DC-Powered Jumping Ring

Cite as: Phys. Teach. 54, 112 (2016); https://doi.org/10.1119/1.4940178

Published Online: 29 January 2016

Rondo N. Jeffery, and Farhang Amiri

ARTICLES YOU MAY BE INTERESTED IN

Optimizing Thomson’s jumping ring

American Journal of Physics 79, 353 (2011); https://doi.org/10.1119/1.3531946

Measurements and mechanisms of Thomson’s jumping ring

American Journal of Physics 68, 238 (2000); https://doi.org/10.1119/1.19407

Thomson’s Jumping Ring Over a Long Coil

The Physics Teacher 56, 176 (2018); https://doi.org/10.1119/1.5025301

Phys. Teach. 54, 112 (2016); https://doi.org/10.1119/1.4940178 54, 112

© 2016 American Association of Physics Teachers.

DC-Powered Jumping Ring
Rondo N. Jeffery and Farhang Amiri, Weber State University, Ogden, UT

T
he classroom jumping ring demonstration is nearly done using a Variac to give 130 VAC. When DC is applied, the
always performed using alternating current (AC), in warm ring jumps a short distance (about 5 cm) and then falls
which the ring jumps or flies off the extended iron core back down. When chilled with LN2 the ring jumps to the top
when the switch is closed. The ring jumps higher when cooled of the core (approximately 30 cm), then falls back down. With
with liquid nitrogen (LN2). We have performed experiments AC the ring flies completely off the core. When the switch
using DC to power the solenoid and find similarities and is opened after the DC jumps, a large arc is observed at the
significant differences from the AC case. In particular, the switch contacts. The arc is not seen with AC.
ring does not fly off the core but rises a short distance and The magnetic field within and around the iron core pro-
then falls back. If the ring jumps high enough, the rising and vides the key to understanding the force on the ring. The
the falling motion of the ring does not follow simple vertical magnetic field lines point radially outward from the iron core
motion of a projectile. This indicates that there are additional above the energizing solenoid as shown with iron filings.11
forces on the ring in each part of its motion. Four possible This iron-filing pattern is identical whether the solenoid is
stages of the motion of the ring with DC are identified, which powered with AC or DC. With AC the ring current is phase-
result from the ring current changing directions during the shifted with respect to the induced emf. When averaged over
jump in response to a changing the entire cycle of the AC power,
magnetic flux through the moving this gives an upward force on the
ring. ring that is essentially continu-
ous.5,11 The DC case is different,
Introduction however, and is more readily un-
The mechanism of the AC jump- derstood in terms of Lenz’s law. In
ing ring has been the subject of the following sections we analyze
numerous textbook illustrations,1-2 the MIT video and other DC vid-
papers,3-14 and YouTube videos.15-19 eos we have made using an auto-
The DC case has been studied, but motive battery.
not as extensively as the AC case.
Tanner et al.20 found that an ener- DC mechanism –
gizing voltage of 700 V was required Changing magnetic flux
to get the ring to fly off. Some text- The current induced in the ring
books give an erroneous impression is the result of a changing magnetic
of the DC situation. For example, flux through the ring, in accord
Serway21 and Walker22 show a bat- Fig. 1. Emf measured in a 10-turn coil of wire placed with Faraday’s law and Lenz’s law.
over the extended iron core after closing the switch. y is
tery connected to a switch and the the height of the bottom of the coil above the ring start
This first happens when the switch
jumping ring coil, and assert that position. is closed and the field builds up
when the switch is closed the ring in the solenoid and extended iron
flies off. They then ask the question, core. This build-up is not instan-
“Why does it do that?” When set up taneous, but has a finite time con-
as shown—with a battery—the ring stant, τ = L/R. We have determined
does not fly off. At best it will lift the build-up of the magnetic field
slightly and then fall back down.23 in our Welch apparatus by measur-
The ring cooled with liquid nitrogen ing the emf produced in a coil of
will jump several centimeters but wire placed over the iron core. As
still does not “fly off.” the field increases following clos-
A YouTube video by the MIT ing the switch, an induced voltage
physics demonstration group shows appears in the coil of wire just as it
jumps powered with both DC and does in the jumping ring. The
AC.16 The DC experiments were PASCO ScienceWorkshop 750
done with 120 VDC and used a 10- interface and software, and the
Ω resistor (heater coil) in series with 1-shot digital storage scope fea-
the solenoid to limit the current and ture, were used to capture the brief
Fig. 2. Magnetic field vs time build-up after the switch
match the impedance of the coil was closed. Data obtained from integrating the emf. event. The coil was made of 10
with 60 Hz AC. The AC jumps are turns of 24 AWG wire wound on a

