Content uploaded by Per-Ludvik Kjendlie
Author content
All content in this area was uploaded by Per-Ludvik Kjendlie on Nov 17, 2014
Content may be subject to copyright.
Rev Port Cien Desp 6(Supl.2) 15–113 75
The velocity meter curves provide some tentative evidence that
the swimmers were able to generate propulsion with their
affected limb, as there was a marked increase in intra-cyclic
speed during the push phase of this limb. This occurred when
the sound arm was either still recovering, entering or in the
non-propulsive glide phase. Not surprisingly, the swimmers
were more effective at increasing their swimming speed with
their sound limb than they were with their affected limb
(Figure 3). The peak swimming speed achieved during the
push phase of the sound limb (1.30 ± 0.17 m.s-1) was signifi-
cantly higher than it was during the push phase of the affected
limb (1.14 ± 0.11 m.s-1).
Inter-swimmer correlations revealed a significant relationship
(r=0.72, p<0.05) between mean swimming speed and stroke
rate. Interestingly, the swimmers who exhibited the highest
stroke rates were not necessarily those who pulled their affect-
ed limb through the water the quickest, as the correlation
between the extension velocity of the affected limb and stroke
rate was non-significant (r=-0.36). Extension velocities of the
affected limb ranged from 8.8 to 12.9 rad.s-1. There was no
relationship between the extension velocity and the peak swim-
ming speed that was produced during the push phase of this
limb. This indicates that factors other than limb speed, such
as the timing and trajectory of the pull, may be more important
in determining the effectiveness of the pull.
CONCLUSION
Swimmers with a uni-lateral arm amputation have demonstrat-
ed that, in the absence of a forearm and hand, it is possible to
use the upper arm to increase swimming speed within the
front crawl stroke cycle, but not as effectively as with the com-
plete arm.
REFERENCES
1. Craig AB, Pendergast DR (1979). Relationships of stroke
rate, distance per stroke, and velocity in competitive swim-
ming. Med Sci Sports, 11 (3): 278-283.
2. Hay JG, Thayer AM (1989). Flow visualization of competi-
tive swimming techniques: The tufts method. J Biomechanics,
22: 11-20.
3. Kent MR, Atha J (1973). Intra-cycle retarding force fluctua-
tions in breaststroke. J Sports Med, 13: 274-279.
4. Toussaint HM, Beek PJ (1992). Biomechanics of competitive
front crawl swimming. Sports Med, 13 (1): 8-24.
THE EFFECT OF THE BREATHING ACTION ON VELOCITY IN FRONT
CRAWL SPRINTING
Tommy Pedersen, Per-Ludvik Kjendlie
Norwegian School of Sport Sciences, Oslo, Norway.
Ten competitive, national level adult swimmers (age 25 ± 3
years (mean ± SD) swam three 25m freestyle sprints with dif-
ferent breathing patterns in randomised order to examine how
breathing actions influence velocity during a 25m front crawl
sprint. Velocity measurements were carried out using a com-
puterized swimming speedometer and data from mid-pool free
swimming (10-20m) was extracted. There was no significant
difference in mean (±SD) velocity (v) between sprinting with
one breath (v=1.74±0.14 m_s-1) compared to no breath
(v=1.73±0.14 m_s-1). There was a significant (p<0.05) reduc-
tion in velocity when breathing every stroke cycle
(v=1.70±0.14 m_s-1), compared to both no breath and one
breath trials. Swimmers should breathe as little as possible
during 50m freestyle races and breathe no more than every 3rd
stroke cycle during a 100m freestyle race.
Key Words: biomechanics, breathing, swimming performance,
freestyle, sprint.
INTRODUCTION
To achieve a high swimming velocity, one main goal for swim-
ming technique is to create optimal propulsion and minimal
resistance (4). For a front crawl swimmer, minimal resistance
winds down to keeping an optimal streamline; the head and
body in a straight line and the body as horizontal as possible.
