Author Options:

Supercapacitor + Solenoid + MOSFET? Answered

I am designing a coilgun with three stages. Each stage has 1mm thick copper wire, with 32 turns across 32mm of length (1000 turns / meter). There are 4 layers with the same wire, length, and number of turns on each stage. Each coil will be powered by a 100F 2.7V supercapacitor (assume charging circuitry has been implemented).

With those values, I have calculated that the resistance of the coils is about 0.1 ohms. With a 2.7V capacitor, I can get almost 25 amps of current through the coils, resulting in a total magnetic field of about 0.1 tesla. If I wire the coils in parallel, I get a resistance of 0.05 ohms and 420 amps through the coil, resulting in a field of strength 2 tesla.

Now, the problem I am running into is shutting off the current at the right time. I found a MOSFET used for 3D printers that says it can handle up to 280A (360W at 12V or 720W at 24V). If I use thinner wire, I can decrease the current to below 280A which I suppose is powerful enough for the coilgun.

Since I will be using an Arduino or similar microcontroller to manage the timing for the coilgun, would I be able to use this MOSFET (linked below) to shut off the current using a signal from the Arduino?



The forums are retiring in 2021 and are now closed for new topics and comments.

1 year ago

I would start by matching your coil system to the capacitor plus the expected losses for connections and power control, like the mosfet.
The goal is to utilise the maximum power from the capacitor with getting it into the overcurrent region.
This is important as otherwise you risk a deadly ringing for the mosfet.
At just 2.7V you also hav the problem that most mosfets will create a far to high loss for properly utilising the stored energy.

What the heck, since everything can be found through the famous search engine if you use the right terms I might just go against better judgement here...
You approach is flawed, not the idea itself though.
For a coil gun you you need extremely high energy levels, the more Joule, the better.
You also want to keep resistance losses as low as possible - the higher the voltage the less you need to morry about a few Ohm here and there...
In you case with a multi stage system you also have the timing problem against you, combined with constantly changing friction losses next to impossible to solve with a little microcontroller and mosfets....
Think out of the box for a minute and try to follow me:
A simple throw away flashligh camera - yes, those still using a roll of film in them - uses a single battery to charge a beefy capacitor to around 400V.
(hint: common electrical wire for your house installation is rated for this)
The energy is kept in the capacitor and the electronics use only little power to keep it topped up.
Once you press the trigger a tiny coil that acts like the ignition coil in your car ionises the flash tube.
This enough for the high voltage to jump from one contact to the other and create the flash as a form of the fast capacitor discharge.
So much for the "low energy" example.
In your case you would need a similar circuit to charge a bank of these high voltage discharge capacitors.
The energy they store should be just over the rated limit and resistance of the coil the supply.
Means if they can provide a 400V depletion with 500A than you certainly need a coil with a matching resistance and thickness for the wire.
The discharge would be quite similar to the flashlight, with a high amp discharge tube, like those used for lightning protection.
Select a long enough one that still fits the voltage requirements as you need to add the ionising wire and "ignition" part to it while making sure said wire is protected against a possible discharge to short out the tube.
For the timing and activation of the ignition system can be done by simple means of IR sensors and detectors, a diode (maybe with a focussing lens) and a IR transistor that switches a stronger one to activate the ignition.

Why so complicated and high voltage?
Timing and power are critial.
Nothing is worse that trying to get a microprocessor in the right mood to provide consistent results here.
Changing the placements of simple IR detector to provide an interrupted beam when the projectile enters is quick, easy, reliable...
For the "flashlight" part there are certainly better options as the discharge tubes still provide a significant resistance.
But they are a really good base to experiment with and focus on the physical desgin and shape of coils, tube and projectile.
Keep in mind that a magnetised projectile stores a magnetic field.
You can either try to utilise this for the next stage in terms of a higher overall magnetic pull between projectile or try compensate for i by making sure the next stage is far enough away so that the magnetic field in the projectiles is depleted.
When the coils are fired in a closed circuit you create an oscillator.
The discharge tube provides a low resistance when it is activated through a high enough voltage to create an arc inside.
Once the voltage goes below the threshold it is like an open circuit.
With the right selection the ringing or oscillation is minimised which means the effect on the projectile is minimised without the need of expensive and often sensitive power electronics.
When the basic desing works suffiently so that the losses of the tubes can be cosidered you have the timing already solved as well as the design kinks.
With that you should be able step up the power level to something that might actually work.
Assuming you made it to the stage where your new setup actually made your projectile fly further than a meter or two:
Take a new approach to the coil design and power distribution.
With the timing already solved you can utilise what is known as a "can or coin crusher".
Of course only they types with re-usable coils, not those where the coils is literally exploding in the process.
Look it up once you are there and know what power levels you really need for the mass of your projectile and the lenght of your system ;)