112 The Physics Teacher ◆ Vol. 54, February 2016 DOI: 10.1119/1.4940178
 Fig. 3. Steady-state magnetic field within and along the
length of the iron core. The red points are measured
with a Hall-effect magnetometer (F.W. Bell Model 4048
Gauss/Tesla meter) along the axis above the core.
cardboard tube that slid on the 43-mm diameter iron core
and is about 5 mm in length. Placing the coil at different
positions along the iron core and switching on the current
resulted in the emf measurements shown in Fig. 1. The emf
results from the changing magnetic flux through the coil, fol-
lowing Faraday’s law:
(1)
where AB is the flux in the core; A is the area and B the mag-
netic field inside the core.
Integrating (summing under) the curves in Fig. 1 gives the
field from the measured emf:
(2)
The resulting B versus time results are shown in Fig. 2.
These empirical results are almost, though not quite, expo-
nential curves. Taking the half-value of these curves gives a
time constant of 80-100 ms. The inductance of our solenoid is
80 mH with the core out 8 cm, and the resistance is 1  (cur-
rent 12 A with 12 V), giving a calculated time constant from
L/R of 80 ms, in good agreement with the curve values.
The steady-state value of the magnetic field decreases
along the length of the iron core, as shown in Fig. 3. These
results are essentially the same as what was found in the AC
case.11 Measurements of e were made every 1.0 cm along the
core. Only four positions are shown in Figs. 1 and 2 for clarity,
but all are shown in Fig. 3.
The time constant for the MIT video can be estimated
from information given in Ref. 16. A 10- heater coil was
used to limit the current with DC to what it was with AC,
about 12 A. This indicates that their solenoid resistance was
about 1 , similar to our Welch coil, and inductance of about
70 mH. This gives a time constant of roughly t = L/R =
70 mH/11 V = 6.4 ms. This is about 10 times faster magnetic
field build-up compared with our Welch setup.
The Physics Teacher ◆ Vol. 54, February 2016
Fig. 4. Setup of Welch jumping ring apparatus with an
adjustable iron core, 5-mm ring falling. The knife switch
connects the solenoid to a 12-V battery that is out of
the picture to the right. This frame is at the moment the
switch was opened, showing the arc.
Four stages of motion
Frame-by-frame analysis of the MIT YouTube video16 has
been made, as well as videos made with the Welch jumping
ring apparatus powered by an automotive battery. Based on
this analysis the motion of the ring can be broken down into
four stages. There is some overlap of these stages, depending
on the time constant, but all four stages are seen in the MIT
videos. One of the striking features of the ring motion with
DC is the terminal velocity as the ring falls from its maximum
height. In addition, the maximum height reached is not the
same as would be expected if the ring were simply tossed
upward by the initial magnetic impulse, and then followed
one-dimensional (1-D) projectile motion under the force of
gravity. Each of these points is discussed below. The present
setup is shown in Fig. 4.
Stage 1 – Transient or impulse
When the switch is closed, the magnetic field within the
iron core rapidly increases. This does not happen instanta-
neously, however, but rises exponentially to its steady-state
value. The large flux change through the ring produces an
induced emf and current in the ring. The current in the ring
is in such a direction that the magnetic field due to the ring
current opposes the dB⁄dt through the ring, in accord with
Lenz’s law.
The mechanism for the force on the ring is said to be either
opposing dipole fields—“north pole facing north pole”1,17
—or as the Lorentz force of the ring current crossed into the
radial magnetic field of the iron core:
FLorentz = I l x Br , (3)
where Br is the radial field at that point, I is the current in
the ring, and l is the circumference of the ring. Evidence that
the force is the Lorentz force is seen in the movies,23 which
113

