Electromagnetic Ring Launcher -- Building a Science Museum Classic: Page 2 of 4

The Jumping Ring or Electromagnetic Ring Launcher experiment is a staple of science museums and physics class rooms. Here's how to build your own right in your home.

driver circuit (Fig.1). The circuit needs two power supplies, a 15V, 0.8A supply to deliver power to the oscillator and MOSFET driver and a 24V, 4A supply is powering up the coil.


Fig.1. Circuit diagram of the ring launcher

To achieve near-50% duty cycle resistor R1 is chosen 180 Ohm, which is a much smaller value compared to R2+R7 (minimum~4.7k). By varying R2 from 100K to 0 Ohm, square wave output is obtained from 700Hz to 18 kHz. This square wave output at pin-3 of 555 timer IC should not be used directly to drive the MOSFET (Q3), for the gate capacitance. A MOSFET driver is implemented using two transistors, Q1 and Q2. To limit the initial high gate current, R5 is used. A high power and high current MOSFET (Q3) is used to drive the coil-capacitor combination. A fast recovery diode, D1 is used to leave the LC circuit to run free during the OFF time of the MOSFET. An inductor of 5 micro Henry (L1) is used to limit the initial high current, when the MOSFET is switched ON. This L1 can be made easily by winding ~40 turns on a piece of plastic pipe of 1cm diameter. When the MOSFET is ON energy is delivered to the LC circuit, when the MOSFET is OFF, the stored energy in the capacitor C and coil L begins to flow in-between L and C.

When the switching frequency of the MOSFET matches the resonance frequency of the LC circuit, minimum energy is used by the LC circuit to sustain the oscillation. In this situation, though low current is drawn from the supply, much higher current flows in the LC circuit. This high current creates intense magnetic field in the ferrite core. The copper ring which runs through the core, acts as a single turn coil of low resistance. The alternating magnetic field in the ferrite core induces a voltage in the copper ring, thereby high current flows through the ring as well. These two interacting field forces the ring jump out of the core. The MOSFET Q3 and diode D1 gets heated after sometime of operation, especially during tuning. Two small heat sinks are needed for these two devices. The PCB layout of the circuit is shown in Fig.2. The 10 Ohm power resister (R8) is not shown in the PCB, as it is a panel mounted type. R8 should be screwed to aluminum enclosure as shown in Fig.3.

Fig.2. PCB Layout


Fig.3. Circuit-PCB inside enclosure


Fig.4. Front panel


The Coil with Ferrite Core:

To make a suitable coil holder and a base for the ferrite cylinder, uncladed FR4 sheet was used. Several pieces were cut and screwed as shown in Fig.5. On the top piece, a hole was made for the ferrite cylinder to go in. As long ferrite rods are becoming uncommon now a days, instead two ferrite cylinders with hole inside were used. Each cylinder is 5 cm long, these are joined by a long nylon stand-off and nylon screws. After the coil is made


Note that with a 15 Volt supply to the 555 and R1 at 180 Ohms, you'll have over 80 milliamps running through R1 into the discharge pin during the discharge cycle. The discharge transistor may well come out of saturation at that current, resulting in substantial power across that transistor. You run the risk of destroying your 555. The ST datasheet for the NE555N shows a worst-case saturation voltage of 480 mV at only 15 mA when operating at Vcc=15V. You have more than five times that current.

Hello William: I think your information is wrong. If you take a closer look at the data sheet of NE555, you will see that at Vcc=15V, Idisch=15 mA, Vsat has a minimum to a maximum value of 480 mV. Furthermore, take a closer look at the footnote, even if you take a lower value for R1, there is an internal current limit. At 80 mA, the collector emitter voltage Vce of the discharge transistor is less than a volt. Now considering the duty cycle calculate the power dissipation and consider the therm

thermal resistance Rthja. you will find the package (DIP8) temperature is well below the maximum allowable range. Practically, I ran the circuit for quite a long time, without overheating or damaging the IC NE555

Does the circuit being resonant limit the rate of current increase in the coil? I've only ever made a launcher (not for hovering) and needed to get the current in the coil to maximum before the ring left the field. I dumped about 30A @ 240V into the coil. My launcher gets the ring 2m to 3m into the air at room temperature. When the (aluminium) ring is cooled in liquid nitrogen then it reaches the top of a three storey atrium - about 10m. It's used for university demonstrations.

Hello John: Nice to know your wonderful experience. The detailed analysis of a similar circuit is published in the journal "Physics Education" Jan., 2016. The use of resonance gives us the advantage that if the supply current is ~2A, actual coil current is ~8 to ~ 10 times. So, we need a smaller DC supply. Furthermore, at higher frequency, the upward force is also higher, as it gives larger phase difference between induced voltage and current in the ring

Wow! I built one of these in 7th grade. I must have been about 13. The core was made of laminations of black stove pipe iron: sheet metal strips varnished so they wouldn't conduct eddy currents. The strips were 1" x 12", laminated into a core 1" x 1" x 12". The ring was a stack of aluminum sheets, about 1.5" ID and 4" OD. The coil was a few hundred turns of AWG 18 or 20 wound between two 1/8" masonite spacers maybe 3" apart on one end of the core. Ring never hovered but did fly off impressively

Hello Laurence, It is nice to know your experience. This jumping ring effect really makes lot of kids curious, even today. I was also fascinated when I was a kid, but never tried anything at that time like you. Thanks for your comments.

Can you post more info on making L1? The photos show a toroidal coil with about 25 turns on it, but the text says 40 turns on a cylindrical core - at least that's what I get out of it.

Hi Mark: I wrote the instruction for an air core coil, it will work, Though, in the photo, you can see I used a toroidal ferrite core with few turns on it. I do not recommend doing it, because most of the toroidal core you buy have high permeability and high hysteresis loss at the frequency of our interest. I used the core, as I have it at my disposal which has a low permeability and low loss. However finding a similar one in the supplier's stock is a time consuming job. Rather, it is better to

make one with air core. Though the number turns are larger, it will not affect the normal operation. If you have a ferrite rod, you can also make the coil with 10-15 turns on it. this is also better than choosing a ferrite toroid. The purpose of the coil is only to prevent transient high current. I have a pretty bad experience with various types of toroids. Most of these heat up and hamper the normal operation by consuming extra power.


Add new comment

By submitting this form, you accept the Mollom privacy policy.