Optimal propulsion means keeping effective propulsive forces,
high propelling efficiency and high power output throughout
the swimming distance. The breathing action in front crawl
swimming is in most cases a movement that inflicts the swim-
mers streamline or propulsion because the head has to move
out of normal swimming position to make inspiration of air
possible. How long the inspiration lasts will also inflict the
swimmers streamline and propulsion (1). Both Cardelli, Lerda
& Chollet (1) and Lerda & Cardelli (2) have found in previous
studies that there is a connection between how good a swim-
mer is to coordinate the breathing action in front crawl swim-
ming and their technical level. More expert swimmers tend to
use shorter time on the inspiration of air compared to less
expert swimmers (1). Furthermore more expert swimmers
were found to have an improved ability to coordinate arm-
strokes and inspiration of air so that body balance and contin-
ued propulsion is more efficient also during the breathing
action (2). Even so swimmers are often instructed to breathe
as little as possible during 50 m sprint swimming, and during a
100 m race swimmers tend to reduce their breathing compared
to longer distances.
The purpose of this study was to examine how breathing actions
influence velocity during a 25m front crawl sprint by using two
different breathing patterns compared to no breathing.
METHODS
Subjects
Ten competitive, Norwegian national level, adult swimmers
volunteered to participate in this study (8 males and 2 females,
mean±SD; age 25±3 years, personal best 50m freestyle
25.15±1.98 sec, season best 50m freestyle 25.62±2.19 sec).
All subjects signed an informed consent after having the proto-
col explained to them both verbally and in writing.
Test protocol
Before start of the trial the subjects conducted a standardized
warm up of about 1500m including four short sprints. The trial
consisted of three 25m freestyle sprints with different breath-
ing patterns conducted in a randomised order: a) 25m sprint
with no breathing b) 25m with one breath after 15m of swim-
ming c) 25m with one breath every stroke cycle. All breathing
was to the subjects’ preferred side. Each 25m sprint started
every 4 minutes, giving the subjects about 3 min and 45 sec
recovery between each sprint. During this recovery they had to
swim one 25m to get back to start, the rest of the recovery was
passive.
SWIMMING BIOMECHANICS
Rev Port Cien Desp 6(Supl.2) 15–113
76
Measurements
Velocity measurements were carried out using a computerized
swimming speedometer, connected to the swimmer via a thin
non elastic line. The speedometer, attached to the pool side,
consisted of the speedometer and a digitizing unit. The
speedometer had a reel for the line which was set to give a
small, but constant resistance on the line to ensure a trouble
free outlet of the line. The line went from the reel via a small
wheel to the hip of the swimmer. The small wheel (9 cm inn
diameter) was connected to the axis of an incremental encoder
(Leine & Linde nr IS630, Strängnes, Sweden) which gave 250
square pulses (0-5V TTL logic) for every rotation of the wheel.
The swimmers pulled the line and the incremental encoder
produced impulses for every turn of the small wheel. These
pulses was digitized in a computer card (DAQ 6024E, National
instruments, USA), and the signal was treated with Digital
acquisition software LabVIEW 7 Express (National
Instruments, USA).
Every impulse from the speedometer gave position data which
the program smoothened by a floating mean of 10 measure-
ments. The velocity was then calculated in the program by a
mean of two positions. Fig. 1. shows an example of the velocity
output vs time. Sampled frequency was 100 Hz. The coefficient
of variation for the equipment used was calculated to <2 %.
A camera (Panasonic GS3, Japan) was used to film the swim-
mers above water while they swam each trial. This film was
later used to find out the number of strokes performed in the
10m distance of the one breath trial, and how many breaths
the swimmers had on the same distance on the breath every
stroke cycle trial.
Data from mid-pool free swimming (10-20m) was extracted
and used in all analyses.
Fig. 1. Example of velocity vs time curve from the speedometer data.
Vertical lines represent right arm entry.
Statistics
All data are presented as mean ± standard deviation. A paired
t-test was used to determine difference between the trials
where p<0.05 was considered significant.
RESULTS
There was no significant difference in mean velocity (v)
between 10m of mid pool sprinting when the swimmers took
one breath compared to no breath. To breathe once every 10
meters equalled about one breath every 3rd stroke cycle for the
swimmers in this study. There was a significant (p<0.05)
reduction in velocity when breathing every stroke cycle, com-
pared to both no breath and one breath trials, see table 1. The
swimmers in this study breathed 5-7 times over 10m of mid
pool sprinting when breathing every stroke cycle.
Table 1: Mean velocity (±SD) from the three trials.