Fig. 5. Two frames of movie showing the ring lifting
on the side closest to the iron core. Note the blurring
of the part of the ring that is lifting up.
show that the side of the ring closest to the iron core lifts first.
The cylindrically symmetrical radial field strength decreases
roughly as 1/r out from the iron core, as measured with a
magnetometer. This makes the Lorentz force greater on the
side of the ring closest to the core, resulting in the lift on that
side of the ring. This effect is shown in Fig. 5.
The direction of the ring current depends on the direction
of the changing magnetic field in the core and will oppose that
change according to Lenz’s law. For instance, if the field direc-
tion is up, then dB⁄dt is also upward in Stage 1. The current
direction in the ring then must be clockwise (CW) looking
down on the ring to oppose that change (right-hand rule).
With an upward field within the core, the radial component
of the field around the core is directed outward. The CW cur-
rent and the radially outward magnetic field give a Lorentz
force on the ring that is upward. This provides the upward
impulse on the ring in Stage 1.
Stage 2 – Upward velocity, steady or declining field
The second stage is if the current has reached or nearly
reached its steady-state value while the ring is still moving
upward. That upward motion of the ring continues until it
reaches its maximum height before falling back. This maxi-
mum jump height over the iron core is one of the significant
differences between the AC and DC cases. With AC the
force is upward and essentially continuous when integrated
over the complete cycle, though the magnitude of the force
decreases as the ring goes higher on the core where the core
field becomes weaker. The result is that the ring acceler-
ates upward continuously and flies off the core. With DC,
it is tempting to treat the ring in this stage, where dB⁄dt ≈ 0,
as simply being a projectile that has been tossed up by the
impulse in Stage 1. Under the influence of gravity, it should
reach a maximum height and then fall back with acceleration
a = –g. However, gravity is not the only force acting during
114 The Physics Teacher ◆ Vol. 54, February 2016
this stage, as is seen by careful analysis of movies of the DC
jump.
If the magnetic field reaches its steady-state value before
the ring reaches its greatest height, there is still a changing
flux through the ring. This is due to the fact that the magnetic
field within the iron core decreases along the length of the
core. (This is shown in some YouTube videos19 and is readily
demonstrated with a coil and light bulb, where the bulb glows
brighter when lower on the core.) The emf versus height
shown in Fig. 3 of Ref. 11 shows how the field within the core
decreases toward the top of the core for the AC case. With DC
it is exactly the same, as shown in Fig. 3. As the ring moves
upward (designated as the +y direction) it thus experiences a
decreasing flux because dB⁄dy is negative. The velocity of the
ring, dy ⁄dt, is positive upward in this stage. The combination
gives a dB⁄dt felt by the ring that is negative, or downward.
This negative flux change felt by the ring results in the current
in the ring reversing direction from what it was in Stage 1, so
that now the Lorentz force is downward. In other words, if
the current direction was clockwise (CW) in Stage 1, it is now
counterclockwise (CCW) in Stage 2. This downward Lorentz
force causes the ring to slow down at a rate greater than if
gravity alone were acting and reaches a maximum height that
is less than the expected maximum height as a projectile if
gravity were the only force acting.
Stage 3 – Downward velocity, steady field
After the ring has reached its greatest jump height and
begins to fall, conditions change again. Initially the only force
on the ring is gravity, but that changes as soon as the ring be-
gins to move. Now, because of the fact that the magnetic field
within the core increases towards the bottom of the core, and
the direction of the velocity of the ring is down, the ring feels
an increasing flux as it moves downward. The current induced
in the ring now reverses again from Stage 2 (back to CW, for
example), resulting in a Lorentz force that is upward. As the
speed of the falling ring increases, this upward force also in-
creases until it balances the downward gravitational force on
the ring. When the two forces are equal, the ring’s downward
speed becomes constant, or terminal velocity. The three MIT
videos of the chilled ring show a terminal velocity of around
–80 cm/s to –90 cm/s. In our videos made with a battery, the
terminal velocity is smaller, around –12 cm/s with the core
out 8.4 cm.
Stage 3 can be simulated by dropping the ring over the
core with power on. An interesting variation is to extend the
drop path with a cardboard tube placed over the core. The
ring falls with acceleration a = –g until it reaches the core, at
which point there is magnetic braking that suddenly slows
the ring to terminal velocity.
Stage 4 – Collapsing field, or “Snap”
If the ring is allowed to fall to the platform at the bottom
of the core, it hits and may bounce slightly. However, if the
switch is opened before the ring hits the platform, it shows a

