic field. Granted magnetic recording tape might not do much since the ferro layer is so thin, but it is often used by coil gunners. Feel free to argue with them.

Coil guns don't often have a single turn in the windings. That tends to give a RLC that is much too short to be useful. Coil guns (at least ones that work "well") rarely have more than a couple dozen turns and less than ten is common.

Coil gun generally have terrible efficiencies, but I suspect any solenoid setup that must give a fast movement with significant energy is going to have awful efficiency no matter how you do it. An AC or DC motor has great efficiency (typically) but I've never seen one that can accelerate anywhere near fast enough for this application. High acceleration (high speed overall) pretty much forces you into crappy performance.

High frequency switch mode power supplies use powdered iron cores for high efficiency. Magnetic recording tape does not. A powdered iron core is designed to NOT retain a magnetic field after the magnetizing force is removed unlike the recording medium which is designed to be magnetized and retain it.

Please don't confuse a low coercivity magnetic material with a high coercivity material used for data recording. They are not interchangeable. The article you linked to specified epoxy and powdered iron mix. I still say magnetic recording materials are a poor substitute for a proper powdered iron core. A proper powdered iron core can be beneficial as the article states.

[joepage2008x2]

Yep this setup had terrible efficiency, it had about 50 or so turns of 1.2mm wire. If anyone wants to build a valve same as this then make sure your gun isnt going to be portable (as capacitor banks are bulky) and use SCR`s for switches, the good thing is that the open time is extremely fast and high pressures can be used, i worked out the force that the solenoid has, mine could have taken over 3000psi - not that the seal may be able to take that.

Someone said that its unusual to have a coilgun with more than 12 turns, well it depends on the voltage, i made a coilgun at 12v with 4 turns at 2kj - it was very ineffiecient and weak. I also made one at 500v at 0.5kj which had around 200 turns - it was far more powerful. Most coilguns are around 300 to 800v and have about 50 to 300 turns, higher voltage - more efficient and less current through switches.

I had started a second solenoid a long time ago, it should have been around 60% efficient compared to the average coilgun thats about 1-2% efficient, this then wouldnt need a large power source.

[ramses]

higher voltage - more efficient and less current through switches.

actually, more current flows for a shorter period of time, but they do tend to work better.