Breath every
No breath One breath stroke cycle
v
10-20
(m·s
-1
)v
10-20
(m·s
-1
)v
10-20
(m·s
-1
)
Mean (±SD) 1.74 (±0.14) 1.73 (±0.14) 1.70
*
(±0.14)
* significant different from both no and one breath trials (p<0.05)
DISCUSSION
The results indicate that swimmers at this performance level
may breathe once every 3rd stroke cycle without loosing veloci-
ty due to breathing actions in front crawl sprint. If swimmers
breathe every stroke cycle they may loose up to about 0.1 sec
pr 10m of mid pool swimming.
Unpublished observations of 50m freestyle for males at the
Norwegian Long course National championship 2004 showed
that all the top 8 swimmers breathed 1, 2 or 3 times with at
least 3 stroke cycles in between each breath in the final. Even
though there was no significant difference between the one and
no breath trial in this study, a difference of only 0.01 m·s-1 as
found here represents a loss of 0.03 sec over 10 m swimming.
Even at this performance level a loss of 0.03 sec because of one
extra breath could mean 2nd place instead of 1st place. There
were individual differences; the highest difference between no-
and one breath trial was 0.04 m·s-1 or 0.15 sec. This indicates
that all swimmers can gain by learning better breathing tech-
nique and breath control, but coaches should know that some
individuals have even more to gain.
Furthermore, observations of the 100m freestyle race for both
females and males in the same National Championship
revealed that 100m freestyle swimmers seemed to vary what
breathing pattern they choose, but most common was to
breathe every 2nd, 3rd or 4th stroke cycle for the first part of the
race, and than increase to every stroke cycle or every 2nd stroke
cycle the last part of the race. Only a few swimmers choose to
breathe as little as every 3rd or 4th stroke cycle throughout the
race, amongst these was the winner of both male and female
100m freestyle. The main reason for swimmers to increase
their breathing pattern the last part of a 100m race is caused
by an urge to breathe more due to a lower partial CO2pressure
in the blood caused by the high intensity of the swimming.
Peyrebrune et al. (3) found no reduced performance based on
physiological markers when swimmers breathed as little as
every 4th stroke cycle, during 55 sec of tethered swimming.
This indicates that the swimmers can choose to breathe as lit-
tle as every 3rd to 4th stroke cycle without loss in performance
due to either physiological factors or biomechanical factors
(breathing action).
CONCLUSION
Coaches should stress breath control both in training and com-
petitions and also teach effective breathing technique to avoid
velocity reductions due to breathing actions. In a 50 m
freestyle sprint the swimmers should breathe as little as possi-
ble, but during 100 m race swimmers must breathe more and
can breath as often as every 3rd stroke cycle without to much
SWIMMING BIOMECHANICS
Rev Port Cien Desp 6(Supl.2) 15–113 77
loss of velocity compared to breathing more often. To give
accurate advice about which breathing patterns to use in 100m
races, both individual differences in technique and physiologi-
cal and metabolic variables must be taken into consideration. A
further investigation in this matter seems necessary, combining
biomechanical and physiological methods.
REFERENCES
1. Cardelli C, Lerda R, Chollet D (2000). Analysis of breathing
in the crawl as a function of skill and stroke characteristics.
Percept Motor Skills, June: 979-987.
2. Lerda R, Cardelli C (2003). Breathing and propelling in
crawl as a function of skill and swim velocity. Int J Sports Med,
24:75-81.
3. Peyrebrune MC, Robinson J, Lakomy HK, Nevill ME (2003).
Effects of controlled frequency breathing on maximal tethered
swimming performance. In: Chatard JC (ed.) Biomechanics and
Medicine In Swimming IX. Saint-Étienne: Publications de
l’Universite de Saint-Étienne, 289-294.
4. Toussaint HM, Beek PJ (1992). Biomechanics of competitive
front crawl swimming. Sports Med, 13:8-24.
BIOMECHANICAL ANALYSIS OF THE TURN IN FRONT CRAWL
SWIMMING
Suzana Pereira, Luciana Araújo, Elinai Freitas, Roberta
Gatti, Graziela Silveira, Helio Roesler
Universidade do Estado de Santa Catarina, Laboratório de Pesquisas
em Biomecânica Aquática. Florianópolis, Santa Catarina, Brasil.
The main purpose of this study was to investigate the contribu-
tion of the dynamic and kinematic variables to the performance
in freestyle. The turns of 38 swimmers were analyzing using an
underwater force platform and two video cameras that sup-
plied. Angle of knee flexion (AK), maximum normalized force
peak (FPn) and contact time (CT) were measured as variables.