Fig. 6. Motion of chilled ring, solenoid powered with 12-V
battery. The switch was closed at t = 0. Blue points are
the average position of the ring (LH scale); red points
are instantaneous velocities (RH scale). The switch
was opened at about 0.42 s after the start, while the
ring was still moving downward at terminal velocity of
–12 cm/s (see Fig. 4). The acceleration of the ring from
0.45 s to 0.65 s during the bounce behaves like 1-D pro-
jectile motion with a = –g. At 0.65 s one edge of the ring
struck the base and the downward motion changed. The
dotted lines mark the approximate divisions between
the four stages of motion as described in the text.
Fig. 7. Second MIT jump of the cold ring with DC. The
vertical dashed lines mark the approximate separation
of the four stages of motion described in the text. The
switch was opened at about 0.50 s when the arc occurs,
followed by the sudden downturn of the position and the
velocity becoming more negative.
sudden “snap” downward and hits the platform hard. The
reason for this is again related to dB⁄dt. When the switch is
opened, the solenoid current is cut off and the magnetic field
in the solenoid collapses. The inductance of the coil prevents
the collapse from being instantaneous, however, and produc-
es a very large back emf at the switch contacts. This is seen by
arcing at the switch contacts, as seen in the MIT videos and in
Fig. 4. The large downward dB⁄dt through the ring results in
The Physics Teacher ◆ Vol. 54, February 2016
a very large ring current that is in the opposite direction from
the current in Stage 3 (now CCW). This large downward
force on the ring produces the “snap.”
In one of our videos, shown in Fig. 6, the ring bounced
after the snap higher than it had jumped in Stages 1 and 2.
In this jump there is overlap between Stage 1 and Stage 2
because of the finite time constant of the field build-up. The
slope of the velocity-time graph from roughly 80 to 160 ms
is about –0.6g. This shows that there is still some residual
upward force on the ring due to the field having not quite
reached its steady-state value. The 1-D projectile motion fol-
lowing the bounce (after 0.42 s) is noticeably different from
that seen in the powered portion of the motion. The accelera-
tion, or slope of the velocity-time graph from 0.45 to 0.65 s,
is –g.
MIT videos
The MIT DC electrical conditions were discussed earlier.
Because their time constant is about one-tenth what we have
with the Welch coil, the steady-state field is reached much
faster. The result is that the four stages of motion are more
clearly separated. Figure 7 shows the analysis of one of the
three MIT DC jumps with the chilled ring. After the initial
impulse (Stage 1) ending at roughly 0.05 s, the slowdown be-
fore it reaches the maximum height has a slope of about –4g.
This shows Stage 2 more clearly than seen in our Fig. 6. Stage
3, as the ring falls from its maximum height, shows the termi-
nal velocity of about –80 cm/s to –90 cm/s in all three videos.
Two of the three MIT chilled-ring videos show the snap
in Stage 4. In the first of the three cold jumps, the switch was
opened just as the ring hit the bottom so no snap occurred.
In the second and third jumps, the switch was opened when
the ring was still falling at terminal velocity, with roughly
one-third of the height yet to fall. The increase of downward
velocity is clearly seen in frame-by-frame analysis.
The approximate accelerations in the two jumps shown in
Figs. 6 and 7 are summarized in Table I. These are rough esti-
mates from the velocity graphs and not accurate second dif-
ferences from the position values. In the case of the MIT data
(Fig. 7), the fact that the movie was only shot at 30 frames per
second limits the accuracy because there are so few points,
particularly in Stages 1 and 4. Nevertheless, it is interesting to
see the correlation between the two sets of data displayed in
this manner.
Table I. Approximate acceleration (∆v/∆t) by stage, relative to
the gravitational acceleration, g.
Stage Fig. 6 Fig. 7
1 +1.0 to +1.7 +12
2 –0.6 to –1.0 –4.4
3 0 0
4 –12 –1.2
Hit (after 4) +20 +20
bounce –1.0 xxx
115
Conclusion
The jumping ring experiment using DC power shows
many interesting features that should be accessible to physics
students from high school to college. The behavior of the DC
ring jump is similar in some ways yet different enough from
the AC case to merit careful study. For one thing, the analysis
is simpler in terms of Faraday’s law and Lenz’s law, which ex-
plain, for example, the terminal velocity of the falling ring in
terms of the spatial variation of the magnetic field within the
iron core.
Movies of the jump have shown that the motion can be
divided into four stages. The first, transient stage overlaps the
second stage somewhat in time due to the finite time it takes
the magnetic field to reach its steady-state value and the elec-
trical characteristics of the setup. Reaching a maximum jump
height and falling with a terminal velocity are features unique
to the DC case, as is the snap if the switch is opened while the
ring is still falling.
A final comment is in order on the conclusion that the
force on the ring is the Lorentz force. The force is due to the
interaction of the induced current in the ring crossed into the
radial component of the magnetic field around the iron core.
This is the same mechanism whether the apparatus is pow-
ered with AC or DC. The lift on the side closest to the core,
shown in Fig. 5, is explained by the fact that the radial mag-
netic field strength decreases as 1/r out from the core, so that
the Lorentz force is greater on that side of the ring.
Acknowledgments
The authors acknowledge the many helpful comments and
suggestions made by the referees.
References
1. R. P. Feynman, R. B. Leighton, and M. Sands, The Feynman
Lectures on Physics (Addison-Wesley, Reading, MA, 1964),
II:16-5. Available on line at: http://www.feynmanlectures.
caltech.edu/II_16.html#Ch16-S3 (accessed April 17, 2015).
2. D. J. Griffiths, Introduction to Electrodynamics, 4th ed. (Pear-
son, Boston, 2014), p. 316.
3. E. J. Churchill and J. D. Noble, “A demonstration of Lenz’ law?”
Am. J. Phys. 39, 285–287 (March 1971).
4. D. J. Sumner and A. K. Thakkrar, “Experiments with a ‘jump-
ing ring’ apparatus (undergraduate exercise),” Phys. Educ. 7 (4),
238–242 (May 1972).
5. A. R. Quinton, “The ac repulsion demonstration of Elihu
Thomson,” Phys. Teach. 17, 40–42 (Jan. 1979).
6. S. Y. Mak and K. Young, “Floating metal ring in an alternating
magnetic field,” Am. J. Phys. 54, 808–811 (Sept. 1986).
7. J. Hall, “Forces on the jumping ring,” Phys. Teach. 35, 80–83
(Feb. 1997).
8. C. S. Schneider and J. P. Ertel, “A classroom jumping ring,” Am.
J. Phys. 66, 686–692 (Aug. 1998).
9. P. J. H. Tjossem and V. Cornejo, “Measurements and mecha-
nisms of Thomson’s jumping ring,” Am. J. Phys. 68, 238–244