Through investigation of the contribution of the variables AK,
PMn and CT to the variable TT it was possible identify that
PMn explains the greatest percentage of variance in turn per-
formance (17,70%). The relation between AK and PMn indicat-
ed that larger values of AK (smaller flexions) tend to provide
larger values of PMn (r = 0,38). Start from the results analysis,
it can be suggested that angles of knee flexion between 110
and 120 degrees tend to provide larger force peaks, smaller
contact times and smaller turn times, reaching the best per-
formance of the crawl stroke turn execution.
Key Words: swimming, turn, biomechanics, flip turn,
dynamometry, kinemetry.
INTRODUCTION
The final times of swimming tests can be influenced from the
turns in up to 20% (6). The process of the biomechanical study
of the turns developed by the Research in Aquatic
Biomechanics Laboratory of the University of the State of Santa
Catarina (UDESC) is described by Roesler (8) and is part of
the studies (1), (5) e (7). The present research complements
previous studies and researched parameters on improving per-
formance of the turns, investigating the relationships between
the variables: Maximum Peak of normalized force (PMn) and
Time of Contact (TC) with the performance in the turn in it I
swim Crawl (TV), through the time of turn in 15m.
METHODS
38 swimmers, integrant of the team of swimming of the Club
“12 de Agosto” of the city of Florianópolis/SC, federated by
Aquatic Federacy of Santa Catarina (FASC), participated in the
research, chosen intentionally once they have domain over the
technique of execution of the flip turn in front crawl swim.
They have an average age of 18,2 years, average body mass of
63,8Kg, and average stature of 1,70m.
For the acquisition of the dynamic data, an underwater strain
gauge platform (9) with sensitivity of 2N and natural frequency
of 60Hz was used. The force plate was associated to a special
support to be fixed to the inside of the turning wall of the
swimming pool, in the vertical plan, and on the opposite side
to the departure blocks, in lane 4. Once the platform cover is
0,2m thick, the black traces in the swimming pool bottom were
modified to adapt to the new configuration, respecting the
same official distance for the accomplishment of the turns.
For the kinematic data acquisition, a video camera Mini-DV
Mega Pixel 3CCD (60Hz) inserted into a water proof box
(camera 1), and a VHS camera with acquisition frequency of
60Hz (camera 2) were used. Camera 1 was located inside the
swimming pool, allowing a underwater sight from the bottom
to the top of the force platform. It was used for the assessment
of the angle of knee flexion (AK). To determine the variable
AK, a set of anthropometric landmark points were used (great
trochanter, lateral epicondilus and lateral maleolus), made with
coloured adhesive ribbon for the ulterior recognition in the
video analysis. For the assessment of the turning time (TV) in
15m, camera 2 was located outside the water, 17,5m of the
departure platforms, allowing a lateral sight of the swimming
pool. The measurement of the turning time was initiated at the
moment where the image of the swimmer’s head reached the
mark of 7,5m in direction to the turning wall, and finished
when the swimmer’s head reached again the mark of 7,5 m,
but after the turn.
Data collection of was carried out during a training session. The
swimmers warmed-up in accordance with the coach, trying suc-
cessive impulses with the feet in the platform, in order to adapt
themselves to the experimental conditions. Each swimmer start-
ed swimming from inside the swimming pool, under the depar-
ture blocks, reaching maximum speed at 12m from he starting
wall, carrying through the turn and keeping the maximal speed
until the 12m. This exercise was repeated 8 times with a resting
interval of 12 minutes between each repetition.
The data obtained through the force platform has been separated
and filed for swimmer, calibrated and filtered through a
Butterworth filter from (30 Hz), and the normalization was con-
ducted dividing the measured force archive by the weight of the
swimmers, both carried through in system SAD 32 (10) supply-
ing the PMn, which are the greater value registered of the force
and the TC, that is the time during which the swimmer keeps
contact with the platform. The swimmers weight was measured
directly with a digital scale, Plenna, model MEA-08128 (0,1kg).
For the assessment of AK, the images of camera 1 were used,
selecting, through the edition images program Adobe Premiere
6,5, the picture where the swimmer carries through the maxi-
mum knee flexion when touching the force platform. The flex-
ion angle was obtained using the program Corel Photo-Paint ver-
sion 10.
SWIMMING BIOMECHANICS