(March 2000).
10. J. M. Bostock-Smith, “The jumping ring and Lenz’s law – An
116 The Physics Teacher ◆ Vol. 54, February 2016
analysis,” Phys. Educ. 43, 265–269 (2008).
11. Rondo N. Jeffery and Farhang Amiri, “The phase shift in the
jumping ring,” Phys. Teach. 46, 350–357 (Sept. 2008).
12. P. J. H. Tjossem and E. C. Brost, “Optimizing Thomson’s jump-
ing ring,” Am. J. Phys. 79, 353–358 (April 2011).
13. F. Waschke, A. Strunz, and J-P. Meyn, “A safe and effective
modification of Thomson’s jumping ring experiment,” Eur. J.
Phys. 33, 1625–1634 (2012).
14. C. L. Ladera and G. Donoso, “Unveiling the physics of the
Thomson jumping ring,” Am. J. Phys. 83, 341–348 (April 2015).
15. University of Manchester Virtual Laboratory video, “Jump-
ing Rings Demonstrations,” https://www.youtube.com/
watch?v=T9PflsLZqY8 (accessed April 17, 2015).
16. MIT physics demonstration video, “Jumping Ring,” http://
video.mit.edu/watch/physics-demo-jumping-ring-2841/ . Also,
https://www.youtube.com/watch?v=Pl7KyVIJ1iE (both ac-
cessed April 17, 2015); also electrical details on Technical Ser-
vices Group website (dated Sept. 3, 1987) — http://tsgphysics.
mit.edu/front/?page=demo.php&letnum=H%2022&show=0
(accessed June 28, 2015).
17. Carleton University (Ottawa, Ontario) Physics Lab Demo11:
Jumping Ring, https://www.youtube.com/watch?v=0_kKXJU-
argU (accessed April 17, 2015).
18. Dan Russell (Penn. State Univ.), “Jumping Ring Induced
Current Demonstration,” https://www.youtube.com/
watch?v=V690VphqTwg (accessed April 17, 2015).
19. James Lincoln AAPT/Films, Video demonstrating
the Pasco Ring Launcher, https://www.youtube.com/
watch?v=G0sTOcyhcFM (accessed April 17, 2015). Shows iron
filing magnetic field pattern and different combinations of
rings—separately, stacked, etc., and with light bulb.
20. P. Tanner, J. Loebach, J. Cook, and H. D. Hallen, “A pulsed
jumping ring apparatus for demonstration of Lenz’s law,” Am. J.
Phys. 69, 911–916 (Aug. 2001).
21. Raymond A. Serway, Physics for Scientists & Engineers, 3rd ed.
(Saunders, Philadelphia, 1990), p. 893; Raymond A. Serway,
Robert J. Beichner, and John W. Jewett, Jr., Physics for Scientists
and Engineers, 5th ed. (Saunders, Philadelphia, 2000), p. 1002.
22. James S. Walker, Physics, 3rd ed. (Pearson/Prentice Hall, Upper
Saddle River, NJ, 2007), p. 795.
23. Rondo N. Jeffery and Farhang Amiri, http://www.jumpingring.
com/research. Video 2 shows repeated jumps of 5-mm ring
with the Welch apparatus, powered by a 12-V battery (accessed
August 24, 2015).
Rondo N. Jeffery is an emeritus professor of physics at Weber State
University. He received BS and MS degrees from Brigham Young University
and the PhD from the University of Illinois/Urbana-Champaign in con-
densed matter physics. In addition to teaching, he has worked in industry,
including a geophysics company that traces ground water using magnetic
fields. He and his co-author, Farhang Amiri, partnered in integrating com-
puters into the physics labs at Weber State.
rnjeffery@msn.com
Farhang Amiri is an emeritus professor of physics at Weber State
University. He received his PhD from Florida State University in elementary
particle physics. His research interests are in undergraduate physics. He
has developed computer animations and videos of physics demonstra-
tions in collaboration with other colleagues. He continues his research
by publishing papers related to topics of interest in the undergraduate
curriculum.